2007-04-01
[21:07:00.0000] <zcorpan_>
<meta http-equiv=content-access-control> should perhaps be allowed, as the html version of <?access-control?>

[23:08:00.0000] <Hixie>
zcorpan_: can't be an element, access control has to be done before you create the root element

[23:10:00.0000] <othermaciej>
you could use a pre-parser, but that's lame 

[23:10:01.0000] <othermaciej>
that reminds me, I owe a review of access-control

[23:27:00.0000] <zcorpan_>
ok

[01:08:00.0000] <zcorpan_>
http://www.brucelawson.co.uk/index.php/2007/wcag-2-released/

[01:13:00.0000] <Lachy>
that links to http://juicystudio.com/article/wcag2surpriserelease.php ...

[01:13:01.0000] <Lachy>
which links to http://accessify.com/news/2007/04/wcag-20-finally-here/ ...

[01:13:02.0000] <Lachy>
and then to ... http://boxofchocolates.ca/archives/2007/04/01/wcag2-a-done-deal ...

[01:13:03.0000] <othermaciej>
looks like they coordinated

[01:13:04.0000] <Lachy>
http://www.einfach-fuer-alle.de/blog/eintraege.php?id=2044_0_1_0 and finally back to Bruce Lawson's

[01:14:00.0000] <zcorpan_>
fishy eh

[01:14:01.0000] <Lachy>
it's an april fools joke

[01:14:02.0000] <zcorpan_>
yup

[01:14:03.0000] <zcorpan_>
othermaciej has another :)

[01:15:00.0000] <Lachy>
I got suspicious of that when I noticed that bruce didn't actually link to the new spec

[01:18:00.0000] <othermaciej>
who, me?

[01:18:01.0000] <othermaciej>
/me looks innocent

[01:19:00.0000] <zcorpan_>
hehe

[01:19:01.0000] <zcorpan_>
/me looks forward to safari 3 using trident

[01:21:00.0000] <Lachy>
excellent! Now we just need Opera and Mozilla to drop their engines and adopt trident to, and we'll have 4 fully interoperable browsers!

[01:21:01.0000] <othermaciej>
apparently mozilla is adopting WebKit: http://burntelectrons.org/index.php?itemid=219

[01:24:00.0000] <Lachy>
so when will apple begin shipping vista on all new macs?

[01:25:00.0000] <zcorpan_>
1 april 2008

[01:25:01.0000] <othermaciej>
I cannot comment on future product releases

[01:25:02.0000] <hsivonen>
/me goes find the latest RFC

[01:29:00.0000] <hsivonen>
doesn't IETF have an April 1 RFC this year at all?

[01:30:00.0000] <othermaciej>
maybe they have not posted it yet

[01:33:00.0000] <Lachy>
http://en.wikipedia.org/wiki/April_1st_RFC - they didn't post one last year either

[01:34:00.0000] <othermaciej>
maybe they decided it wasn't funny any more

[01:36:00.0000] <othermaciej>
haha, UTF-9

[01:36:01.0000] <Lachy>
we should post something on whatwg blog.  hmm. any ideas?

[01:37:00.0000] <othermaciej>
adopting XHTML2 might be too obvious

[01:37:01.0000] <Lachy>
that's what I thought

[01:37:02.0000] <othermaciej>
how about a new RDF serialization of HTML5?

[01:37:03.0000] <othermaciej>
probably also too obvious

[01:37:04.0000] <Lachy>
adopting the role attribute?

[01:38:00.0000] <othermaciej>
that might not be obvious enough :-)

[01:39:00.0000] <hsivonen>
a few years back Jukka Korpela wrote about a new version of HTML on April 1st, but, in reality, XSL-FO was pretty much like his joke...

[01:40:00.0000] <Lachy>
yeah, XHTML 3.0!

[01:40:01.0000] <hsivonen>
http://www.cs.tut.fi/~jkorpela/html/xhtml3.html

[01:42:00.0000] <hsivonen>
an N3 serialization for HTML5 would take the RDF idea a step further

[01:44:00.0000] <hsivonen>
(in general, I'm not a big fan of April Fools jokes polluting the information space)

[01:52:00.0000] <Hixie>
if someone posts an april 1st blog entry, they better have an april 2nd blog entry lined up

[01:53:00.0000] <Lachy>
why? To explain that the first was a joke?

[01:54:00.0000] <Hixie>
no, to not confuse people when april fools is over

[01:54:01.0000] <Lachy>
ok

[01:54:02.0000] <Hixie>
:-)

[02:26:00.0000] <ROBOd>
will someone post an April fools joke? :)

[02:26:01.0000] <Lachy>
if someone comes up with a good idea for one

[02:27:00.0000] <mjshtml>
Hixie to be replaced as HTML5 editor by Chris Wilson

[02:27:01.0000] <Lachy>
LOL!

[02:27:02.0000] <ROBOd>
"The WHATWG will disolve given that Chris Wilson has taken over doing a great job" (scary joke)

[02:28:00.0000] <Lachy>
who wants to write it?

[02:28:01.0000] <mjshtml>
or Steven Pemberton, take your pick

[02:28:02.0000] <mjshtml>
maybe that can be the 4/2 clarification

[02:29:00.0000] <Lachy>
nah, Steve's doing a great job with XHTML2.  Chris is more believable since he's actually joining the HTMLWG

[02:30:00.0000] <mjshtml>
"doing a great job with XHTML2"

[02:30:01.0000] <mjshtml>
parse error

[02:31:00.0000] <hsivonen>
people may take a Chris Wilson joke too seriously

[02:31:01.0000] <hsivonen>
RDF fun is safer

[02:32:00.0000] <mjshtml>
Chris Wilson may take a Chrisl Wilson joke too seriously

[02:32:01.0000] <Lachy>
I don't know RDF well enough to write anything about it

[02:32:02.0000] <ROBOd>
"Due to many complaints we've decided rewrite the entire HTML 5 specification, from scratch"

[02:32:03.0000] <mjshtml>
I think it would be funny if written right, you just have to get more outrageous as you go along

[02:32:04.0000] <mjshtml>
start believable and work your way up

[02:33:00.0000] <ROBOd>
mjshtml: exactly

[02:36:00.0000] <Lachy>
what about a reason for Hixie no longer doing it?  Lack of time? motivation? support? illness? something else?

[02:36:01.0000] <ROBOd>
another joke .. but this might be also taken too seriously and it's quite subtle: start from this email http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2004-August/002069.html

[02:38:00.0000] <hsivonen>
I think writing the spec from scratch with outrageous technical details is the best one so far (I'd avoid stuff that might start serious PR problem rumors. like Hixie getting sick or legal stuff)

[02:38:01.0000] <ROBOd>
yes

[02:38:02.0000] <Lachy>
how about adopting XForms Transitional?

[02:39:00.0000] <ROBOd>
Lachy: just looked yesterday at XForms Transitional... uhm... that's soo "very" Web Forms2 (IIANM)

[02:39:01.0000] <ROBOd>
(besides being still a work in progress...)

[02:40:00.0000] <hsivonen>
XForms Transitional, XSL-FO, RDF, XHTML 2.0, W3C XML Schema, SOAP, ...

[02:40:01.0000] <mjshtml>
forget XForms Transitional, how about straight-up XForms?

[02:40:02.0000] <Lachy>
XForms transitional does seem to be an effort to rewrite WF2 from scratch

[02:40:03.0000] <ROBOd>
Lachy: what's the use?

[02:40:04.0000] <mjshtml>
the <style> element will no longer contain CSS, instead it will include an inline XSLT stylesheet

[02:40:05.0000] <Lachy>
the use of what?

[02:41:00.0000] <mjshtml>
the global href attribute from XHTML2 will be replaced by a global xlink:href attribtue from XLink

[02:41:01.0000] <ROBOd>
Lachy: of rewriting from scratch WF2?

[02:41:02.0000] <Lachy>
I don't know, ask DanC

[02:41:03.0000] <mjshtml>
the global src attribute will also be replaced by a global xlink:href attribute from XLink

[02:41:04.0000] <ROBOd>
Lachy: he's not around :)

[02:41:05.0000] <mjshtml>
it writes itself

[02:41:06.0000] <mjshtml>
actually, DaveR would be the one to ask

[02:42:00.0000] <Lachy>
oh, yes, sorry

[02:42:01.0000] <ROBOd>
mjshtml: yes, Dave Ragget?

[02:42:02.0000] <mjshtml>
yes

[02:42:03.0000] <othermaciej>
he invented it

[02:42:04.0000] <hsivonen>
HTML5 will be rewritten from scratch to separate semantics (RDF) from presentation (XSL-FO) and behavior (the XML script thing from XForms). the normative schema will be in XSD and the for enterprise-readiness, the transport will be SOAP with WS-Security for making it bullet-proof

[02:44:00.0000] <ROBOd>
hsivonen: for it to be a good joke, you must reference some Microsoft proprietary format :)

[02:44:01.0000] <hsivonen>
ROBOd: it can be deployed today thanks to WPF/E?

[02:44:02.0000] <othermaciej>
hah

[02:45:00.0000] <ROBOd>
"HTML 5 will be based off the new ISO standard by Microsoft: Office Open XML"

[02:45:01.0000] <othermaciej>
haha

[02:45:02.0000] <othermaciej>
the problem with redesigning HTML in a bad way is it's hard to fit in all the possible bad technologies

[02:45:03.0000] <ROBOd>
90% of the users will have support for viewing the documents

[02:46:00.0000] <ROBOd>
othermaciej: exactly, so pick one really bad. OOXML :)

[02:48:00.0000] <ROBOd>
advantages must also be stated clearly: "1. market penetration (even your secretary knows how to view and edit such OOXML documents).  2. we firmly believe that this ISO standard is written for better interoperability between implementations, making life easier for developers and users."

[02:48:01.0000] <ROBOd>
and 3. we do not really want to reinvent the wheel :)

[02:49:00.0000] <othermaciej>
hah

[02:50:00.0000] <hsivonen>
ROBOd: you can have the best of both worlds by defining an XSLT mapping between OOXML and RDF/XSL-FO

[02:50:01.0000] <ROBOd>
hsivonen: hahahaha, exactly that's what i was writing!

[02:50:02.0000] <ROBOd>
lol

[02:51:00.0000] <Lachy>
ROBOd, are you writing the post?

[02:51:01.0000] <ROBOd>
the greatest advantage is you can use only one can XSLT to convert your HTML4 (error on purpose)  to OOXML HTML5, and vice-versa

[02:51:02.0000] <othermaciej>
maybe it should use a combination of GRDDL and XSLT

[02:51:03.0000] <othermaciej>
GRDDL to generate RDF, and XSLT to generate XSL-FO

[02:52:00.0000] <ROBOd>
Lachy: yeah :). how can I do this?

[02:52:01.0000] <ROBOd>
if you want i can write it

[02:52:02.0000] <ROBOd>
(i am not *writing* it as we speak, of course)

[02:52:03.0000] <Lachy>
go for it

[02:53:00.0000] <ROBOd>
Lachy: do i have to signup somewhere and login? or i shall write the HTML document and send it to you via email?

[02:54:00.0000] <Lachy>
just write it, put a draft somewhere so we can review it, then I can publish the final version for you

[02:54:01.0000] <ROBOd>
Lachy: ok, will do

[02:54:02.0000] <Lachy>
or you can just register on blog.whatwg.org and do it yourself

[02:54:03.0000] <ROBOd>
i suppose it doesn't have to be long :P (the blog post)

[02:55:00.0000] <Lachy>
no, just 2 or 3 paragrahs should be sufficient

[02:56:00.0000] <ROBOd>
yes

[02:57:00.0000] <Lachy>
is anyone able to understand this? http://www.w3.org/mid/711a73df0704010055i13fe8c38vc710cb1016f6e344⊙mgc

[02:57:01.0000] <hsivonen>
the serialization could be OOXML but the *architectural model* could be RDF/XSL-FO/XForms-script-thing

[02:58:00.0000] <othermaciej>
hah

[02:58:01.0000] <othermaciej>
make sure to work in the word "backplane"

[03:00:00.0000] <hsivonen>
Lachy: assuming that Dave P isn't joking today, I am rather surprised by his position

[03:04:00.0000] <Lachy>
I just don't understand why downstream processijng of xml requires validation, or what versioning has to do with that

[03:05:00.0000] <hsivonen>
Lachy: it requires validation if your processing code barfs on invalid stuff

[03:06:00.0000] <hsivonen>
Lachy: what I don't understand is how versioning saves you in that case

[03:06:01.0000] <hsivonen>
Lachy: basically, validating first allows for less robust Java/Python/whatever code

[03:06:02.0000] <Lachy>
I also don't understand what that has to do with HTML and CSS, which the thread is about

[03:06:03.0000] <hsivonen>
Lachy: since the code can assume the doc met the constraints already

[03:07:00.0000] <hsivonen>
Lachy: Dave P is an XML guy, so perhaps he isn't thinking about a browser use case but a semi-private system use case

[03:09:00.0000] <ROBOd>
done

[03:09:01.0000] <ROBOd>
the document is available locally... i don't like pasting links on the channel (it's logged)

[03:09:02.0000] <ROBOd>
Lachy: can I PM you?

[03:09:03.0000] <Lachy>
sure

[03:09:04.0000] <ROBOd>
thanks

[03:10:00.0000] <hsivonen>
Lachy: can I see it as a draft in WP?

[03:10:01.0000] <Lachy>
yeah

[03:11:00.0000] <ROBOd>
erm, there are some typos and grammar errors :)

[03:13:00.0000] <hsivonen>
whoa: http://www.w3.org/TR/backplane/

[03:13:01.0000] <Lachy>
Should we call it HTML6 intead?

[03:14:00.0000] <othermaciej>
did you mention the backplane?

[03:14:01.0000] <othermaciej>
it's essential for maximum ludicrousness value

[03:14:02.0000] <hsivonen>
enterprise-strength backplane

[03:14:03.0000] <Lachy>
people might find it hard to believe that we're dropping everything we've already done, it would be more believable if they think it's the future replacement

[03:14:04.0000] <ROBOd>
othermaciej: change the article, if you have access

[03:14:05.0000] <othermaciej>
people will believe anything

[03:15:00.0000] <othermaciej>
I don't

[03:15:01.0000] <ROBOd>
Lachy: good point

[03:16:00.0000] <hsivonen>
Lachy: dropping everything is good

[03:16:01.0000] <ROBOd>
name it HTML 6 ... "we are now planning for the future HTML 6, since ... HTML 5 is already being worked on by the W3C"

[03:16:02.0000] <hsivonen>
Lachy: otherwise HTML5 needs to become HTML6 Transitional

[03:16:03.0000] <Lachy>
hsivonen, that would be DaveR's job

[03:16:04.0000] <ROBOd>
:)

[03:17:00.0000] <hsivonen>
the good thing with global warming is that ocean boiling requires less additional energy

[03:18:00.0000] <ROBOd>
Lachy: let me know when the joke is posted :)

[03:19:00.0000] <Lachy>
I'm just making a few modifications to call it HTML6 and change the first paragraph accordingly

[03:19:01.0000] <hsivonen>
/me hopes the X people won't get too bitterly offended

[03:19:02.0000] <Lachy>
othermaciej, I don't know what to say about the backplane

[03:19:03.0000] <Lachy>
hi annevk

[03:19:04.0000] <ROBOd>
Lachy: just say something about the b*l*ackplane

[03:19:05.0000] <ROBOd>
:)

[03:19:06.0000] <hsivonen>
Lachy: I can try to work the backplane in

[03:20:00.0000] <Lachy>
ok, I'll save the draft

[03:21:00.0000] <hsivonen>
/me looks

[03:21:01.0000] <annevk>
maybe we should have something about the UN negotiating a truce between the W3C and WHATWG?

[03:21:02.0000] <othermaciej>
Lachy: what you say doesn't have to mean anything

[03:21:03.0000] <othermaciej>
that's the key

[03:21:04.0000] <othermaciej>
"backplane" means whatever you want it to mean

[03:21:05.0000] <annevk>
othermaciej, btw, review the dev.w3.org version of access-control, not the thing published on /tr/

[03:22:00.0000] <hsivonen>
ok if I work in a bit more X stuff along the lines I said earlier?

[03:22:01.0000] <ROBOd>
Lachy: in the advantages say ...

[03:22:02.0000] <annevk>
"backplane" is the XForms data model + more of where that came from

[03:22:03.0000] <annevk>
/me isn't sure what the XForms data model is, exactly

[03:23:00.0000] <ROBOd>
"Having XML as the backplane, developers will see the immediate benefit of using their existing XML parsers and serializers"

[03:23:01.0000] <annevk>
you want "XML toolchain" I think

[03:24:00.0000] <ROBOd>
annevk: xforms doesn't fit into the OOXML archtecture model

[03:24:01.0000] <annevk>
yeah, well, what does?

[03:24:02.0000] <Lachy>
annevk, HTML6 does, apparently :-)

[03:24:03.0000] <ROBOd>
annevk: yeah, whatever :) ... the XML toolchain, saw fit to use parser and serializer

[03:25:00.0000] <annevk>
maybe we should also state something that's not a joke

[03:25:01.0000] <annevk>
after HTML5 we'll focus on SVG5

[03:28:00.0000] <ROBOd>
"We also sent a proposal to the CSS WG: we would like CSS 4 to marry the syntax of XSLT and the power of CSS" (lol)

[03:28:01.0000] <ROBOd>
css3-syntax-sugar :)

[03:29:00.0000] <ROBOd>
(which is available as a css 3 module; future revisions becoming css 4)

[03:30:00.0000] <hsivonen>
still editing

[03:31:00.0000] <ROBOd>
:)

[03:33:00.0000] <annevk>
you mean the syntax of XSL-FO

[03:33:01.0000] <annevk>
XSLT is sort of orthogonal to CSS at this point

[03:34:00.0000] <ROBOd>
annevk: hence... the idea :)

[03:35:00.0000] <ROBOd>
annevk: i also stated that XSLT can be used to transform HTML4 to HTML 6 OOXML :)

[03:35:01.0000] <ROBOd>
a simple XSLT document, that is

[03:35:02.0000] <hsivonen>
ok. I worked in some more buzzwords

[03:36:00.0000] <ROBOd>
but keep it believable :) at least for newbies :)

[03:36:01.0000] <hsivonen>
minor edit still

[03:36:02.0000] <hsivonen>
Lachy: did you intend to publish take two tomorrow?

[03:37:00.0000] <hsivonen>
still a bit more editing

[03:38:00.0000] <Lachy>
oh, oops, I meant to take that out of the draft.  ROBOd added that for tomorrows post, but I don't think we need it.

[03:38:01.0000] <hsivonen>
done

[03:38:02.0000] <Lachy>
we could just publish something on a more serious topic

[03:38:03.0000] <Lachy>
like <video>

[03:39:00.0000] <hsivonen>
Lachy: do you approve of my edits? are they offensive?

[03:39:01.0000] <annevk>
seen http://www.google.com/tisp/ already?

[03:39:02.0000] <ROBOd>
annevk: yeah, kinda ... weak

[03:39:03.0000] <Lachy>
/me looks

[03:40:00.0000] <ROBOd>
i liked http://www.google.com/tisp/notfound.html

[03:42:00.0000] <annevk>
heh http://groups.google.com/group/google-tisp/

[03:43:00.0000] <hsivonen>
Lachy: FWIW, I think what I wrote is pretty tame compared to real suggestions. the potentially offensive part is framing stuff that reads like real suggestions as a joke

[03:45:00.0000] <Lachy>
it's funny.  Most people will find it believable because they have no idea what on earth you're talking about :-)

[03:46:00.0000] <annevk>
do we have multiple posts?

[03:46:01.0000] <Lachy>
I took out the take two section aswell

[03:46:02.0000] <Lachy>
no, just the one on blog.whatwg.org

[03:47:00.0000] <hsivonen>
Lachy: are you editing still? I suggest chaging "integration with SOAP-based" to "binding with..."

[03:47:01.0000] <othermaciej>
/me looks forward to seeing it

[03:47:02.0000] <annevk>
/me too

[03:47:03.0000] <Lachy>
hsivonen, fixed

[03:48:00.0000] <hsivonen>
Lachy: thanks

[03:48:01.0000] <annevk>
"Founded in 1998 by Stanford Ph.D. wannabes Larry Page and Sergey Brin,"

[03:51:00.0000] <Lachy>
I think it's ready to publish, unless you have any last minute changes

[03:51:01.0000] <Lachy>
I changed the title to Plans for HTML6

[03:51:02.0000] <hsivonen>
Lachy: the second to last paragraph isn't funny enough, but I don't have good fixes

[03:53:00.0000] <hsivonen>
afk.

[03:53:01.0000] <othermaciej>
if someone sends by privmsg I can make suggestions

[03:53:02.0000] <othermaciej>
or y'all can just post it, I'm sure it's amusing enough

[03:53:03.0000] <ROBOd>
othermaciej: agreed

[03:53:04.0000] <Lachy>
I'll put it online temporarily for those of you without blog accts...

[03:54:00.0000] <ROBOd>
registered

[03:55:00.0000] <Lachy>
you probably won't be able to see it cause your not its editor or an admin

[03:55:01.0000] <annevk>
you can also just publish it and make changes later...

[03:56:00.0000] <ROBOd>
ah, yes

[03:57:00.0000] <othermaciej>
the only change I'd make is to remove "The above being said, "

[03:58:00.0000] <othermaciej>
it's pretty amusing

[03:58:01.0000] <ROBOd>
it's pretty good

[03:58:02.0000] <Lachy>
othermaciej, done

[03:58:03.0000] <ROBOd>
good additions

[04:06:00.0000] <ROBOd>
still editing?

[04:06:01.0000] <hsivonen>
Using a fairly simplistic XSLT should probably be "Using a fairly simplistic XSLT style sheet"

[04:09:00.0000] <hsivonen>
let's ship it!

[04:09:01.0000] <Lachy>
hsivonen, done.

[04:10:00.0000] <hsivonen>
Lachy: thanks

[04:11:00.0000] <othermaciej>
is it there yet?

[04:13:00.0000] <Lachy>
just making a few minor changes that annevk PM'd me

[04:14:00.0000] <Lachy>
http://blog.whatwg.org/html6-plan

[04:14:01.0000] <annevk>
I think we should have something like "reveal our behind-closed-doors-decided plans ..."

[04:14:02.0000] <hsivonen>
thanks to everyone who participated

[04:16:00.0000] <ROBOd>
:)

[04:16:01.0000] <annevk>
/me submits a comment

[04:16:02.0000] <othermaciej>
lol

[04:16:03.0000] <ROBOd>
annevk: already did

[04:17:00.0000] <ROBOd>
seriously now... any ETA?

[04:17:01.0000] <ROBOd>
lol

[04:17:02.0000] <Lachy>
annevk, lol!

[04:18:00.0000] <ROBOd>
i think the use of WHATTF name adds to the point :)

[04:18:01.0000] <hsivonen>
http://whattf.org/

[04:18:02.0000] <Lachy>
oh crap, we got the expansion wrong for WHATF

[04:18:03.0000] <Lachy>
*WHATTF

[04:18:04.0000] <ROBOd>
hsivonen: precisely

[04:21:00.0000] <annevk>
/me changes topic to 'WHATWG (HTML6) -- http://www.whattf.org/ -- Logs: http://krijnhoetmer.nl/irc-logs/ -- Please leave your sense of logic at the door, thanks!'

[04:21:01.0000] <annevk>
/me changes topic to 'WHATTF (HTML6) -- http://www.whattf.org/ -- Logs: http://krijnhoetmer.nl/irc-logs/ -- Please leave your sense of logic at the door, thanks!'

[04:21:02.0000] <ROBOd>
:)

[04:46:00.0000] <ROBOd>
we have the HTML 6 plans blog post, the HTML 6 topic, .... we only need a HTML 6 thread on the mailing list

[04:47:00.0000] <hsivonen>
I think the blog comment work better today than email

[04:48:00.0000] <ROBOd>
someone digg&slashdot the article

[04:51:00.0000] <ROBOd>
i have to go now, i'll be back later

[04:57:00.0000] <Lachy>
it would be funny if it made slashdot and people thought it was sersious :-)

[05:51:00.0000] <Dashiva>
"A++++ would implement again" 

[05:51:01.0000] <Lachy>
what?

[05:52:00.0000] <Dashiva>
I was just thinking of a comment for the HTML6 post

[05:56:00.0000] <met_>
/me just read about HTML 6

[06:00:00.0000] <hsivonen>
hmm. I have 21 slides for a 15-minute presentation about HTML5 conformance checking

[06:01:00.0000] <hsivonen>
have to practice some more to squeeze it a bit

[06:01:01.0000] <Lachy>
that should be enough

[06:01:02.0000] <Lachy>
how long are you spending on each slide?

[06:01:03.0000] <hsivonen>
slightly less than a minute on average. my last practice run was 18 minutes

[06:02:00.0000] <annevk>
wow, that's a lot of slides

[06:02:01.0000] <hsivonen>
yes :-(

[06:02:02.0000] <annevk>
/me had eight / nine for a twenty minute presentation

[06:02:03.0000] <hsivonen>
but these are Keynote slides not PowerPoint slides

[06:03:00.0000] <annevk>
what does that mean? one word per slide?

[06:03:01.0000] <hsivonen>
annevk: something like that :-)

[06:03:02.0000] <hsivonen>
I have 5 slides up front about what HTML5 and WHATWG are

[06:03:03.0000] <hsivonen>
I think I should try to squeeze that part

[06:04:00.0000] <Lachy>
can we see the slides?

[06:04:01.0000] <jgraham>
/me has managed both 40 minute talks on <30 slides and 17 minute talks on 25 slides...

[06:04:02.0000] <Lachy>
oh, never mind, I don't have keynote anyway

[06:05:00.0000] <hsivonen>
Lachy: http://hsivonen.iki.fi/thesis/dippaesitelma.pdf

[06:05:01.0000] <Lachy>
hmm. Apparently a trial version of keynote came with my mac.  Nice!

[06:07:00.0000] <jgraham>
hsivonen: Do you know the projector you're using? We have one with an odd issue with white-on-black (move your head and you see rainbows)

[06:07:01.0000] <hsivonen>
jgraham: I don't

[06:07:02.0000] <hsivonen>
jgraham: I could make an inverted backup

[06:08:00.0000] <jgraham>
I've only ever seen the problem with this one projector but it is quite annoying...

[06:10:00.0000] <hsivonen>
perhaps I should get rid of references to the WHATWG and not explain the context on that level

[06:13:00.0000] <hsivonen>
pruned to 18 slides

[06:13:01.0000] <Lachy>
which ones did you take out?

[06:13:02.0000] <hsivonen>
the ones about the WHATWG and new and old features

[06:14:00.0000] <hsivonen>
not that relevant to the implementation of conformance checking

[06:14:01.0000] <Lachy>
yeah, I was going to suggest those because they're not relevant

[06:35:00.0000] <hsivonen>
when I had the presentation about Gecko bug 18333, I made sure that I don't go overtime and then I was too quick...

[06:47:00.0000] <hsivonen>
I wonder if there a drop-in color invert filter for Quartz...

[12:20:00.0000] <Hixie>
i think i know why i'm uncomfortable with the proposed ready states

[12:20:01.0000] <Hixie>
they don't all apply to audio

[12:25:00.0000] <Hixie>
also, i don't like the idea of a readyState-like API not only incrementing

[12:43:00.0000] <virtuelv>
Hixie: huh? care to elaborate (about readyStates incrementing)

[12:43:01.0000] <Hixie>
for <video>

[12:43:02.0000] <Hixie>
http://webkit.org/specs/HTML_Timed_Media_Elements.html#mediastatus

[12:44:00.0000] <virtuelv>
I got that part, but why are you uncomfortable with a non-incremental value for readyState?

[12:48:00.0000] <virtuelv>
(if anything, I'm a bit uncomfortable not being able to predict when the buffer will be empty beforehand

[12:49:00.0000] <Hixie>
the proposal from apple has the state going 0..2..3..4..2..3..2..3..4..5..2..3..4..2..6

[12:49:01.0000] <Hixie>
instead of 0..1..2..3..4..55

[12:50:00.0000] <Hixie>
http://www.whatwg.org/specs/web-apps/current-work/#media is what i'm thinking of using

[12:50:01.0000] <Hixie>
(not yet specced out)

[12:50:02.0000] <Hixie>
(just in the idl)

[12:53:00.0000] <annevk>
loaded is 10 because you expect many more states at some point?

[12:53:01.0000] <Hixie>
actually loaded was 10 in order to have the buffered and networking states in sync

[12:53:02.0000] <Hixie>
but screw that

[12:54:00.0000] <Hixie>
we don't need loaded and partial under buffering

[12:56:00.0000] <annevk>
so which states don't make sense for audio?

[12:57:00.0000] <Hixie>
pausable

[12:57:01.0000] <Hixie>
but i think we'll deal with it

[12:58:00.0000] <annevk>
also, how about replacing .seek() with a read/write .position

[12:59:00.0000] <virtuelv>
again, what worries me a bit is that there's insufficient information in both suggestions to determine when to start playing back and be guaranteed a skip-free playback

[12:59:01.0000] <annevk>
that will also allow you to do the theoretical .advance() easily

[12:59:02.0000] <annevk>
virtuelv, i think ENDABLE is for that

[13:00:00.0000] <virtuelv>
annevk: that is not neccesarily always what you want

[13:00:01.0000] <virtuelv>
the latency for that might be too long, depending on the length of the media

[13:01:00.0000] <Hixie>
yes seek() is going away

[13:01:01.0000] <Hixie>
one thing at a time

[13:03:00.0000] <Hixie>
virtuelv: the state will become ENDABLE when the UA believes that at the current buffering rate, you'll be able to reach the end just as the end is downloaded.

[13:03:01.0000] <Hixie>
is that what you meant?

[13:09:00.0000] <virtuelv>
it is what I meant, but you can't always guarantee that endable is a definite state

[13:10:00.0000] <Hixie>
could you elaborate?

[13:10:01.0000] <virtuelv>
someone might, for instance, start a download at the same time, at which point the movie will no longer be endable

[13:10:02.0000] <Hixie>
yup, then the state will no longer be that

[13:12:00.0000] <Hixie>
what information would you want instead?

[13:14:00.0000] <virtuelv>
usually, someone streaming media will know the size and duration of the media, and knowing exactly how much is in the browser's buffer/cache and the current download rate would make more sense to me

[13:15:00.0000] <Hixie>
that information is already available in the api

[13:42:00.0000] <othermaciej>
Hixie: what you came up with seems pretty complicated

[13:42:01.0000] <othermaciej>
Hixie: three separate state enums?

[13:43:00.0000] <othermaciej>
plus a bunch of booleans

[13:43:01.0000] <othermaciej>
I think the only one of Apple's states that might not apply to pure audio is PRESENTABLE

[13:49:00.0000] <Hixie>
i removed a bunch of hte booleans that were there before actually

[13:50:00.0000] <Hixie>
the only real difference between this and the apple proposal is the splitting of the 'state' into two, one for network state, and one for the ready state

[13:50:01.0000] <Hixie>
which are orthogonal anyway

[13:50:02.0000] <Hixie>
since 'ready' state can be affected by things like seeking even when the whole file is downloaded

[13:51:00.0000] <othermaciej>
that and you made playback state a quad-state instead of a boolean

[13:51:01.0000] <Hixie>
well that was there before

[13:51:02.0000] <Hixie>
you need it to be quadstate, otherwise you don't know when to show the "i'm _trying_ to play!" spinner

[13:52:00.0000] <othermaciej>
you could just say paused reflects only the user-chosen playback state

[13:53:00.0000] <othermaciej>
I'm not sure there's a need to reflect EMPTY vs. PAUSED (in purely playback terms, as opposed to the EMPTY state in the two other concepts of state) or PLAYING vs. WAITING

[13:54:00.0000] <othermaciej>
wouldn't the state of the controls be exactly the same in case of the latter two?

[13:55:00.0000] <Hixie>
the controls would be, but in waiting you'd have a spinner

[13:55:01.0000] <Hixie>
this is something that the quicktime and frontrow renderers do really badly actually

[13:55:02.0000] <othermaciej>
you would?

[13:55:03.0000] <Hixie>
they just freeze when they're seeking or buffering, with no indication to the user that they're working and not locked up

[13:56:00.0000] <Hixie>
obviously there's no requirement that the author provide such a ui

[13:56:01.0000] <Hixie>
but it should at least be possible

[13:56:02.0000] <othermaciej>
as does YouTube

[13:56:03.0000] <othermaciej>
and Google Video

[13:56:04.0000] <othermaciej>
does anyone's actual UI show a spinner in such a case?

[13:57:00.0000] <Hixie>
those two at least shows the buffered bar, so there's some indication of why it's paused

[13:57:01.0000] <othermaciej>
yes, so does QuickTime

[13:58:00.0000] <othermaciej>
that's kind of the de facto standard for such things

[13:58:01.0000] <Hixie>
well for some reason i am much more often confused as to whether quicktime is locked up or not

[13:58:02.0000] <Hixie>
in any case, i don't see the disadvantage of exposing this information. Is it hard to implement or something?

[13:59:00.0000] <othermaciej>
no, it's just more complicated to use

[13:59:01.0000] <othermaciej>
since you have extra states that don't relate to anything you'd do in the UI

[13:59:02.0000] <Hixie>
how about if i make paused = 0 and playing and waiting 1 and 2 respectively

[13:59:03.0000] <Hixie>
then you can do if (video.playing)

[13:59:04.0000] <Hixie>
or playbackState or whatever it's called

[14:01:00.0000] <othermaciej>
the fact that "waiting" is useful only for a UI idea you had that no one actually uses makes me doubt it should be overloaded onto the same variable that tells you playing vs. paused

[14:01:01.0000] <Hixie>
i suggested having two booleans and you were against that too

[14:03:00.0000] <othermaciej>
well, Apple's proposal gives you enough info to tell if you are really advancing in time without having a specific state value or bool for it, since it is of marginal value for reflection in the UI

[14:04:00.0000] <othermaciej>
I'm not sure if your version still does, since I don't know what the states mean or what the allowed transitions are

[14:04:01.0000] <othermaciej>
but I gotta say that three separate state enums, all of which are primarily useful to tell you what actions are allowed now, triggers my complexity overload alert

[14:05:00.0000] <Hixie>
well it's a complicated problem

[14:05:01.0000] <Hixie>
the three states are orthogonal

[14:06:00.0000] <Hixie>
(regarding the rate thing -- that part of the apple proposal wasn't really clear to me)

[14:06:01.0000] <Hixie>
(i couldn't work out what values the currentRate would have other than 0 and the playbackRate, at which point it is exactly the same as a boolean)

[14:06:02.0000] <Dashiva>
As long as a "basic user" can ignore most of the states and still get a sensible interface, the complexity isn't necessarily bad

[14:06:03.0000] <othermaciej>
Hixie: you could tell even without looking at the rate, since you can see whether your current state is playable, and if you are not paused; only if both are true are you actually advancing

[14:06:04.0000] <Hixie>
(and i don't see that two booleans, playing and 'playbackRate', which can never be false and true respectvely, is better than a three-state attribute)

[14:07:00.0000] <Hixie>
"(playable || endable) && (!paused)"

[14:08:00.0000] <Hixie>
and even that's not true, because in the autoplay case you'll wait for endable, not playable

[14:08:01.0000] <othermaciej>
video.readyState < PLAYABLE && !video.paused is your WAITING state

[14:08:02.0000] <othermaciej>
in the autoplay case, you are paused until autoplay starts

[14:08:03.0000] <Hixie>
so paused _isn't_ the user's state.

[14:08:04.0000] <othermaciej>
(that's how it should be anyway, since the user should still be able to hit "play" early)

[14:08:05.0000] <Hixie>
how can the user prevent autoplay then?

[14:09:00.0000] <othermaciej>
you can't

[14:09:01.0000] <Hixie>
...

[14:09:02.0000] <othermaciej>
you have to hit pause after it autoplays

[14:09:03.0000] <othermaciej>
unless there is some special control for it

[14:09:04.0000] <Hixie>
what would the control do?

[14:09:05.0000] <othermaciej>
that's how things that autoplay work now

[14:09:06.0000] <othermaciej>
no idea

[14:09:07.0000] <othermaciej>
remove the "autoplay" attribute?

[14:10:00.0000] <Hixie>
so autoplay would be implemented as a default action of the transition to endable?

[14:10:01.0000] <Hixie>
i guess that could work

[14:10:02.0000] <othermaciej>
I really don't like endable as a name either

[14:10:03.0000] <Hixie>
(i'm not sure designing the api around current UIs to the exclusion of better UIs is good design, though)

[14:10:04.0000] <othermaciej>
it implies the author or user could take the action of ending in that state

[14:10:05.0000] <Hixie>
the names can be sorted out later

[14:11:00.0000] <othermaciej>
well, some of the names actively confuse me

[14:11:01.0000] <othermaciej>
I don't know what pausable means

[14:11:02.0000] <Hixie>
same as your presentable

[14:11:03.0000] <othermaciej>
I don't know what the distinction between "loading" and "buffering" is (aren't the two words basically synonymous in normal computer use?)

[14:11:04.0000] <Hixie>
loading = we don't have metadata, buffering = we do have metadata

[14:12:00.0000] <Hixie>
i'm open to any name changes

[14:13:00.0000] <othermaciej>
let's set aside the name changes right now and talk about the actual states and transitions first

[14:13:01.0000] <gsnedders>
does anything apart from Gecko trunk support getElementsByClassName?

[14:14:00.0000] <othermaciej>
my net connection is too fast because even on the QuickTime trailers site I can't find in-page video that's large enough that it doesn't autoplay promptly

[14:14:01.0000] <othermaciej>
was gonna give a link to an example of such

[14:15:00.0000] <Hixie>
oh i've seen that ui often

[14:15:01.0000] <Hixie>
drives me crazy

[14:15:02.0000] <Hixie>
i can't go through opening a bunch of videos in tabs and go through and pause them all before they play

[14:16:00.0000] <Hixie>
because if i hit pause, they play instead.

[14:16:01.0000] <Hixie>
but i agree that that's ok

[14:16:02.0000] <Hixie>
we can do autoplay as a default action from the paused state when you transition to 'endable'

[14:16:03.0000] <othermaciej>
I think that is true of all autoplaying UIs (though I don't have a dialup connection handy to test it for youtube)

[14:17:00.0000] <othermaciej>
in Safari we don't start plugins in background tabs so they won't play until you switch to the tab anyway

[14:17:01.0000] <Hixie>
they won't start downloading either

[14:17:02.0000] <Hixie>
so i have to go to the tab to start the download, then start them, then pause them

[14:17:03.0000] <Hixie>
it's annoying

[14:17:04.0000] <othermaciej>
that is true

[14:17:05.0000] <Hixie>
then when i'm ready i have to rewind and play

[14:18:00.0000] <othermaciej>
it would be nice to preload videos in tabs w/o them playing

[14:18:01.0000] <Hixie>
instead of what i'd like to do, which is just, when i'm ready, switch the tab and hit play

[14:18:02.0000] <othermaciej>
I often want that when I'm gonna ride the train

[14:18:03.0000] <Hixie>
but <video> will help with this

[14:18:04.0000] <othermaciej>
it could load but not autoplay

[14:18:05.0000] <othermaciej>
in a background tab

[14:18:06.0000] <Hixie>
since we can have user prefs to disable autoplay

[14:18:07.0000] <othermaciej>
until you switch to the tab

[14:18:08.0000] <othermaciej>
or user prefs

[14:18:09.0000] <Hixie>
or that

[14:18:10.0000] <othermaciej>
but we don't tend to go to prefs as the first thing to solve UI problems

[14:19:00.0000] <Hixie>
yeah the automatic thing is better

[14:19:01.0000] <Hixie>
just not autoplaying in bg tabs

[14:20:00.0000] <Hixie>
anyway

[14:20:01.0000] <Hixie>
so i've collapsed the currentRate 'boolean' and the playing boolean to just the playing three-state thing

[14:21:00.0000] <Hixie>
and split the state in the apple proposal into two states, since the apple proposal didn't allow for the case where even though everyhting is loaded you might not have the current frame ready yet (seeking)

[14:21:01.0000] <Hixie>
those don't seem like an increase in complexity, they seem like a lateral move

[14:21:02.0000] <Hixie>
and an improvement

[14:22:00.0000] <Hixie>
(reload the spec for the latest strawman: http://www.whatwg.org/specs/web-apps/current-work/#media )

[14:24:00.0000] <othermaciej>
I think CAN_PLAY_THROUGH makes more sense than ENDABLE -- first thought was CAN_PLAY_TO_END, but the end isn't even relevant on an infinite stream

[14:24:01.0000] <othermaciej>
and yet you can be in this state (but only if your transfer rate is fast enough)

[14:25:00.0000] <othermaciej>
I think dropping LOADED might be better than this split

[14:25:01.0000] <othermaciej>
LOADED is only useful in theory to tell you that not only can you play through everywhere and do everything, but also that can't possibly change

[14:26:00.0000] <othermaciej>
but if seeking can temporarily change that, then it doesn't have much value

[14:27:00.0000] <othermaciej>
I still don't think WAITING is justified and I don't think you have really countered my argument against it (can discover from other info and UI value is questionable)

[14:27:01.0000] <Hixie>
i disagree, i think you need is so that controllers who join part way can establish what the state of the network connection is

[14:27:02.0000] <Hixie>
(re loaded)

[14:27:03.0000] <Hixie>
since they won't get the progress events

[14:28:00.0000] <othermaciej>
hmm

[14:29:00.0000] <othermaciej>
the thing is,  you might not currently be using the network even when not in LOADED state

[14:30:00.0000] <othermaciej>
if an implementation caches the remote resource in chunks for instance

[14:30:01.0000] <othermaciej>
and won't load the next ever if you pause in the middle of the current one

[14:30:02.0000] <othermaciej>
and it may be impossible to achieve LOADED

[14:30:03.0000] <Hixie>
once you're LOADED, you can go offline without any troubles ever

[14:31:00.0000] <Hixie>
if you're not in LOADED, going offline could show a warning

[14:31:01.0000] <othermaciej>
if you are presenting an rtsp stream, or if your implementation discards buffers

[14:31:02.0000] <othermaciej>
ok, that seems like a valid use

[14:31:03.0000] <othermaciej>
this makes me think LOADING vs. BUFFERING distinction really belongs in the other state, since it is not about network progress but about what you can currently do with the data

[14:32:00.0000] <othermaciej>
(you may be in BUFFERING state when in fact not currently doing a network transfer or planning to)

[14:32:01.0000] <othermaciej>
so perhaps it should be EMPTY, LOADING, LOADED, ERROR for one

[14:32:02.0000] <othermaciej>
and EMPTY, HAVE_METADATA, HAVE_FRAME, PLAYABLE, CAN_PLAY_THROUGH for the other

[14:33:00.0000] <Hixie>
could do

[14:33:01.0000] <Hixie>
not sure it's more useful though

[14:33:02.0000] <othermaciej>
well, HAVE_METADATA vs. HAVE_FRAME seems more clear than LOADING vs. BUFFERING

[14:34:00.0000] <Hixie>
you mean the names?

[14:34:01.0000] <Hixie>
the names can change

[14:34:02.0000] <othermaciej>
and LOADING vs. BUFFERING is not about the state of networking, it is about what level of presentation is available, which is what the readyState is about

[14:35:00.0000] <Hixie>
well it's kind of networking -- LOADING means you haven't received enough data yet to know if you'll go to ERROR or not

[14:35:01.0000] <Hixie>
BUFFERING means you've received enough data to know you're happy and to have initialised the rest of the object

[14:37:00.0000] <Hixie>
the problem with putting METADATA in the readyState is that you then have this awkward EMPTY METADATA HAVE_FRAME distinction

[14:37:01.0000] <Hixie>
where EMPTY becomes effectively useless

[14:37:02.0000] <othermaciej>
EMPTY means you can't do anything

[14:37:03.0000] <Hixie>
since you'll never be EMPTY once you've recieved METADATA

[14:38:00.0000] <othermaciej>
yeah, it's an initial state you can't come back to

[14:38:01.0000] <othermaciej>
much like the network-related EMPTY state

[14:38:02.0000] <Hixie>
conceptually, the current strawman has one state that always increases in value, and the readyState which can switch randomly across all states

[14:38:03.0000] <othermaciej>
you can switch back to EMPTY?

[14:38:04.0000] <Hixie>
i don't really like making the second one parially random-access but partially ordered

[14:39:00.0000] <Hixie>
sure, EMPTY is just "i don't have a frame or anything for this playback position"

[14:39:01.0000] <othermaciej>
I think it is overly simplistic to say state machines must either be linear or always allow transition to all states

[14:39:02.0000] <othermaciej>
I mean, you couldn't build a coke machine with those constraints

[14:39:03.0000] <Hixie>
they don't have to

[14:39:04.0000] <Hixie>
but it makes it way simpler if they do :-)

[14:39:05.0000] <othermaciej>
and that's the canonical example of a simple state machine

[14:39:06.0000] <othermaciej>
simpler for who?

[14:39:07.0000] <othermaciej>
the spec author, who can get out of drawing a state transition diagram?

[14:40:00.0000] <Hixie>
authors to understand, me to spec, implementors to follow...

[14:40:01.0000] <gsnedders>
it's easier to understand when implementing

[14:40:02.0000] <othermaciej>
when implementing, you have to understand every single transition arc

[14:40:03.0000] <Hixie>
right, so it's simpler if those are easyto enumerate

[14:41:00.0000] <othermaciej>
if you can go from any state to any other, that's n^2-n transition arcs

[14:41:01.0000] <Hixie>
there are no more transitions in your proposal (METADATA in readyState) than mine (METADATA in networkState)

[14:41:02.0000] <othermaciej>
that is in fact maximal complexity for an n-state machine, not minimal

[14:42:00.0000] <Hixie>
our proposals have the same number of transitions

[14:42:01.0000] <othermaciej>
sure, I'm just debating the premise that a "random access" state machine is somehow good for implementors

[14:42:02.0000] <Hixie>
i'm not saying that

[14:43:00.0000] <Hixie>
i'm saying that two state machines one of which is purely linear and the other of which is purely random-access is better than two state machines where one is linear and the other is bigger than the random access one, adding one extra arc that isn't random access

[14:44:00.0000] <othermaciej>
and I think if you want to partition state, you should do it by how the state will be used and what it represents, not by what the state transitions look like

[14:45:00.0000] <Hixie>
that's actually what i said originally when you originally suggested that

[14:45:01.0000] <Hixie>
i think having METADATA in the network state is more useful

[14:45:02.0000] <othermaciej>
[ EMPTY --> LOADING --> METADATA --> LOADED ] doesn't make logical sense

[14:46:00.0000] <Hixie>
[ EMPTY -> opening connection -> received metadata -> receiving data -> received all ] makes logical sense

[14:46:01.0000] <Hixie>
however, one argument against this is you'd never stay in the metadata state

[14:46:02.0000] <othermaciej>
it would if "receiving metadata" was actually a separate step

[14:47:00.0000] <othermaciej>
but having the metadata is a side effect of being in the "I'm transferring bytes now" step, not a separate operation

[14:47:01.0000] <Dashiva>
But the "all metadata available" event would be useful, so it has to get into there somehow

[14:48:00.0000] <Hixie>
that's a point

[14:48:01.0000] <Hixie>
having it in the readyState means you'll get that message more often

[14:48:02.0000] <Hixie>
though

[14:48:03.0000] <Hixie>
no

[14:48:04.0000] <othermaciej>
you'll get it any time you seek to a place where you don't have a frame ready immediately

[14:48:05.0000] <Hixie>
we could just say that when you don't get the BUFFERING state until you have the metadata

[14:48:06.0000] <Hixie>
s/when//

[14:48:07.0000] <othermaciej>
but otherwise you would have gotten the readyState EMPTY state then

[14:49:00.0000] <Dashiva>
Or make metadataLoaded a separate boolean with associated event... mhrr

[14:49:01.0000] <Hixie>
actually we can get rid of this step altogether if we have the states split....

[14:50:00.0000] <Hixie>
you just define BUFFERING to happen only after you have metadata

[14:50:01.0000] <Hixie>
then you're set

[14:50:02.0000] <othermaciej>
I'm not sure what that means or how it is different than what the doc says now

[14:51:00.0000] <Hixie>
/me gets himself really confused

[14:51:01.0000] <Hixie>
i just went through three arguments in a row that put me straight back to where i was initially

[14:51:02.0000] <Hixie>
i am not dizzy

[14:51:03.0000] <Hixie>
now

[14:51:04.0000] <othermaciej>
you have [ EMPTY -> LOADING -> BUFFERING -> LOADED ] right now

[14:52:00.0000] <Hixie>
right. that's what i think we want. EMPTY = nothing done. LOADING = sent request. BUFFERING = got back metadata, other fields now are initialised. LOADED = all data now downloaded, you can go offline safely.

[14:52:01.0000] <othermaciej>
I think LOADING vs. BUFFERING is descriptively poor as a way to say whether you have the metadata, and I think having the metadata belongs logically with presentability state, not networking state

[14:52:02.0000] <Hixie>
LOADING doesn't mean you have the metadata

[14:52:03.0000] <Hixie>
LOADING means you don't have anything, but you've requested it

[14:52:04.0000] <othermaciej>
LOADING means you don't have the metadata

[14:52:05.0000] <Hixie>
SENT maybe would be better

[14:52:06.0000] <othermaciej>
BUFFERING means you do

[14:53:00.0000] <othermaciej>
that is the only difference

[14:53:01.0000] <othermaciej>
otherwise they both mean you are receiving some octets over the network

[14:53:02.0000] <Hixie>
right.

[14:53:03.0000] <Hixie>
so what's wrong with that?

[14:54:00.0000] <othermaciej>
the names of the states have no relation to this distinction

[14:54:01.0000] <othermaciej>
and in fact sound like synonyms

[14:54:02.0000] <Hixie>
i'm open to new names

[14:54:03.0000] <othermaciej>
(because from a purely networking POV, they are)

[14:55:00.0000] <Dashiva>
So you will be in BUFFERING state even if you're playing back with fast enough download to never enter buffering mode before end?

[14:55:01.0000] <Hixie>
i thought we weren't discussing names yet :-)

[14:55:02.0000] <othermaciej>
I think the problem with the names is because they are trying to draw a distinction that, from the point of view of networking, does not exist

[14:55:03.0000] <Hixie>
/me changes the names to EMPTY SENT LOADING LOADED

[14:55:04.0000] <Dashiva>
FINISHED :)

[14:56:00.0000] <Hixie>
finished, like language, initialised, and various other words, are bad in APIs because people mistype them too much

[14:56:01.0000] <othermaciej>
SENT vs. LOADING is misleading, since you might take it to mean that in SENT state you have not received any bytes yet

[14:56:02.0000] <othermaciej>
which is not the distinction it is trying to draw

[14:57:00.0000] <othermaciej>
seriously, having the metadata or not is a presnetability issue not a networking issue

[14:57:01.0000] <othermaciej>
putting it into the readyState immediately makes it super clear, for this reason

[14:57:02.0000] <Hixie>
UNSENT, SENT, RECEIVING, DONE? that's what XHR uses.

[14:57:03.0000] <othermaciej>
it's true that you can't go back to not having metadata

[14:57:04.0000] <Hixie>
it totally is a network issue

[14:58:00.0000] <Hixie>
until you have this part of the data, nothing else makes any sense

[14:58:01.0000] <othermaciej>
it is no more a network issue than having the first frame is a network issue

[14:58:02.0000] <Hixie>
except that you can go and lose the first frame

[14:58:03.0000] <Dashiva>
/me wonders about mistyping finished

[14:58:04.0000] <Hixie>
but you can never lose the metadata

[14:59:00.0000] <othermaciej>
whether you can lose that level of presentability is not in any way due to networking

[14:59:01.0000] <othermaciej>
it's because having a frame is dependent on where you are on the timeline

[15:00:00.0000] <Hixie>
granted, but i don't see why that matters. how is having it in the attribute that tells you the state of frame playbackability useful?

[15:01:00.0000] <Hixie>
it seems to me that having the metadata is closer to the network state (it's a milestone along the download) than it is to the seeking and decoding state

[15:01:01.0000] <othermaciej>
that's the state you normally need to track to know what state to put the controls in

[15:02:00.0000] <Hixie>
controls don't disable themselves just because you're not able to play the current frame in current UIs

[15:03:00.0000] <Hixie>
they enable themselves once, when you have enough data to know wtf you're looking at

[15:03:01.0000] <othermaciej>
it's really only the first transition to having a frame that significantly matters, really

[15:04:00.0000] <othermaciej>
since most things only show placeholder UI before you have the first frame, when you seek they usually stick on the last frame seen

[15:04:01.0000] <Hixie>
agreed

[15:05:00.0000] <Hixie>
/me ponders on this

[15:08:00.0000] <Hixie>
(on names, HAVE_FRAME doesn't really work for audio)

[15:08:01.0000] <othermaciej>
that state isn't really relevant to pure audio

[15:09:00.0000] <othermaciej>
since audio is more purely time-based, there's no such thing as presenting a static view of the current time

[15:10:00.0000] <Hixie>
there could be (e.g. embedded static images, captions, visualisation)

[15:12:00.0000] <othermaciej>
a static image along the lines of cover art would be considered metadata, not a playable track

[15:12:01.0000] <othermaciej>
if there is a text track displayed as captions, it's probably reasonable to consider each piece of text a "frame"

[15:13:00.0000] <othermaciej>
probably better than calling the state CAN_SHOW_STATIC_VIEW_OF_CURRENT_TIME

[15:13:01.0000] <om_coffee>
will be back

[15:45:00.0000] <hsivonen>
what's the process of getting something published in w3.org/TR/?

[15:45:01.0000] <hsivonen>
I mean: why is the backplane stuff there?

[15:49:00.0000] <karlUshi>
http://www.w3.org/TR/2006/NOTE-backplane-20061116/

[15:49:01.0000] <karlUshi>
*W3C Coordination Group Note*

[15:50:00.0000] <karlUshi>
"Publication as a Coordination Group Note does not imply endorsement by the W3C Membership. "

[15:51:00.0000] <karlUshi>
ok time to prepare to take a shower 

[15:56:00.0000] <othermaciej>
hsivonen: a W3C Working Group can publish just about anything it wants to as a W3C Note

[15:59:00.0000] <Hixie>
othermaciej: i updated the strawman proposal

[15:59:01.0000] <Hixie>
got rid of WAITING

[15:59:02.0000] <Hixie>
went in the other direction than you were pushing me for the networkState vs readyState though

[15:59:03.0000] <Hixie>
since that gives us useful one-shot events for free

[16:00:00.0000] <Dashiva>
Is the strawman available online?

[16:00:01.0000] <Hixie>
http://www.whatwg.org/specs/web-apps/current-work/#media

[16:00:02.0000] <Hixie>
in the IDL

[16:00:03.0000] <othermaciej>
Hixie: looking

[16:19:00.0000] <othermaciej>
Hixie: I like that somewhat better

[16:19:01.0000] <Hixie>
ooo, progress :-)

[16:19:02.0000] <mpt>
"can display text to the user informing them of how to access the video contents"

[16:19:03.0000] <mpt>
... or an alternative, such as a slideshow or animated GIF

[16:20:00.0000] <othermaciej>
Hixie: I'll have to mull on it a bit

[16:20:01.0000] <Hixie>
k

[16:20:02.0000] <Hixie>
i have to go offline now anyway

[16:20:03.0000] <othermaciej>
ok

[16:20:04.0000] <othermaciej>
ttyl

[16:46:00.0000] <othermaciej>
Hixie: you should update the copyright on Web Apps 1.0 to say "Apple Inc." instead of "Apple Computer, Inc."


2007-04-02
[17:28:00.0000] <Lachy_>
aargh!  that HTML6 post got spammed already! :-(

[17:41:00.0000] <hsivonen>
the HTML6 post didn't make it to slashdot. did it make it to digg?

[17:42:00.0000] <othermaciej>
the webkit post didn't either, but it apparently actually fooled some people at least for a while

[03:03:00.0000] <annevk>
zcorpan_, I fixed the abort() tests here: http://tc.labs.opera.com/apis/XMLHttpRequest/abort/

[03:03:01.0000] <annevk>
zcorpan_, I also fixed the specification to deal with when abort() should not trigger a state change to DONE

[03:04:00.0000] <zcorpan_>
annevk: ok

[03:08:00.0000] <annevk>
maybe RECEIVING should become LOADING

[03:08:01.0000] <annevk>
and SENT should prolly be SENDING

[03:29:00.0000] <annevk>
zcorpan_, regarding the proxy thing, I agree...

[03:30:00.0000] <annevk>
zcorpan_, I didn't draft that text... What I think it means is that some UAs let the user specify some proxy to use...

[03:35:00.0000] <zcorpan_>
ok

[03:36:00.0000] <zcorpan_>
/me found an obscure opera bug regarding large negative 'left' plus some negative 'margin-left' on absolutely positioned elements appearing to the far _right_

[03:37:00.0000] <annevk>
might just be that we round something...

[03:37:01.0000] <annevk>
to save storage

[03:38:00.0000] <zcorpan_>
http://simon.html5.org/test/opera/negative-absolute.htm

[03:41:00.0000] <hasather>
zcorpan_: I don't get a FAIL anywhere

[03:42:00.0000] <zcorpan_>
hasather: what build are you using?

[03:42:01.0000] <hasather>
latest weekly for Linux, build 628

[03:42:02.0000] <zcorpan_>
ok

[03:43:00.0000] <zcorpan_>
i have 8767 windows

[03:43:01.0000] <hasather>
oops, sorry, completely missed the scrollbar

[03:43:02.0000] <zcorpan_>
ah

[03:55:00.0000] <zcorpan_>
/me filed #258854

[03:58:00.0000] <hasather>
zcorpan_: when was your job interview?

[03:58:01.0000] <hasather>
tomorrow?

[03:58:02.0000] <zcorpan_>
yes

[03:59:00.0000] <hasather>
ok

[03:59:01.0000] <hasather>
what kind of job is it? QA?

[03:59:02.0000] <zcorpan_>
i said that i was interested in working on a test suite for html5

[03:59:03.0000] <zcorpan_>
we'll see what they say

[03:59:04.0000] <hasather>
aha

[04:04:00.0000] <krijnh>
'Well one thing could be done would be to close the WHATWG list temporarily.'

[04:07:00.0000] <zcorpan_>
lol

[04:08:00.0000] <zcorpan_>
well one thing could be done would be for the htmlwg members to join the whatwg too

[04:15:00.0000] <krijnh>
Imho there should be just one list

[04:16:00.0000] <krijnh>
Just one spec, and just one editor

[04:17:00.0000] <krijnh>
Although, that one editor is subscribed to more than one list, so he could as well follow public-html as well :)

[04:17:01.0000] <krijnh>
as well as well, yeah, makes sense

[04:17:02.0000] <krijnh>
Think McFly!

[04:52:00.0000] <krijnh>
align="char" isn't in current drafts, is it going to be dropped or is something better coming up?

[04:53:00.0000] <annevk>
dropped in favor of text-align:<string> iirc

[04:53:01.0000] <Lachy>
krijnh, no browser ever supported it

[04:53:02.0000] <krijnh>
Lachy: I know

[04:53:03.0000] <krijnh>
annevk: Ah, k

[04:53:04.0000] <Lachy>
so if it's not even remotes supported, then there's no way it's getting included in HTML5

[04:53:05.0000] <Lachy>
unless it's a new feature

[04:54:00.0000] <krijnh>
I didn't know about css3 text-align: <string>

[04:54:01.0000] <zcorpan_>
it was in css 2.0 too iirc

[04:54:02.0000] <zcorpan_>
dropped in 2.1

[04:54:03.0000] <annevk>
yeah, what zcorpan_ said

[04:55:00.0000] <Lachy>
I wonder if it will ever be supported

[04:55:01.0000] <annevk>
well, if Dave Raggett is right and spreadsheets are the next thing...

[04:56:00.0000] <Lachy>
This has got to be close to the quote of the day: "Surely we're not trying to ensure that a Web page is presented the same in every browsing environment? What would be the use of that?"

[04:56:01.0000] <Lachy>
:-)

[04:56:02.0000] <annevk>
April Fools day is over Lachy

[04:56:03.0000] <krijnh>
Hehe

[04:58:00.0000] <annevk>
www-html is funny too

[04:58:01.0000] <annevk>
apparently XHTML 1.1 allows text/html as MIME type now

[04:58:02.0000] <annevk>
(the second edition that is)

[04:58:03.0000] <krijnh>
No April fool joke?

[04:58:04.0000] <Lachy>
ha! They published one: http://tools.ietf.org/html/rfc4824

[04:59:00.0000] <krijnh>
What a waste of time :/

[04:59:01.0000] <Lachy>
I noticed that XHTML 1.1 post today :-)

[05:04:00.0000] <Dashiva>
So is the text/html xhtml a joke or not?

[05:04:01.0000] <zcorpan_>
pointer?

[05:04:02.0000] <annevk>
The XHTML2 WG is close to a joke

[05:05:00.0000] <annevk>
... and they're the ones doing this

[05:05:01.0000] <Dashiva>
http://www.w3.org/TR/xhtml11/conformance.html#uaconf

[05:05:02.0000] <zcorpan_>
oh, i thought there was a blog post or something

[05:06:00.0000] <Lachy>
there was a post to www-html

[05:06:01.0000] <Lachy>
http://lists.w3.org/Archives/Public/www-html/2007Apr/0005.html

[05:06:02.0000] <Lachy>
there's a few responses as well

[05:08:00.0000] <annevk>
why doesn't the HTML WG get all the XHTML 1.x stuff btw?

[05:08:01.0000] <annevk>
so we can obsolete them all and say that they are irrelevant?

[05:08:02.0000] <annevk>
/me thinks that would make more sense

[05:08:03.0000] <Lachy>
raise the issue on public-html

[05:09:00.0000] <Lachy>
the XHTML2 WG should just get XHTML2.  we should get all XHTML Basic, Print, 1.x, Mobile and everything else

[05:10:00.0000] <zcorpan_>
does it matter?

[05:11:00.0000] <annevk>
no, it would prolly mean more work for us

[05:11:01.0000] <annevk>
otoh, not having them might too

[05:11:02.0000] <annevk>
as silly people will start pointing out obvious incompatibilities

[05:12:00.0000] <annevk>
(I would, for instance, if I didn't know the story behind it.)

[05:12:01.0000] <zcorpan_>
we could just say that they are deliverables from the xhtml2 wg and can be ignored, or something

[05:12:02.0000] <annevk>
Anyone who feels that the RECEIVING constant should not be renamed to LOADING?

[05:12:03.0000] <annevk>
(for XMLHttpRequest)

[05:13:00.0000] <gsnedders>
is there any sort of feed for the SVN?

[05:14:00.0000] <annevk>
not until you contribute a patch to generate one

[05:14:01.0000] <gsnedders>
where is the SVN located anyway?

[05:14:02.0000] <zcorpan_>
for html5?

[05:14:03.0000] <gsnedders>
yes

[05:15:00.0000] <zcorpan_>
http://svn.whatwg.org/

[05:15:01.0000] <annevk>
noted at the top of the draft... svn.whatwg.org/webapps/

[05:15:02.0000] <gsnedders>
ah

[05:15:03.0000] <gsnedders>
I'd been trying to avoid downloading the entire draft again, looking around on the site

[05:16:00.0000] <zcorpan_>
annevk: i don't feel strongly about it, but currently HTMLMediaElement has LOADING

[05:16:01.0000] <gsnedders>
any real preference as to what language it is written in?

[05:16:02.0000] <zcorpan_>
gsnedders: there is a tracker written in python

[05:17:00.0000] <zcorpan_>
adding a feed to that should be trivial

[05:17:01.0000] <annevk>
the code for that tracker is online

[05:17:02.0000] <MikeSmith>
annevk, gsnedders - with some minimal config, you could set up a feed for the project SVN though CIA

[05:17:03.0000] <MikeSmith>
http://cia.vc/

[05:17:04.0000] <MikeSmith>
WebKit is using that

[05:17:05.0000] <annevk>
http://code.google.com/p/html5/ has it somewhere

[05:18:00.0000] <MikeSmith>
I mean the WebKit project

[05:18:01.0000] <annevk>
oh, maybe we could let Hixie install that then...

[05:18:02.0000] <gsnedders>
/me has never actually written anything much in Python, though used code as reference many times

[05:18:03.0000] <annevk>
on tracker.whatwg.org or something

[05:19:00.0000] <zcorpan_>
or svn.whatwg.org

[05:19:01.0000] <gsnedders>
where is the code for the tracker? I can't see a link anyway… maybe I'm just not doing well today :)

[05:20:00.0000] <annevk>
http://html5.googlecode.com/svn/trunk/web-apps-tracker/

[05:21:00.0000] <annevk>
zcorpan_, yeah, I noticed, LOADING it is

[07:43:00.0000] <Lachy>
The XHTML2 WG finally got something right! http://www.w3.org/mid/4611023C.3020607⊙ac :-)

[07:44:00.0000] <Lachy>
well, almost

[07:46:00.0000] <zcorpan>
oh. i really thought they wanted to allow text/html

[07:46:01.0000] <Lachy>
the only problem is that there is still no MUST level requirement

[07:47:00.0000] <Lachy>
it should say something like "... MUST use an XML MIME type, and SHOULD use app/xhtml+xml"

[07:47:01.0000] <zcorpan>
/me doesn't care about the 1.1 spec anyway

[07:47:02.0000] <Lachy>
yeah, but we should encourage them to get something right

[07:48:00.0000] <zcorpan>
go ahead :)

[07:48:01.0000] <krijnh>
/me has to care about the 1.1 spec

[07:48:02.0000] <krijnh>
I get clients requesting 'XHTML 1.1' valid markup from time to time

[07:49:00.0000] <krijnh>
No you don't you silly client, you really want to support IE

[07:49:01.0000] <Lachy>
krijnh, the correct response to that is "No."  The markup will be conforming, but it will not be XHTML 1.1 due to compatibiley reasons"

[07:50:00.0000] <krijnh>
Yeah, but they are going to change the spec now, I can't say that anymore ;)

[07:50:01.0000] <krijnh>
but if

[07:50:02.0000] <Lachy>
"... I am using conforming HTML 4.01 (or 5), which will yeild identical results"

[07:50:03.0000] <Lachy>
s/identical/better/

[07:50:04.0000] <zcorpan>
yeah, sell them html5 instead

[07:50:05.0000] <krijnh>
That's what my CMS uses, yes

[07:50:06.0000] <krijnh>
I am :)

[07:51:00.0000] <Lachy>
they might object to HTML5's maturity level

[07:51:01.0000] <krijnh>
HTML5 is more mature than XHTML 1.x imho

[07:51:02.0000] <zcorpan>
/me usually points out that html5 is more in line with what browsers do, and it's possible to use a subset of it

[07:52:00.0000] <krijnh>
Off topic; typo3 sucks

[07:52:01.0000] <krijnh>
Tee hee

[07:52:02.0000] <zcorpan>
not that i do a lot of freelance work, but anyway

[07:52:03.0000] <krijnh>
zcorpan: what do you do for work then?

[07:52:04.0000] <zcorpan>
i study

[07:53:00.0000] <krijnh>
Ah

[07:53:01.0000] <krijnh>
Me too :)

[07:59:00.0000] <zcorpan>
the html60 guy found public-html

[08:02:00.0000] <Lachy>
I encountered several parse errors while reading his e-mail

[08:04:00.0000] <Lachy>
a file preview feature might be a good feature for browsers to consider implementing, but it certainly doesn't need a new element

[08:05:00.0000] <zcorpan>
i don't understand any of his proposals, frankly

[08:06:00.0000] <Lachy>
all he wants is a way for a browser to show a preview of a picture being uploaded

[08:06:01.0000] <Lachy>
everything else he wrote about octogos and new input controls is irrelevent

[08:07:00.0000] <Lachy>
*octogons

[08:07:01.0000] <zcorpan>
he has a lot of other proposals too on his site

[08:09:00.0000] <Dashiva>
Isn't it a bit early to start on html 60 while we're still doing 5?

[08:09:01.0000] <zcorpan>
/me will work on html 7000

[08:12:00.0000] <Lachy>
Dashiva, it's good to start working on future versions in parallel with the current version :-)

[08:20:00.0000] <Lachy>
aargh! he sent that post 3 times :-(

[08:37:00.0000] <hsivonen>
I wonder why the HTML 6.0 guy/gal doesn't sign the proposals with a real name

[08:38:00.0000] <Lachy>
if I wrote proposals like that, I'd want to be anonymous too!

[08:38:01.0000] <krijnh>
Hehe

[08:38:02.0000] <Dashiva>
/me doesn't see any proposals

[08:38:03.0000] <krijnh>
Perhaps for a native Russian this is very good :)

[08:45:00.0000] <Lachy>
new Polish translation of the FAQ http://blog.whatwg.org/faq/pl/

[09:06:00.0000] <annevk>
seems like the W3C is not going to fix <?xml-stylesheet?>: http://lists.w3.org/Archives/Public/public-xml-core-wg/2007Apr/0001.html

[09:06:01.0000] <annevk>
silly

[09:06:02.0000] <annevk>
otoh, it was already on my XML5 todo list

[09:07:00.0000] <krijnh>
XML5?

[09:07:01.0000] <annevk>
that or XML2

[09:07:02.0000] <Lachy>
krijnh, we're building Web 5.0!

[09:07:03.0000] <hsivonen>
what's the deal with RELAX NG fragments in the spec? surely it isn't Hixie's idea?

[09:07:04.0000] <annevk>
dunno, no

[09:07:05.0000] <Lachy>
what RelaxNG fragments?

[09:08:00.0000] <krijnh>
Lachy: you have a logo already? ;)

[09:08:01.0000] <hsivonen>
Lachy: not in the spec now but said in email as if there would be

[09:08:02.0000] <Lachy>
/me looks...

[09:08:03.0000] <krijnh>
"Since having informal Relax-NG schema fragments in the HTML specification has been discussed, I think there might be room to discuss the inclusion of default CSS (perhaps even Aural CSS) fragments in the specification as well."

[09:09:00.0000] <Lachy>
found the email

[09:09:01.0000] <annevk>
/me suspects that people who archive e-mail put that in "xml-pipedream"

[09:09:02.0000] <krijnh>
hsivonen: perhaps people misread your mails about your thesis

[09:10:00.0000] <hsivonen>
krijnh: that would be a rather radical misreading, but radical misreadings happen

[09:10:01.0000] <annevk>
btw, is anyone going to post the 2 april post?

[09:10:02.0000] <annevk>
or was there no such thing?

[09:11:00.0000] <Lachy>
oh crap, I was going to write something about <video>

[09:11:01.0000] <hsivonen>
Lachy: did you have a <video> post coming up?

[09:11:02.0000] <hsivonen>
right

[09:11:03.0000] <Lachy>
oh well, it's too late now

[09:11:04.0000] <Lachy>
it's getting late and I don't have time now

[09:11:05.0000] <annevk>
it's still April 2

[09:11:06.0000] <Lachy>
anyone, feel free to write it if you want

[09:11:07.0000] <krijnh>
Not in Australia

[09:11:08.0000] <zcorpan>
what were you going to write?

[09:11:09.0000] <Lachy>
yeah, get with the times!  It's April 3 already

[09:12:00.0000] <annevk>
it's e-April 2 damn it

[09:12:01.0000] <Lachy>
I was just going to do an overview of the new video element

[09:12:02.0000] <annevk>
oh ok

[09:12:03.0000] <Lachy>
yeah, it is in UTC, so I'll accept that

[09:12:04.0000] <gavin_>
what's special about April 2?

[09:12:05.0000] <gavin_>
International Children's Book Day?

[09:12:06.0000] <annevk>
day after April 1

[09:13:00.0000] <annevk>
I thought the idea was to make another joke this day to confuse people

[09:13:01.0000] <Lachy>
no, Hixie wanted another post to go up after the april 1 post just so people didn't get confused aboutit any longer

[09:13:02.0000] <annevk>
oh, he wanted the opposite

[09:13:03.0000] <annevk>
that's not fun

[09:13:04.0000] <Lachy>
yeah, but it'd be funny to confuse people more

[09:13:05.0000] <Lachy>
post about HTML 60!

[09:14:00.0000] <zcorpan>
let's ask the russian guy to blog

[09:14:01.0000] <gavin_>
was there an April 1 gag on the WHATWG list that I missed?

[09:14:02.0000] <Lachy>
on the blog

[09:14:03.0000] <gavin_>
oh

[09:14:04.0000] <krijnh>
And it was hilarious.

[09:15:00.0000] <gavin_>
I don't read the blog

[09:15:01.0000] <gavin_>
maybe I should

[09:15:02.0000] <Lachy>
http://blog.whatwg.org/html6-plan

[09:16:00.0000] <annevk>
"To avoid any confusion about the estimated time of arrival HTML6 has been renamed to HTML60. The idea is that when all twenty year somethings currently working on the specification turn sixty something it will be done."

[09:17:00.0000] <annevk>
s/year//

[09:17:01.0000] <hsivonen>
a sociologist could have a field day comparing the dynamics of the WHATWG, AtomPub and HTML WG mailing lists

[09:19:00.0000] <Lachy>
I'm hoping public-html will become more like whatwg once we actually have both Chairs and everyone agrees to work on HTML5, keeping the specs in sync

[09:20:00.0000] <hsivonen>
yeah

[09:20:01.0000] <zcorpan>
and the people who can't manage the large amount of emails have left ;)

[09:20:02.0000] <zcorpan>
or at least don't say so much

[09:20:03.0000] <Lachy>
at the moment, public-html just seems to be suffering from a few growing pains.  To many newbies all joined at once, without any clear direction given

[09:20:04.0000] <annevk>
woha, 2 days 60 e-mails

[09:21:00.0000] <zcorpan>
Lachy: indeed

[09:21:01.0000] <annevk>
that's 1800 e-mails for one month

[09:21:02.0000] <annevk>
and most of the crap was posted in the weekend!

[09:22:00.0000] <annevk>
/me is already at 25 messages to public-html since the start...

[09:22:01.0000] <Lachy>
If this rate keeps up, we'll get another 800 e-mails on public-html this month

[09:23:00.0000] <annevk>
of course, nothing compared to a certain Mike Schinkel

[09:23:01.0000] <Lachy>
or, actually, I'd expect more because weekends tend to be the quieter times

[09:23:02.0000] <annevk>
I just said, 1800

[09:24:00.0000] <Lachy>
Mike has a total of 56

[09:24:01.0000] <krijnh>
Lachy: didn't you send him a mail?

[09:24:02.0000] <Lachy>
how did you calculate 1800?

[09:25:00.0000] <Lachy>
krijnh, yes, but that wasn't conted

[09:25:01.0000] <Lachy>
counted

[09:25:02.0000] <krijnh>
But what did you tell him?

[09:25:03.0000] <annevk>
nm

[09:25:04.0000] <annevk>
I should have said 900...

[09:26:00.0000] <annevk>
/me did 30*60 instead of 30*30 or 15*60...

[09:27:00.0000] <Lachy>
krijnh, I wrote a polite e-mail informing him to stop flooding the list

[09:27:01.0000] <Lachy>
I can forward you a copy if you like

[09:28:00.0000] <Lachy>
he responed, and I'm drafting my response to that too

[09:29:00.0000] <krijnh>
Perhaps he didn't get it

[09:29:01.0000] <Lachy>
oh, he didn't.  That was obvious from his reply

[09:29:02.0000] <annevk>
/me changes topic to 'WHATTF (HTML60) -- http://www.whattf.org/ -- Logs: http://krijnhoetmer.nl/irc-logs/ -- Please leave your sense of logic at the door, thanks!'

[09:31:00.0000] <Lachy>
I also wrote to him about the issue numbers in the wiki, to which he responded "The numbers are the "killer feature"" :-)

[09:31:01.0000] <Lachy>
Hopefully the thread on public-html will convince him otherwise

[09:32:00.0000] <krijnh>
The killer feature

[09:32:01.0000] <krijnh>
Okay :)

[09:42:00.0000] <annevk>
/me missed http://mail.google.com/mail/help/paper/

[09:46:00.0000] <zcorpan>
lol

[10:17:00.0000] <Philip`>
Hmm, my <canvas> tests have found a bug in a pre-XPSP2 Win32 string processing function, which was somewhat unexpected and irritating since it's wasted half a day - but tests are meant to find bugs, so I suppose I shouldn't complain...

[10:20:00.0000] <zcorpan>
Philip`: perhaps you should document it somewhere so others don't waste half a day too

[10:22:00.0000] <Philip`>
I'll complain to Microsoft that their documentation is wrong (since it looks like they updated it to account for the bug, but incorrectly), and get Mozilla's code fixed to work around it, and then I'll be happy :-)

[10:23:00.0000] <zcorpan>
ok

[10:23:01.0000] <zcorpan>
something opera would have to work around too?

[10:27:00.0000] <Philip`>
Only if they use Uniscribe's ScriptItemize function (and expect it to stay within the buffer length it's given) - as far as I can tell from Opera.dll, they don't use that function, so it shouldn't be a problem

[11:14:00.0000] <zcorpan>
i didn't read the <q> mail, but whatever it says it suggests to me it's time to drop it :)

[11:16:00.0000] <annevk>
<q src=uri cite=string> is the suggestion

[11:16:01.0000] <annevk>
which is a) not backwards compatible, b) silly and c) too late to count as a April 1 joke

[11:32:00.0000] <KevinMarks>
/me is listening to the yahoo pipes talk at google

[11:33:00.0000] <KevinMarks>
just asked for a way to say that a layer doesn't get events

[11:39:00.0000] <othermaciej>
does "layer" mean something special in the context of pipes, or do you mean a CSS layer?

[11:39:01.0000] <KevinMarks>
they meant a css layer (though they are using canvas too)

[11:40:00.0000] <annevk>
opacity:0 in case of IE

[11:40:01.0000] <annevk>
well, something equivalent to that

[12:09:00.0000] <othermaciej>
KevinMarks: there's no standard way (and no way at all in most browsers) to make something that is visible but can be clicked through

[12:13:00.0000] <Hixie>
if people want to write a funny but not wrong post for april 2nd, i'm sure they can find plenty of material in amongst the tag, xhtml2, xforms, svg, and other wgs and drafts...

[12:19:00.0000] <othermaciej>
Hixie: would Thursday work for you for lunch?

[12:19:01.0000] <Hixie>
i was about to say "sure!", but let me check my calendar first!

[12:20:00.0000] <Hixie>
i have a meeting up til 11:30

[12:20:01.0000] <Hixie>
but after that i'm free

[12:21:00.0000] <othermaciej>
all right, let's do 11:30

[12:21:01.0000] <Hixie>
cool

[12:21:02.0000] <Hixie>
you guys coming to the 'plex?

[12:22:00.0000] <othermaciej>
yes

[12:22:01.0000] <othermaciej>
(hopefully carpooling so we can park)

[12:23:00.0000] <Hixie>
cool

[13:34:00.0000] <hsivonen>
WordPress could you a UI for marking multiple comments as spam in one go

[13:35:00.0000] <hsivonen>
Lachy: btw, are you keeping an eye on spam on the whatwg blog?

[13:35:01.0000] <hsivonen>
I just got rid of a bunch of SEO spam

[14:48:00.0000] <Hixie>
othermaciej: btw, did you have a chance to mull over the api proposal?

[14:48:01.0000] <Hixie>
http://www.whatwg.org/specs/web-apps/current-work/#media

[15:01:00.0000] <othermaciej>
Hixie: CAN_SHOW_STATIC_VIEW_OF_CURRENT_TIME seems like an overly long name, although it is very precise

[15:01:01.0000] <Hixie>
it was your proposal :-P

[15:01:02.0000] <othermaciej>
LOADED_FRAME might be better named LOADED_INITIAL_FRAME or something

[15:01:03.0000] <othermaciej>
Hixie: it was not intended seriously

[15:02:00.0000] <Hixie>
oh ok

[15:02:01.0000] <Hixie>
i don't really have a better name

[15:02:02.0000] <othermaciej>
CAN_SHOW_CURRENT_FRAME

[15:03:00.0000] <Hixie>
i suppose that's clear enough even in the audio case, yeah

[15:03:01.0000] <Hixie>
regenning the spec...

[15:04:00.0000] <othermaciej>
the fact that EMPTY is used twice is a little awkward, not sure what to call the second EMPTY instead

[15:05:00.0000] <othermaciej>
CANT_PLAY is a little inaccurate

[15:05:01.0000] <Hixie>
UNAVAILABLE ?

[15:09:00.0000] <othermaciej>
sounds good

[15:09:01.0000] <Hixie>
ok

[15:42:00.0000] <hsivonen>
looks like (legal issues aside) in terms of spec writing terms Theora complies with WHATWG philosophy but H.264 does not

[15:43:00.0000] <hsivonen>
s/writing terms/writing,/

[15:43:01.0000] <Hixie>
the whatwg philosophy is pragmatism

[15:43:02.0000] <Hixie>
if one of the major browser vendors doesn't want to do ogg, ogg isn't an option. same with h.264.

[15:44:00.0000] <Hixie>
this leaves us in an interesting conundrum out of which i see no solution.

[15:44:01.0000] <Hixie>
(the arguments in favour of h.264 are compelling, as much as those in favour of ogg.)

[15:45:00.0000] <Dashiva>
What about system codecs, have those been taken into the equation?

[15:45:01.0000] <hsivonen>
the big practical problem with H.264 is that saying "H.264" is pointless. Saying "the H.264 subset that QuickTime supports" is more realistic, but producing that stuff with non-Apple encoders is *hard*

[15:46:00.0000] <hsivonen>
Dashiva: doesn't that degenerate to the issue of different installations having different plug-ins

[15:46:01.0000] <hsivonen>
?

[15:46:02.0000] <hsivonen>
they'd just be different plug-ins

[15:47:00.0000] <hsivonen>
I have advocated H.264 on standards grounds elsewhere. 

[15:47:01.0000] <Dashiva>
Not as much, since the API would be the same in all cases

[15:48:00.0000] <hsivonen>
but when people ask me, what x264 flags to use, I don't know what to tell them

[15:48:01.0000] <hsivonen>
Dashiva: same as in QuickTime, gstreamer and whatever the Windows Media API is called

[15:49:00.0000] <Dashiva>
I'm not sure what you mean

[15:50:00.0000] <hsivonen>
Dashiva: every platform has a different codec extensibility API

[15:51:00.0000] <Dashiva>
So the browsers need to program to one API per OS, but they still present a single media API in the browser.

[15:51:01.0000] <Hixie>
hsivonen: the proposal is H.264 Baseline Profile, which is supposedly patent-safe (though MPEG-LA claim otherwise)

[15:51:02.0000] <Hixie>
hsivonen: that's what people mean when they say H.264 should be the minimum required.

[15:54:00.0000] <hsivonen>
Hixie: good luck with authors figuring out how to clamp an encoder to that profile

[15:55:00.0000] <hsivonen>
Hixie: Baseline JPEG works because authors don't have non-baseline software available to them

[15:55:01.0000] <Hixie>
i'm not the one to convince.

[15:55:02.0000] <Hixie>
(note how the spec already says Ogg Theora)


2007-04-03
[17:32:00.0000] <Hixie>
we need a better term than readyState

[17:32:01.0000] <Hixie>
for the CAN_SHOW_/CAN_PLAY_ states

[17:36:00.0000] <Hixie>
readyState would be fine except that XHR and IE use it for network status

[17:37:00.0000] <othermaciej>
yeah

[17:37:01.0000] <othermaciej>
the accurate but too long name would be presentabilityState

[17:38:00.0000] <othermaciej>
or playabilityState if you wanna be a bit less accurate

[17:38:01.0000] <othermaciej>
or just playability

[18:45:00.0000] <Hixie>
is an event that fires every few 100ms while playing useful, or should position ui be updated via timeouts?

[18:47:00.0000] <othermaciej>
timeouts + notification of reasons for rate change seem sufficient (that being pause or becoming unplayable)

[18:47:01.0000] <Hixie>
ok.

[18:49:00.0000] <Hixie>
oh another thing, do we want paused to be a mutable attribute, or do we prefer play()/pause() ?

[18:49:01.0000] <Hixie>
this baby is going to have six bazillion events, jesus

[18:49:02.0000] <Hixie>
/me has the events up on his whiteboard

[18:49:03.0000] <othermaciej>
hmm

[18:50:00.0000] <othermaciej>
play() / pause() seems a lot more natural, even though normally I'd be more inclined to a read/write attribute

[18:50:01.0000] <Hixie>
that was my conclusion too

[18:50:02.0000] <KevinMarks>
don't they map to setplayrate anyway?

[18:51:00.0000] <Hixie>
it's not clear what we're going to do with the proposed playRate stuff

[18:51:01.0000] <Hixie>
but that's not today's problem anyway

[18:51:02.0000] <Hixie>
/me is trying to spec out all the states first

[18:51:03.0000] <KevinMarks>
play() == setPlayRate(1.0) pause == setPlayRate(0) rewind=setPlayRate(-1)

[18:51:04.0000] <KevinMarks>
etc

[18:52:00.0000] <KevinMarks>
if you're doing a property approach, that fits

[18:53:00.0000] <Hixie>
/me ponders whether we need an event specifically for when playback stops despite it not being paused, in addition to the event we have for entering the readyState UNAVAILABLE state (which can also fire for other reasons, e.g. seeking to unbuffered data)

[18:54:00.0000] <othermaciej>
KevinMarks: Apple's proposal didn't work quite like that

[18:56:00.0000] <Hixie>
i'd use 'stalled' but we're using that for indicating lack of network progress

[18:56:01.0000] <Hixie>
maybe 'onwaiting'?

[18:56:02.0000] <Hixie>
i'll call it 'waiting' for now and we can work out what to do later.

[18:57:00.0000] <Hixie>
ok that's 20 events for <video> and <audio>

[18:58:00.0000] <othermaciej>
I think you should read Kevin Calhoun's message about networking models

[18:58:01.0000] <othermaciej>
I'm not sure 'stalled' makes sense, depending on your access strategy

[18:58:02.0000] <othermaciej>
(or things like 'bufferRate')

[18:58:03.0000] <othermaciej>
but he said it better

[18:58:04.0000] <Hixie>
the one he sent earlier today?

[18:58:05.0000] <othermaciej>
yes

[18:59:00.0000] <Hixie>
i've already rolled most of his feedback into the strawman proposal

[18:59:01.0000] <Hixie>
stalled just means that the UA isn't getting data but is expecting it

[18:59:02.0000] <Hixie>
provides an easy way to give a "your connection may be broken" message

[19:00:00.0000] <Hixie>
(or for greasemonkey scripts, e.g., to go in and provide that)

[19:00:01.0000] <Hixie>
the "seekable" attribute was added in response to his message, btw

[19:00:02.0000] <Hixie>
i'll reply to it in detail when i go through the e-mail feedback though

[19:04:00.0000] <othermaciej>
I think "readonly attribute TimeRanges seekable" may be more general than needed, in practice I would expect you have only a single range at the beginning or you pretend any point is available

[19:04:01.0000] <othermaciej>
I think keeping a list of buffered ranges is just not applicable to some loading strategies so maybe we should think about what problem it is trying to solve

[19:05:00.0000] <othermaciej>
and I think bufferingRate isn't useful 

[19:06:00.0000] <othermaciej>
if you want to estimate bytes per second, there's always progress events for that

[19:06:01.0000] <othermaciej>
although I'm not really sure how progress events would apply to an RTSP stream, or something that buffers incrementally using byte range requests

[19:22:00.0000] <KevinMarks>
othermaciej: think about the bittorrent case, where you may have multiple blocks cached but not all

[19:23:00.0000] <othermaciej>
KevinMarks: bittorrent isn't the only case where that can happen

[19:23:01.0000] <KevinMarks>
I know, I'm using it as an example

[19:23:02.0000] <othermaciej>
KevinMarks: but bittorrent isn't something you'd use as a protocol to back video playback, since it gives you the data in random order

[19:23:03.0000] <othermaciej>
which is useless unless you are willing to download the whole thing up front

[19:24:00.0000] <KevinMarks>
no, you can treat it statistically like a fast-start

[19:25:00.0000] <KevinMarks>
when you have enough contiguous pieces from the start to play for longer then expected remaining download time, you can play

[19:27:00.0000] <Hixie>
hm, there are two error states really

[19:27:01.0000] <Hixie>
one where the media is still partly playable

[19:27:02.0000] <Hixie>
and one where it's not

[19:27:03.0000] <KevinMarks>
Kevin Calhoun wrote a usenet datahandler that would collect chunks from alt.binaries postings

[19:28:00.0000] <othermaciej>
the thing is, with bittorrent, the timewise first chunk could be the last you get

[19:29:00.0000] <KevinMarks>
yes, so at that point you do need to wait for the whole thing

[19:31:00.0000] <othermaciej>
so yeah, you could make it work, it just wouldn't be advisable to serve page-embedded video by bittorrent

[19:32:00.0000] <othermaciej>
unless it is small enough to d/l the whole thing quickly

[19:32:01.0000] <othermaciej>
in which case, why bother w/ torrent

[19:33:00.0000] <KevinMarks>
I find torrents about the same speed as pulling videos from itunes store via akamai at home

[19:34:00.0000] <KevinMarks>
so it's a little moot

[19:36:00.0000] <KevinMarks>
for a fast-start, you know it is mostly coming in order, so you reach the expected completion time sooner, but for a torrent or sliced usenet download you can still reach it before download is complete

[19:38:00.0000] <othermaciej>
yeah the torrent problem is just that there's no way to bias it towards timewise earlier chunks and it is fine-grained enough that this is likely to hose you

[19:40:00.0000] <KevinMarks>
well, there is, as which chunk you request is under client control, so you could favour earlier ones between otherwise equivalent ones (the next chunk rule is usually a function of rarity and others estimated rates)

[20:04:00.0000] <Hixie>
i'm not sure how to handle this error problem

[20:04:01.0000] <Hixie>
short of having two error states

[20:04:02.0000] <Hixie>
or three, even

[20:05:00.0000] <Hixie>
/me changes topic to 'WHATWG (HTML5) -- http://www.whatwg.org/ -- Logs: http://krijnhoetmer.nl/irc-logs/ -- Please leave your sense of logic at the door, thanks!'

[20:12:00.0000] <Hixie>
hm, the apple proposal doesn't handle this at all, it just has an error state for the early part

[20:12:01.0000] <Hixie>
not for the later part

[22:43:00.0000] <Lachy_>
yay! XHTML+RDFa :-) http://www.w3.org/MarkUp/2007/ED-xhtml-rdfa-20070402/

[22:44:00.0000] <othermaciej>
it's a day late

[22:45:00.0000] <Lachy_>
yeah, but got to give them credit for stealing our HTML6 ideas so quickly

[22:47:00.0000] <othermaciej>
I see exactly three user agent conformance requirements

[22:47:01.0000] <othermaciej>
all of which are stupid

[22:51:00.0000] <Lachy_>
I found 3 MUST level and 1 SHOULD level UA requirement

[22:52:00.0000] <othermaciej>
ah, I missed the SHOULD

[22:55:00.0000] <Lachy_>
the whole thing is impossible to implement anyway, since it reuses the XHTML1 NS in incompatible ways

[22:55:01.0000] <othermaciej>
is this supposed to be part of XHTML2?

[22:55:02.0000] <Lachy_>
it appears to be XHTML1

[22:56:00.0000] <Lachy_>
"This document is an internal editors draft for development purposes. However, its content is based upon mature materials from [XHTML2] and is therefore considered nearly complete."

[22:56:01.0000] <othermaciej>
it seems to assume href on every element

[22:56:02.0000] <Lachy_>
that explains the problem

[22:57:00.0000] <othermaciej>
well, XHTML5 presumably won't have modularization so we will be safe

[23:16:00.0000] <Lachy_>
I just love how easy DTDs make things --> http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2007Apr/0017.html

[23:17:00.0000] <Hixie>
introducing a new status to mean "not completely loaded but partialy loaded and we're not going to bother getting more" seems like more work than it's worth

[23:17:01.0000] <Hixie>
(for implementors and authors)

[23:17:02.0000] <Hixie>
but the alternative is to have the 'error' event send to either ERROR or LOADED depending on whether we passed LOADED_INITIAL_FRAME or not

[23:17:03.0000] <Hixie>
which seems poor

[23:23:00.0000] <Hixie>
this sucks

[23:23:01.0000] <Hixie>
having a network problem or user cancelation nearly arbitrarily result in a LOADED or an ERROR is stupid

[23:24:00.0000] <Hixie>
in the former case there's no way to catch the error if joining late, e.g.

[23:24:01.0000] <Hixie>
but we can't use the same code as otherwise you can't tell if the other attributes are safe

[23:24:02.0000] <Hixie>
maybe we should just have a lastError attribute

[23:24:03.0000] <Hixie>
which is reset when you get a successful transition

[23:24:04.0000] <Hixie>
or rather, is reset by load()

[23:25:00.0000] <Hixie>
or maybe when we have an error or abort we should leave the state in its last state and set an error flag

[23:25:01.0000] <othermaciej>
I'm not sure you should go to LOADED on a user cancel or late error - probably better to stay in current state with alternate way to get the error (like lastError)

[23:25:02.0000] <Hixie>
that way you can see where the error occured

[23:26:00.0000] <othermaciej>
that might obviate the need for an ERROR state I guess

[23:26:01.0000] <Hixie>
yeah

[23:54:00.0000] <Hixie>
man, the list of ways in which playback can stop when the media isn't paused is getting longer by the hour

[23:55:00.0000] <othermaciej>
oh yeah?

[23:55:01.0000] <othermaciej>
what besides entering non-playable state or reaching the end?

[23:57:00.0000] <Hixie>
currently: seeking (UNAVAILABLE and paused=false and seeking=true), buffering (UNAVAILABLE or CAN_SHOW_CURRENT_FRAME and paused=false and seeking=false, playback position isn't in the buffered range), ended is true, and a fatal error occured, either network error or during decoding.

[23:58:00.0000] <Hixie>
currently the first two will fire a 'waiting' event when they occur, the latter two won't.

[23:58:01.0000] <othermaciej>
fatal error partway should arguably count as truncating the duration

[23:59:00.0000] <KevinMarks>
/me remembers all those arcane QuickTime chapters on clocks

[23:59:01.0000] <Hixie>
fatal errors might occur decoding the stream half-way, with a later part of the stream still playable if seeked to (i see this on DVDs often, sadly)

[23:59:02.0000] <Hixie>
not sure we care about that though

[00:55:00.0000] <Hixie>
should i reuse 'reset' for when we reset the <video> element when load() is invoked on an already-running <video>?

[00:58:00.0000] <othermaciej>
what else is 'reset' used for?

[00:58:01.0000] <Hixie>
forms

[01:05:00.0000] <Hixie>
hm, 'reset' bubbles for forms

[01:06:00.0000] <Hixie>
i guess i could make one here that doesn't but that seems like it would just confuse matters.

[01:07:00.0000] <Hixie>
'emptied' will do for now

[01:36:00.0000] <Hixie>
http://www.whatwg.org/specs/web-apps/current-work/source#mediaevents may be of interest

[01:41:00.0000] <othermaciej>
definition of "stalled" in the prose seems wrong (though description in the table is ok)

[01:41:01.0000] <othermaciej>
"If at any point the user agent has received no data for more than about three seconds, the user agent must fire a stalled event at the element."

[01:42:00.0000] <Hixie>
the prose isn't done yet

[01:42:01.0000] <Hixie>
i just did the table

[01:42:02.0000] <Hixie>
most of the prose is unchanged from the original checkin before the apple proposal was sent to the list

[01:43:00.0000] <othermaciej>
event names seem to be a mix of past and present tense

[01:44:00.0000] <othermaciej>
that is indeed a whole lot of events

[01:44:01.0000] <Hixie>
yeah at some point we need to do a cleanup of the names in this section

[01:44:02.0000] <Hixie>
events, constants, methods, attributes

[01:44:03.0000] <Hixie>
right now i just want to get the thing described

[01:45:00.0000] <Hixie>
the names can be discussed in the html wg :-P

[01:46:00.0000] <hsivonen>
ExternalizeBikeshedding as a design principle? :-)

[01:46:01.0000] <othermaciej>
hah

[01:46:02.0000] <Hixie>
:-)

[01:48:00.0000] <othermaciej>
is Murray Maloney a known person in the web standards world?

[01:49:00.0000] <Hixie>
othermaciej: yes

[01:49:01.0000] <othermaciej>
I think I upset him by implying that the fact that he agreed or disagreed with something, with no further info, was uninteresting

[01:49:02.0000] <othermaciej>
(by private email)

[01:50:00.0000] <Hixie>
hah

[01:51:00.0000] <Hixie>
http://www.xml.com/pub/au/56 describes for what he's known pretty succintly

[03:00:00.0000] <MikeSmith>
Hixie, om_sleep - I think Murray was also active back in the day in the IETF HTML WG

[03:00:01.0000] <MikeSmith>
https://listserv.heanet.ie/cgi-bin/wa?A0=html-wg

[03:01:00.0000] <om_sleep>
/me obviously has no respect for his elders

[03:01:01.0000] <om_sleep>
I gave Dave Raggett a hard time too

[03:04:00.0000] <MikeSmith>
om_sleep - Dave's got thick skin. I don't know Murray personally, but I'd guess he's been involved enough in mailing list discussions to not get upset and take stuff personally

[03:05:00.0000] <MikeSmith>
and I think it's good for him to know how annoying people find those multiple +1 messages

[06:26:00.0000] <hsivonen>
clearly, http://www.w3.org/TR/2007/WD-curie-20070307/ has not been developed with backwards compat in mind

[06:28:00.0000] <Lachy>
did you expect it to be?

[06:28:01.0000] <hsivonen>
Lachy: I didn't. 

[06:29:00.0000] <hsivonen>
but if it becomes a REC, some people might ask it to be supported for the XHTML 1.x namespace. :-(

[06:30:00.0000] <hsivonen>
qNames in content is already an anti-pattern

[06:31:00.0000] <Lachy>
I haven't actually read curie yet, how is it supposed to work?

[06:32:00.0000] <Lachy>
I'll just go read it, since it's quite short

[06:32:01.0000] <hsivonen>
Lachy: IRIs that start with [ and end with ] are treated as bracketed qNames that are resolved relative to namespace URIs

[06:58:00.0000] <Lachy>
I just finished reading the spec.  It seems curies could be potentially useful in an authoring tool or CMS that processes them and outputs the full URIs to the web

[07:44:00.0000] <hsivonen>
Hixie: btw, do you have a reference to H.264 Baseline being free of enforceable patents?

[07:46:00.0000] <hsivonen>
I find it unbelievable that Baseline would be unencumbered. I would have expected to have heard the details by now elsewhere if it was.

[10:11:00.0000] <zcorpan>
/me was in linköping today

[10:13:00.0000] <tylerr>
Did you say hello to the Cardigans? :-)

[10:13:01.0000] <hasather>
zcorpan: how did it go?

[10:14:00.0000] <hasather>
tylerr: they are from Jönköping

[10:14:01.0000] <hasather>
if I'm not mistaken

[10:14:02.0000] <zcorpan>
hasather: it went well

[10:14:03.0000] <tylerr>
Oh! I need to check my sources then.

[10:14:04.0000] <hasather>
zcorpan: you got the job?

[10:14:05.0000] <zcorpan>
don't know yet

[10:14:06.0000] <zcorpan>
but it seems likely

[10:14:07.0000] <tylerr>
Ah yes you're right hasather, Jönköping. I get my pings confused. :-)

[10:14:08.0000] <hasather>
ok, but great that it went well

[10:15:00.0000] <hasather>
tylerr: :)

[10:16:00.0000] <zcorpan>
they knew more about html5 than last time i was there. talked to two other guys this time (although only one of them seemed to know stuff about html5)

[10:16:01.0000] <zcorpan>
last time i was there (in august) i talked to an svg guy or something

[10:16:02.0000] <hasather>
ok

[10:18:00.0000] <hasather>
zcorpan: did they like your idea about a HTML5 test suite?

[10:19:00.0000] <zcorpan>
yeah

[10:19:01.0000] <hasather>
cool

[10:19:02.0000] <zcorpan>
although they could come up with other stuff i could work on too

[10:19:03.0000] <zcorpan>
like internal test suites or something

[11:07:00.0000] <zcorpan>
anyone know what "interoperable" is in swedish?

[11:08:00.0000] <zcorpan>
samfungerande?

[11:14:00.0000] <zcorpan>
or simply "interopererande"?

[11:18:00.0000] <hasather>
zcorpan: both could probably work, don't know a good word for it myself

[11:20:00.0000] <zcorpan>
http://sv.wikipedia.org/wiki/Interoperabilitet

[11:20:01.0000] <zcorpan>
/me will go with the latter

[11:20:02.0000] <hasather>
yea, found that

[11:21:00.0000] <hasather>
zcorpan: ineroperabel maybe

[11:21:01.0000] <hasather>
+t

[11:22:00.0000] <hasather>
not a common Swedish word though

[11:29:00.0000] <zcorpan>
interoperable it is

[11:29:01.0000] <zcorpan>
er

[11:30:00.0000] <zcorpan>
interoperabel

[13:14:00.0000] <Hixie>
hsivonen: no, nobody seems willing to make that kind of claim on record :-(

[13:22:00.0000] <markp>
all right, this is probably a FAQ, but what's up with accesskey?

[13:23:00.0000] <markp>
it's mentioned repeatedly in wf2 but not at all in html5

[13:26:00.0000] <Hixie>
haven't yet worked out wtf to do with it

[13:27:00.0000] <Hixie>
it's so fundamentally broken in so many ways...

[13:27:01.0000] <markp>
documenting current practice isn't sufficient?

[13:27:02.0000] <markp>
i understand the extent of its brokenness

[13:28:00.0000] <Hixie>
currnet practice varies dramatically and is fundamentally broken

[13:28:01.0000] <markp>
that has never stopped you before...

[13:30:00.0000] <Hixie>
markp: the difference is that for accesskey i don't know how to fix it :-(

[13:31:00.0000] <markp>
i see

[13:36:00.0000] <markp>
argh!

[13:36:01.0000] <markp>
http://msdn2.microsoft.com/en-us/library/ms535181.aspx

[13:36:02.0000] <markp>
"This example uses the address element to italicize text."

[13:36:03.0000] <markp>
just shoot me

[13:49:00.0000] <markp>
http://www.whatwg.org/specs/web-apps/current-work/#area does not define the nohref attribute

[13:49:01.0000] <Hixie>
what does the nohref attribute do?

[13:49:02.0000] <markp>
described here: http://www.w3.org/TR/html401/struct/objects.html#adef-nohref

[13:50:00.0000] <Hixie>
but what does the nohref attribute do?

[13:50:01.0000] <Hixie>
(there are few UA conformance requirements in the HTML4 spec, and none for nohref, so it's not a good source of information for what things in HTML do)

[13:51:00.0000] <markp>
i understand

[13:52:00.0000] <markp>
i believe it's a way of marking that the <area>'s lack of an @href attribute is intentional

[13:52:01.0000] <Hixie>
so it does nothing?

[13:52:02.0000] <markp>
i.e. for validation

[13:52:03.0000] <markp>
checking firefox code

[13:53:00.0000] <markp>
hang on

[13:53:01.0000] <Hixie>
(and if we can't trust that the lack of an href attribute is intentional, why can we trust that the presence of a nohref attribute _is_?)

[13:54:00.0000] <markp>
firefox does precisely nothing with it

[13:54:01.0000] <markp>
no changes in behavior, nothing

[13:54:02.0000] <Hixie>
ok then

[13:54:03.0000] <markp>
http://msdn.microsoft.com/workshop/author/dhtml/reference/properties/nohref.asp says "Sets or retrieves whether clicks in this region cause action."

[13:54:04.0000] <markp>
which is slightly tantalizing

[13:54:05.0000] <Hixie>
msdn is wrong. :-)

[13:55:00.0000] <Hixie>
i tested it a lot, couldn't find that the attribute did anything, so i dropped it

[13:55:01.0000] <Hixie>
it seemed singularily pointless.

[13:55:02.0000] <markp>
aha

[13:55:03.0000] <markp>
http://www.idocs.com/tags/images/_AREA_NOHREF.html

[13:55:04.0000] <markp>
it is for dead zones in shaped <area>s

[13:56:00.0000] <markp>
see the idocs link

[13:58:00.0000] <Hixie>
yes but how does that differ from the lack of an href attribute?

[14:00:00.0000] <markp>
hacking the idocs test case reveals that there is no difference

[14:00:01.0000] <markp>
i have discovered html's most useless attribute

[14:00:02.0000] <Hixie>
pretty much

[14:00:03.0000] <markp>
or, more likely, rediscovered

[14:18:00.0000] <aja>
hi all...offering some experience re: 512 scanning for encoding.  GRDDL parsing seems to be depending on profile for each microformat...and if one defines each official and draft uF, 512 is rapidly approached.  might be something to work out with the GRDDL/hGRDDL WG folks  

[14:19:00.0000] <aja>
2nd concern: if multiple media and/or alternate stylesheet PI's are used, 512 can become a concern, too

[14:49:00.0000] <KevinMarks>
what's the significance of 512?

[14:51:00.0000] <a-ja>
KevinMarks: only first 512 bytes scanned for content encoding

[14:52:00.0000] <KevinMarks>
you mean explicit content-encoding, or for statistical guessing?

[14:52:01.0000] <KevinMarks>
oops

[14:52:02.0000] <a-ja>
explicit

[14:52:03.0000] <KevinMarks>
so the recommendation is to put the profiles after the content header?

[14:54:00.0000] <a-ja>
guess putting them on individuaal uF-containing elements rather than on <html> would do....but is that supported already?

[14:56:00.0000] <Hixie>
a-ja: we've dropped profile="" from HTML5 at the moment

[14:56:01.0000] <Hixie>
a-ja: based on the principle that nobody on the web was actually using it

[14:56:02.0000] <a-ja>
tell the GRDDL WG !

[14:57:00.0000] <a-ja>
tks, Hixie...wasn't aware of that.  did see some related discussion back in Dec though

[14:57:01.0000] <Hixie>
nobody uses GRDDL either :-)

[14:58:00.0000] <a-ja>
heheh....certainly not yet anyway

[14:58:01.0000] <Dashiva>
GRDDL doesn't sound cool enough

[14:59:00.0000] <a-ja>
doesn't have the same ring to it that RDFa has?

[14:59:01.0000] <Dashiva>
It's missing at least one X

[15:01:00.0000] <Hixie>
well more importantly, there doesn't really seem to be much demand for a microformats-to-RDF convertor

[15:01:01.0000] <Hixie>
most people who consume microformats will just have microformat-specific parsers

[15:02:00.0000] <a-ja>
true enuff....and i haven't seen one yet that looked for a profile

[15:03:00.0000] <a-ja>
"in the wild" anyway

[15:03:01.0000] <KevinMarks>
as publishing profiles is a major pain

[15:03:02.0000] <KevinMarks>
and grepping for the container class is easier

[15:08:00.0000] <Hixie>
when someone changes the <video src=""> and invokes .load() again, what should happen?

[15:09:00.0000] <Hixie>
should everything reset to initial values?

[15:09:01.0000] <Hixie>
should events be fired for each attribute that's reset?

[15:19:00.0000] <Dashiva>
Looking at xhr, some kind of abort/unload event coupled with state changes to 0/empty/uninit. More than that could get really messy

[15:44:00.0000] <othermaciej>
Hixie: sorry for not being in the right place

[15:44:01.0000] <Hixie>
no worries

[15:44:02.0000] <Hixie>
i was confused

[15:44:03.0000] <Hixie>
anyway

[15:44:04.0000] <Hixie>
23:17 <         Hixie>|when someone changes the <video src=""> and invokes .load() again, what should happen?

[15:44:05.0000] <Hixie>
23:17 <         Hixie>|should everything reset to initial values?

[15:44:06.0000] <Hixie>
23:17 <         Hixie>|should events be fired for each attribute that's reset?

[15:46:00.0000] <Hixie>
in particular, right load you invoke a load() method to restart the video element with a new source

[15:46:01.0000] <Hixie>
but that needs to fire a whole sequence of events: abort on the previous thing, emptied because we're going back to zero, paused if it was previously playing, unavailable if we had a frame before, begin to indicate we've started a new download

[15:47:00.0000] <Hixie>
and i'm not really sure what order those events should fire in, and whether they should fire before or after load() returns

[15:48:00.0000] <Hixie>
it feels like 'abort' should fire after it returns, but it also feels like the attributes should be reset before it returns, and it feels like they should be reset after abort is fired

[15:48:01.0000] <Hixie>
which is an overconstrained situation

[15:48:02.0000] <Hixie>
we could just say that you can't re-use a <video> but that seems dumb

[15:48:03.0000] <Hixie>
i think i'm going to think on this while cycling to work. feel free to describe your thoughts here and i'll read them when i get in.

[15:52:00.0000] <KevinMarks>
what happened to sequencing?

[15:54:00.0000] <othermaciej>
Hixie: I think everything should reset but probably a single "emptied" event is ok to express, not sure offhand

[15:54:01.0000] <othermaciej>
KevinMarks: I don't think sequencing is in yet

[15:56:00.0000] <KevinMarks>
this kind of stuff is why trying to handle sequencing in the js layer is hard

[15:57:00.0000] <Hixie>
othermaciej: so not even an 'abort' event?

[15:57:01.0000] <othermaciej>
Hixie: I'm not up to speed on the current state of the events

[15:58:00.0000] <hsivonen>
Hixie: I see (Re: no enforceable patents of Baseline)

[15:58:01.0000] <hsivonen>
markp: I'm glad to see you can finally talk with us. Welcome.

[15:59:00.0000] <hsivonen>
markp: can you say what your job at Google is? are you going to work on Web Forms 2.0 accessibility in Gecko?

[15:59:01.0000] <othermaciej>
KevinMarks: the main reason for built-in sequencing is the same as looping, that you can better avoid glitches by appropriately pre-buffering

[16:00:00.0000] <Hixie>
othermaciej: 'abort' would be the progress event, from DOM Level 0

[16:01:00.0000] <othermaciej>
Hixie: would probably make sense to have that, since it has more general purposes than UI control state

[16:01:01.0000] <Hixie>
yeah, just not sure when to fire it

[16:04:00.0000] <KevinMarks>
yes

[16:08:00.0000] <markp>
hsivonen: hi; yes, i can talk to you (on company time, even!); my official title is "lead technical writer"; my position involves some of these things: http://www.google.com/support/jobs/bin/answer.py?answer=53090 ; i am no longer working on firefox or accessibility issues

[16:13:00.0000] <hsivonen>
markp: ok


2007-04-04
[17:11:00.0000] <Hixie>
ok we're gonna synchronously fire an 'abort' event while load() is running, followed by an 'emptied' event.

[17:11:01.0000] <Hixie>
the next questios whether to fire the 'begin' event before or after the method returns

[17:11:02.0000] <Hixie>
and whether to set the state to LOADING before or after the method returns

[17:11:03.0000] <Hixie>
i'm pretty sure we want to set LOADING before.

[17:12:00.0000] <Hixie>
but if we fire 'begin' afterwards, that would mean there's a period of time where it's LOADING and there is script running even though the 'begin' event hasn't fired yet

[17:12:01.0000] <Hixie>
i suppose now that we have 'abort' firing synchronously we can do the same with 'bugin'

[17:12:02.0000] <Hixie>
'begin' even

[17:28:00.0000] <Hixie>
hm, we're gonna have to change the html parsing spec to not drop <option> elements to support web forms 2 <datalist>

[21:52:00.0000] <zcorpan>
morning

[21:53:00.0000] <Lachy>
hey zcorpan 

[21:53:01.0000] <Lachy>
how did your interview at Opera go?

[22:02:00.0000] <zcorpan_>
Lachy: it went well (see logs)

[22:02:01.0000] <Lachy>
in this channel or #html-wg?

[22:02:02.0000] <Lachy>
and on what date?

[22:03:00.0000] <othermaciej>
/me wonders if he could get a req approved to hire a standards expert so he can order others to flame on mailing lists and go to useless meetings instead of doing it himself

[22:04:00.0000] <zcorpan__>
18:19 20070403 #whatwg

[22:08:00.0000] <Lachy>
where is Jönköping?

[22:08:01.0000] <Lachy>
/me looks it up in google maps...

[22:09:00.0000] <Lachy>
ah, Sweeden

[22:11:00.0000] <zcorpan__>
content authors will continue to use flash until <video> works with one codec across the browsers the author cares about

[22:12:00.0000] <zcorpan__>
then they might reconsider

[22:16:00.0000] <Lachy>
and they will only switch from flash to <video> if there are clear benefits to do so

[22:17:00.0000] <Hixie>
othermaciej: if you do do that, make sure it's someone who actually knows what they're talking about

[22:17:01.0000] <Hixie>
othermaciej: a lot of companies will hire ivory tower people to do that and that's how you end up in the mess we have now :-)

[22:18:00.0000] <othermaciej>
Hixie: I would only hire someone for such a post whose standards-related work I was personally familiar with, most likely

[22:18:01.0000] <Hixie>
that works

[22:19:00.0000] <othermaciej>
I think I know which subset of those reflects a good set of web standards values

[22:20:00.0000] <KevinMarks>
do what?

[22:21:00.0000] <othermaciej>
KevinMarks: that's a convoluted way of saying "people who are not idiots and whose opinions about web stuff are reasonably similar to my own"

[22:23:00.0000] <zcorpan>
what will the hired someone do?

[22:24:00.0000] <othermaciej>
participate in standards groups, help develop standards, help write specs for apple proposed standard features, make standards-driven test cases for both official standard body use and for webkit regression testing use

[22:24:01.0000] <othermaciej>
but I'm not yet sure if I need a special person to do that or can get people I work with to do more of that kind of stuff

[22:25:00.0000] <Lachy>
do you have anyone in mind who you'd like to hire. if you could?

[22:27:00.0000] <othermaciej>
I have a likely short list in mind, but probably not worth discussing further until the position is less hypothetical

[22:27:01.0000] <Lachy>
fair enough

[22:27:02.0000] <KevinMarks>
would that include iTunes baroque RSS support or is this a pure webkit position?

[22:28:00.0000] <othermaciej>
mainly web standards that are relevant to Safari/WebKit though I'd hope there could be internal advocacy to other teams at Apple when appropriate

[22:28:01.0000] <othermaciej>
(lord knows I don't have the time for such things)

[22:59:00.0000] <karlUshi>
othermaciej: I think it is a mistake to hire someone for a standard group as it disconnects the person from the real work in the team. It would work only if the person was really working with all engineers on a daily basis.

[23:00:00.0000] <othermaciej>
karlUshi: I would expect such a person to work actively with the team, though probably more on testing than implementation

[23:00:01.0000] <karlUshi>
though someone specialist of Web stuff advocating to all teams developing things related to html inside the company could indeed be a good thing

[23:00:02.0000] <karlUshi>
othermaciej: yes testing seems a good option

[23:01:00.0000] <karlUshi>
for spec writing we lack often of good technical writers too, or at least technical writers for reviewing the spec.

[23:02:00.0000] <karlUshi>
hmmm thinking of it.  Apple might have in-house technical writers who could read stuff

[23:03:00.0000] <othermaciej>
we have tech writers but I think at best they could help with copyediting

[23:22:00.0000] <zcorpan>
with firefox 3, i can finally start to use display:inline-block. :) floats can be such a pita when you really just want inline-block

[01:07:00.0000] <Hixie>
othermaciej: when talking about the CAN_SHOW_this or that states, complete this sentence: "Media elements have a state which describes..."

[01:08:00.0000] <othermaciej>
"... their current degree of presentability"

[01:08:01.0000] <othermaciej>
"... their current degree of presentability at the current time position"

[01:08:02.0000] <Hixie>
that works

[01:08:03.0000] <Hixie>
thanks

[03:36:00.0000] <Hixie>
this <video> patch to introduce the new state systems is gonna be a whopper.

[03:36:01.0000] <Hixie>
it's over 1500 lines already

[05:29:00.0000] <hsivonen>
It'll be interesting to see what Nokia ships when Opera is ready to offer them Theora support as part of the engine

[05:31:00.0000] <MikeSmith>
hsivonen - didn't (somebody) say that there was no way that Nokia would ship Theora?

[05:33:00.0000] <hsivonen>
MikeSmith: timeless said on the list that they wouldn't. And the manager of the whole Maemo operation earlier said that they wouldn't ship Vorbis without their own people vetting it first (without indication of whether any vetting is actually taking place).

[05:37:00.0000] <hsivonen>
taking out features from Presto due no an OEM legal policy would be uncool, though

[05:38:00.0000] <hsivonen>
s/no/to/

[06:04:00.0000] <MikeSmith>
hsivonen - yeah, certainly would not be the best thing for end users

[06:49:00.0000] <Dorward>
/me hears a rumour that Hixie is pushing for clients to be allowed to treat text/plain documents as HTML and points out that when Safari did that to http://www.hixie.ch/advocacy/xhtml it became rather difficult to read. 

[06:50:00.0000] <Dashiva>
Rumors aren't always that reliable

[06:50:01.0000] <Dorward>
Dashiva: No, which is why I <em>ed that it was a rumour rather then acting as though it was fact.

[06:58:00.0000] <zcorpan>
now we need another google project: a blog using genchi

[06:59:00.0000] <zcorpan>
should be straight-forward to migrate from WP or other popular blogs to that hypothetical blog

[07:00:00.0000] <zcorpan>
and it should integrate with hsivonens validator

[07:18:00.0000] <Dashiva>
"The src content attribute on media elements gives the address of the video to show."

[07:19:00.0000] <Dashiva>
Hixie: Maybe that should say media instead of video?

[09:56:00.0000] <annevk>
hello

[10:07:00.0000] <zcorpan>
hello annevk

[10:08:00.0000] <annevk>
/me though that HTMLImageElement.complete was some new feature but apparently browsers already support it in incompatible ways

[10:09:00.0000] <annevk>
I suppose "in incompatible ways" goes without saying...

[10:12:00.0000] <Philip`>
I've used that in Firefox and Opera with no problems (to poll for image loading, since I didn't want to use events), though my code was gratuitously incompatible with IE anyway so I have no idea how it handles complete

[10:13:00.0000] <annevk>
Philip`, what did you use it for?

[10:13:01.0000] <Philip`>
But "The DOM attribute complete must return true if the user agent has downloaded the image specified in the src attribute, and it is a valid image." doesn't seem to agree with what I found - I'm fairly sure I had complete==true once the image had failed to load (e.g. with a 404 error)

[10:13:02.0000] <Philip`>
(so I tested width!=0 to see if it had actually loaded successfully, and replaced it with an error image otherwise)

[10:14:00.0000] <annevk>
/me used "javascript:alert((new Image()).complete)" as testcase

[10:15:00.0000] <Philip`>
That was just in a canvas game, which loaded images while running and only wanted to check for updated status between frames

[10:19:00.0000] <annevk>
/me will make some testcases later to test interop a bit better

[10:19:01.0000] <annevk>
/me is going to fetch a Wii now!

[10:20:00.0000] <Philip`>
Hmm, I can't work out what I was doing that made it look like it worked, or whether I was just imagining the whole thing...

[10:22:00.0000] <Philip`>
Oh wait, that's because I'm trying .completed instead of .complete, hence it always returning undefined...

[10:24:00.0000] <Philip`>
data:text/html,<script>var i=new Image(); i.src="http://microsoft.com/404"; alert(i.complete); setTimeout(function(){alert(i.complete)}, 1000);</script>

[10:24:01.0000] <Philip`>
That does what I expected - False at first, then True after it's got the error response

[10:24:02.0000] <Philip`>
but only in FF/Opera - IE returns False/False

[10:45:00.0000] <annevk>
IE makes more sense imo

[10:48:00.0000] <gavin_>
/me quotes annevk out of context

[10:48:01.0000] <gavin_>
I can see it on Digg now

[10:48:02.0000] <annevk>
:p

[10:53:00.0000] <Philip`>
IE isn't as useful since it only allows the polling-equivalent of onload and you can't poll for the same conditions as onerror, whereas FF/O sort of let you tell the difference between complete-successful and complete-error

[10:53:01.0000] <Philip`>
but IE does make more sense for something called "complete"

[10:54:00.0000] <annevk>
afaict the attribute just reflects whether the load event has been dispatched on the element or not

[10:55:00.0000] <annevk>
i suppose what FF/O is let it reflect whether load or error has been dispatched

[10:55:01.0000] <annevk>
(but that may be what you just said, I'm a bit dense)

[10:57:00.0000] <Philip`>
That was what I was thinking (and what seems to happen when I test it), though I'm not sure if what I actually said made sense :-)

[11:01:00.0000] <annevk>
That doesn't seem to be true either

[11:02:00.0000] <annevk>
alerting complete in a load event handler returns false in IE

[11:03:00.0000] <annevk>
Philip`, it's a security risk... consider an image from an intranet

[11:05:00.0000] <annevk>
.complete does return true after the load event of the window object is dispatched, but not before...

[11:05:01.0000] <annevk>
/me wonders if that can be considered a bug in IE

[11:06:00.0000] <Philip`>
For data URLs? I was thinking that if you are able to call canvas.toDataURL, you are the same person who was calling canvas.drawImage, so you had the Image, so you could read image.src and get the image data out. But I don't know how much cross-domain scripting is allowed, and whether one of those assumptions is wrong

[11:09:00.0000] <annevk>
Philip`, the only way to get the image data is through <canvas>

[11:09:01.0000] <annevk>
Philip`, hence the restriction

[11:10:00.0000] <annevk>
(the security model of the web is rather odd, but we have to deal with it :))

[11:11:00.0000] <Philip`>
But when the image is loaded from a data: URL (e.g. because it came from another call to canvas.toDataURL), you can read .src and get the image data without using the canvas at all

[11:12:00.0000] <Philip`>
like by writing a PNG decoder in JavaScript, which I guess is a little painful but theoretically possible

[11:12:01.0000] <annevk>
I'm not up to speed with data: URIs and security, sorry

[11:14:00.0000] <annevk>
One thing that is problematic with data: URIs if I remember correctly is when they are the result of a redirect

[11:17:00.0000] <Philip`>
Ah, sounds like that kind of thing makes it awkward

[11:24:00.0000] <annevk>
http://tc.labs.opera.com/html/img/ has four tests which show that no browser implements the spec

[11:29:00.0000] <Philip`>
And none misimplement it in the same way

[11:29:01.0000] <Philip`>
   Test: 1 2 3 4

[11:29:02.0000] <Philip`>
Firefox: F P P F

[11:29:03.0000] <Philip`>
Opera:   F P P DNR

[11:29:04.0000] <Philip`>
IE:      P F F P

[11:29:05.0000] <Philip`>
Safari:  F F F DNR

[11:34:00.0000] <zcorpan>
 another <figure> with <credit>: http://www.tellusgame.com/Tellusfilm/2.%20Finished/artikelsidan.html

[11:36:00.0000] <zcorpan>
i've noticed that whatever html5 feature i use (usually type=email &c), as soon as i pass the code to someone else it disappears. probably their editors or tools remove them or complain about them

[11:37:00.0000] <zcorpan>
or they are dicks and want to pass validator.w3.org

[11:45:00.0000] <zcorpan>
http://www.tellusgame.com/Tellusfilm/2.%20Finished/kronikorsidan.html has only <credit> and no caption for the figures

[11:50:00.0000] <zcorpan>
firefox seems to be buggy about display:table-cell. sometimes they get rendered on their own line. on reload it renders correctly

[11:50:01.0000] <zcorpan>
/me will try to reproduce with a simpler test case

[12:04:00.0000] <annevk>
Microsoft just joined the HTML WG

[12:04:01.0000] <annevk>
With Chris Wilson as their sole participant

[12:07:00.0000] <annevk>
if Martin Atkins is around, my apologies...

[12:11:00.0000] <Dashiva>
How so?

[12:11:01.0000] <annevk>
for asking he question he apparently already addressed

[12:12:00.0000] <annevk>
s/he question/a question/

[12:46:00.0000] <zcorpan>
hsivonen: would it be possible to generate a typesetted pdf of the spec dynamically? would be nice to offer a printable version of the latest spec without having people to typeset it manually

[13:06:00.0000] <hsivonen>
zcorpan: yes, it would be reasonable. however, I think I'd overstep my Prince license if I did that

[13:06:01.0000] <hsivonen>
zcorpan: but with a Server license, no technical problem

[13:06:02.0000] <annevk>
I suppose we might be able to get one...

[13:09:00.0000] <hsivonen>
zcorpan: I guess the best way to do it would be installing Prince (under a server license) on Dreamhost and running it in an SVN post-commit hook each time Hixie commits a new version

[13:09:01.0000] <hsivonen>
it should probably be run twice: A4 and Letter

[13:13:00.0000] <hsivonen>
hmm. Perhaps if Hixie is the licensee and invokes Prince as part of his commit script, a Single-User license with Hixie as the licensee would work

[13:15:00.0000] <annevk>
There's quite a few Prince people on the WHATWG list

[13:18:00.0000] <hsivonen>
annevk: I know. More the reason to comply. :-) but OTOH, more chance that the Prince people might help with such a setup

[13:18:01.0000] <annevk>
/me points molly to http://simon.html5.org/test/ie7b2-bugs/

[13:22:00.0000] <zcorpan>
/me should probably mark the ones already fixed as obsolete

[13:23:00.0000] <annevk>
you could add a readm.txt mentioning all the fixed tests

[13:23:01.0000] <annevk>
readme*

[13:24:00.0000] <zcorpan>
you know off-hand which they are?

[13:24:01.0000] <annevk>
nope

[13:25:00.0000] <zcorpan>
ok

[14:43:00.0000] <annevk>
http://twitter.com/Hixie/statuses/14588161

[14:43:01.0000] <annevk>
seems Hixie adopted twitter

[14:43:02.0000] <othermaciej>
!

[14:44:00.0000] <hober>
yay!

[14:46:00.0000] <othermaciej>
haha, I know what group that is

[14:47:00.0000] <Hixie>
blame tantek

[14:47:01.0000] <Hixie>
for me using twitter

[14:47:02.0000] <Hixie>
i wish it worked though

[14:47:03.0000] <othermaciej>
wow, that's a whole lotta twittering

[14:47:04.0000] <Hixie>
the jabber server never sends me updates

[14:48:00.0000] <Hixie>
/me uses it as an IM away message

[14:48:01.0000] <annevk>
oh yeah, I just picked a funny line

[14:48:02.0000] <tantek>
Hixie, check your prefs and make sure you set (*) IM

[14:48:03.0000] <Hixie>
i have

[14:48:04.0000] <Hixie>
and i've said 'on'

[14:48:05.0000] <tantek>
and then i believe you have to appear "online"

[14:48:06.0000] <tantek>
rather than "away"

[14:48:07.0000] <Hixie>
i do

[14:48:08.0000] <tantek>
otherwise i think it stops sending them to your jabber

[14:48:09.0000] <tantek>
er, im

[14:48:10.0000] <tantek>
odd

[14:49:00.0000] <Hixie>
i'm present, logged in, enabled it, confirmed it with the code, said 'on', the works

[14:49:01.0000] <Hixie>
never gotten it to send any of my friends' messages to me

[14:49:02.0000] <Dashiva>
The twitter server seems slow like the css 2.1 specification process

[14:50:00.0000] <Dashiva>
/me notes that Hixie himself only sends lggwg moves during like 1/4 of the day

[14:51:00.0000] <Hixie>
more like 3/4

[14:51:01.0000] <Hixie>
but sure

[14:51:02.0000] <othermaciej>
if twitter starts containing useful information, I might have to sign up :-(

[14:52:00.0000] <annevk>
/me was wondering about the same thing

[14:52:01.0000] <Hixie>
twitter, as far as i've experienced, is a write-only medium

[14:52:02.0000] <Hixie>
note that my blog shows my status and updates it in real time if you leave the page open

[14:53:00.0000] <hsivonen>
twitter on ADSL is slower than Flickr on GPRS

[14:56:00.0000] <tantek>
btw, some folks hangout sometimes in #twitter

[14:57:00.0000] <tantek>
and at least you can twitter about any problems you have on twitter

[14:57:01.0000] <Dashiva>
We could make a twitter clone that selects random lines from IRC instead :)

[14:58:00.0000] <annevk>
krijnh, could modify his bot so we could do "twitter: blah" and it would appear in some #whatwg twitter

[14:58:01.0000] <Dashiva>
twitter: Hi mom

[14:58:02.0000] <Hixie>
tantek: i've done so several times :-)

[14:59:00.0000] <Hixie>
tantek: i really have no idea why it doesn't work

[15:01:00.0000] <tantek>
How strange - I thought I had added you and for some reason when I went to your profile I had not.

[15:01:01.0000] <othermaciej>
the twitter web site seems to be pretty slow right now

[15:01:02.0000] <tantek>
re-adding twitter.com/hixie

[15:01:03.0000] <tantek>
ah there are your bug reports

[15:01:04.0000] <tantek>
ok, let's try this...

[15:02:00.0000] <hsivonen>
aaagh. Atom abuse by twitter: double escaped titles

[15:03:00.0000] <hsivonen>
getting rid of this problem is why Atom exists

[15:04:00.0000] <Dashiva>
Making it easier to do it right just means more energy left for doing it wrong!

[15:04:01.0000] <annevk>
i suppose that shows that inventing something new doesn't solve problems

[15:08:00.0000] <tantek>
hsivonen - I use XHTML titles and very few Atom consumers decode them - instead I see literal: "<div> Tantek's Updates</div>"  as the title

[15:09:00.0000] <tantek>
e.g. http://blogsearch.google.com/blogsearch?q=link:tantek.com&scoring=d

[15:09:01.0000] <annevk>
nowadays i think that Atom might have been a mistake

[15:10:00.0000] <tantek>
Of course Technorati's Atom parsing is compliant (as one of the earlier consumers: http://www.intertwingly.net/wiki/pie/KnownAtomConsumers )

[15:10:01.0000] <KevinMarks>
the problem is the pain people went through for RSS lingers in their atom parsers

[15:10:02.0000] <zcorpan>
is http://simon.html5.org/test/ie7b2-bugs/037.html valid or not?

[15:10:03.0000] <tantek>
annevk, as a publisher, I'm very happy about Atom being there - semantics are much more well defined

[15:11:00.0000] <annevk>
tantek, we could have fixed RSS to make semantics well defined

[15:11:01.0000] <KevinMarks>
and as a parser, it is way better

[15:11:02.0000] <annevk>
zcorpan, it's valid per HTML5, yes

[15:11:03.0000] <othermaciej>
it would have been better to fix RSS

[15:11:04.0000] <tantek>
and rel="enclosure" made it trivial to mark media items that I had already been linking to as podcast items

[15:11:05.0000] <zcorpan>
annevk: ok

[15:11:06.0000] <KevinMarks>
that was made impossible

[15:11:07.0000] <KevinMarks>
really

[15:11:08.0000] <tantek>
as opposed to dealing with yet another random new element

[15:11:09.0000] <hober>
I don't think fixing RSS was, politically, an option

[15:11:10.0000] <annevk>
in due course someone still has to fix RSS and Atom

[15:11:11.0000] <KevinMarks>
Atom is fixed

[15:12:00.0000] <annevk>
the Atom specification doesn't say what consumers should do with invalid Atom

[15:12:01.0000] <tantek>
annevk - no, people tried fixing RSS to make its semantics well defined and failed.  That's why Atom happened.  See: Sam Ruby, Tim Bray.

[15:12:02.0000] <KevinMarks>
the feed validaotr makes invalid atom very clear

[15:12:03.0000] <tantek>
Atom is much more easily "fixed" in that regard (interoperable error handling) than RSS.

[15:13:00.0000] <annevk>
the HTML WG tried to fix HTML at some point to and then went on with XHTML2 because it was considered too hard

[15:13:01.0000] <KevinMarks>
with RSS the error message are more like 'this is a matter of debate'

[15:13:02.0000] <annevk>
RSS should be "trivial" compared to HTML

[15:13:03.0000] <tantek>
annevk, no that's not the reason

[15:13:04.0000] <tantek>
there was political motivation/pressure to do a "pure XML" version of HTML

[15:13:05.0000] <KevinMarks>
which became RSS 1.0, the RDF flavour

[15:13:06.0000] <tantek>
that was greater than the pressure to "fix" HTML

[15:13:07.0000] <othermaciej>
Atom does seem like an XHTML2-style solution

[15:13:08.0000] <annevk>
ok, maybe the former XForms WG then

[15:13:09.0000] <tantek>
some folks had the delusion that "pure XML" = "fix"

[15:14:00.0000] <KevinMarks>
RSS 1.0 was the xhtml2-like solution

[15:14:01.0000] <annevk>
probably the XForms people, yes

[15:14:02.0000] <hober>
KevinMarks: exactly

[15:14:03.0000] <tantek>
KevinMarks FTW

[15:14:04.0000] <annevk>
Atom is yet another one

[15:14:05.0000] <KevinMarks>
no

[15:14:06.0000] <tantek>
Atom doesn't have that much reinvention.

[15:14:07.0000] <KevinMarks>
I suppose hAtom is the HTML solution...

[15:15:00.0000] <annevk>
in two flavors, 0.3 and 1.0

[15:15:01.0000] <hsivonen>
tantek: I use XHTML titles as well. No complaints. However, many Atom consumers don't get namespace-prefixed XHTML right.

[15:15:02.0000] <annevk>
tantek, it does, it has all the same features under different element names

[15:15:03.0000] <tantek>
hsivonen, I use @xmlns rather than namespace-prefixed tags

[15:15:04.0000] <KevinMarks>
as opposed to the 9 or 12 flavours of RSS

[15:15:05.0000] <tantek>
annevk, not quite - it has more and more precise semantics

[15:15:06.0000] <tantek>
(atom vs. rss)

[15:16:00.0000] <annevk>
KevinMarks, well, you have to support all those flavors already

[15:16:01.0000] <othermaciej>
fortunately Atom and RSS are both much simpler than HTML

[15:16:02.0000] <KevinMarks>
it specifies which elements can contain HTML

[15:16:03.0000] <annevk>
KevinMarks, so better fix those than introduce even more

[15:16:04.0000] <hsivonen>
tantek: me too for the feed that I actually intend to be read. however, to point out brokenness I have implemented a non-canonical feed elsewhere

[15:16:05.0000] <othermaciej>
so an incompatible break is merely inordinately expensive as opposed to impossible

[15:16:06.0000] <hober>
Fortunately, Mark Pilgrim et al. already did all the heavy lifting, by writing the UFP :)

[15:16:07.0000] <tantek>
/me leaves it to KevinMarks to provide the rest of the history of Atom lesson.

[15:17:00.0000] <annevk>
/me was part of the Atom WG

[15:17:01.0000] <KevinMarks>
/me was too

[15:17:02.0000] <annevk>
/me doesn't need history lessons about it

[15:17:03.0000] <hsivonen>
Atom wasn't like XHTML 2.0. HTML didn't have a Winer.

[15:17:04.0000] <KevinMarks>
so how would you propose fixing RSS?

[15:17:05.0000] <hsivonen>
/me was also part of the Atom WG

[15:17:06.0000] <othermaciej>
making a whatwg style community breakaway spec for RSS might have, in retrospect, been better

[15:18:00.0000] <tantek>
I thought that was what was tried, by Cadenhead

[15:18:01.0000] <hober>
othermaciej: in some sense, that's what the RSS AB is

[15:18:02.0000] <hober>
tantek: right

[15:18:03.0000] <tantek>
with the RSS committee or WTF it was called

[15:18:04.0000] <annevk>
much the same way HTML has been fixed

[15:18:05.0000] <hober>
the rss advisory board

[15:18:06.0000] <tantek>
but that blew up in a bunch of blog drama AFAIK

[15:18:07.0000] <tantek>
yeah that thing

[15:18:08.0000] <annevk>
I think that was a pretty silly effort

[15:18:09.0000] <annevk>
not much community involvement at all and it focused on paragraph level changes as opposed to actually digging up all the issues etc.

[15:20:00.0000] <othermaciej>
anyway, it's true that staging a hostile takeover of a broken spec is harder than making a new one

[15:20:01.0000] <zcorpan>
/me added some notes to http://simon.html5.org/test/ie7b2-bugs/

[15:20:02.0000] <annevk>
problem is that the new one doesn't define much error handling either

[15:20:03.0000] <hober>
annevk: I'd say that that's an underlying problem with xml

[15:21:00.0000] <hober>
seems to me the html5lib liberalxmlparser is showing the way in that case

[15:21:01.0000] <annevk>
hober, I don't mean namespace-well-formedness error handling

[15:21:02.0000] <hober>
I know what you mean

[15:22:00.0000] <annevk>
what to do with a feed <feed xmlns="atomns"></feed>

[15:22:01.0000] <tantek>
othermaciej - good point

[15:22:02.0000] <tantek>
however, if any group *could* do it, i think whatwg could

[15:22:03.0000] <tantek>
and certainly would get a lot of, ahem, press for it

[15:22:04.0000] <annevk>
we already defined sniffing for feeds

[15:22:05.0000] <tantek>
because you can almost guarantee a fairly high level of blog drama

[15:22:06.0000] <annevk>
well, Hixie did

[15:23:00.0000] <tantek>
is there an RSS 2.1 spec?

[15:23:01.0000] <othermaciej>
I think the takeover of html is keeping everyone sufficiently busy

[15:23:02.0000] <hober>
othermaciej++

[15:23:03.0000] <tantek>
agreed.  fixing HTML is much more important than fixing RSS

[15:23:04.0000] <annevk>
hober, I plan to remove namespace-well-formed from XML btw

[15:23:05.0000] <Philip`>
Maybe work on RSS5 can be slotted between HTML5 and XML5

[15:23:06.0000] <kingryan>
why don't we just obsolete RSS?

[15:24:00.0000] <hober>
kingryan: RFC4287 already did :)

[15:24:01.0000] <kingryan>
/me points at http://microformats.org/wiki/hatom

[15:24:02.0000] <annevk>
kingryan, obsolete in what sense?

[15:24:03.0000] <hsivonen>
I think an RSS5 would be destructive to the atmosphere of the WHATWG list

[15:24:04.0000] <annevk>
rss-whatwg

[15:24:05.0000] <tantek>
especially since "S" and "5" look too similar in many fonts

[15:24:06.0000] <Dashiva>
It would be much more effective to aim for legislation making RSS illegal

[15:25:00.0000] <hsivonen>
people are nicer on the WHATWG list than on the AtomPub list

[15:25:01.0000] <othermaciej>
things that are more important to fix than RSS include:

[15:25:02.0000] <tantek>
when RSS is outlawed, only outlaws will (be in a) syndicate ?

[15:25:03.0000] <kingryan>
annevk: create something more useful and easier to us

[15:25:04.0000] <kingryan>
use*

[15:25:05.0000] <othermaciej>
DOM Core, CSS, SVG

[15:25:06.0000] <hsivonen>
syndication in general and RSS in particular make people act in non-nice ways

[15:25:07.0000] <othermaciej>
MathML is probably less important to fix than RSS

[15:25:08.0000] <tantek>
SVG? that needs something much more like Atomization treatment

[15:26:00.0000] <annevk>
VML!

[15:27:00.0000] <tantek>
annevk, no no no, you really want VRML.

[15:27:01.0000] <zcorpan>
TIME

[15:27:02.0000] <tantek>
keeps on ticking

[15:27:03.0000] <zcorpan>
or, SMIL

[15:28:00.0000] <annevk>
othermaciej, HTTP

[15:28:01.0000] <annevk>
XML

[15:28:02.0000] <annevk>
video codecs :)

[15:28:03.0000] <hsivonen>
WS-*5

[15:28:04.0000] <annevk>
heh

[15:29:00.0000] <zcorpan>
RDF5

[15:29:01.0000] <othermaciej>
HTTP spec does need fixing

[15:29:02.0000] <othermaciej>
I don't know what is wrong with XML offhand

[15:29:03.0000] <annevk>
/me wants graceful error handling

[15:29:04.0000] <othermaciej>
HTTP isn't in nearly as bad shape as SVG

[15:29:05.0000] <hober>
I'm all for interoperable error handling -- if someone wants to write up "what the UFP does" in spec form, that would be potentially good

[15:29:06.0000] <othermaciej>
WS-* and RDF are of no interest to me

[15:30:00.0000] <tantek>
hsivonen, othermaciej - I believe the proper term is WS-death-*

[15:30:01.0000] <othermaciej>
annevk: an error-recovering version of XML would be a significant incompatible break - not sure it's worth it

[15:30:02.0000] <annevk>
lots of mobile content already breaks in Opera because we don't do it

[15:31:00.0000] <annevk>
I think Webkit shipped in Nokia has therefore disabled XML for XHTML...

[15:31:01.0000] <tantek>
ah, "mobile content"

[15:31:02.0000] <tantek>
is that another oxymoron?

[15:31:03.0000] <annevk>
and feeds too

[15:31:04.0000] <annevk>
although with feeds it's mostly character encoding which most browsers seem to ignore...

[15:31:05.0000] <tantek>
btw, Hixie, you may want to de-add yourself from yourself: http://twitter.com/Hixie - that might be causing problems.

[15:31:06.0000] <annevk>
(on the HTTP level most implementations of XML are non-compliant)

[15:32:00.0000] <zcorpan>
it could possible to define graceful error handling for xml in a way that is compatible with xml 1.0 handling, although it wouldn't be possible to drop the internal subset (other than making it non-conforming)

[15:32:01.0000] <zcorpan>
/me thinks

[15:32:02.0000] <annevk>
yeah, internal subset is most of the complexity of XML

[15:33:00.0000] <zcorpan>
if new processors silently ignored internal subsets then they wouldn't be compatible with existing xml documents that rely on the internal subset

[15:33:01.0000] <zcorpan>
a lot of svg images do

[15:34:00.0000] <annevk>
it has to be supported obviously

[15:34:01.0000] <annevk>
all XML 1.0 content needs to remain working

[15:34:02.0000] <zcorpan>
indeed

[15:34:03.0000] <annevk>
and all wanabe XML 1.0 content will start working

[15:34:04.0000] <annevk>
in some way

[15:35:00.0000] <zcorpan>
an xml 1.0 processor should also be a conforming xml5 processor (i.e. error recovery should be optional)

[15:35:01.0000] <zcorpan>
(imho)

[15:36:00.0000] <annevk>
xml5 would be a superset

[15:36:01.0000] <annevk>
as xml1.1 will be mostly integrated as well

[15:36:02.0000] <annevk>
and namespaces for xml too

[15:37:00.0000] <othermaciej>
this all sounds like a lot of work

[15:37:01.0000] <othermaciej>
it's a wonder the web works at all

[15:37:02.0000] <zcorpan>
but things that are parse errors in xml 1.0 (and namespaces in 1.0) surely should continue to be (so that processors are allowed to stop)?

[15:38:00.0000] <annevk>
that would mean that a processor is allowed to stop at <?xml version="1.1"?>

[15:38:01.0000] <zcorpan>
yes

[15:38:02.0000] <annevk>
i'm not really sure what the value in that is

[15:39:00.0000] <hsivonen>
annevk: the value is creating an atmosphere of doubt about guarantees of non-XML 1.0 XML5 working

[15:39:01.0000] <hsivonen>
annevk: so people would still have an incentive to try to get things right

[15:39:02.0000] <hsivonen>
annevk: which would make the upgrade less damaging to users of XML 1.0 processors

[15:40:00.0000] <annevk>
i suppose

[15:40:01.0000] <hsivonen>
nn

[15:40:02.0000] <annevk>
i'll let the spec lawyering to you :)

[15:40:03.0000] <zcorpan>
hsivonen: nn

[15:52:00.0000] <KevinMarks>
SMIL got broken by the second rev 2 - the first rev was pretty good

[15:55:00.0000] <KevinMarks>
hober, what the UFP does is try really hard to clean things up - the 2500+ tests define what it does pretty well

[15:55:01.0000] <KevinMarks>
back-porting them to a spec would be difficult

[15:57:00.0000] <hober>
agreed

[15:58:00.0000] <KevinMarks>
the feed validator is probably a better practical tool than a spec

[15:58:01.0000] <othermaciej>
validators are useful to producers only if consumers match its behavior

[15:58:02.0000] <othermaciej>
and for that you need a spec and a test suite to throw at consumers

[15:59:00.0000] <KevinMarks>
well, UFP provides the test suite

[16:01:00.0000] <hober>
What I was thinking was that the wa1 parsing algorithm is a reverse-engineer of what is more-or-less the existing interoperable error recovery for parsing html.

[16:02:00.0000] <hober>
there's less in terms of existing interoperable behavior in the feed parsing world, so I'd say reverse-engineering the UFP's behavior into a "liberal feed parsing algorithm" a-la wa1 is a possibility

[16:05:00.0000] <Dashiva>
Atom is a bit easier to implement than XHTML2, though (and orthogonal to rss, too)

[16:15:00.0000] <Hixie>
tantek: Hixie1 and Hixie are different accounts

[16:22:00.0000] <tantek>
Hixie, but your twitter.com/Hixie appears to be subscribed to twitter.com/Hixie - Hixie1 had no friends/followers (until I added it moments ago)

[16:23:00.0000] <Hixie>
i can't seem to remove Hixie from Hixie

[16:23:01.0000] <Hixie>
not sure how that happened or how to remove it

[16:23:02.0000] <Hixie>
it doesn't appear on my friends list

[16:27:00.0000] <Hixie>
do we have a page somewhere that explains in detail why version syntax isn't needed?

[16:27:01.0000] <Hixie>
hsivonen, Lachy_?

[16:31:00.0000] <tantek>
Hixie, shouldn't the burden of proof be the opposite?

[16:31:01.0000] <tantek>
nothing is needed unless it can be demonstrated that it is

[16:31:02.0000] <zcorpan>
http://dbaron.org/log/2007-03#e20070325a

[16:32:00.0000] <Dashiva>
tantek: The widespread use of versioning in other cases will probably be used for that

[16:34:00.0000] <Hixie>
tantek: people give all kinds of arguments in favour of version syntax

[16:34:01.0000] <Hixie>
tantek: they're all bogus imho, but it would be nice to be able to point to a page that explains why isn't of having to rehash hsivonen's e-mail each time

[16:34:02.0000] <tantek>
then you need debunking of those arguments, not a general attempt at a proof against

[16:34:03.0000] <Hixie>
right

[16:34:04.0000] <tantek>
perhaps in FAQ form

[16:35:00.0000] <tantek>
we need the same thing for namespaces in general

[16:35:01.0000] <tantek>
debunking thereof

[16:51:00.0000] <zcorpan>
nn

[16:58:00.0000] <Hixie>
so a lot of spammers have started using whatwg stuff to seed their spam blogs and referral spam campaigns

[16:58:01.0000] <othermaciej>
weird

[16:59:00.0000] <Hixie>
what's really funny however is that i work for google

[16:59:01.0000] <Hixie>
and i report them internally

[16:59:02.0000] <Hixie>
which (a) kills their page rank, not only on that site, but on all related ones, and (b) gets their spam blogs deleted from sites like blogger.


2007-04-05
[17:00:00.0000] <Hixie>
i hope they continue

[17:00:01.0000] <Hixie>
makes our life hella easier

[17:03:00.0000] <tantek>
Hixie, indeed.  It strikes me as odd when spammers copy content that references my name (a reasonably unique search term) into their spam blogs, not realizing that I'm likely to find and kill their spam blogs minutes after they post (since I work at Technorati).

[17:07:00.0000] <tylerr>
Oh what a little research would do to the world of spammers.

[17:07:01.0000] <tylerr>
Rather, what a little research on spammers part would have on them not being so easily found out.

[17:09:00.0000] <Dashiva>
Rather, what a little research by average users would have done on making spam less profitable :)

[17:09:01.0000] <tylerr>
/me chuckles.

[17:09:02.0000] <tylerr>
Too true.

[17:10:00.0000] <tylerr>
I'd use my surname as a URL but it would be hard for people to remember.

[17:10:01.0000] <Hixie>
there's a small number of people who actually like spam and use it

[17:10:02.0000] <Hixie>
and that's all it takes for it to be profitable

[17:11:00.0000] <tylerr>
I'm still hunting for a decent domain name.

[17:11:01.0000] <tylerr>
I suppose I'm too much of a perfectionist though and the quest will continue forever on.

[17:39:00.0000] <Lachy>
Hi

[17:45:00.0000] <Lachy>
I'm finding spammers are actually getting creative and posting comments that are somewhat relevant to the post.  If it wasn't for their sites clearly being commercial and not related to web dev, it'd be hard to tell

[17:45:01.0000] <tantek>
Hixie, FWIW, I am getting your twitters

[17:54:00.0000] <Hixie>
tantek: yeah, i just don't get anyone else's.

[17:56:00.0000] <Hixie>
is there a way to see who has added you as a friend?

[17:56:01.0000] <Hixie>
the "27 Followers" line in my profile isn't a link

[17:59:00.0000] <KevinMarks>
http://twitter.com/Hixie/with_friends shows them

[18:00:00.0000] <Hixie>
ah thanks

[18:00:01.0000] <KevinMarks>
you sure you're logged in? the followers link is active for me whenI am

[18:00:02.0000] <Hixie>
definitely logged in

[18:00:03.0000] <Hixie>
and that page doesn't show a list of followers

[18:00:04.0000] <Hixie>
just the Friends

[18:00:05.0000] <hober>
fwiw, I see links for favorites & friends, but not followers

[18:02:00.0000] <Hixie>
oh hey

[18:02:01.0000] <Hixie>
i see a link to Followers when i go to the Favorites page

[18:02:02.0000] <Hixie>
man this site is so flaky

[05:30:00.0000] <zcorpan>
on a page listing news, would one use one <article> to wrap all news and then an <article> for each news item, or would one use a <section> to wrap all news then an <article> for each news item?

[05:55:00.0000] <zcorpan_>
http://simon.html5.org/temp/w3c-home-in-html5.html

[05:56:00.0000] <hsivonen>
zcorpan_: cool!

[05:57:00.0000] <zcorpan_>
styling doesn't work very well in firefox :(

[06:01:00.0000] <zcorpan_>
hsivonen: running it through your validator seems to reveal some bugs

[06:02:00.0000] <hsivonen>
The header element must have exactly one h1–h6 descendant.

[06:02:01.0000] <hsivonen>
yeah, that's a bug

[06:03:00.0000] <hsivonen>
I thought I had fixed it already

[06:03:01.0000] <zcorpan_>
also it complains about <nav><a> and <footer><address>

[06:03:02.0000] <hsivonen>
zcorpan_: perhaps the content model of <nav> has changed

[06:03:03.0000] <zcorpan_>
/me looks

[06:04:00.0000] <hsivonen>
I guess I might as well do the header/footer refactoring over easter

[06:04:01.0000] <hsivonen>
that should fix the address bug

[06:05:00.0000] <zcorpan>
Zero or more block-level elements, or inline-level content (but not both).

[06:05:01.0000] <hsivonen>
zcorpan_: I am rather convinced that it was block only when I worked on that part

[06:06:00.0000] <zcorpan>
ok

[06:45:00.0000] <annevk>
zcorpan, nice

[06:59:00.0000] <annevk>
zcorpan, maybe make an XHTML5 one for the styling?

[07:00:00.0000] <annevk>
http://simon.html5.org/temp/valid-html5.png lol!

[07:04:00.0000] <hsivonen>
yeah, that was fun even though in general I don't like validity badges

[07:13:00.0000] <zcorpan>
i don't like them either :)

[07:14:00.0000] <zcorpan>
annevk: i'd rather mozilla fixed their html parser

[07:15:00.0000] <annevk>
fair enough

[07:16:00.0000] <hsivonen>
Hixie: what's the deal with "one or more block-level elements" in the content model of <details>? usually you have zero or more

[07:17:00.0000] <hsivonen>
fantasai's framework assumes that you always have zero or more, so I have a schema-wide refactoring ahead if you really meant one or more

[07:17:01.0000] <annevk>
/me wonders what the use of zero would be

[07:17:02.0000] <annevk>
in that case

[07:17:03.0000] <hsivonen>
annevk: making placeholders conforming

[07:19:00.0000] <hsivonen>
although the refactoring would be good to schema aesthetics anyway 

[07:21:00.0000] <hsivonen>
/me realizes that zero block level elements counts as inline level empty string

[07:22:00.0000] <zcorpan>
there is a concept of inline level empty string?

[07:23:00.0000] <hsivonen>
zcorpan: I meant in terms of deriving the document tree from the grammar

[07:24:00.0000] <zcorpan>
ok

[07:24:01.0000] <hsivonen>
zcorpan: but the empty string counts as inline-level content

[07:24:02.0000] <hsivonen>
not a spec concept, though

[07:25:00.0000] <Dashiva>
So <p></p> contains a single empty string textnode, while <p><b></b></p> contains a single <b> element, which in turn contains a single empty string textnode?

[07:25:01.0000] <hsivonen>
Dashiva: they do not contain text nodes in the DOM sense

[07:26:00.0000] <Dashiva>
So document tree != the final DOM?

[07:26:01.0000] <hsivonen>
Dashiva: but as far as the spec's notion of inline content and the RELAX NG notion of text is concerned, yes

[07:26:02.0000] <zcorpan>
why does tbody take One or more tr elements, instead of 0 or more? http://www.whatwg.org/demos/repeat-01/ has an empty tbody if you click Remove

[07:27:00.0000] <hsivonen>
zcorpan: excellent question

[07:27:01.0000] <zcorpan>
/me will bug the list

[07:29:00.0000] <Dashiva>
+1 ;)

[07:32:00.0000] <zcorpan>
should perhaps apply to <tr>s as direct children to table too

[11:57:00.0000] <KevinMarks>
CSS Naked day is a good day fro a whatwg meet

[11:57:01.0000] <KevinMarks>
http://naked.dustindiaz.com/

[12:31:00.0000] <annevk>
http://www.markbaker.ca/blog/2007/04/05/faith/

[12:31:01.0000] <zcorpan>
ha

[13:20:00.0000] <zcorpan_>
and does someone know how opera have treated wml historically?

[13:26:00.0000] <hasather>
zcorpan_: I only know that it has been supported for quite some time

[13:28:00.0000] <zcorpan_>
i've seen hixie say that wml content generally isn't well-formed xml, and wml parsers aren't drocanian xml parsers

[13:28:01.0000] <zcorpan_>
but i can't find it back

[13:28:02.0000] <hasather>
zcorpan_: ok. Downliad Opera 4 and test it :)

[13:29:00.0000] <zcorpan_>
/me sighs

[15:09:00.0000] <zcorpan_>
chris didn't top-post! let's hope he continues that way :)

[15:34:00.0000] <hasather>
noticed that. text/plain too :)

[15:43:00.0000] <ianloic>
coffee good

[16:07:00.0000] <othermaciej>
I <heart> coffee

[16:12:00.0000] <zcorpan_>
/me ♥ unicode

[16:12:01.0000] <zcorpan_>
(and utf-8 is implied in this room ;) )

[16:41:00.0000] <Philip`>
Hmm, the example code under http://www.whatwg.org/specs/web-apps/current-work/#outlines seems to be missing some brackets in the second and third if blocks

[16:43:00.0000] <Hixie>
indeed

[16:44:00.0000] <zcorpan_>
would be nice to have that script in a working demo page

[16:45:00.0000] <zcorpan_>
/me tried it before but didn't get it working


2007-04-06
[19:12:00.0000] <ajnewbold>
is it just me or does inline style not work when using the proposed <!DOCTYPE HTML> doctype?

[19:12:01.0000] <Dashiva>
Define does not work, and which browser?

[19:13:00.0000] <ajnewbold>
I was using firefox at work earlier, and now Safari here at home, and it seems like <div style="whatever"> is just being ignored

[19:13:01.0000] <ajnewbold>
the style within the quotes isn't applied to the element

[19:13:02.0000] <ajnewbold>
/me double-checks

[19:14:00.0000] <ajnewbold>
yeah, it's just ignored

[19:15:00.0000] <ajnewbold>
I've never used any doctype other than the html4 strict one, but with that the usual behavior was that any inline style overrode everything else

[19:16:00.0000] <Dashiva>
I'm not getting any problems at http://folk.ntnu.no/magnusrk/test/inline5.html

[19:16:01.0000] <ajnewbold>
hmm damn

[19:16:02.0000] <Hixie>
ajnewbold: i doubt it's the doctype doing it

[19:16:03.0000] <Hixie>
ajnewbold: do you have a test page we can look at?

[19:16:04.0000] <ajnewbold>
yep, http://newbold.name/

[19:16:05.0000] <ajnewbold>
where it says "test"

[19:17:00.0000] <Dashiva>
size?

[19:17:01.0000] <ajnewbold>
it should be 5em unless I've done something stupid

[19:17:02.0000] <ajnewbold>
yeah

[19:17:03.0000] <ajnewbold>
I could have sworn I used to be able to do that

[19:17:04.0000] <Hixie>
'size' isn't a css property :-)

[19:17:05.0000] <Hixie>
you want font-size i think

[19:17:06.0000] <ajnewbold>
oh

[19:17:07.0000] <ajnewbold>
OH

[19:17:08.0000] <ajnewbold>
BLAST

[19:17:09.0000] <ajnewbold>
sorry for the false alarm, then :D

[19:17:10.0000] <ajnewbold>
this is what happens when I go away from html for a while :(

[19:17:11.0000] <ajnewbold>
I come back thinking things have broken, when it's really just me

[19:18:00.0000] <Hixie>
hehe

[19:18:01.0000] <Dashiva>
http://jigsaw.w3.org/css-validator/validator?uri=http%3A%2F%2Fnewbold.name%2F&warning=1&profile=css21&usermedium=all

[19:18:02.0000] <Dashiva>
For later

[19:19:00.0000] <Dashiva>
I'm postively surprised they support more than just CSS2 now

[19:19:01.0000] <ajnewbold>
woo no errors

[19:19:02.0000] <ajnewbold>
go me

[19:24:00.0000] <ajnewbold>
so what's the future look like for html 5?

[19:24:01.0000] <ajnewbold>
will the w3c accept it?

[19:26:00.0000] <Hixie>
hasn't really been discussed yet

[19:26:01.0000] <ajnewbold>
I know this might sound like the stupidest comment ever, but..

[19:26:02.0000] <ajnewbold>
I just want the w3c's validator to support it, once it's final :)

[19:27:00.0000] <Hixie>
yeah

[19:27:01.0000] <Hixie>
hsivonen is working on a validator

[19:27:02.0000] <ajnewbold>
ah, that takes care of that then

[19:28:00.0000] <ajnewbold>
hsivonen: will it do cool things like give helpful suggestions about how to fix problems?

[19:28:01.0000] <ajnewbold>
I remember an older version of the w3c's validator that was just downright unhelpful

[19:28:02.0000] <ajnewbold>
it'd be all "you forgot to close a tag somewhere, lol, good luck!"

[19:34:00.0000] <deltab>
it should check that its suggestions make sense, though

[19:35:00.0000] <Dashiva>
Well, if you have a non-closing <em> (automatic closing aside), is the checker supposed to analyze your text and estimate where you'd want to end it? :)

[19:37:00.0000] <deltab>
e.g. if you try to put a block in an inline context, it shouldn't say that you can only do that by putting it inside a button element, as the validator has done

[19:40:00.0000] <deltab>
nor close the inline context and forget about it, later finding and ignoring the end tag a couple of blocks down

[19:42:00.0000] <deltab>
i.e. treating <b><p>...</p><p>...</p></b>  as  <b></b><p>...</p><p>...</p>

[19:43:00.0000] <ajnewbold>
is <b><p>...</p><p>...</p></b> legal?

[19:44:00.0000] <deltab>
not in HTML4

[19:44:01.0000] <ajnewbold>
is it in html 5?

[19:44:02.0000] <deltab>
I don't know

[19:44:03.0000] <ajnewbold>
it should be, because it's effin' cool

[19:44:04.0000] <deltab>
I realize the HTML5 parser algorithm tries to handle that sort of thing; it'd be useful for the validator to explain how the parsing went

[19:45:00.0000] <ajnewbold>
"Parsing your document went kind of poorly.  You write HTML like it'd 1994."

[19:45:01.0000] <ajnewbold>
it's*

[19:45:02.0000] <deltab>
and, where possible, suggest *useful* (and not just *possible*) fixes

[19:45:03.0000] <ajnewbold>
I think it'd be great if it suggested books you could read

[19:45:04.0000] <ajnewbold>
or maybe even alternate career choices

[19:45:05.0000] <deltab>
heh

[19:46:00.0000] <deltab>
^useful^sensible

[19:47:00.0000] <ajnewbold>
"Finished checking document: 782 errors found in 1,092 lines.  Have you considered the very rewarding position of Waffle House line cook?"

[19:47:01.0000] <Dashiva>
On a scale from 1 to 10, you rate as Dreamweaver

[19:47:02.0000] <ajnewbold>
ouch!

[19:48:00.0000] <ajnewbold>
well, could be worse

[19:48:01.0000] <ajnewbold>
FrontPage. :P

[19:48:02.0000] <Lachy>
ajnewbold: <b><p>...</p></b> isn't legal because it wouldn't be completely compatible with the parsing requirements in all cases

[19:49:00.0000] <ajnewbold>
Lachy: aww :(

[19:49:01.0000] <ajnewbold>
it looked like fun

[19:49:02.0000] <ajnewbold>
or rather it reminded me of my first attempts at html, which were fun

[19:49:03.0000] <Lachy>
unfortunately, it's because of the way browsers have to handle it when the </p> is omitted

[19:50:00.0000] <ajnewbold>
I never understood the whole </p> omission thing

[19:50:01.0000] <ajnewbold>
I know it's legal, but damn does it feel weird

[19:50:02.0000] <Lachy>
see the adoption agency algorithm in the spec for more detail

[19:50:03.0000] <ajnewbold>
I've never brought myself to do it

[19:50:04.0000] <Lachy>
why?  It's a convenient shorthand

[19:50:05.0000] <ajnewbold>
I can't really explain it

[19:50:06.0000] <karlUshi>
ajnewbold: in the early history of p semantics it was designed as a separator

[19:50:07.0000] <Lachy>
why should authors have to type</p> when its not necessary?

[19:50:08.0000] <ajnewbold>
Lachy: I don't know :(

[19:51:00.0000] <ajnewbold>
It's just an odd personal feeling of mine

[19:51:01.0000] <Lachy>
ajnewbold: it's probably because you've been seduced by the dark side of XHTML

[19:51:02.0000] <ajnewbold>
ironically, I haven't!

[19:51:03.0000] <karlUshi>
Lachy: it confuses people. :) it's because you are an old HTML cow 

[19:51:04.0000] <karlUshi>
:p

[19:51:05.0000] <ajnewbold>
I despise X*

[19:51:06.0000] <Dashiva>
Typing </p> is like saying 'over' on the phone. It clarifies, but usually isn't necessary :)

[19:51:07.0000] <ajnewbold>
Dashiva: good analogy

[19:51:08.0000] <ajnewbold>
I still feel totally like I've screwed up if I leave it out, though :(

[19:51:09.0000] <karlUshi>
/me prefers to close his tags also because it helps me to understand my documents.

[19:52:00.0000] <karlUshi>
for my personal use, it is really a usability question

[19:52:01.0000] <Dashiva>
I always close my tags unless I'm doing a throw-away testcase or somesuch

[19:52:02.0000] <Dashiva>
But that's just the neat freak in me

[02:49:00.0000] <annevk>
"<b><p>test</p><p>test</p></b>" is handled just fine by HTML5

[02:54:00.0000] <zcorpan>
omitting optional tags might take some getting used to at first, but i find it very effective. :) in particular it's faster to write markup. i've never made friends with auto-close features in editors

[02:55:00.0000] <annevk>
whoa, howcome posted an experimental version of Opera with <video> support online

[02:55:01.0000] <annevk>
cool

[02:55:02.0000] <Hixie>
been up for a while

[02:55:03.0000] <Hixie>
since last thursday

[02:55:04.0000] <annevk>
oh, didn't know

[02:59:00.0000] <zcorpan>
/me thinks css, xbl and media queries should cover all use-cases of onbeforeprint

[03:00:00.0000] <annevk>
they also cover all use cases for onmouseover

[03:02:00.0000] <zcorpan>
could be, but changing things for print is always presentational, is it not?

[03:03:00.0000] <zcorpan>
not that i'm strongly opposed to onbeforeprint or anything, though

[03:04:00.0000] <annevk>
onmouseover hopefully too

[03:04:01.0000] <annevk>
(but it's not)

[03:04:02.0000] <annevk>
/me doesn't really have a strong opinion either

[03:05:00.0000] <hsivonen>
It seems to me that the best realistic initial outcome with <video> is that Opera and Mozilla ship Ogg Theora support, Apple ships a generic QuickTime hook, Microsoft ships a generic hook for the Window Media framework and Xiph ships QuickTime and Windows Media framework extensions for Ogg Theora

[03:07:00.0000] <zcorpan>
that could work

[03:07:01.0000] <hsivonen>
/me doesn't believe that MS and Apple would ship Theora support initially

[03:08:00.0000] <zcorpan>
/me is off to Robertsfors to visit his mother for the weekend

[03:08:01.0000] <annevk>
see you

[03:08:02.0000] <zcorpan>
bye

[03:08:03.0000] <hsivonen>
bye

[03:21:00.0000] <MikeSmith>
hsivonen - would be interesting to hear from other browser vendors and device manufacturers about this

[03:21:01.0000] <annevk>
Nokia said no

[03:21:02.0000] <MikeSmith>
Access NetFront, Openwave, Teleca Obigo, Infraware

[03:47:00.0000] <hsivonen>
MikeSmith: yes.

[03:55:00.0000] <MikeSmith>
lots of mobile browsers out there ... 

[03:56:00.0000] <MikeSmith>
Infraware is on 70 percent of the handsets in Korea

[03:56:01.0000] <MikeSmith>
NetFront is in PSP, lots of other devices

[04:03:00.0000] <hsivonen>
I was rather unimpressed by the PSP browser

[04:04:00.0000] <MikeSmith>
UI compared to Nintendo DS is not so great

[04:04:01.0000] <hasather>
I did some testing of the PSP browser when it came out. I was also very unimpressed

[04:05:00.0000] <MikeSmith>
because unilike DS, PSP has no pen input/touchscreen

[04:05:01.0000] <MikeSmith>
pen makes browsing a lot easier

[04:05:02.0000] <MikeSmith>
on that kind of device

[13:22:00.0000] <Hixie>
who's running the forums?

[13:23:00.0000] <fax_machine>
there are forums?

[13:24:00.0000] <fax_machine>
oh, there ARE forums

[13:24:01.0000] <fax_machine>
/me realizes he's never actually seen the front page of the WHATWG web site before

[13:24:02.0000] <Hixie>
heh

[13:25:00.0000] <fax_machine>
nice colors, btw

[13:25:01.0000] <Hixie>
hehe

[13:25:02.0000] <fax_machine>
er

[13:25:03.0000] <fax_machine>
is the word "The" supposed to be behind the word "Web"?

[13:26:00.0000] <fax_machine>
top-left corner of the front page

[13:26:01.0000] <Hixie>
yes

[13:26:02.0000] <fax_machine>
ah ok

[13:26:03.0000] <fax_machine>
good then :)

[13:26:04.0000] <Hixie>
though it's funny how many people ask that

[13:26:05.0000] <fax_machine>
well, to us lay people with no design skills, it almost looks like a mistake

[13:26:06.0000] <gsnedders>
I mean, no part of how HTML works is a mistake :P

[13:27:00.0000] <fax_machine>
and of course the first thought is "omg they made a mistake lol" or "oh crap does my browser suck?"

[13:27:01.0000] <fax_machine>
it's interesting

[13:27:02.0000] <fax_machine>
in Opera, it looks best

[13:27:03.0000] <fax_machine>
there's enough of the "The" poking up above the "Web" that the effect looks really good

[13:28:00.0000] <fax_machine>
in Safari and Camino it looks like a mistake :P

[13:28:01.0000] <gsnedders>
probably due to Opera rounding down numbers

[13:28:02.0000] <fax_machine>
heh, Opera always makes things look better :)

[13:28:03.0000] <fax_machine>
I should use it more

[13:34:00.0000] <fax_machine>
heh, does Anne van Kesteren ever come here?

[13:34:01.0000] <gsnedders>
as annevk

[13:34:02.0000] <gsnedders>
often

[13:34:03.0000] <fax_machine>
cool

[13:34:04.0000] <fax_machine>
I totally need to tell her that I love that pic on her site

[13:34:05.0000] <fax_machine>
the "Standards Suck" guy

[13:34:06.0000] <gsnedders>
*his site

[13:35:00.0000] <gsnedders>
*tell him

[13:35:01.0000] <fax_machine>
ah yeah

[13:35:02.0000] <fax_machine>
first I have to remove my foot from my mouth

[13:35:03.0000] <gsnedders>
he goes as anne on irc.w3.org

[13:35:04.0000] <fax_machine>
and become exposed to a bit more global culture, evidently

[13:35:05.0000] <gsnedders>
even in en, Anne is a gender neutral name

[13:35:06.0000] <fax_machine>
in en, maybe, but not in the USA

[13:36:00.0000] <fax_machine>
a man named Anne here wouldn't make it past 1st grade in school :(

[13:36:01.0000] <gsnedders>
I mean in the English language

[13:36:02.0000] <fax_machine>
I know

[13:36:03.0000] <gsnedders>
It's rare, but it exists.

[13:36:04.0000] <fax_machine>
but as you surely know, we don't really use the English language over here :P

[13:36:05.0000] <gsnedders>
of course :)

[13:36:06.0000] <fax_machine>
we speak in vomit-like tongues, a mixture of oblivious richness and SUV exhaust

[13:37:00.0000] <gsnedders>
it's most common in North Germanic languages as a male name

[13:38:00.0000] <gsnedders>
well, nowadays

[13:38:01.0000] <fax_machine>
interesting

[13:38:02.0000] <gsnedders>
there've been times when it's been predominantly male in en

[13:38:03.0000] <fax_machine>
I believe it

[13:38:04.0000] <fax_machine>
there've been times when men were the predominant knitters, as well :)

[13:38:05.0000] <fax_machine>
(early 1600s)

[13:39:00.0000] <gsnedders>
there've been times when 14 year olds are involved in developing standards…

[13:39:01.0000] <gsnedders>
wait, that's now.

[13:40:00.0000] <fax_machine>
heh

[13:41:00.0000] <fax_machine>
I know of a 12-year old who contributed significantly to PHP's earlier version 4 releases

[13:41:01.0000] <gsnedders>
heh

[13:41:02.0000] <gsnedders>
/me has a strong dislike of PHP

[13:41:03.0000] <gsnedders>
parts because of how the developers behave

[13:41:04.0000] <fax_machine>
yeah :(

[13:42:00.0000] <gsnedders>
I had to report a bug twice to get it fixed, once told that what PHP did following the spec, despite me saying it didn't

[13:42:01.0000] <gsnedders>
the second time I cited the specific sections of the spec

[13:42:02.0000] <fax_machine>
hehe

[13:42:03.0000] <fax_machine>
yeah, you shouldn't have to do that

[13:42:04.0000] <gsnedders>
the devs should go look the spec up themselves when they are told it breaks the spec.

[13:42:05.0000] <fax_machine>
I hate it when developers get defensive

[13:42:06.0000] <gsnedders>
I shouldn't have to.

[13:42:07.0000] <fax_machine>
exactly

[13:42:08.0000] <fax_machine>
well, they just assumed that they were right

[13:43:00.0000] <fax_machine>
it's a very arrogant attitude

[13:43:01.0000] <gsnedders>
F/OSS run as  a dictatorship only works with a good strong dictator

[13:43:02.0000] <fax_machine>
:)

[13:44:00.0000] <gsnedders>
/me doesn't want to put the current release of SimplePie through all the new unit tests

[13:44:01.0000] <gsnedders>
I'd hate to see how badly it does

[13:45:00.0000] <fax_machine>
what's SimplePie?

[13:45:01.0000] <gsnedders>
<http://simplepie.org/>

[13:45:02.0000] <gsnedders>
written in PHP, ironically :)

[13:45:03.0000] <fax_machine>
ah

[13:45:04.0000] <fax_machine>
that looks neat

[13:47:00.0000] <gsnedders>
http://diveintomark.org/archives/2004/08/16/specs – sometimes I think I fall a bit too much into the "Asshole" category

[13:48:00.0000] <fax_machine>
/me is a "moron"

[13:48:01.0000] <gsnedders>
/me used to be a total one

[13:49:00.0000] <gsnedders>
specs! peh! they make no sense! all too technical!

[13:49:01.0000] <gsnedders>
now I can read specs without thinking about it :P

[13:50:00.0000] <fax_machine>
I guess I might be a super-moron

[13:50:01.0000] <fax_machine>
not only have I traditionally ignored specs, but I've ignored standards too :P

[13:50:02.0000] <fax_machine>
only recently have I taken an interest in standards

[13:51:00.0000] <gsnedders>
/me sadly got caught by those who say XHTML > HTML

[13:51:01.0000] <gsnedders>
now as someone who has actually read specs and understood them, it's just wrong.

[13:52:00.0000] <fax_machine>
I've only seen maybe one or two sites done in XHTML that (a) used the right content-type, and (b) really actually made use of any of xhtml's features that aren't available in plain old html

[13:52:01.0000] <gsnedders>
/me sighs…

[13:52:02.0000] <fax_machine>
my theory is that 99% of developers who use xhtml do so simply because they think it's the newer, and therefore better, markup

[13:53:00.0000] <gsnedders>
I have a whole book of English past paper beside me

[13:53:01.0000] <gsnedders>
I'm really tempted to do one of the writing ones

[13:53:02.0000] <fax_machine>
does Mark Pilgrim ever chat in here

[13:54:00.0000] <Dashiva>
I'm sure some of them just let the editor/cms/whatever add the doctype

[13:54:01.0000] <hasather>
fax_machine: it happens

[13:54:02.0000] <gsnedders>
as markp

[13:54:03.0000] <fax_machine>
hsivonen: is he as clever and funny here as he is on his blog?

[13:55:00.0000] <hasather>
fax_machine: I think you meant that for me. Dunno really, haven't seen him talk much

[13:55:01.0000] <gsnedders>
/me tries not to laugh at one of the tasks – "Television shows such as Pop Idol do more harm than good. Discuss."

[13:55:02.0000] <gsnedders>
he barely talks

[13:56:00.0000] <gsnedders>
I've talked to him once, but all about rather complex stuff with both of us busy, little time for jokes

[13:56:01.0000] <fax_machine>
hasather: heh yes, I did mean that for you :)

[13:58:00.0000] <gsnedders>
time to write yet another essay, this time fiction, unlike the last <http://geoffers.uni.cc/archives/2007/04/06/new-people-in-new-surroundings/>

[13:59:00.0000] <fax_machine>
"Questioning me as to what colour her bra is, I answer honestly. It’s black, and rather visible through her thin white shirt."

[13:59:01.0000] <fax_machine>
/me keeps reading :)

[14:00:00.0000] <gsnedders>
fax_machine: it's completely true. It's written in the first person with reason. :)

[14:00:01.0000] <fax_machine>
hehe

[14:00:02.0000] <fax_machine>
dude, you told her no :(

[14:00:03.0000] <gsnedders>
I didn't then

[14:00:04.0000] <fax_machine>
you could have totally scored

[14:00:05.0000] <gsnedders>
too ill then anyway

[14:01:00.0000] <fax_machine>
ah :(

[14:01:01.0000] <gsnedders>
only really been well enough since Dec

[14:01:02.0000] <fax_machine>
well

[14:01:03.0000] <gsnedders>
just been too shy and cowardly since

[14:01:04.0000] <fax_machine>
what are you waiting for?

[14:01:05.0000] <fax_machine>
ah

[14:01:06.0000] <fax_machine>
you can fix that

[14:02:00.0000] <gsnedders>
/me starts writing an essay about a homosexual relationship, non-fiction, this time

[14:02:01.0000] <fax_machine>
try very hard to get a glimpse of the sort of live you might envision having that would involve her, and then ask yourself how much more of your life you're going to waste before you make that vision a reality :)

[14:02:02.0000] <fax_machine>
s/live/life/

[14:02:03.0000] <gsnedders>
she's so much older than me, though…

[14:02:04.0000] <fax_machine>
pfft, meaningless

[14:03:00.0000] <fax_machine>
it's only an issue if you make it one

[14:03:01.0000] <gsnedders>
if I overheard what her friends were saying once in another class, she's over two years older than me

[14:03:02.0000] <fax_machine>
how old are you?

[14:03:03.0000] <gsnedders>
14, 15 in 2 weeks

[14:03:04.0000] <fax_machine>
heh :)

[14:03:05.0000] <gsnedders>
she's already 17, since Mar

[14:03:06.0000] <fax_machine>
My wife and I met in high school

[14:04:00.0000] <fax_machine>
and when I was 14, she was 17 (but I turned 15 shortly after that)

[14:04:01.0000] <fax_machine>
we're 2 years and 2 months apart in age

[14:04:02.0000] <gsnedders>
I've been moved up a year, her, from what I heard, down one (or maybe she started a year late)

[14:04:03.0000] <fax_machine>
now I'm 27 and she's 29 and we're as happy as anyone could ever be :)

[14:05:00.0000] <fax_machine>
when you're 14/15, 2 years can feel like a really big deal

[14:05:01.0000] <gsnedders>
/me has been ill for a quarter  of his life, missed years of school (literally), and still manages to be moved up a year and do very very well

[14:05:02.0000] <fax_machine>
but trust me, with every year that passes, it becomes less significant

[14:05:03.0000] <fax_machine>
gsnedders: that's awesome

[14:05:04.0000] <fax_machine>
you've definitely got something great going on there

[14:05:05.0000] <othermaciej>
the HTML WG is continuing to get less fun

[14:06:00.0000] <gsnedders>
/me has been in web dev since age… 11 (and I mean by that actually hand coding)

[14:06:01.0000] <fax_machine>
now, get the girl and live happily ever after :)

[14:06:02.0000] <gsnedders>
othermaciej: totally.

[14:06:03.0000] <gsnedders>
othermaciej: codec arguments, questions about <canvas>… ugh.

[14:06:04.0000] <fax_machine>
sounds like #web

[14:07:00.0000] <gsnedders>
just on a far more technical level

[14:07:01.0000] <Dashiva>
othermaciej: Is it reality setting in? 

[14:07:02.0000] <othermaciej>
Dashiva: it's more of the parties who do not share the WHATWG rough consensus coming to the table

[14:08:00.0000] <Dashiva>
I also feel there's an air of "we're at the w3c now, and we do things this way" in some of the dialogue

[14:08:01.0000] <gsnedders>
I think the questions about <canvas> are pointless. It already has three stable implementations.

[14:09:00.0000] <Hixie>
haven't got to those yet

[14:09:01.0000] <Hixie>
i'm still trying to deal with the versioning issue

[14:09:02.0000] <Philip`>
There are at least two more (Rhino Canvas and KHTML), though I have no idea whether they count as stable

[14:09:03.0000] <Hixie>
ex-canvas, too

[14:10:00.0000] <Philip`>
Oh, that too

[14:11:00.0000] <gsnedders>
fax_machine: pm?

[14:11:01.0000] <fax_machine>
gsnedders: sure, go for it

[14:12:00.0000] <Philip`>
though it doesn't seem fair to call it stable either, since most features are buggy or missing (and presumably impossible with the limits of VML)

[14:19:00.0000] <Philip`>
Given the lack of extreme enthusiasm for <canvas>, it would seem sensible to avoid bringing up the issue of an OpenGL canvas in the HTML WG - that's probably best left to implementors and the WHATWG for now :-)

[14:20:00.0000] <Hixie>
opengl canvas is only missing one thing

[14:20:01.0000] <Hixie>
someone who knows opengl, who knows how to write neat simple js apis, and who can write specs

[14:21:00.0000] <othermaciej>
I am not sure such a mythical person exists

[14:22:00.0000] <Hixie>
well i can do the last one, and i'm not too bad at the middle one when i have people reviewing the apis to point out improvements, the first one is my problem. failing all else, i'll have to go on a 3d/opengl crash course.

[14:23:00.0000] <Philip`>
At least all the complicated graphical stuff can be left in the OpenGL spec - probably the only place where the 3D canvas interacts with actual graphics is when loading textures, and the rest is a wrapper API

[14:23:01.0000] <Philip`>
(unlike the 2D canvas, where the rendering has to be defined)

[14:24:00.0000] <othermaciej>
OpenGL is not a very friendly API to use directly

[14:24:01.0000] <othermaciej>
so a very literal-mided wrapper would probably not work so well

[14:25:00.0000] <Philip`>
Would it be sensible to leave user-friendly APIs to other non-standards people, where they can just implement wrappers in JavaScript and users can decide which one works best?

[14:28:00.0000] <Philip`>
(though I think the canvas should at least provide a way to efficiently load arrays of numbers into GL vertex-arrays, since that was the slowest bit when I tried loading models into the Firefox canvas3d extension some time ago - so that's probably the kind of place where the design gets more complex)

[14:32:00.0000] <hsivonen>
I guess it is hard to get people to agree on anything higher-level than OpenGL

[14:32:01.0000] <Hixie>
nah

[14:33:00.0000] <hsivonen>
Considering that the 2D abstraction is basically filling and stroking bézier paths, I don't think OpenGL is too low-level

[14:33:01.0000] <Hixie>
who runs the forums, anyone remember?

[14:33:02.0000] <Hixie>
i have a support request from someone

[14:33:03.0000] <Hixie>
was it zcorpan?

[14:33:04.0000] <hsivonen>
Hixie: it's zcorpan and he left

[14:33:05.0000] <hsivonen>
(for the weekend, that is)

[14:33:06.0000] <Hixie>
k

[14:34:00.0000] <Hixie>
i'll forward it to him

[14:34:01.0000] <hsivonen>
fax_machine: Regarding the conformance checker question last night: No, the conformance checker is not helpful in its current version. I do have plans on how to make it more helpful, though.

[14:34:02.0000] <fax_machine>
hsivonen: sweet

[14:35:00.0000] <othermaciej>
hsivonen: I don't think the basic ideas are too low-level, but some of the details of OpenGL APIs are

[14:35:01.0000] <othermaciej>
dunno, it's been a couple of years since I did any OpenGL programming

[14:36:00.0000] <hsivonen>
I haven't done OpenGL programming since 2002

[14:36:01.0000] <hsivonen>
I'm assuming that whatever was available on SGI O2 then is now available in OpenGL ES :-)

[14:39:00.0000] <othermaciej>
hey KevinMarks 

[14:41:00.0000] <Philip`>
OpenGL ES has been stripped of lots of the old OpenGL stuff - even basic things like immediate-mode Begin/Vertex/End and display lists - so it's probably not quite like old OpenGL programmers would remember :-)

[14:43:00.0000] <KevinMarks>
hi om

[14:47:00.0000] <hsivonen>
Philip`: ok. I haven't really taken a close look at OpenGL ES.

[15:26:00.0000] <fax_machine>
sheesh

[15:26:01.0000] <fax_machine>
my cable company sucks


2007-04-07
[06:17:00.0000] <gsnedders>
http://geoffers.uni.cc/archives/2007/04/07/why-xhtml-as-texthtml-is-evil/ – anyone want to check I haven't made some really stupid mistake in that?

[06:21:00.0000] <hsivonen>
s/as HTML is a SGML application/as HTML 4 purports to be an SGML application/

[06:22:00.0000] <hsivonen>
"This is rendered as such in a browser that the first <img> and the final “>” is displayed." umm. not in the real non-SGML world

[06:23:00.0000] <hsivonen>
s/reference engineering/reverse engineering/

[06:23:01.0000] <hsivonen>
"a few UAs that still use SGML parsers, such as feed readers implementing autodiscovery" Huh?

[06:24:00.0000] <Dashiva>
Might want to point a reference on that one

[06:24:01.0000] <ajnewbold>
s/(*.)/xhtml sucks/

[06:24:02.0000] <hsivonen>
"thereby not being a delta specification, based upon SGML" the comma seems to give the sentence exactly the wrong meaning

[06:24:03.0000] <Dashiva>
I thought the / in the tags would appear as text, but that could be my mistake

[06:24:04.0000] <Philip`>
s/\(\*\.\)/(.*)/

[06:24:05.0000] <ajnewbold>
Philip`: drat

[06:24:06.0000] <ajnewbold>
I fail :)

[06:25:00.0000] <moeffju[Headway]>
Dashiva, no, the slash is either an invalid attribute name or a NET-enabling marker

[06:25:01.0000] <hsivonen>
typo: guaranty

[06:25:02.0000] <ajnewbold>
hsivonen: he's british

[06:25:03.0000] <ajnewbold>
so it's ok

[06:25:04.0000] <ajnewbold>
I, too, prefer "guarantee"

[06:26:00.0000] <hsivonen>
gsnedders: well-formed as a concept does not apply to HTML

[06:26:01.0000] <hsivonen>
ajnewbold: ok. I wasn't familiar with the alternative spelling

[06:28:00.0000] <ajnewbold>
I wish he found an excuse to use "bloody" in there

[06:28:01.0000] <gsnedders>
maybe I wrote it a bit too quickly :P

[06:29:00.0000] <ajnewbold>
I also wish Safari would properly match the background-color to the color used in the background image of the main section :(

[06:30:00.0000] <gsnedders>
ajnewbold: I never got around to doing anything about that

[06:30:01.0000] <hsivonen>
my crystal ball tells me you are talking about this: http://hsivonen.iki.fi/png-gamma/

[06:30:02.0000] <ajnewbold>
yeah, that's the issue

[06:30:03.0000] <Dashiva>
I sometimes fear that if too many people go writing "xhtml sucks" articles, the weaker ones will be used as an excuse to derail the real issue

[06:31:00.0000] <ajnewbold>
Dashiva: yeah, but it would sure save on bandwidth

[06:31:01.0000] <ajnewbold>
;P

[06:32:00.0000] <gsnedders>
hsivonen: surely the "this is rendered" in that context refers to the DOM tree?

[06:32:01.0000] <Philip`>
The background colours don't match in Opera either, at least if you zoom in/out

[06:32:02.0000] <hsivonen>
gsnedders: well, you might want to emphasize that it is a hypothetical situation

[06:32:03.0000] <hsivonen>
gsnedders: you do mention browsers

[06:35:00.0000] <gsnedders>
hsivonen: "How then do real world browsers create the same DOM tree in both parsing modes?…" doesn't that paragraph cover that?

[06:35:01.0000] <hsivonen>
does anyone remember off-hand how <legend>foo <ul><li>bar</li><ul></legend> parses as HTML?

[06:36:00.0000] <hsivonen>
gsnedders: yes, but I considering how easy it is to start bogus rumors, I suggest emphasizing the hypothetical nature of the example

[06:41:00.0000] <Dashiva>
Heh

[06:42:00.0000] <Dashiva>
Someone is blasting anne's mail about comments because "they are not compatible with current or previous UAs"

[06:47:00.0000] <gsnedders>
hsivonen: emphasising it how?

[06:49:00.0000] <gsnedders>
hsivonen: "(thereby in the real world making it irrelevant)"?

[06:53:00.0000] <hsivonen>
"if brewsers used real SGML parsers" or something to that effect 

[06:53:01.0000] <hsivonen>
browsers

[08:44:00.0000] <annevk>
hsivonen, that becomes html > [head, body > legend > ["foo", ul > [li > "bar", ul]]]

[08:44:01.0000] <annevk>
if you meant to write </ul> the syntax would match the tree

[08:45:00.0000] <annevk>
(per html5lib anyway)

[08:54:00.0000] <hsivonen>
annevk: excellent. thanks

[08:54:01.0000] <hsivonen>
annevk: the result means that struct-inline works in <legend> as text/html

[08:55:00.0000] <hsivonen>
annevk: and yes, I meant </ul>

[08:56:00.0000] <annevk>
someone should just make some simple applets and document <applet>

[08:58:00.0000] <annevk>
I thought of doing it, but I'm not sure I want to promote client-side Java usage 

[09:01:00.0000] <hsivonen>
annevk: I don't want to promote applet usage, either, but I don't want conformance checking to implicitly promote <object> contortions, either.

[09:01:01.0000] <webben>
What's our reason for wishing to discourage applet usage?

[09:02:00.0000] <webben>
s/applet/client-side Java/

[09:03:00.0000] <hsivonen>
webben: client-side Java is fine as far as I am concerned, but applets suck really badly

[09:03:01.0000] <webben>
i thought all client-side Java was applets?

[09:03:02.0000] <webben>
or do you mean <applet> sucks?

[09:04:00.0000] <hsivonen>
webben: there's Web Start and there are permanently installed apps

[09:04:01.0000] <hsivonen>
webben: I meant Java applet suck no matter how you embed them

[09:04:02.0000] <webben>
ah okay. i thought <applet> relied on Web Start

[09:05:00.0000] <hsivonen>
it doesn't

[09:05:01.0000] <hsivonen>
predates Web Start by several years

[09:05:02.0000] <webben>
I see.

[10:20:00.0000] <annevk>
hsivonen, right, <object> contortians?

[10:20:01.0000] <annevk>
<embed type=application/java code=foo.class>

[11:03:00.0000] <hsivonen>
annevk: the Sun <object> examples are rather ugly compared to <applet>

[12:58:00.0000] <annevk>
hsivonen, eventually one of those <object> cases will have to work cross browser

[13:05:00.0000] <annevk>
What's IPTC?

[13:05:01.0000] <hsivonen>
annevk: isn't that the pre-XMP way of putting metadata in bitmaps in Photoshop?

[13:05:02.0000] <annevk>
see http://www.w3.org/mid/4617ECB7.4050308⊙ac

[13:06:00.0000] <annevk>
International Press Telecommunications Counsel

[13:07:00.0000] <hsivonen>
seems to be the organization that defined the metadata field that Photoshop uses

[13:07:01.0000] <hsivonen>
so I wasn't totally off-mark


2007-04-08
[18:39:00.0000] <Dashiva>
Oh wow

[18:40:00.0000] <Dashiva>
Email suggesting to replace <canvas>fallback</canvas> with <img src="fallback.gif">

[18:42:00.0000] <Lachy>
Dashiva: it's best to just ignore Andrew's emails.  He rarely says anything sensible

[18:43:00.0000] <Dashiva>
... Isee

[18:47:00.0000] <ajnewbold>
woo! utterly useless:  http://newbold.name/gradient/?start=339966&end=CCFF99&height=300

[18:48:00.0000] <ajnewbold>
anyone know GD functions and want to help me turn that into something that makes an image?

[18:48:01.0000] <ajnewbold>
I'd ask in ##php but I'm banned, heh

[18:49:00.0000] <Dashiva>
That doesn't look very 300px to me

[18:49:01.0000] <ajnewbold>
hmm

[18:49:02.0000] <ajnewbold>
what browser?

[18:50:00.0000] <Dashiva>
IE and Opera

[18:50:01.0000] <ajnewbold>
hmm

[18:50:02.0000] <ajnewbold>
does it look higher or lower?

[18:50:03.0000] <Dashiva>
It's more like 3000px

[18:50:04.0000] <ajnewbold>
OH HOLY CRAP

[18:50:05.0000] <ajnewbold>
yeah

[18:50:06.0000] <ajnewbold>
I just looked at it in Opera

[18:51:00.0000] <ajnewbold>
now this is interesting

[18:51:01.0000] <ajnewbold>
it's a bunch of divs, each 1px high

[18:51:02.0000] <ajnewbold>
300 of them

[18:51:03.0000] <ajnewbold>
why would that stack up to more than 300px?

[18:51:04.0000] <ajnewbold>
it still looks smooth, interesting

[18:51:05.0000] <Dashiva>
Because they aren't actually 1px high

[18:51:06.0000] <ajnewbold>
oh

[18:51:07.0000] <ajnewbold>
ohhhhh

[18:51:08.0000] <ajnewbold>
nvm, I see what I did

[18:51:09.0000] <ajnewbold>
/me growls at himself

[18:52:00.0000] <ajnewbold>
Dashiva: try again now :)

[18:52:01.0000] <Dashiva>
Still 3000 in IE, but 300 in Opera

[18:54:00.0000] <ajnewbold>
hmmm

[18:54:01.0000] <ajnewbold>
Dashiva: I don't have IE :(

[18:54:02.0000] <ajnewbold>
any idea what IE's problem might be?

[18:54:03.0000] <ajnewbold>
is it freaking out that I don't have a doctype or something?

[18:55:00.0000] <ajnewbold>
or... <body> ... heh

[18:55:01.0000] <ajnewbold>
/me cleans things up

[18:57:00.0000] <Dashiva>
You might want to outsource height to a stylesheet, and the width is somewhat redundant

[18:58:00.0000] <Dashiva>
Okay, workie in IE

[18:59:00.0000] <ajnewbold>
Dashiva: heh, funny

[18:59:01.0000] <ajnewbold>
so all IE wanted was valid, well-formed source

[18:59:02.0000] <ajnewbold>
that's a new one

[18:59:03.0000] <Dashiva>
There's a reason they call it quirks mode

[18:59:04.0000] <ajnewbold>
ah

[19:00:00.0000] <ajnewbold>
well, what I'd like to do is generate an image file

[19:00:01.0000] <ajnewbold>
I'm sure lots of people out there just want simple gradients to use as backgrounds, but don't want to mess with graphics apps just to make them

[19:01:00.0000] <ajnewbold>
ah, nevermind

[19:01:01.0000] <ajnewbold>
http://calebegg.com/grad/   someone's already done it

[19:01:02.0000] <Dashiva>
Have the server run opera, firefox or safari, generate the gradient on a canvas, toDataUrl the canvas into a new window, "save as" that image, and serve it to the user

[19:02:00.0000] <ajnewbold>
canvas, hmm

[19:02:01.0000] <ajnewbold>
I really need to learn more about that

[19:02:02.0000] <Dashiva>
(The above is not to be taken as a serious suggestion)

[19:02:03.0000] <ajnewbold>
oh

[19:02:04.0000] <ajnewbold>
I take everything seriously

[19:03:00.0000] <ajnewbold>
it's a horrible personality flaw

[19:03:01.0000] <Dashiva>
There are many ways to do it, really

[02:19:00.0000] <met_>
http://vocamus.net/dave/?p=39

[02:21:00.0000] <annevk>
interesting

[02:21:01.0000] <Hixie>
hey

[02:21:02.0000] <annevk>
morning

[02:22:00.0000] <hsivonen>
morning

[02:22:01.0000] <met_>
good morning

[02:22:02.0000] <met_>
and the ussage in the example http://zenit.senecac.on.ca/~bts630_071g07/canvas/BargraphExample.xhtml

[02:22:03.0000] <met_>
altough not valid of course

[03:35:00.0000] <Lachy>
Hey Hixie

[03:50:00.0000] <Hixie>
hey

[04:20:00.0000] <annevk>
Does Chris Wilson mean that the CSS fixes they will be doing will solely apply to Really We Mean it this Time Standards Mode?

[04:21:00.0000] <annevk>
I'm not really sure I understand his example with floats as other browsers likely render those sites ok...

[04:24:00.0000] <hsivonen>
looks like Dreamhost is having an outage again

[04:24:01.0000] <annevk>
yeah

[04:25:00.0000] <annevk>
whatwg.org down, html5.org down

[04:25:01.0000] <annevk>
etc.

[04:28:00.0000] <hsivonen>
once the easter holidays are over, someone has to raise the issue of adopting HTML5 in the HTML WG. the current situation is unproductive

[04:29:00.0000] <annevk>
/me wonders how that will go

[04:29:01.0000] <hsivonen>
I have a draft email

[04:32:00.0000] <met_>
annevk, any way of voting? (dunno who can start voting in html wg)

[04:33:00.0000] <annevk>
I suppose voting could work

[04:33:01.0000] <webben>
would the main bar to that be Microsoft?

[04:33:02.0000] <webben>
or the xforms-lite folks?

[04:34:00.0000] <annevk>
I'm not sure what the position of either is

[04:34:01.0000] <annevk>
Doesn't really matter much at this point

[04:35:00.0000] <annevk>
It starts getting interesting once another draft + editor is being put forward

[04:35:01.0000] <annevk>
but I don't really see that happening

[04:35:02.0000] <webben>
how come? just general pessimism about the process?

[04:36:00.0000] <webben>
oh wait you mean another as in additional to whatwg's html5?

[04:36:01.0000] <annevk>
Yes

[04:44:00.0000] <hendry>
dreamhost are a tad unreliable :/

[05:21:00.0000] <Dashiva>
Considering even the expert group making up the WG can't get acronyms and initialisms right (or even agree what is right), why are we still chasing after the idea that average authors will?

[05:52:00.0000] <Lachy>
Hey Hixie, I finally started playing around with irssi.

[05:52:01.0000] <Lachy>
it's not as hard as I initially thought it would be, but there are still many things I need to learn about it.

[06:57:00.0000] <annevk>
<figure> should work inside <p>

[06:57:01.0000] <annevk>
much like <random> works

[08:20:00.0000] <hsivonen>
annevk: nice for HTML5. less nice for the schema handling of struct-inline and text/html...

[08:21:00.0000] <hsivonen>
annevk: I'll comment on the list. thanks

[09:03:00.0000] <annevk>
http://www.dreamhoststatus.com/2007/04/08/dns-issues-causing-possible-site-and-email-downtime/

[11:48:00.0000] <annevk>
http://www.quirksmode.org/blog/archives/2007/04/html_5.html

[16:47:00.0000] <Lachy>
the idea for the generic script data attr is interesting.

[16:48:00.0000] <Lachy>
I've used class="" in the past for such things

[16:48:01.0000] <Dashiva>
Yeah, and I predict that if role is ever made common, people will use role for it too

[16:49:00.0000] <Lachy>
If they did, it would be abuse

[16:49:01.0000] <Lachy>
but I doubt role will ever become successful

[16:50:00.0000] <Dashiva>
As long as the validator didn't complain, that's all that would matter

[16:50:01.0000] <Lachy>
I'd like to know more about the use cases ppk has in mind for it

[16:51:00.0000] <Lachy>
I'll ask him.

[16:51:01.0000] <Dashiva>
An example would be declarative form validation before wf2

[16:51:02.0000] <webben>
Lachy, Isn't role already becoming successful for a tech that hasn't even been standardized yet?

[16:51:03.0000] <webben>
(leaving aside the Q of whether role is well specced)

[16:52:00.0000] <Dashiva>
You could either make up attributes and face invalidation (scary) or use class

[16:52:01.0000] <Lachy>
only in a very limited market

[16:53:00.0000] <Lachy>
I haven't seen it used beyond a few experimental pages so far

[16:53:01.0000] <webben>
I think Dojo are either using it or planning on using it.

[16:53:02.0000] <Lachy>
what for?

[16:54:00.0000] <webben>
Lachy, WAI

[16:54:01.0000] <webben>
(ARIA support in FF and major screen readers)

[16:54:02.0000] <Lachy>
for what purpose?

[16:54:03.0000] <webben>
increased accessibility of weird widgets

[16:54:04.0000] <webben>
and (I suppose) live regions

[16:55:00.0000] <Lachy>
I don't understand what it's for.  HTML5 elements and form controls seem to cover everything role does, and more

[16:55:01.0000] <webben>
see e.g. http://dojotoolkit.org/node/549

[16:55:02.0000] <webben>
Lachy, I'm not sure HTML5 yet has a way of expressing the priority of live updates?

[16:56:00.0000] <webben>
Lachy, might as well ask what Dojo is for

[16:56:01.0000] <webben>
the important thing is that Dojo apps if built be made accessible, i think

[16:58:00.0000] <webben>
Lachy, I suspect at least some of this reflects authors' desire to style form widgets

[16:58:01.0000] <webben>
but I don't really know

[16:59:00.0000] <Lachy>
styled form controls are a problem that CSS will solve

[16:59:01.0000] <webben>
do we have HTML5 tree widgets?


2007-04-09
[17:00:00.0000] <webben>
Lachy, I thought CSS was abstaining from defining how UAs should render form widgets

[17:01:00.0000] <Lachy>
http://www.w3.org/TR/css3-ui/

[17:02:00.0000] <Lachy>
and also see Web Controls 1.0

[17:02:01.0000] <webben>
Is there anything to indicate that Safari is likely to implement those?

[17:03:00.0000] <webben>
(I single out Safari for being the greatest stickler for native-looking widgets)

[17:04:00.0000] <Lachy>
no idea

[17:06:00.0000] <Dashiva>
Safari looks like the OS extension IE is ;)

[17:07:00.0000] <webben>
because if not ... people are likely to continue to use divs and spans

[17:07:01.0000] <webben>
especially as those work in IE6-7

[17:07:02.0000] <Hixie>
safari does a lot of css3-ui already

[17:08:00.0000] <webben>
really? I'll have to try that out. (not that i particularly like styling widgets to look non-native, but oh well)

[17:16:00.0000] <Lachy>
http://www.quirksmode.org/blog/archives/2007/04/html_5.html (see comments 7 and 8)

[17:18:00.0000] <Lachy>
/me needs to blog about why xhtml2 is not the future

[17:59:00.0000] <deltab>
Lachy: it's the flying-car future

[01:37:00.0000] <annevk>
We could have a state= attribute on certain elements like <xbl:div state>

[01:37:01.0000] <annevk>
for scripts

[01:38:00.0000] <hsivonen>
would "private" conflict with ECMA reserved words?

[01:39:00.0000] <annevk>
iirc, yes

[01:40:00.0000] <annevk>
I'm not entirely convinced it's need though given XBL2

[01:42:00.0000] <hsivonen>
http://www.w3.org/2007/uwa/

[01:42:01.0000] <hsivonen>
how (in)applicable is Web Apps 1.0 when it comes to home monitoring and control, home entertainment, office

[01:42:02.0000] <hsivonen>
equipment, mobile and automotive

[01:42:03.0000] <hsivonen>
?

[01:43:00.0000] <annevk>
I think that group is about putting device specific APIs into a generic framwork or something

[01:43:01.0000] <annevk>
/me isn't sure how that helps anyone

[01:57:00.0000] <hsivonen>
made a lot of progress: http://hsivonen.iki.fi/validator/html5/?doc=http%3A%2F%2Fsimon.html5.org%2Ftemp%2Fw3c-home-in-html5.html

[02:26:00.0000] <Hixie>
hsivonen: what are we looking for, progress wise?

[02:27:00.0000] <hsivonen>
Hixie: compared to the time when zcorpan pasted that URL here, there are a lot less incorrect error messages

[02:27:01.0000] <Hixie>
ah cool!

[02:27:02.0000] <hsivonen>
Hixie: I updated the schema layer

[02:42:00.0000] <MikeSmith>
hsivonen, annevk - the UWA WG is basically the Device Independece WG, renamed

[02:42:01.0000] <MikeSmith>
as far as what part of it might be applicable ... 

[02:43:00.0000] <MikeSmith>
http://www.w3.org/2006/10/uwa-charter.html#dci

[02:43:01.0000] <MikeSmith>
http://www.w3.org/2006/10/uwa-charter.html#device-coordination

[02:44:00.0000] <annevk>
I didn't care much for their specs either

[02:47:00.0000] <MikeSmith>
annevk - some criticize them as being overengineered

[02:47:01.0000] <annevk>
yeah, though that goes for most of the W3C :)

[02:48:00.0000] <hsivonen>
http://hsivonen.iki.fi/validator/html5/?doc=http%3A%2F%2Fsimon.html5.org%2Ftemp%2Fw3c-home-in-html5.html

[02:48:01.0000] <hsivonen>
yay

[02:50:00.0000] <annevk>
the "conforming 

[02:50:01.0000] <annevk>
message should be clearer

[02:51:00.0000] <MikeSmith>
annevk - perhaps so (about W3C). and perhaps more people are learning that or have learned it -- the hard way, by not seeing their specs get implemented more widely

[02:51:01.0000] <hsivonen>
annevk: intentional weasel words at this point :-)

[02:51:02.0000] <hsivonen>
annevk: or, rather, intentional in the parentheses

[02:52:00.0000] <hsivonen>
annevk: what would you suggest before the parentheses?

[02:53:00.0000] <Hixie>
i love the "valid html5" icon he made

[02:53:01.0000] <Hixie>
that's awesome

[02:53:02.0000] <annevk>
"Congratulations, your document is conforming HTML5"

[02:53:03.0000] <annevk>
or something

[02:53:04.0000] <annevk>
with some !!11! following it...

[02:57:00.0000] <annevk>
hsivonen, http://hsivonen.iki.fi/validator/html5/?doc=http%3A%2F%2Fannevankesteren.nl%2F I don't get the "bad value for attribute href" messages...

[02:57:01.0000] <annevk>
hsivonen, is that because my href has http:// at the start and @ somewhere in it?

[03:01:00.0000] <hsivonen>
annevk: the line numbers are weird. do you mix Unix and Windows line breaks?

[03:01:01.0000] <annevk>
maybe, dunno

[03:02:00.0000] <annevk>
I'm including some standalone files that may or may not have UNIX line endings...

[03:02:01.0000] <hsivonen>
anyway, TextWrangler and my parser have a different notion of line numbers

[03:03:00.0000] <hsivonen>
but yeah, chances are that @ is the problem

[03:03:01.0000] <annevk>
(use it as a testcase)

[03:03:02.0000] <annevk>
per HTML5 line endings shouldn't matter

[03:04:00.0000] <hsivonen>
either @ isn't allowed unescaped in HTTP IRIs or the IRI library has a bug in it

[03:05:00.0000] <hsivonen>
annevk: my parser should decide the nature of a line break on a line-by-line basis, but TextWrangler sniffs the top and assumes the same kind of line breaks throughout the file

[03:08:00.0000] <annevk>
time to update HTTP IRIs then :)

[03:08:01.0000] <hsivonen>
annevk: did you already figure out if it is a library bug or an RFC bug?

[03:09:00.0000] <annevk>
oh no

[03:10:00.0000] <hsivonen>
annevk: can't be just @. you have more than two instances of such IRIs

[03:11:00.0000] <annevk>
ipchar seems to allow a literal @

[03:11:01.0000] <annevk>
what are the IRIs that trigger it?

[03:12:00.0000] <hsivonen>
annevk: I don't know, because the line numbers are all weird

[03:12:01.0000] <annevk>
the line numbers from the validator are correct?

[03:13:00.0000] <annevk>
ah, it's "irc://irc.w3.org:80/html-wg"

[03:13:01.0000] <hsivonen>
annevk: hopefully. :-)

[03:13:02.0000] <hsivonen>
now, *that's* weird

[03:13:03.0000] <annevk>
and "irc://irc.freenode.net/whatwg"

[03:14:00.0000] <hsivonen>
the library should apply generic IRI processing to irc: IRIs

[03:14:01.0000] <hsivonen>
and obviously those should be fine as far as the generic processing goes

[03:14:02.0000] <hsivonen>
irc on port 80?

[03:14:03.0000] <annevk>
yeah

[03:15:00.0000] <hsivonen>
firewall thing?

[03:15:01.0000] <Hixie>
and people tell _me_ off for misusing port 80. pah.

[03:15:02.0000] <annevk>
to work around certain firewalls that block on port rather than functionality

[03:16:00.0000] <hsivonen>
annevk: added to my todo list

[03:16:01.0000] <hsivonen>
annevk: thanks

[03:18:00.0000] <annevk>
it also complains about accesskeys and my doctype which apparently triggers quirks in an unused browser 

[03:18:01.0000] <hsivonen>
/me expects a bug report email when the first person hits the transparent content model of <video> inside <figure>

[03:18:02.0000] <hsivonen>
annevk: are accesskeys in HTML5 already?

[03:19:00.0000] <Hixie>
hsivonen: yeah that's a bug

[03:19:01.0000] <Hixie>
in the spec

[03:19:02.0000] <annevk>
I suppose they aren't

[03:19:03.0000] <Hixie>
need to rework how <figure> works to allow <ol>s, though, so i'll fix it later

[03:19:04.0000] <annevk>
and <svg:svg>

[03:19:05.0000] <Hixie>
accesskeys aren't in because nobody has yet explained how to fix them

[03:20:00.0000] <hsivonen>
Hixie: <ol>s? 

[03:20:01.0000] <Hixie>
and <ul>s, and <pre>, and whatever else people have argued should be figurable

[03:20:02.0000] <Hixie>
assuming we want to do that

[03:20:03.0000] <hsivonen>
oh

[03:22:00.0000] <annevk>
would <figure> <legend/> <object> <p>test </p> <p> test </p> </object> </figure> be ok?

[03:22:01.0000] <Hixie>
who are you asking?

[03:23:00.0000] <hsivonen>
annevk: IIRC, yes, provided that you put a required attribute on <object>

[03:23:01.0000] <annevk>
k

[03:24:00.0000] <hsivonen>
annevk: also, I haven't updated the significant inline stuff to deal with legend

[03:26:00.0000] <MikeSmith>
Hixie - what's broken about accesskey?

[03:28:00.0000] <Hixie>
what's not broken

[03:28:01.0000] <Hixie>
it has never been implemented in a usable way

[03:28:02.0000] <Hixie>
it isn't device independent

[03:28:03.0000] <Hixie>
it isn't i18n-compatible

[03:28:04.0000] <Hixie>
it isn't well specified

[03:29:00.0000] <annevk>
chaals claims Opera now has a somewhat useful impl

[03:29:01.0000] <Hixie>
shipped?

[03:29:02.0000] <Hixie>
does he mean the modal toggle?

[03:29:03.0000] <annevk>
yeah, I believe we expose the access keys of a page in a sidebar the user can invoke in some way

[03:29:04.0000] <annevk>
maybe

[03:29:05.0000] <Hixie>
steps to reproduce?

[03:30:00.0000] <annevk>
maybe later, haven't played with it myself

[03:32:00.0000] <Hixie>
well in opera 9.00 it isn't usable

[03:34:00.0000] <annevk>
ah, Shift+Esc

[03:34:01.0000] <annevk>
try that on my homepage for instance

[03:35:00.0000] <MikeSmith>
Hixie - I think it's implemented is some mobile browsers, including Openwave's browser, and very widely used in mobile-specific sites

[03:36:00.0000] <Hixie>
oh it's used

[03:36:01.0000] <Hixie>
but all uses suffer from the problems i described

[03:36:02.0000] <MikeSmith>
yeah, true

[03:36:03.0000] <Hixie>
it's not too bad in walled-garden device-specific scenarios

[03:36:04.0000] <Hixie>
but of course that's hardly what we care for here

[03:36:05.0000] <Hixie>
annevk: again, not usable

[03:37:00.0000] <Hixie>
anyway

[03:37:01.0000] <Hixie>
bed time

[03:42:00.0000] <annevk>
g'night

[03:53:00.0000] <MikeSmith>
unfortunate fact about accesskey as far as mobile users go is that because it is something that's so widely used in mobile sites, if a content provider wants to move his site from being a walled garden/WAP site to being a Web site, lack of accesskey support is basically a regression in application behavior as far as they are concerned

[03:53:01.0000] <MikeSmith>
or as far as users of their site will be

[03:54:00.0000] <MikeSmith>
that's true for inputmode too (which most non-WAP browsers don't support)

[03:57:00.0000] <MikeSmith>
at least here in Japan, sites use accesskey a lot, and users expect it on sites

[03:58:00.0000] <MikeSmith>
Konqueror has accesskey support

[03:59:00.0000] <MikeSmith>
if you press and release Ctrl, it shows tooltip text over each link, with letter of corresponding access key

[04:01:00.0000] <MikeSmith>
I think if a site doesn't provide accesskey info in the markup, Konq actually generates accesskeys for the page

[04:02:00.0000] <hasather>
MikeSmith: yea, that seems to be the case

[04:03:00.0000] <hasather>
wow, that's cool

[04:09:00.0000] <MikeSmith>
Konqueror has a lot nice features

[09:00:00.0000] <annevk>
zcorpan, <map> no longer has name=

[11:26:00.0000] <Philip`>
Since someone mentioned setTimeout: Mozilla passes the timeout handlers a ""secret" final argument that indicates timeout lateness in milliseconds". Is that already known, and is it at all relevant/interesting to anything, like whether it contradicts the spec and should be removed, or whether it should be adopted by the spec, or whether it's fine if nothing changes?

[11:34:00.0000] <othermaciej>
that's pretty weird

[11:39:00.0000] <deltab>
it's not mentioned on MDC

[11:44:00.0000] <Philip`>
http://lxr.mozilla.org/mozilla/source/dom/src/base/nsGlobalWindow.cpp#6818 is the only place I've seen it mentioned

[11:46:00.0000] <gavin_>
I'm sure you'll be interested to know that that "secret argument" was in the original revision of nsGlobalWindow, 1.1 <kipp⊙nc> 1998-07-15 18:16

[11:47:00.0000] <gavin_>
good luck on finding the reason why it was added! :)

[11:47:01.0000] <othermaciej>
does anything actually use it?

[11:48:00.0000] <othermaciej>
seems like checking the current time will almost always be better than a lateness value

[11:48:01.0000] <gavin_>
per rule #432, someone surely must :)

[11:49:00.0000] <gavin_>
it's relatively well hidden, though, so maybe we could remove it

[11:50:00.0000] <gavin_>
I'm not sure it's very useful

[11:52:00.0000] <deltab>
oh, it is mentioned on MDC, on a talk page: http://developer.mozilla.org/en/docs/Talk:DOM:window.setTimeout

[11:53:00.0000] <gavin_>
https://bugzilla.mozilla.org/show_bug.cgi?id=10637 has some background info

[11:54:00.0000] <gavin_>
"It was very useful to *us* at Netscape in performance testing of animation

[11:54:01.0000] <gavin_>
using intervals, since a script could detect when the CPU became saturated, i.e.

[11:54:02.0000] <gavin_>
the timeout lateness was increasing monotonically with each timeout."

[11:55:00.0000] <Dashiva>
A feature nobody knows about that was last used in NS4?

[11:56:00.0000] <deltab>
and that complicates bindings to non-JS languages

[12:24:00.0000] <Philip`>
According to bug 263945 it is "an underdocumented feature", which seems like an understatement since the only documentation is the source code and the comments on two bug reports

[14:32:00.0000] <hsivonen>
http://wiki.whatwg.org/wiki/Video_type_parameters in case anyone cares to continue

[14:33:00.0000] <Hixie>
cool, thanks

[14:34:00.0000] <annevk>
ouch

[14:34:01.0000] <annevk>
looks amazingly complicated

[14:35:00.0000] <Hixie>
welcome to video

[14:35:01.0000] <annevk>
when I was welcomed there was <video>, three methods and a couple of .ogg files :)

[14:35:02.0000] <Hixie>
hehe

[14:37:00.0000] <hsivonen>
frankly, I think the codec RFC is way too hard for authors. if the codecs parameters is supposed to work, we need sniffer tools that take a file and print out the appropriate MIME type with the codecs parameter

[14:38:00.0000] <hsivonen>
annevk: and I'm only covering sane and contemporary container/codec combos...

[14:38:01.0000] <annevk>
hopefully we just have <video src=blah> for the majority of cases

[14:39:00.0000] <annevk>
maybe the wiki page should document the signature of the files as well?

[14:39:01.0000] <hsivonen>
annevk: feel free to edit. I'm never fully grokked how MP4-derived magic numbers are supposed to work

[14:40:00.0000] <Hixie>
i'm not a fan of the rfc either

[14:40:01.0000] <Hixie>
but do we have something better?

[14:41:00.0000] <annevk>
RFC42815

[14:41:01.0000] <Hixie>
?

[14:41:02.0000] <hsivonen>
Hixie: nothing better that all the relevant vendors would agree to, no

[14:41:03.0000] <annevk>
Hixie, last digit

[14:41:04.0000] <Hixie>
what do you have that's better which vendors wouldn't agree to?

[14:41:05.0000] <Hixie>
annevk: ?

[14:41:06.0000] <annevk>
Hixie, oh, and it's a joke

[14:42:00.0000] <hsivonen>
Hixie: mandated Ogg :-)

[14:42:01.0000] <annevk>
and  RFC4281 is the current RFC

[14:42:02.0000] <Hixie>
annevk: oh i see

[14:42:03.0000] <Hixie>
annevk: well even for that, we'd have to have something better to put _in_ that spec

[14:42:04.0000] <annevk>
fair enough

[14:42:05.0000] <Hixie>
hsivonen: no i mean for codec description. even if we mandate codec support, we can't disallow other codecs.

[14:42:06.0000] <Hixie>
hsivonen: so ogg doesn't solve this problem.

[14:43:00.0000] <annevk>
and we want authors to be able to provide multiple versions?

[14:43:01.0000] <hsivonen>
Hixie: I know. but having one true codec would. but not gonna happen

[14:43:02.0000] <Hixie>
christ, people in the csswg again trying to open new issues in things that aren't problems.

[14:44:00.0000] <Hixie>
hsivonen: one true codec would not ever be a solution. there's no one codec that solves all problems. e.g. a codec that's good for screencaps isn't going to be good for slow moving video.

[14:44:01.0000] <Hixie>
if there's a solution that really is better than 4281, i'm open to it

[14:45:00.0000] <annevk>
so the answer to my question is yes?

[14:45:01.0000] <Hixie>
yes

[14:45:02.0000] <othermaciej>
hsivonen: now that I know more about it, it does sound way too hard

[14:45:03.0000] <Hixie>
we'll be adding a media="" attribute to <source> anyway

[14:45:04.0000] <Hixie>
which takes a media query

[14:45:05.0000] <hsivonen>
Hixie: well, killing all the ISO profiles and using plain FourCCs in as codec identifiers would make things easier. but not gonna happen, either

[14:46:00.0000] <hsivonen>
s/in as/as/

[14:46:01.0000] <Hixie>
hsivonen: why wouldn't it happen? (i have no idea what FourCC is)

[14:46:02.0000] <annevk>
Hixie, based on the beforeprint discussion?

[14:46:03.0000] <othermaciej>
hsivonen: the codecs need to have comprehensible names

[14:46:04.0000] <Hixie>
annevk: ? no

[14:46:05.0000] <othermaciej>
FourCC is a four-character code

[14:46:06.0000] <annevk>
Hixie, a use case for that came up there

[14:46:07.0000] <Hixie>
annevk: onbeforeprint will be added in due course.

[14:46:08.0000] <hsivonen>
Hixie: because ISO has gone profile-crazy. too late to undo it, I think

[14:46:09.0000] <Hixie>
hsivonen: ah

[14:46:10.0000] <Hixie>
hsivonen: well if you have any solutions that actually would work, let me know :-)

[14:47:00.0000] <hsivonen>
Hixie: a FourCC is a four-character identifier that you put in a container to identify the type of a data stream

[14:47:01.0000] <annevk>
actually, if media="" is added to source you might want to add start end, etc. to <source> as well

[14:47:02.0000] <annevk>
so for non interactive media you can start at a specific location

[14:47:03.0000] <Hixie>
annevk: we're not trying to encourage different media being seent, if they have different timelines that's bad

[14:47:04.0000] <hsivonen>
Hixie: so FourCC is kind of like an intra-file magic or HFS type code

[14:48:00.0000] <annevk>
or show a specific frame, so to say (which is the use case that came up)

[14:48:01.0000] <Hixie>
annevk: oh you're misunderstanding. this wouldn't be for screen vs print. it would be for highres vs lowres, or b&w vs color, etc

[14:48:02.0000] <Hixie>
hsivonen: ah

[14:48:03.0000] <annevk>
Hixie, k

[14:49:00.0000] <annevk>
/me goes to read the RFC

[14:49:01.0000] <hsivonen>
If I had the power to fix MPEG-4, I'd have two video codecs: Visual Simple Profile without any levels or subprofiles and H.264 without any levels or subprofiles

[14:50:00.0000] <Hixie>
hsivonen: so would you also introduce a new version every year as hardware improved?

[14:51:00.0000] <hsivonen>
Hixie: At least I'd give the new version a different marketing name and FourCC instead of defining it as an Advanced Extended High 10:2:2 profile level 5.2 of something pre-existing

[14:52:00.0000] <Hixie>
fair enough

[14:52:01.0000] <Hixie>
so we'd have H.264, H.265, H.266, etc?

[14:52:02.0000] <hsivonen>
Hixie: for example

[14:52:03.0000] <Hixie>
makes sense

[14:52:04.0000] <Hixie>
oh well

[14:52:05.0000] <Hixie>
sadly we are not the ISO

[14:54:00.0000] <gsnedders>
MPEG-5 anyone? :P

[14:54:01.0000] <Hixie>
that's one that i will definitely NOT be ever doing

[14:55:00.0000] <gsnedders>
/me has absolutely no clue about the technical details of codecs

[14:56:00.0000] <annevk>
at some point bandwidth will be less of an issue and all the linked files can simply be inspected...

[14:57:00.0000] <Hixie>
i doubt that point will come any time soon

[14:57:01.0000] <Hixie>
at least not in the next two or three decades

[14:59:00.0000] <annevk>
isn't the required data not in the first few KiBs?

[15:00:00.0000] <Hixie>
now scale that up by a billion and imagine what it would do to a site like youtube.

[15:01:00.0000] <Philip`>
Would latency still be an issue, even if the bandwidth is insignificant?

[15:01:01.0000] <Hixie>
that too

[15:02:00.0000] <hsivonen>
what annevk said is more likely to work, though, than requiring authors to write codec identifiers

[15:02:01.0000] <Philip`>
(unless you had a server-side script that quickly inspected all the files and then told the browser what they all were)

[15:03:00.0000] <hsivonen>
I wouldn't dismiss magic-based sniffing and HTTP request closing as crazy when type parameters aren't available

[15:03:01.0000] <hsivonen>
YouTube has staff to figure out RFCs

[15:03:02.0000] <Hixie>
i don't think this will be used much on sites that don't have staff.

[15:03:03.0000] <Hixie>
if you don't have staff, you'll likely just use one file.

[15:04:00.0000] <hsivonen>
Hixie: that's true

[15:04:01.0000] <hsivonen>
though I don't have staff and I've provided dual videos: Theora and H.264

[15:05:00.0000] <Hixie>
that's an easy one

[15:05:01.0000] <Hixie>
don't even need codec="" for that

[15:05:02.0000] <hsivonen>
right

[15:05:03.0000] <hsivonen>
actually, you do

[15:05:04.0000] <Hixie>
why? just stick ogg first

[15:05:05.0000] <hsivonen>
what if Ogg contains Dirac and the UA supports Theora and H.264?

[15:06:00.0000] <Hixie>
nobody supports dirac. it's not even done yet.

[15:06:01.0000] <hsivonen>
or, rather, how's the UA going to know that the Ogg file contains Theora, not Theora

[15:06:02.0000] <Hixie>
and why would you put dirac in an ogg container?

[15:06:03.0000] <annevk>
because authors are silly?

[15:06:04.0000] <hsivonen>
Hixie: I thought the plan was to put Dirac in Ogg

[15:06:05.0000] <Hixie>
oh.

[15:07:00.0000] <Hixie>
well then you use the codec value

[15:07:01.0000] <Philip`>
http://wiki.xiph.org/index.php/OggDirac

[15:07:02.0000] <ericcarlson>
I don't think this will really be such a big issue for authors

[15:07:03.0000] <hsivonen>
Hixie: does that mean that you are going to define default codecs for containers?

[15:07:04.0000] <annevk>
can't you have other things inside Ogg as well?

[15:07:05.0000] <Hixie>
i'm open to better schemes, but requiring the browser to sniff the data stream is simply not a workable solution.

[15:07:06.0000] <Hixie>
hsivonen: i'll probably put your wiki page in the spec in due course (or refer to it), but i'd rather just have a better solution.

[15:08:00.0000] <hsivonen>
ericcarlson: I think it will be. How do I even figure out what AVC level QuickTime made when I encode?

[15:09:00.0000] <ericcarlson>
People compressing media for the web export to a "profile" so they

[15:09:01.0000] <ericcarlson>
know a class of user/devide will be able receive and decode it

[15:09:02.0000] <annevk>
another problem with those codec= stuff is that authors will just copy and paste it and change the src= value and sometimes it will work and sometimes it won't (though arguably this is a minor issue)

[15:09:03.0000] <ericcarlson>
When exporting you choose the profile, and that defines the parameters passed to the encoder

[15:09:04.0000] <hsivonen>
ericcarlson: OK, if I check Baseline in QuickTime, how do I figure out what to put in the AVC level "xx" placeholder in Dave Singer's email?

[15:10:00.0000] <Hixie>
for the people arguing that the current type="" codec= RFC4281 solution is bad: i agree. giving more reasons why it's not an optimal solution is not required. What's required is a more optimal solution.

[15:10:01.0000] <ericcarlson>
and it defines the characteristics of the file that comes out the other end (bitrate, B-frames or not, size, etc)

[15:11:00.0000] <hsivonen>
Hixie: I'm not arguing why it is bad. I'm trying to discover useful data that authors will need when they use it.

[15:11:01.0000] <Hixie>
i was mostly talking to anne :-)

[15:11:02.0000] <Hixie>
/me would like to know the answer to your question too

[15:11:03.0000] <hsivonen>
ericcarlson: How do I map QuickTime export dialog values to the RFC indentifiers?

[15:15:00.0000] <ericcarlson>
hsivonen: which dialog values?

[15:15:01.0000] <hsivonen>
Movie to MPEG-4, Options...

[15:16:00.0000] <Hixie>
ericcarlson: if he checks Baseline in QuickTime, how can he figure out what to put in the AVC level "xx" placeholder in Dave Singer's email?

[15:16:01.0000] <hsivonen>
dialog titled MPEG-4 Export Settings

[15:17:00.0000] <hsivonen>
ericcarlson: there's no setting for AVC level

[15:17:01.0000] <ericcarlson>
That information has to come out of the encoded file, so someone (me ;-) will have to write a utility

[15:17:02.0000] <hsivonen>
ericcarlson: ok :-)

[15:17:03.0000] <Hixie>
ew

[15:18:00.0000] <hsivonen>
presumably Movie to iPod will lower the AVC level to a magic value (still without telling me)

[15:19:00.0000] <ericcarlson>
Yes, but the level is chosen for you based on the "dial-able" parameters.

[15:20:00.0000] <ericcarlson>
hsivonen: Yes, any preset that doesn't allow any parameter tweaking will be fixed.

[15:23:00.0000] <Hixie>
i'm about to check in a big set of media/video/audio changes, if anyone wants to give it a look-see while i go grab some food to see if there are any mistakes, i'll fix them before checking in

[15:23:01.0000] <Hixie>
page will be genned in about 60 seconds

[15:24:00.0000] <annevk>
some of my reported bugs haven't yet been fixed

[15:24:01.0000] <annevk>
no apparent visual bugs though

[15:29:00.0000] <Hixie>
oh?

[15:29:01.0000] <annevk>
HTMLSourceElement member names in the IDL

[15:29:02.0000] <annevk>
haven't checked others

[15:29:03.0000] <Hixie>
ooh, good catch

[15:30:00.0000] <annevk>
attribute float currentLoop -> unsigned long currentLoop

[15:30:01.0000] <Hixie>
cool

[15:30:02.0000] <Hixie>
any others?

[15:31:00.0000] <annevk>
not right away

[15:31:01.0000] <Hixie>
hm, i forgot to define seekable

[15:31:02.0000] <ericcarlson>
I don't see anything obvious

[15:32:00.0000] <annevk>
in that case, you also forgot to define some events such as volumechange

[15:32:01.0000] <ericcarlson>
Well, "position" still doesn't have the right name.

[15:32:02.0000] <Hixie>
yeah name changes and events are next

[15:32:03.0000] <ericcarlson>
OK

[15:32:04.0000] <annevk>
/me wonders what's wrong with position

[15:33:00.0000] <Hixie>
apple prefers currentTime, and nobody has argued anything else

[15:33:01.0000] <annevk>
currentTime is hard to type

[15:33:02.0000] <annevk>
/me misspelled it while typing it in, even!

[15:33:03.0000] <Hixie>
true, i often type it as currenTTime

[15:34:00.0000] <annevk>
I forgot the first t

[15:34:01.0000] <Hixie>
ericcarlson: what were the arguments for currentTime and against position, again?

[15:35:00.0000] <ericcarlson>
It is a temporal property, and  "position" doesn't  really have meaning in the time domain.

[15:35:01.0000] <othermaciej>
Hixie: "position" is more vague and sounds like it would be spacial, not temporal

[15:35:02.0000] <Hixie>
ah yes, there you go

[15:36:00.0000] <annevk>
I suppose "location" has the same issues?

[15:36:01.0000] <annevk>
maybe more

[15:36:02.0000] <othermaciej>
location sounds like it should be a URL

[15:36:03.0000] <Hixie>
even worse, yeah

[15:36:04.0000] <Hixie>
btw did we say that seeking to outside the current loop boundaries should clamp?

[15:37:00.0000] <ericcarlson>
I think it should

[15:37:01.0000] <ericcarlson>
For whatever that is worth.

[15:37:02.0000] <annevk>
and how about position -> time, currentLoop -> loop

[15:38:00.0000] <Hixie>
that might work

[15:38:01.0000] <ericcarlson>
"loop" sounds like a boolean

[15:40:00.0000] <ericcarlson>
The video element still fills with black

[15:41:00.0000] <Hixie>
i haven't done any of the changes from our lunch

[15:41:01.0000] <othermaciej>
time might be doable though it is a little more vague

[15:41:02.0000] <annevk>
what should it be? transparent black?

[15:41:03.0000] <Hixie>
annevk: should just use css background

[15:41:04.0000] <annevk>
makes sense

[15:42:00.0000] <Hixie>
though i think we should have video { background: black } in the UA stylesheet

[15:42:01.0000] <Hixie>
personally

[15:42:02.0000] <Hixie>
but that might just be me

[15:43:00.0000] <ericcarlson>
sounds good for a page with a black background :-)

[15:45:00.0000] <annevk>
I still think .position is a good pick as everybody will understand what it means and it's way easier to remember and type than .currentTime

[15:45:01.0000] <Philip`>
Is it possible for the video content to be (semi-)transparent, so you'd need that background colour behind the video rather than just in the areas that don't contain the video?

[15:45:02.0000] <Hixie>
jesus, just this update to use the new states rather than the old state system is a 2000 line patch

[15:46:00.0000] <Hixie>
Philip`: if you count animated gifs as video, i guess

[15:46:01.0000] <othermaciej>
Philip`: there are such things as codecs with alpha

[15:46:02.0000] <annevk>
Philip`, it's possible with Flash

[15:46:03.0000] <annevk>
prolly also with FLV

[15:47:00.0000] <annevk>
Hixie, I suppose it's less than 2000 for /source?

[15:47:01.0000] <Hixie>
no that was /source :-/

[15:47:02.0000] <Philip`>
Okay - I just saw the currently-online version talks about "Areas of the element's playback area that do not contain the video should be filled with pure black", which doesn't seem to cover that case, though I don't know if that's been changed already

[15:48:00.0000] <Hixie>
it's going to change

[15:48:01.0000] <Hixie>
but later

[15:48:02.0000] <Hixie>
:-)

[15:48:03.0000] <othermaciej>
compositing onto the (possibly transparent) background seems like the right thing for alpha codecs

[15:49:00.0000] <Hixie>
yeah

[15:49:01.0000] <annevk>
HTMLSourceElement is still not changed fwiw

[15:49:02.0000] <Hixie>
haven't regenned yet

[15:49:03.0000] <annevk>
(I say this because currentLoop is)

[15:49:04.0000] <Hixie>
oh

[15:49:05.0000] <Hixie>
wait, what?

[15:49:06.0000] <Hixie>
/me looks again

[15:49:07.0000] <Hixie>
oh, there was more broken than i thought

[15:49:08.0000] <Hixie>
oops

[15:50:00.0000] <Hixie>
oh heh

[15:50:01.0000] <Hixie>
i fixed it the wrong way -_-

[15:50:02.0000] <KevinMarks>
I've said for a while that Apple needs to invent a marketing name for h264

[15:50:03.0000] <KevinMarks>
they're good at that

[15:50:04.0000] <othermaciej>
iCodec

[15:50:05.0000] <othermaciej>
(or Final Codec Pro?)

[15:51:00.0000] <KevinMarks>
Pixlet Pro?

[15:51:01.0000] <annevk>
iVideo

[15:51:02.0000] <KevinMarks>
AVC/H264 is such a non-Apple name

[15:51:03.0000] <SimonW>
hi all... I'm putting together a "what the heck is HTML 5" talk for a local geek event here in Oxford... is there anywhere online that explains the implications of the Apple <canvas> patent e-mail?

[15:51:04.0000] <KevinMarks>
especially as the original QT codec from 1990 was called AVC

[15:51:05.0000] <SimonW>
or is it safe to just ignore that bit?

[15:52:00.0000] <annevk>
Some people explained it to me mean that when <canvas> goes to the W3C Apple will play nicely

[15:52:01.0000] <annevk>
But I'm no lawyer

[15:52:02.0000] <Hixie>
SimonW: i'd ignore it and then if you get a question about it just say that now the work is under the patent policy at w3c, there's no problem.

[15:53:00.0000] <SimonW>
cool, shall do

[15:53:01.0000] <Philip`>
http://tech.groups.yahoo.com/group/canvas-developers/message/371 has a comment

[16:20:00.0000] <Hixie>
right that's checked in

[16:20:01.0000] <annevk>
Hixie, I suppose you want a feature that turns that commit log into a <ul> or something?

[16:21:00.0000] <Hixie>
hah no

[16:21:01.0000] <Hixie>
i hope i never check in so much at once again

[16:22:00.0000] <Hixie>
oh wow it feels so much better to not have such a huge checkin pending

[16:35:00.0000] <annevk>
it should prolly be defined when an element is in a document

[16:35:01.0000] <annevk>
for <source>

[16:36:00.0000] <annevk>
another thing that seems to be lacking is requirements on what happens when a media element is inserted or removed

[16:36:01.0000] <Hixie>
it's there

[16:36:02.0000] <annevk>
<source> causes implicit loading, but <video src> doesn't seem to do so

[16:36:03.0000] <Hixie>
yes it does

[16:36:04.0000] <Hixie>
it's there somewhere

[16:37:00.0000] <Hixie>
i wrote it! i swear!

[16:37:01.0000] <Hixie>
:-)

[16:37:02.0000] <Hixie>
"in a document" is a problem i don't want to look at too closely yet

[16:37:03.0000] <Hixie>
"If a media element whose networkState has the value EMPTY is inserted into a document, user agents must implicitly invoke the load() method on the media element as soon as all other scripts have finished executing. Any exceptions raised must be ignored."

[16:38:00.0000] <annevk>
i see

[16:38:01.0000] <annevk>
cheers

[16:53:00.0000] <ianloic>
hey

[16:53:01.0000] <ianloic>
if we want to expose functionality in our user agent that doesn't match any existing APIs what should we do?

[16:53:02.0000] <ianloic>
right now we're planning on just exposing a global object

[16:54:00.0000] <Hixie>
depends on the api

[16:54:01.0000] <ianloic>
I'm trying to work out the Right way to expose events.

[16:54:02.0000] <Hixie>
whatever you do make sure to prefix the api names with the name of your vendor

[16:54:03.0000] <Hixie>
what's the api, if i may ask?

[16:54:04.0000] <ianloic>
well, we're a music player - initially it'll be play/pause/playURL and an interface to get the currently playing track

[16:55:00.0000] <ianloic>
later it'll be richer integration

[16:55:01.0000] <ianloic>
(this is in Songbird)

[16:55:02.0000] <Hixie>
sounds like something for Window, yeah. You probably want window.songbirdPlay(), or window.songbird.play(), etc al

[16:55:03.0000] <Hixie>
et al, even

[16:56:00.0000] <ianloic>
the goal is to allow web sites to implement something with the functionality of the iTunes music store, or Pandora against your own library or applications like that

[16:56:01.0000] <ianloic>
we'd ideally like to produce something that could be adopted by other vendors. I'm a little hesitant to push our product name into the API

[16:57:00.0000] <Hixie>
(using your vendor name is important to prevent future official apis from clashing with you, and giving you problems)

[16:57:01.0000] <Hixie>
once you have implementation experience, you would come to a standards organisation and propose it as a standard api

[16:57:02.0000] <Hixie>
and it would change in incompatible ways, probably

[16:58:00.0000] <Hixie>
at which point you'd be screwed if you hadn't used vendor prefixes first

[16:58:01.0000] <ianloic>
true dat

[16:59:00.0000] <ianloic>
I just resubscribed to the whatwg list since there seems to be a bunch of audio/video work going on and we have some different requirements/interests in that than a normal browsery user agent

[16:59:01.0000] <Hixie>
your input would be very welcome

[16:59:02.0000] <ianloic>
Hixie, so where would you attach events? To the window?


2007-04-10
[17:00:00.0000] <Hixie>
as i'm actively working on that right now

[17:00:01.0000] <ianloic>
very good!

[17:00:02.0000] <Hixie>
events that affect the entire environment would be dispatched at the Window object or the root element or both, yeah

[17:00:03.0000] <ianloic>
ok cool

[17:00:04.0000] <ianloic>
the storage event is on the body tag though isn't it?

[17:01:00.0000] <othermaciej>
ianloic: you can (in theory) make any arbitrary object an event target

[17:02:00.0000] <Hixie>
firing an event at the body element means that it'll fire at the body element, and probably bubble up to the window (if it's a bubbling event), and if there is no body, it's instead fired straight at the Document and bubbles to the Window

[17:02:01.0000] <ianloic>
othermaciej, that's true. I wonder how well that's supported by the various js libraries people are using.

[17:03:00.0000] <ianloic>
/me nods.

[17:06:00.0000] <Lachy>
good morning

[17:06:01.0000] <Hixie>
hey

[17:07:00.0000] <ianloic>
jeebus these <video> threads are long, and I don't even care about codec choices

[17:07:01.0000] <Hixie>
http://www.whatwg.org/specs/web-apps/current-work/#video is the spec so far

[17:10:00.0000] <ianloic>
hmm, I can't remember what it was I felt was missing when I read that before

[17:10:01.0000] <ianloic>
I'll have to think about this a bit

[17:18:00.0000] <othermaciej>
it's becoming much more thorough than the original version

[17:18:01.0000] <ianloic>
cool

[17:19:00.0000] <othermaciej>
you may also find this interesting: http://webkit.org/specs/HTML_Timed_Media_Elements.html

[17:19:01.0000] <othermaciej>
although Hixie is rapidly folding much of that content into the whatwg spec, generally improving it in the process

[17:20:00.0000] <othermaciej>
most of the mailing list discussion has been a waste of electrons

[17:20:01.0000] <Hixie>
most of the mailing list discussion has been about codecs

[17:20:02.0000] <ianloic>
othermaciej, yeah, that does look interesting

[17:21:00.0000] <ianloic>
yeah, I really really don't care about codecs

[17:21:01.0000] <ianloic>
the fact is that everyone will ship with mpeg 1/2/4 support

[17:21:02.0000] <Dashiva>
um

[17:21:03.0000] <KevinMarks>
for some value of 4

[17:22:00.0000] <ianloic>
KevinMarks, well, yeah

[17:22:01.0000] <othermaciej>
and for some value of ship, which may include "not ship"

[17:23:00.0000] <ianloic>
ok, so let me rephrase that a little: user agents will just build on top of the platform's media playback system which will support all the common formats

[17:23:01.0000] <Dashiva>
Which requires the user to acquire codecs

[17:23:02.0000] <ianloic>
Dashiva, which is basically automatic these days isn't it?

[17:23:03.0000] <Dashiva>
Not quite

[17:24:00.0000] <ianloic>
ok - I'm only really familiar with the Linux world.

[17:24:01.0000] <Lachy>
I stopped reading that codec discussion about 160 messages ago.  I assume no _new_ arguments were introduced since then?

[17:24:02.0000] <Dashiva>
I think they've mostly stopped arguing

[17:24:03.0000] <Dashiva>
Now they're talking about defining the codec parameters and somesuch

[17:24:04.0000] <ianloic>
Lachy, all I read was that maciej is conspiring against us all...

[17:25:00.0000] <othermaciej>
ianloic: Firefox currently appears to plan to ship native support for Ogg/Theora/Vorbis, and may later also support OS-native codec stacks

[17:25:01.0000] <Philip`>
ianloic: The only way I could find to play a MPEG-4 (actually Xvid) video on my university's Linux computers was by compiling mplayer from source, which wasn't particularly automatic

[17:25:02.0000] <othermaciej>
yes, Gervase has finally unmasked my evil for all to see

[17:26:00.0000] <Dashiva>
Next week it'll be revealed that othermaciej and chris are really identical twins

[17:26:01.0000] <othermaciej>
am I the evil twin?

[17:26:02.0000] <ianloic>
othermaciej, oh bizare! I wonder why they'd want to ship ogg. I guess they must believe the patent claims.

[17:26:03.0000] <Dashiva>
That's what the debate will be about

[17:27:00.0000] <ianloic>
Philip`, oh really? weird.

[17:27:01.0000] <ianloic>
I think the baseline codec should be animated image/gif

[17:27:02.0000] <ianloic>
that's something we can all agree on

[17:27:03.0000] <othermaciej>
ianloic: they consider it reasonably patent-safe, and like the lack of known patents that cost money to license

[17:27:04.0000] <Philip`>
I think it should be animated ASCII art

[17:27:05.0000] <Dashiva>
Someone already suggested avi/jpg or whatever it's called

[17:28:00.0000] <othermaciej>
animated gif should probably be supported, but it's not actually a *useful* baseline

[17:28:01.0000] <othermaciej>
I mean, it's not gonna get people to switch from Flash

[17:28:02.0000] <ianloic>
true

[17:28:03.0000] <Dashiva>
Might be useful to establish if that should be a guideline or not: should <video> outperform flash?

[17:29:00.0000] <othermaciej>
what do you mean by "outperform"

[17:29:01.0000] <othermaciej>
smaller bitrate for the same video quality?

[17:29:02.0000] <Dashiva>
Be considered a generally superior alternative by relevant metrics

[17:29:03.0000] <Philip`>
What version of Flash? (I remember someone saying newer versions had a better codec, but people don't use it because it's not as widely installed; but it's far more widely installed than <video>)

[17:29:04.0000] <Dashiva>
Not just codec, but ease of use, distribution, content creation, market penetration, etc

[17:29:05.0000] <ianloic>
Philip`, modern flash uses On2 VP6 which is essentially equivalent to mpeg4

[17:30:00.0000] <ianloic>
Dashiva, flv probably has the best market penetration of any codec

[17:30:01.0000] <othermaciej>
equivalent?

[17:30:02.0000] <othermaciej>
in what sense?

[17:30:03.0000] <ianloic>
othermaciej, in that you can do a lossless transform between them (for some profile of mpeg4)

[17:30:04.0000] <othermaciej>
the bitstream is not compatible, and its compression ratio is not as good as H.264

[17:31:00.0000] <Dashiva>
I think many people mean h263 or older when they say mpeg, and explicitly h264 for that

[17:31:01.0000] <ianloic>
othermaciej, oh really? I must be confused :(

[17:32:00.0000] <ianloic>
othermaciej, there are tools to do that transformation (flv to some form of mpeg4)

[17:32:01.0000] <ianloic>
othermaciej, http://vixynet.svn.sourceforge.net/viewvc/vixynet/

[17:32:02.0000] <othermaciej>
flv is usually h.263

[17:33:00.0000] <othermaciej>
http://en.wikipedia.org/wiki/FLV

[17:33:01.0000] <othermaciej>
well, Sorensen Spark, which is a slight variant of H.263

[17:33:02.0000] <Dashiva>
Are there fears of patent issues related to containers (not the codecs and contents)?

[17:33:03.0000] <othermaciej>
Flash 8 supports VP6 but most flash video on the web does not use that

[17:33:04.0000] <ianloic>
oh really?

[17:34:00.0000] <othermaciej>
anyway, codec conversation == boring

[17:34:01.0000] <ianloic>
I've been assuming that most of the web is VP6 not sorenson

[17:34:02.0000] <ianloic>
true

[17:34:03.0000] <ianloic>
we could spice it up by inventing new acronyms

[17:34:04.0000] <ianloic>
but there are more entertaining alternatives

[17:36:00.0000] <othermaciej>
having One True Codec would be great but seems fairly unpossible at this stage

[17:38:00.0000] <KevinMarks>
youtube is the h263 variant

[18:47:00.0000] <Hixie>
hm, go figure

[18:47:01.0000] <Hixie>
someone actually likes SMIL

[18:48:00.0000] <othermaciej>
yeah, I was surprised to hear someone describe it as great for authors

[18:48:01.0000] <Hixie>
certainly isn't what i've heard from authors

[19:01:00.0000] <KevinMarks>
the first SMIL spec wasn't bad; it went stray after that

[19:02:00.0000] <Hixie>
even the first is a bit over the top

[19:02:01.0000] <Hixie>
at least for a web spec that's expected to interact with html, etc

[19:02:02.0000] <Hixie>
which may not have been the intent

[19:04:00.0000] <KevinMarks>
I don't think it was meant to interact with HTML; it defined a 'text' element, not an 'html' element

[19:07:00.0000] <Hixie>
that's fine, but people are trying to use it with HTML (and SVG)

[19:07:01.0000] <othermaciej>
I think in the original SMIL era people did not realize how much HTML would become the basis for all sorts of content, not just what people thought of as documents

[19:08:00.0000] <KevinMarks>
yes, I think that was true

[19:09:00.0000] <KevinMarks>
compare the goals of SMIL 1.0 with SMIL 2.1

[19:11:00.0000] <Hixie>
Lachy: yt?

[19:12:00.0000] <Lachy>
yo!

[19:12:01.0000] <Hixie>
Lachy: can you grab the text for the following questions on the whatwg.org front page and stick them in the FAQ?:

[19:12:02.0000] <Hixie>
What are "Web Applications"?

[19:12:03.0000] <Hixie>
Aren't "Web Applications" already possible?

[19:12:04.0000] <Lachy>
ok

[19:12:05.0000] <Hixie>
and What exactly are you working on?

[19:12:06.0000] <Hixie>
feel free to fix them or change them or whatever

[19:12:07.0000] <Hixie>
i'm about to nuke them front the front page

[19:12:08.0000] <ajnewbold>
I made a web application today

[19:13:00.0000] <ajnewbold>
it's a brain! http://newbold.name/brain/

[19:14:00.0000] <Hixie>
lachy: if you haven't got them yet they're now at http://www.whatwg.org/index.old

[19:14:01.0000] <Lachy>
I just copied them

[19:14:02.0000] <Hixie>
k

[19:14:03.0000] <othermaciej>
/me wonders how the "changes from HTML4" page is going

[19:17:00.0000] <othermaciej>
hmm, looks pretty good

[19:20:00.0000] <Lachy>
Hixie, done http://blog.whatwg.org/faq/#web-apps

[19:20:01.0000] <Lachy>
I'll review and edit them later

[19:22:00.0000] <Hixie>
cool thanks

[19:23:00.0000] <Philip`>
Hmm, it wasn't at all obvious that the "FAQ: WHATWG i HTML 5" link was Polish until I tried it - I thought the "i" in the middle was just a mistake

[19:23:01.0000] <othermaciej>
I wouldn't have realized it either

[19:26:00.0000] <othermaciej>
hmm, I can think of lots of things that could go in http://wiki.whatwg.org/wiki/Changes_from_HTML4 but I am hesitating to do all the work to update it nicely

[19:28:00.0000] <Lachy>
Philip`: I just noticed that and I'm going to adjust the title to say (Polish) at the end

[19:29:00.0000] <Lachy>
done

[19:30:00.0000] <Philip`>
Okay, thanks :-)

[19:31:00.0000] <Hixie>
othermaciej: dump your ideas into the discussion page

[19:31:01.0000] <Hixie>
ok, WHATWG has a twitter account now

[19:31:02.0000] <Hixie>
cunningly named WHATWG

[19:31:03.0000] <Philip`>
(The links in the "What exactly are you working on?" section don't work)

[19:32:00.0000] <Hixie>
the whatwg.org front page automatically updates from the twitter status

[19:32:01.0000] <Lachy>
/me adds WHATWG as a twitter friend

[19:32:02.0000] <Hixie>
and my commit script automatically twitters the commit message when i check in

[19:32:03.0000] <Lachy>
oh, awesome!

[19:36:00.0000] <othermaciej>
Hixie: ah, there's a good idea

[19:41:00.0000] <Hixie>
/me kills the tabs on the front page

[19:42:00.0000] <Lachy>
Hey Hixie (and anyone else who wants to), could you please spare a few minutes to review my responses to some interview questions from the Web Standards Group? http://lachy.id.au/temp/wsg

[19:42:01.0000] <Hixie>
i can do that after dinner, sure

[19:42:02.0000] <Hixie>
bbiab

[19:42:03.0000] <Lachy>
ok, sure.

[20:08:00.0000] <Hixie>
right let's see

[20:10:00.0000] <Hixie>
XHTML2 only started in 2002?

[20:10:01.0000] <Hixie>
wow

[20:10:02.0000] <Hixie>
lachy: you have a sentence that goes "However, ..., but ..." which is awkward

[20:10:03.0000] <Lachy>
did it start earlier than that?  I think that was the first draft i found

[20:11:00.0000] <Lachy>
yeah, I should fix that

[20:11:01.0000] <Hixie>
dunno, you have internal access now :-)

[20:13:00.0000] <Hixie>
"there are over 300 participants already" is ambiguous about which group you mean

[20:13:01.0000] <Hixie>
i'd write "there are over 300 participants in the W3C group already"

[20:13:02.0000] <Hixie>
or some such

[20:13:03.0000] <Hixie>
though if you say that you might want to also point out that there are over 700 in the whatwg :-)

[20:17:00.0000] <Hixie>
"These days, the W3C process is quite strict." the process is... doesn't mean much has changed though

[20:18:00.0000] <Hixie>
other than that looks good

[20:24:00.0000] <Lachy>
Fixed the participant issue and the "however/but" issue

[20:25:00.0000] <Lachy>
should I change anything about the W3C process?

[20:27:00.0000] <othermaciej>
your answers seemed generally good

[20:28:00.0000] <Lachy>
the first public draft of XHTMl2 is 20020805, I'm trying to find anything in the member section that would indicate when they actually started

[20:28:01.0000] <Hixie>
look in w3c-html-wg around that time maybe

[20:33:00.0000] <Lachy>
that's where I'm looking

[20:33:01.0000] <Lachy>
I found a few brief mentions of XHTML2 in a few threads, but nothing linking to an editors draft or anything

[20:34:00.0000] <Lachy>
I might just say it began around 2001-2002

[20:38:00.0000] <Hixie>
/me updates the whatwg.org front page

[20:38:01.0000] <Hixie>
it's been dragged kicking and screaming into 2007

[20:38:02.0000] <Lachy>
I changed it to "XHTML2 has been in <em>secret</em> development since around 2001–2002 and..."

[20:38:03.0000] <Hixie>
hah

[20:38:04.0000] <Hixie>
nice

[20:41:00.0000] <othermaciej>
the new layout doesn't look right in Safari 2.0.4

[20:41:01.0000] <Hixie>
that's ironic, given that i used safari trunk to do it

[20:41:02.0000] <Hixie>
can't win, i guess :-P

[20:41:03.0000] <Hixie>
maybe you should add opt-in versioning in your browser, i hear that solves this kind of problem

[20:42:00.0000] <Hixie>
/me fixes

[20:42:01.0000] <othermaciej>
/me snickers

[20:43:00.0000] <othermaciej>
we'll only render it right if you specifically request IE8 mode

[20:45:00.0000] <MikeSmith>
Lachy - as far as the "They include representatives from 4 major browser vendors: Mozilla, Opera, Safari and IE" part, Nokia also makes a browser and is a member of the HTML WG.

[20:45:01.0000] <Hixie>
i wouldn't call nokia a "major browser vendor"

[20:45:02.0000] <othermaciej>
so does AOL, if you want to get technical

[20:45:03.0000] <Hixie>
and the w3c...

[20:46:00.0000] <Lachy>
they're the 4 that most people will care about.  There's no need to list every single particpant

[20:46:01.0000] <ianloic>
I wonder if Nokia or Apple have shipped more browsers...

[20:46:02.0000] <Lachy>
although, I could link to the participant list

[20:46:03.0000] <Hixie>
ianloic: better question, which has had the most page views of public web content with their browsers?

[20:46:04.0000] <othermaciej>
Nokia has probably shipped more total browser units, but likely fewer of browsers actually written by them rather than licensed

[20:47:00.0000] <othermaciej>
but all the browsers they shipped probably add up to less use share than Safari

[20:47:01.0000] <othermaciej>
if web stats can be at all trusted

[20:47:02.0000] <othermaciej>
gotta roll

[20:47:03.0000] <othermaciej>
later folks

[20:47:04.0000] <Hixie>
later

[20:47:05.0000] <ianloic>
Hixie, yeah, they're both probably important metrics

[20:47:06.0000] <ianloic>
othermaciej, later mate

[20:47:07.0000] <Lachy>
bye

[20:47:08.0000] <Hixie>
install base isn't a particularily interesting metric

[20:47:09.0000] <Hixie>
by install base, lynx has huge numbers.

[20:48:00.0000] <ianloic>
Hixie, not that huge - but probably similar to Safari

[20:48:01.0000] <Hixie>
every linux server out there has lynx on it

[20:49:00.0000] <ianloic>
Hixie, it's not installed by default on Debian & Ubuntu...

[20:49:01.0000] <Hixie>
is links?

[20:49:02.0000] <ianloic>
Hixie, nope

[20:49:03.0000] <Hixie>
elinks?

[20:49:04.0000] <ianloic>
nada

[20:49:05.0000] <Hixie>
Emacs/W3?

[20:49:06.0000] <ianloic>
not even w3m

[20:49:07.0000] <ianloic>
not even wget

[20:49:08.0000] <Hixie>
wow, how are you supposed to browse the web through a shell on those boxes? how retarded.

[20:49:09.0000] <ianloic>
I'm not even sure about telnet

[20:49:10.0000] <ianloic>
"apt-get install lynx"

[20:50:00.0000] <Hixie>
in any case

[20:50:01.0000] <Hixie>
usage is what matters, not install base

[20:52:00.0000] <ianloic>
I think install base can be an interesting number. Perhaps Nokia would have interesting ideas about why people aren't using the browsers they're shipping. Certainly ui, network speed and cost are factors but perhaps there are issues at the protocol and formats level that affect it.

[20:52:01.0000] <ianloic>
And that's only kind of a devil's advocate position

[20:52:02.0000] <Lachy>
Hixie, indeed.  If installed base mattered, then IE would have 100% of all Windows machines, but we know that there are many windows users that use other browsers

[20:53:00.0000] <ianloic>
Lachy, but Nokia has customers who chose to buy a product with web browsing as a feature, but for some reason they're not using it.

[20:53:01.0000] <Lachy>
+ a small percentage of Mac and Linux web that run IE through virtualisation or other means

[20:53:02.0000] <ianloic>
Lachy, those people are interesting

[20:53:03.0000] <MikeSmith>
support for real Web applications is also what matters. Nokia's Webkit-based browser has CSS, Javascript/XHR support.

[20:54:00.0000] <ianloic>
oh nice!

[20:54:01.0000] <ianloic>
my mobile browser didn't do any of that stuff :(

[20:54:02.0000] <MikeSmith>
and I'm not sure why it's important to make the distinction of "major" browser vendors

[20:54:03.0000] <Hixie>
btw if anyone wants to twitter to the whatwg account for some reason (e.g. if you want to hack the blog so it twitters when there's a new post), let me know

[20:55:00.0000] <Hixie>
i don't want to put the password out, since then we could lose control of the account if anyone wants to be nasty

[20:55:01.0000] <Lachy>
we could set the blog up so that the excerpt gets twittered

[20:55:02.0000] <MikeSmith>
ianloic - you need to a handset you can run Opera Mobile or Nokia's S60 browser on :)

[20:56:00.0000] <MikeSmith>
you still using a Sidekick?

[20:56:01.0000] <ianloic>
MikeSmith, indeed - I'm hooked on qwerty

[20:57:00.0000] <Lachy>
I wonder if there's a WP-Twitter plugin

[20:57:01.0000] <MikeSmith>
ianloic - Nokia N61 has qwerty

[20:59:00.0000] <Lachy>
there seem to be plugins for publishing twitter status on the blog, but can't see any for updating twitter status

[20:59:01.0000] <ianloic>
MikeSmith, my carrier doesn't seem to carry it - perhaps in a couple of years :(

[21:01:00.0000] <ianloic>
MikeSmith, plus for a lot of the stuff I do danger's always on model works really well

[21:02:00.0000] <MikeSmith>
ianloic - yeah, I know those are nice devices. they should just have better browsers on them :)

[21:04:00.0000] <Lachy>
awesome! http://blog.victoriac.net/?p=87

[21:04:01.0000] <ianloic>
MikeSmith, true

[21:04:02.0000] <ianloic>
MikeSmith, also, other handsets should behave more like computers on broadband than computers on dialup :(

[21:04:03.0000] <Lachy>
I'll install that later.  Hixie if you could provide the info I need, that would be great

[21:07:00.0000] <Hixie>
 /msg'ed

[21:45:00.0000] <Hixie>
wtf is http 405

[21:45:01.0000] <Hixie>
oh

[21:45:02.0000] <Hixie>
method not allowed

[21:45:03.0000] <Hixie>
why did i get that

[21:45:04.0000] <Hixie>
oh i know

[21:57:00.0000] <Hixie>
anyone have a version of http://www.bloglines.com/images/spinner.gif which they know the provenance of?

[21:57:01.0000] <Hixie>
and/or that they know is public domain?

[22:01:00.0000] <Lachy>
Hixie, the Firefox web dev toolbar has one released under the GPL

[22:01:01.0000] <Lachy>
if you have it installed, see chrome://webdeveloper/content/images/content/loading.gif

[22:01:02.0000] <Hixie>
ooo, that works

[22:01:03.0000] <Hixie>
i don't

[22:05:00.0000] <Lachy>
do you want me to send you the image then?

[22:05:01.0000] <Hixie>
sure

[22:07:00.0000] <Lachy>
emailed it

[22:07:01.0000] <Hixie>
thanks

[22:10:00.0000] <karlUshi>
fwiw: I like smil :) implementations are far to be perfect but it's quite easy to create slideshow with it for examples.

[22:11:00.0000] <Hixie>
right, whatwg's twitter status can now be set from the whatwg home page

[22:12:00.0000] <Lachy>
cool.  btw, you told me it would take 10 minutes to complete, about 1 hour ago :-)

[22:13:00.0000] <Hixie>
i did other things at the same time :-)

[22:29:00.0000] <ianloic>
/me curses the squishy unstable nature of XUL.

[22:29:01.0000] <ianloic>
or should I say, the XUL implementation

[22:53:00.0000] <Hixie>
why wouldn't the script on whatwg.org to update twitter work on firefox?

[22:53:01.0000] <Hixie>
i see nothing that would break it

[22:53:02.0000] <Hixie>
i'm confused

[22:54:00.0000] <ianloic>
Hixie, where is it?

[22:54:01.0000] <Hixie>
search for updateTwitter on whatwg.org

[22:55:00.0000] <Hixie>
for some reason the server isn't getting the status=foo stuff

[22:56:00.0000] <ianloic>
I don't see it on www.whatwg.org

[22:56:01.0000] <Hixie>
oh i have to set the content type i bet

[22:56:02.0000] <Hixie>
never mind

[22:59:00.0000] <Hixie>
ok it works

[22:59:01.0000] <Hixie>
all good

[23:11:00.0000] <Hixie>
well, everyone seems in favour of using the whatwg spec and having me continue as editor so far

[23:15:00.0000] <Lachy>
I'm sure there will be someone who says no.  I'm interested to see what kind of arguments they raise against it and what they put forth as an alternative

[00:56:00.0000] <othermaciej>
I added some items to possibly add to the discussion page for http://wiki.whatwg.org/wiki/Changes_from_HTML4

[00:56:01.0000] <othermaciej>
hey annevk 

[00:57:00.0000] <annevk>
morning

[00:57:01.0000] <othermaciej>
HTML5 has now officially been proposed to the HTML WG

[01:00:00.0000] <annevk>
I saw, cool

[01:00:01.0000] <annevk>
and now we wait

[01:00:02.0000] <othermaciej>
perhaps Chris will have to write yet another essay

[01:03:00.0000] <annevk>
he mentioned somewhere that he wants to move forward with XML

[01:03:01.0000] <othermaciej>
I liked his message where he said XHTML only failed because 80% of browsers didn't support it

[01:04:00.0000] <othermaciej>
I wonder which 80% that could be

[01:05:00.0000] <annevk>
there's an old HTML+Time proposal from Microsoft btw that covers SMIL in HTML

[01:13:00.0000] <annevk>
Re: changes from HTML4 page, I'll take a look later today. <input> needs to be covered indeed and the others as well probably

[01:39:00.0000] <annevk>
bah, currenttime

[01:48:00.0000] <zcorpan>
home again, at last

[01:49:00.0000] <hsivonen>
zcorpan: http://hsivonen.iki.fi/validator/html5/?doc=http%3A%2F%2Fsimon.html5.org%2Ftemp%2Fw3c-home-in-html5.html

[01:49:01.0000] <zcorpan>
hsivonen: nice :)

[01:52:00.0000] <zcorpan>
http://forums.whatwg.org/viewtopic.php?p=112#112 anything inaccurate?

[01:54:00.0000] <annevk>
DOCTYPEs can also just give a public identifier

[01:54:01.0000] <annevk>
if it's assumed that identifier is supported by the client that reads the document

[01:54:02.0000] <zcorpan>
it's still a reference to a dtd

[01:55:00.0000] <annevk>
I'm not sure you answered his second question though

[01:55:01.0000] <annevk>
a DTD defines the valid structure of a document (roughly)

[01:55:02.0000] <annevk>
oh nm

[01:55:03.0000] <zcorpan>
i presumed that he really meant doctype declaration

[01:56:00.0000] <zcorpan>
although i think i covered both :)

[02:00:00.0000] <hsivonen>
Hixie: http://syntax.whatwg.org/ could use a note that the SGML stuff has been abandoned

[02:03:00.0000] <Hixie>
syntax is under fantasai's account i think

[11:02:00.0000] <ianloic>
w3c's mailing list archive format sucks

[11:02:01.0000] <annevk>
/me rather likes it

[11:03:00.0000] <ianloic>
annevk, it's not threaded so it's hard to weave through conversations in the way I like to

[11:03:01.0000] <ianloic>
annevk, basically now I have to go subscribe to html-public

[11:04:00.0000] <annevk>
you'd have to become a member first

[11:04:01.0000] <ianloic>
oh really?

[11:04:02.0000] <ianloic>
is membership still open?

[11:05:00.0000] <annevk>
certainly

[11:05:01.0000] <annevk>
see the steps here: http://blog.whatwg.org/w3c-restarts-html-effort

[11:05:02.0000] <ianloic>
annevk, thanks!

[11:10:00.0000] <gsnedders>
/me would rather anyone could subscribe to the W3C lists, even if not post

[11:12:00.0000] <annevk>
gsnedders, you could make feature requests...

[11:14:00.0000] <othermaciej>
hah, sounds like Chris Wilson *is* working on another essay

[11:14:01.0000] <othermaciej>
I feel bad now, giving him all these writing assignment

[11:14:02.0000] <othermaciej>
s

[11:14:03.0000] <annevk>
maybe you should teach

[11:57:00.0000] <zcorpan>
/me updates /html5-elements...

[11:58:00.0000] <annevk>
http://lists.w3.org/Archives/Public/www-archive/2007Apr/att-0018/entities.htm has an entity test based on IE7

[11:58:01.0000] <annevk>
It seems IE7 is doing reparsing though

[11:58:02.0000] <annevk>
&notin "becomes" &not;in and &notin; stays the "same"

[11:59:00.0000] <zcorpan>
http://simon.html5.org/test/html/parsing/entities/trailing-semicolon/

[12:01:00.0000] <annevk>
ah cool

[12:01:01.0000] <annevk>
zcorpan, you got my <map> comment?

[12:02:00.0000] <zcorpan>
where?

[12:02:01.0000] <zcorpan>
can't recall anything about <map>

[12:02:02.0000] <annevk>
name is dropped

[12:02:03.0000] <zcorpan>
ah

[12:02:04.0000] <zcorpan>
right

[12:03:00.0000] <zcorpan>
but only for conformance

[12:03:01.0000] <zcorpan>
document conformance*

[12:04:00.0000] <annevk>
your list of elements is about document conformance right?

[12:04:01.0000] <zcorpan>
yes

[12:04:02.0000] <zcorpan>
thanks

[12:05:00.0000] <zcorpan>
i have some demos on <map id name> that could be turned into test cases also

[12:05:01.0000] <annevk>
<map id=bar> <map name=bar>

[12:06:00.0000] <annevk>
(both nested and as siblings might be interesting...)

[12:06:01.0000] <zcorpan>
ok

[12:07:00.0000] <zcorpan>
although isn't multiple ids generally undefined?

[12:07:01.0000] <annevk>
it doesn't define it in terms of getElementById()

[12:08:00.0000] <zcorpan>
i'll look closer at the <map> spec then

[12:18:00.0000] <zcorpan>
html5-elements updated

[12:26:00.0000] <zcorpan>
the ids not being permanent is annoying. it should be possible to link to a part of the spec and assume that it will point to the same thing so long as the spec doesn't change other than editorially... why aren't the title=""s used as ids (perhaps with spaces replaced with dashes or something)?

[12:27:00.0000] <zcorpan>
it's not only about my list of elements -- people link to a part of the spec when discussing

[12:29:00.0000] <zcorpan>
(minor but still annoying ;) )

[13:45:00.0000] <othermaciej>
hi everyone

[13:45:01.0000] <hsivonen>
hi

[14:58:00.0000] <Hixie>
so can someone explain the use case for this "data attribute" idea?

[15:00:00.0000] <Dashiva>
declarative approach to data

[15:00:01.0000] <Dashiva>
Some pre-wf2 scripts had homebrew attributes for required, type, etc

[15:01:00.0000] <hsivonen>
Hixie: the idea is that a Web app wants to annotate a DOM node with app-specific data *and* send the in the serialization so that custom properties on the JS-level don't work

[15:02:00.0000] <hsivonen>
IIRC, PPK has an ALA article with a concrete case

[15:03:00.0000] <Hixie>
and "class" doesn't work for this because the data isn't boolean?

[15:03:01.0000] <Hixie>
or?

[15:04:00.0000] <hsivonen>
Hixie: hmm. his use cases are boolean...

[15:04:01.0000] <hsivonen>
Hixie: http://alistapart.com/articles/scripttriggers/

[15:05:00.0000] <Hixie>
well then

[15:05:01.0000] <Hixie>
use the class attribute

[15:05:02.0000] <Hixie>
moving on...

[15:05:03.0000] <hasather>
hmm, anyone else gets 403's on ALA. I've had it for severla months. Not that I care much, but it's weird

[15:06:00.0000] <Dashiva>
Not all booleans, no

[15:07:00.0000] <Dashiva>
He uses key=value to hack around it, e.g. class="large required maxlength=300"

[15:07:01.0000] <othermaciej>
Hixie, hsivonen: sounds like some of the cases aren't boolean

[15:08:00.0000] <Hixie>
ah ok

[15:08:01.0000] <Hixie>
well then we'll have to investigate

[15:11:00.0000] <Philip`>
I used a non-boolean data attribute some time ago, with a list of things somewhat like <a title="text for a tooltip/popup message" _formatted="alternative HTML-formatted text" onmouseover="scripted_tooltip_window(this)">Short uninteresting link text</a>, so it would have sensible non-JS (and non-CSS) fallback but the script could make use of more interesting data to display in its non-standard tooltip-like-thing, without having to store all that data a

[15:11:01.0000] <Philip`>
... a long way away

[15:14:00.0000] <Philip`>
(I can probably think of better non-nonstandard ways of doing that now, but they'd seem more convoluted than just sticking an extra attribute on the element, and I couldn't think of anything better back then)

[15:16:00.0000] <met_>
sounds like in XUL, no restrictions in new and new attributes

[15:17:00.0000] <Dashiva>
Well, it's a significant step from one script data attribute to allowing arbitrary attributes with a "local" type prefix

[15:20:00.0000] <met_>
prefix?

[15:20:01.0000] <met_>
the underscore?

[15:24:00.0000] <Philip`>
If there's only one script data attribute, perhaps people would start doing <textarea scriptdata="maxlength=300"> (which is much clearer than scriptdata="300") or <textarea scriptdata='{"maxlength":300,"minlength":100}'>, and make their own custom parsers, which seems kind of pointless when you've already got an HTML parser that could do all the work for you

[15:25:00.0000] <moeffju>
use $ as a prefix?

[15:26:00.0000] <moeffju>
<foo $data="whee" $more="hurray">

[15:30:00.0000] <met_>
is $ valid in attribute name?

[15:30:01.0000] <hsivonen>
met_: in xml, no

[15:30:02.0000] <met_>
this is what i mean

[15:31:00.0000] <Philip`>
I think I have a mild natural distaste for using $ but that's probably just because I'm too used to Perl and code like |print qq{<foo \$data="$data">}| is not especially beautiful, which isn't a particularly valid concern...

[15:33:00.0000] <Philip`>
Are people trying to make HTML5 and XHTML5 as equivalently powerful as possible, and trying to minimise the exceptions like <p><ul/></p>?

[15:35:00.0000] <Philip`>
(If so, presumably that would rule out $ (if it's not allowed in XML) and : (if it'd mean you couldn't do a lossless HTML5->XHTML5 conversion of <div foo:bar=x>))

[15:39:00.0000] <moeffju>
which brings us back to underscore

[15:44:00.0000] <deltab>
$ would likely interfere with some templating systems

[15:59:00.0000] <Hixie>
hm

[16:00:00.0000] <Hixie>
.defaultPlaybackRate and .playbackRate

[16:34:00.0000] <Hixie>
should we have people do video.playbackRate = video.defaultPlaybackRate, or should we have a video.resetPlaybackRate() that does that?

[16:34:01.0000] <Hixie>
or should .play() do it?

[16:36:00.0000] <moeffju>
playbackRate should init to defaultPlaybackRate, and if people want to reset it, an assignment is fine, IMO

[16:37:00.0000] <moeffju>
play() should not automatically do it though

[16:40:00.0000] <Dashiva>
Is the playbackrate linked to being in a playing state?

[16:40:01.0000] <Dashiva>
i.e. is it 0 when paused?

[16:42:00.0000] <Hixie>
no

[16:43:00.0000] <Dashiva>
Then I see no reason play() should change the playbackrate. And having a reset() just to save an assignment seems a bit redundant

[16:44:00.0000] <Hixie>
so .defaultPlaybacRate is basically a mutable attribute with no effect, then, and just needed to note what to set the rate back to after ffwding

[16:46:00.0000] <Dashiva>
I'd liken it to defaultValue, defaultChecked, etc in forms. It's used for the user to restore defaults

[16:47:00.0000] <nickshanks>
is there a published default stylesheet for braile media? specifically I want to know which elements display without contraction by default


2007-04-11
[17:11:00.0000] <KevinMarks>
/me reads scrollback

[17:11:01.0000] <KevinMarks>
why is scriptdata not just <script> and declare some variables?

[17:12:00.0000] <Dashiva>
Because it's linked to specific elements

[17:12:01.0000] <othermaciej>
KevinMarks: the idea is to attach information to an element in the markup

[17:12:02.0000] <Hixie>
othermaciej: any opinion on the default rate thing?

[17:12:03.0000] <othermaciej>
you want it as hooks on specific elements for script to attach

[17:12:04.0000] <othermaciej>
Hixie: have not had a chance to read yet, and in meeting at the moment

[17:12:05.0000] <Hixie>
k

[17:16:00.0000] <KevinMarks>
ah, so a replacement for the 'add piles of arbitrary attributes with ugly prefixes' model that certain js toolkits use?

[17:17:00.0000] <KevinMarks>
*cough* dojo *cough*

[17:17:01.0000] <Dashiva>
Yeah, among other things :)

[17:40:00.0000] <deltab>
some reliable way of referring to the current script element would be useful

[22:59:00.0000] <Hixie>
Lachy_: for jabber, i recommend bitlbee

[23:00:00.0000] <Lachy_>
ok, I'm looking it up

[23:08:00.0000] <Lachy_>
hmm. Could I have bitlbee running on my computer at home and then just connect to it with any IRC client, anywhere?

[23:20:00.0000] <Hixie>
i assume so

[23:20:01.0000] <Hixie>
i have someone run it for me on the server i run irssi on

[23:21:00.0000] <Hixie>
and so irssi is always connected to it

[23:21:01.0000] <Hixie>
so i'm always on jabber (and aim and msn and icq) as well as being always on irc

[23:21:02.0000] <gavin_>
interesting

[23:21:03.0000] <Hixie>
and since twitter uses jabber, i'm always on twitter too

[23:21:04.0000] <Hixie>
and since it all runs under screen, i can connect to it multiple times from as many machines as i like

[23:22:00.0000] <Hixie>
without affecting my irc connections

[23:23:00.0000] <gavin_>
I don't really mind only being on aim/msn/jabber when I'm actually at my computer

[23:23:01.0000] <gavin_>
IRC is different, since I keep up with other people's discussions so I always need to be conected

[23:24:00.0000] <Lachy_>
awesome! There are public servers I could connect to, so I don't have to set it up myself

[23:24:01.0000] <gavin_>
do you trust those public servers with all your IM/IRC traffic? :)

[23:31:00.0000] <dolphinling_>
So is it just me, or does perl's CGI module not even understand that html 4 exists?

[23:31:01.0000] <dolphinling_>
In the dtd handling, it checks for html 2.0 and html 3.2 and knows not to put an xmlns attribute in, but it doesn't check for html 4.

[23:36:00.0000] <Hixie>
wow

[23:36:01.0000] <Hixie>
check out the ocmments on http://ajaxian.com/archives/proposal-for-the-w3c-to-adopt-html-5

[23:36:02.0000] <Hixie>
that's the only place i've seen negative comments, and irrational ones at that

[23:36:03.0000] <Hixie>
weird

[23:38:00.0000] <gavin_>
"The so called "œHTM 5" is a rip off from all the community work on HTML"

[23:39:00.0000] <gavin_>
nice

[23:40:00.0000] <Hixie>
they get even weirder near the end

[23:41:00.0000] <gavin_>
indeed, just finished reading

[23:42:00.0000] <gavin_>
I'm not really sure how someone can end up reaching the last commenter's conclusions

[23:45:00.0000] <Lachy_>
ha! "... coupled with asshole personalities with big ego like Ian Hickson ..." :-)

[23:45:01.0000] <Hixie>
hey maciej, check this out: http://ajaxian.com/archives/proposal-for-the-w3c-to-adopt-html-5

[23:45:02.0000] <Hixie>
the comments, specifically

[23:45:03.0000] <Hixie>
btw the really really funny thing about this post is that the guy who posted it is on my team at work

[23:46:00.0000] <othermaciej>
/me clicks the link

[23:54:00.0000] <Lachy_>
/me is confused by those comments

[23:55:00.0000] <Lachy_>
They're happy that the W3C has formed the HTMLWG and that they will be working on HTML, but seem to object to the work we did on HTML

[23:56:00.0000] <Lachy_>
it seems to be just because it doesn't come with a nice, shiny W3C logo at the top

[23:57:00.0000] <Lachy_>
I bet that as soon as the HTMLWG accepts HTML5 and begins work on it, all those people who object to the WHATWG will suddendly applaud the W3C for doing such great work

[23:59:00.0000] <othermaciej>
Hixie: nice - although I kinda hope my marketing people don't see it and yell at me for getting embroiled in controversy; gonna read comments now

[00:01:00.0000] <Lachy_>
"Just to point out that web forms 2.0 has been reviewed by some W3C working goups already and rejected." -- which groups rejected it?  AIUI, the XForms folks just want to take it and destroy it for themselves, not reject it

[00:02:00.0000] <othermaciej>
I'm curious who Mitch Niel is

[00:02:01.0000] <othermaciej>
Lachy_: I think it was once submitted as a Member Submission and rejected for standards track, before the WAF WG later picked it up

[00:03:00.0000] <othermaciej>
hah, I love the comment that blames WHATWG for reinventing things needlessly, and then cites SVG and XForms

[00:05:00.0000] <othermaciej>
those comments seem suspiciously pro-W3C and pro-Microsoft

[00:06:00.0000] <othermaciej>
hmm, Mark Arrington uses techcrunch as his URL, perhaps hoping people will mistake him for Mike Arrington

[00:11:00.0000] <othermaciej>
I commented, perhaps foolishly

[00:15:00.0000] <Hixie>
so did i

[00:16:00.0000] <Hixie>
othermaciej: it wasn't ever rejected for standards track, the submission wasn't intended for standards track

[00:16:01.0000] <Hixie>
wonder who this Mike Arrington guy is

[00:16:02.0000] <Hixie>
Mark Arrington even

[00:17:00.0000] <othermaciej>
Hixie: everyone knows who Mike Arrington is, given the misleading URL, I'd guess Mark Arrington is a pseudonym

[00:23:00.0000] <Hixie>
othermaciej: your comment is funny. "Hey! It's not just an Opera conspiracy! We're part of it too!"

[00:25:00.0000] <othermaciej>
Hixie: yeah, I didn't want Opera to get all the credit

[00:42:00.0000] <tylerr>
it

[02:56:00.0000] <othermaciej>
I'm surprised <http://www.digital-web.com/articles/html5_xhtml2_and_the_future_of_the_web/> didn't discuss HTML WG and the possibility of it adopting HTML5 more prominently

[03:12:00.0000] <MikeSmith>
othermaciej - the "W3C and HTML" section seems to do a pretty good job of making it clear.. or I guess you just mean that section should be somewhere else in the document?

[03:12:01.0000] <othermaciej>
the earlier parts of the document make it sound like a W3C vs. WHATWG battle

[05:35:00.0000] <SimonW>
Hello all... quick question

[05:36:00.0000] <SimonW>
is the plan with HTML 5 still to get it working on legacy browsers using scripting?

[05:42:00.0000] <met_>
SimonW, there are some projects http://wiki.whatwg.org/wiki/Implementations_in_Web_browsers

[05:42:01.0000] <met_>
SimonW, eg http://sourceforge.net/projects/wf2/

[05:43:00.0000] <SimonW>
do you know if it's still an official stated aim? I remember when What WG started a few years ago they got Dean Edwards on board for exactly that

[05:43:01.0000] <SimonW>
thanks for the links

[05:44:00.0000] <SimonW>
it looks like the wf2 codebase hasn't been touched in 19 months

[05:45:00.0000] <hsivonen>
SimonW: in particular, making WF 2.0 work in IE6 using as script library was part of the design

[05:46:00.0000] <SimonW>
thanks - I'm putting a talk together for this evening, so I might throw a few more questions around later if that's OK

[05:46:01.0000] <hsivonen>
SimonW: to me, it seems that it is less of a design requirement for the rest of HTML5.

[05:46:02.0000] <hsivonen>
SimonW: however, you should recheck when Hixie is around

[05:46:03.0000] <SimonW>
oh - one other thing, is it fair to say that What WG was created partly in response to the <canvas> element?

[05:46:04.0000] <SimonW>
or was that just coincidental?

[05:47:00.0000] <hsivonen>
SimonW: my understanding is that Apple created <canvas> and the WHATWG were formed independently and when Mozilla and Opera showed interest in <canvas>, Apple contributed the spec

[05:48:00.0000] <hsivonen>
(all the recheck when Hixie is around disclaimers apply)

[05:48:01.0000] <SimonW>
I'll do that, thanks

[05:49:00.0000] <SimonW>
last thing for the moment: where did the name "HTML 5" come from? Is it just a nickname for the combination of WF2 and WA1?

[05:51:00.0000] <hsivonen>
SimonW: it is a nickname and HTML5 comes after HTML 4

[05:51:01.0000] <hsivonen>
SimonW: DOM5, XML5, SVG5, etc. are joke names that have the same "5" pattern

[05:51:02.0000] <met_>
HTML5 is also more comprehensive than WF2 and other

[05:52:00.0000] <hsivonen>
(and they are hypothetics that pop up on IRC every now and then)

[05:52:01.0000] <SimonW>
So a good answer to the question "What is HTML 5?" is "It's the nickname for a set of specifications that originated with the WHATWG" ?

[05:52:02.0000] <hsivonen>
SimonW: nickname for Web Apps 1.0 plus Web Form 2.0

[05:53:00.0000] <hsivonen>
afk

[05:53:01.0000] <SimonW>
Web Controls 1.0 too?

[05:56:00.0000] <hsivonen>
SimonW: no, Web Controls 1.0 is vaporware is practice at this point

[05:56:01.0000] <SimonW>
OK

[06:54:00.0000] <SimonW>
How does the existence of the new HTML Working Group affect the activity around WHATWG?

[06:55:00.0000] <virtuelv>
SimonW: does it? in what sense?

[06:56:00.0000] <SimonW>
I don't know, that's why I'm asking :)

[06:57:00.0000] <gsnedders>
in what way do you mean activity? the number of people on the mailing lists?

[06:57:01.0000] <SimonW>
no, the actual work done by WHATWG

[06:57:02.0000] <SimonW>
will a lot of it move over to HTML WG, or will it still be developed separately and suggested over there once it's matured a bit?

[06:57:03.0000] <SimonW>
especially since the HTML 5 has been offered to HTML WG as a starting point

[06:58:00.0000] <gsnedders>
whether it moves over to the HTML WG is up to the HTML WG.

[06:58:01.0000] <virtuelv>
Personally, I think there is room for both

[06:58:02.0000] <SimonW>
(I'm asking as a relative outsider to WHATWG)

[06:58:03.0000] <virtuelv>
whatwg has been a useful testing ground

[06:58:04.0000] <gsnedders>
if Hixie is editor of both, they will both continue, and be generated from the same document

[06:59:00.0000] <met_>
SimonW, it is maybe on opposite way, WHATWG affects HTML WG, it depends on results of this proposal http://lists.w3.org/Archives/Public/public-html/2007Apr/0429.html

[06:59:01.0000] <gsnedders>
if the HTML WG starts from somewhere else, WHATWG's specs will remain at most a superset, and compatible

[06:59:02.0000] <gsnedders>
(unless of course the HTML WG loses it and goes away from developing something backwards compatible and therefore isn't relevant)

[07:00:00.0000] <gsnedders>
http://lists.w3.org/Archives/Public/public-html/2007Apr/0025.html

[07:00:01.0000] <gsnedders>
http://lists.w3.org/Archives/Public/public-html/2007JanMar/0052.html

[07:01:00.0000] <gsnedders>
that really answers the question

[07:01:01.0000] <gsnedders>
s/that/those

[07:01:02.0000] <SimonW>
thanks

[07:01:03.0000] <SimonW>
so basically, the answer to my question is pretty up in the air

[07:02:00.0000] <SimonW>
but it should all shake out in the next month or so

[07:02:01.0000] <gsnedders>
quicker than that, I expect

[07:02:02.0000] <SimonW>
 "Therefore, as a safety net, if you will, the WHATWG 

[07:02:03.0000] <SimonW>
specs will continue to be developed for the time being."

[07:07:00.0000] <SimonW>
Was XHTML 2 being developed on a non-public mailing list?

[07:13:00.0000] <gsnedders>
IIRC www-html, a public list, was the HTML WG's mailing list, till the XHTML2 and HTML WG were re-chartered

[07:13:01.0000] <gsnedders>
public-xhtml2 is now the XHTML2 WG's list

[07:16:00.0000] <Lachy>
SimonW, XHTML2 was being developed behind closed doors on member-html-wg

[07:17:00.0000] <SimonW>
but all development on the new HTML WG is in public?

[07:17:01.0000] <Lachy>
yes

[07:17:02.0000] <SimonW>
thanks

[07:18:00.0000] <Lachy>
XHTML2 should also be moving entirely to the public list soon, but there still appears to be some activity on the member list

[07:18:01.0000] <gsnedders>
www-html was therefore only for public suggestions?

[07:18:02.0000] <Lachy>
actually, w3c-html-wg and w3c-html-cg were the member lists

[07:18:03.0000] <gsnedders>
/me almost completely ignored the HTML WG before the re-chartering

[10:39:00.0000] <annevk>
<script async> and <script defer> are mutually exclusive right?

[11:37:00.0000] <zcorpan>
hsivonen: you validator doesn't complain about <base> not being first (ignoring <meta charset>), it seems

[11:38:00.0000] <zcorpan>
(<head>'s content model doesn't agree with <meta>'s or <base>'s definition of where they are allowed, though)

[11:59:00.0000] <annevk>
"I have to say, some of the article is questionable at best, laughable at worst. I'm sorry but to try and introduce a standard called WA1 for web applications is just plain silly. We already have standards laid out in xhtml and dom so why add another level of obfustication around both of them and give it an almost identical name as the accessibility specification WAI simply shows a lack of knowledge of the client side."

[12:00:00.0000] <annevk>
from http://www.digital-web.com/articles/html5_xhtml2_and_the_future_of_the_web/comments/

[12:02:00.0000] <jdandrea>
... and Ric's point ... is ...

[12:02:01.0000] <annevk>
that we need a new name supposedly

[12:02:02.0000] <jdandrea>
:)

[12:02:03.0000] <annevk>
hopefully we can rename it HTML 5 or HTML5 in a few weeks

[12:04:00.0000] <kingryan>
and hopefull the html-wg will adopt it

[12:04:01.0000] <kingryan>
hopefully*

[12:04:02.0000] <annevk>
well, that would be required :)

[12:05:00.0000] <kingryan>
from the above article: "HTML 4.01 may be a good, stable ground for developers to stand on..."

[12:05:01.0000] <kingryan>
I don't really understand that. HTML4 as a spec isn't that useful, it's only the semi-interoperable implementations that are stable

[12:06:00.0000] <zcorpan>
perhaps it's useful for authors

[12:06:01.0000] <zcorpan>
although not really either, you can get 10 interprentations of how to correctly use <h1>-<h6> for instance

[12:07:00.0000] <zcorpan>
and it contradicts itself on some things

[12:07:01.0000] <annevk>
I doubt typical authors look up HTML4 at all

[12:07:02.0000] <zcorpan>
they don't

[12:08:00.0000] <zcorpan>
typical author doesn't even use a validator

[12:08:01.0000] <kingryan>
zcorpan: it may be useful for authors, but it's not sufficient

[12:08:02.0000] <zcorpan>
kingryan: indeed

[12:08:03.0000] <zcorpan>
i didn't say it was a good spec

[12:09:00.0000] <kingryan>
for me, html4 is only useful for supporting material when I have to make arguments supporting microformats against those who're more pedantic than I am :D

[12:17:00.0000] <annevk>
Philip` just mentioned the following in #html-wg: http://canvex.lazyilluminati.com/wa1/index.xhtml

[12:18:00.0000] <annevk>
He wrote a script in Python (not completely done yet I  believe) that generates individual pages from the HTML5 proposal and keeps the cross references working

[12:58:00.0000] <zcorpan>
http://simon.html5.org/temp/duke3d%202007-04-11%2020-56-58-93.mpeg

[13:02:00.0000] <zcorpan>
/me has completed all 4 episodes at damn i'm good :D

[13:13:00.0000] <annevk2>
KevinMarks, is your proposal to give each element a datastore?

[13:13:01.0000] <annevk2>
/me kind of likes that idea

[13:17:00.0000] <annevk>
Philip`, get an account from Hixie on http://code.google.com/p/html5/ and put the code there

[13:22:00.0000] <Lachy_>
http://blog.jclark.com/2007/04/validation-not-necessarily-harmful.html

[13:23:00.0000] <Lachy_>
the first half of that is quite good, but then he reaches a conclusion that doesn't quite follow

[13:23:01.0000] <annevk>
i suppose you can do validation for non-complicated languages

[13:53:00.0000] <hsivonen>
zcorpan: base placement should be fixed now. thanks.

[13:53:01.0000] <a-ja>
hi all...don't know if this is a w3c validator or xhtml5 conformance checker issue...basically, for w3c mobile doctypes the w3c validator chokes on xml:base whereas xhtml5 checker chokes on html base href=. 

[13:53:02.0000] <hsivonen>
a-ja: base href= is non-conforming in XHTML5

[13:54:00.0000] <hsivonen>
(conforming in HTML5)

[13:54:01.0000] <annevk>
currently non-conforming*

[13:54:02.0000] <hsivonen>
annevk: well, everything in HTML5 is currently

[13:55:00.0000] <a-ja>
apparently mobile doctypes require xhtml but not xml:base...so shouldn't conformance checker issue warning rather than an error?

[13:55:01.0000] <annevk>
well, I raised an issue about this one :)

[13:56:00.0000] <a-ja>
/me has no idea what old mobile devices actually implement

[13:56:01.0000] <annevk>
HTML most likely

[13:56:02.0000] <annevk>
certainly no xml:base fancyness

[13:56:03.0000] <gsnedders>
a-ja: xml:base is invalid under XHTML Mobile because the DTD doesn't allow it, and is a normative part of the standard

[13:57:00.0000] <a-ja>
that's the impression i get, too, annevk

[13:58:00.0000] <hsivonen>
a-ja: mobile profiles are not supported by the HTML5 conformance checker (unless you bring your own schema and use the generic facet)

[13:59:00.0000] <a-ja>
so, basically the call at this point is that there's not enuff existant mobile content to be a major concern? 

[13:59:01.0000] <a-ja>
/me is only asking in the spirit of "don't break the web"...not arguing that mobile doctypes aren't brain-dead

[14:00:00.0000] <hsivonen>
a-ja: I think mobile profiles are a mistake, so I am not focusing my effort on them. Instead, I am focusing on (X)HTML5 and I intend to revise the XHTML 1.x schemas sometime in more or less distant future.

[14:00:01.0000] <hsivonen>
a-ja: Mobile walled gardens aren't really part of the Web that we are concerned about not breaking

[14:01:00.0000] <a-ja>
hsivonen: gotcha...bigger fish to fry

[14:03:00.0000] <hsivonen>
a-ja: also, document conformance requirements are allowed to make existing practices non-conforming without being considered breaking the Web. Browser behavior could break the Web.

[14:03:01.0000] <Philip`>
http://canvex.lazyilluminati.com/wa1/specsplit.py and http://canvex.lazyilluminati.com/wa1/toxhtml5.py (please forgive/fix any ugliness and tabs) are what's generating http://canvex.lazyilluminati.com/wa1/

[14:04:00.0000] <Philip`>
(Maybe you could combine those scripts, but the html5lib minidom implementation seems incompatible with the normal XML one)

[14:04:01.0000] <annevk>
nice (the output)

[14:08:00.0000] <a-ja>
hsivonen: so the handling of xml:base and base href= in xhtml documents by UA's belongs in wa1 spec, rather than conformance checker, eh? 

[14:08:01.0000] <KevinMarks>
annevk: catching up - what Sam + Mark did for RSS validation is an interesting approach

[14:13:00.0000] <a-ja>
hsivonen: or just plain ignored in spec for walled garden reasons?

[14:14:00.0000] <a-ja>
/me is abandoning all hope of making sites both html5 and mobileOK conformant :)

[14:14:01.0000] <hober>
it amuses me greatly that specifically this blog post YSODs in Firefox: http://tom.opiumfield.com/blog/2007/04/11#When:08:50:10

[14:14:02.0000] <zcorpan>
mobileOK is bogus

[14:17:00.0000] <hsivonen>
a-ja: the conformance checker, at the moment, only checks if xml:base or base href= are allowed at a given spot

[14:17:01.0000] <annevk>
mobile is bogus

[14:17:02.0000] <annevk>
oh wait

[14:17:03.0000] <a-ja>
zcorpan: true enough...was gonna be majorly difficult to get pages under it's size limits in any event.   

[14:18:00.0000] <hsivonen>
a-ja: currently, no conformance requirement check requires URI dereferencing, so the conformance checker doesn't actully implement base URI-based URI resolution

[14:18:01.0000] <KevinMarks>
annevk: re private data store - not sure if that is a workable model, I was just thinking of keeping the script in <script> and setting up a way to associate with the  element. I'll defer to you browser implementers about how to do it well

[14:19:00.0000] <hsivonen>
a-ja: well, in general, WHATWG specs are about *the* Web, which you may browser using Minimo, the S60 WebKit-based browser, Opera or Opera Mini (or any emerging serious mobile browser) on a mobile device

[14:19:01.0000] <annevk>
/me likes the .param approach

[14:19:02.0000] <KevinMarks>
I can see an advantage in having a datablob that can be shared between multiple elements that way

[14:19:03.0000] <hsivonen>
s/may browser/may browse/

[14:19:04.0000] <annevk>
except that I'm not sure if you want to type JSON into an attribute value

[14:20:00.0000] <hsivonen>
annevk: I'd say it is up to scripts to run eval() on JSON if they want to put JSON there, but we shouldn't require JSON as the private data

[14:20:01.0000] <zcorpan>
Philip`: you could perhaps include the trailing <script> on all pages, to get the status annotations

[14:20:02.0000] <annevk>
KevinMarks, the problem with your proposal is that authors have to both set up the attribute and set some default values in a <script> element somewhere

[14:21:00.0000] <hsivonen>
KevinMarks: btw, the HTML5 conformance checker technology preview uses a hybrid model that involves both schemas and Feed Validator -like SAX handlers

[14:21:01.0000] <annevk>
KevinMarks, one of the use cases is (i think) that an author sets some attributes with values and an external script implements the functionality on top of them (not controlled by the author)

[14:23:00.0000] <KevinMarks>
Right, thats the technique I call 'javascript decorators' - where you have some semantic HTML and a classname that the js looks for to interpret it in a new way. That gives nice fallback behaviour.

[14:23:01.0000] <annevk>
class only address booleans

[14:24:00.0000] <annevk>
the discussion ensued to find a way to address the non-boolean issues

[14:24:01.0000] <KevinMarks>
no, classes can mark out content for further manipulation

[14:24:02.0000] <KevinMarks>
eg hCard

[14:24:03.0000] <KevinMarks>
you have a containing class="vcard" and then classes on the other elements to mark up the data

[14:25:00.0000] <KevinMarks>
that way you have DRY built-in

[14:25:01.0000] <annevk>
<input class="maxlength=300"> was one example

[14:25:02.0000] <annevk>
(arguably that's better solved in another way now, but I think that's besides the point)

[14:25:03.0000] <KevinMarks>
the only issue is when you want some extra non-human-readable data

[14:26:00.0000] <Philip`>
zcorpan: Ah, I missed that - adding it now

[14:27:00.0000] <jgraham>
Philip`: What's the problem that requires the serializer in a different file?

[14:30:00.0000] <Philip`>
jgraham: It's mostly because html5lib takes ages to parse the spec, so it's quicker (during development) to just convert it to XML once and then I can repeatedly re-parse it later quickly

[14:31:00.0000] <Philip`>
During non-development that doesn't matter since people won't run the second script more than once, but if I use the html5lib minidom then it throws some exception when I try calling cloneNode

[14:31:01.0000] <jgraham>
Oh I see. So it's not a problem with html5lib (other than perf.)?

[14:32:00.0000] <KevinMarks>
annevk: here's something I wrote before:  http://microformats.org/discuss/mail/microformats-discuss/2006-October/006820.html

[14:32:01.0000] <jgraham>
Oh. Can you file the exception as a bug when you have a moment

[14:32:02.0000] <annevk>
I need some free weeks to implement this stuff in C

[14:32:03.0000] <KevinMarks>
which came out of here:  http://burningbird.net/learning-javascript/ajax-myth-busting/

[14:32:04.0000] <jgraham>
(the functionality in the script is very cool fwiw)

[14:33:00.0000] <annevk>
Philip`, could you "AddType text/plain .py"?

[14:35:00.0000] <annevk>
KevinMarks, interesting

[14:36:00.0000] <annevk>
KevinMarks, unfortunately I'm not that into mf

[14:37:00.0000] <annevk>
actually, I get it, nm

[14:37:01.0000] <Philip`>
jgraham: Okay, I'll try to find where the problem is and report it

[14:37:02.0000] <KevinMarks>
the basic need is having variables for script access, so having them in <script> makes sense

[14:37:03.0000] <Philip`>
annevk: Added, I think

[14:40:00.0000] <annevk>
cheers

[14:40:01.0000] <Philip`>
jgraham: Oh, just realised it's because I'm calling doc.cloneNode(True) where doc is the root object returned by the parser

[14:41:00.0000] <annevk>
KevinMarks, I don't think it's the most convenient for authors though

[14:41:01.0000] <Philip`>
(which seems to be a legitimate thing to do, I hope)

[14:42:00.0000] <annevk>
it would be pretty awesome if HTML5 was generated with tools it defines itself

[14:43:00.0000] <Philip`>
...and only when the document in doc has a doctype - so maybe that's what it can't clone

[14:44:00.0000] <KevinMarks>
putting JSON in attributes is much less convenient; at least by keeping it in script you have a decent chance of validating it, and being more readily backwards compatible

[14:44:01.0000] <jdandrea>
/me beams at the prospect of HTML5 generated w/self-defined tools

[14:45:00.0000] <annevk>
KevinMarks, I don't think your approach is more backwards compatible than the JSON in attribute value approach...

[14:45:01.0000] <annevk>
KevinMarks, but I agree that JSON in an attribute value is not very nice

[14:46:00.0000] <KevinMarks>
I think both that and the 'just add a big pile of our own attributes' approach that say, dojo, uses are both dodgy

[14:46:01.0000] <Philip`>
annevk: One problem with that approach is that it'll discourage changes to the HTML5 spec when those changes will break the tools that are used to write the spec :-)

[14:47:00.0000] <jdandrea>
Philip: Hmm. Didn't it work for C compilers though? "There's got to be a better way!"

[14:47:01.0000] <KevinMarks>
sounds like pypy

[14:48:00.0000] <jdandrea>
Ahh - http://codespeak.net/pypy/

[14:50:00.0000] <Philip`>
jgraham: http://code.google.com/p/html5lib/issues/detail?id=33 should be it

[14:50:01.0000] <jgraham>
Thnaks!

[14:57:00.0000] <Philip`>
Hmm, the annotation script doesn't work because it loads annotate-data.xml relative to the document, but that file doesn't exist relative to my version of the document

[14:58:00.0000] <annevk>
/me doesn't think the annotation is currently very useful

[14:59:00.0000] <jgraham>
Philip`: I don't see that exception here. Do you see something similar with:

[14:59:01.0000] <jgraham>
from xml.dom import minidom

[14:59:02.0000] <jgraham>
d = minidom.parseString("<body></body>")

[15:00:00.0000] <jgraham>
d.cloneNode(True)

[15:00:01.0000] <Philip`>
(Easy solution to annotation problem: just upload annotate-data.xml to my site, and don't forget to update it in the future)

[15:01:00.0000] <Philip`>
jgraham: That code works correctly with no exception

[15:01:01.0000] <annevk>
(in the future the whole thing should be in Hixie's publishing toolchain)

[15:01:02.0000] <Philip`>
(I'm using Python 2.5, in case that matters)

[15:03:00.0000] <jgraham>
Hmm. I tried 2.5 and 2.4 but didn't get an exception in either case. Can you just confirm what type of object doc is bound to in your example (i.e. what type of object are you getting from parser.parse('&lt;!DOCTYPE HTML&gt;'))

[15:04:00.0000] <Philip`>
parser.parse returns <xml.dom.minidom.Document instance at 0x00C71A08>

[15:05:00.0000] <Philip`>
(Also, this is with the 0.9 release download, rather than the latest from SVN)

[15:05:01.0000] <annevk>
oh, we usually fix the bugs after the release :)

[15:06:00.0000] <jgraham>
annevk: I can't see we'd have touched anything that affected DOM though.

[15:17:00.0000] <gsnedders>
ergh. more XHTML2 4ever people on fora :\

[15:17:01.0000] <othermaciej>
we really should make T-shirts that say "5 > 2"

[15:17:02.0000] <gsnedders>
totally.

[15:18:00.0000] <annevk>
lol

[15:18:01.0000] <gsnedders>
and put no mention of (X)HTML on them, so they aren't overly geeky

[15:18:02.0000] <gsnedders>
well, obviously geeky

[15:18:03.0000] <annevk>
5 > 2 is pretty geeky

[15:19:00.0000] <zcorpan>
/me wants such a t-shirt

[15:19:01.0000] <gsnedders>
but not obviously

[15:19:02.0000] <gsnedders>
to someone who doesn't know what it's about, it isn't.

[15:19:03.0000] <gsnedders>
to those who know, very.

[15:19:04.0000] <gsnedders>
discreetness is nice, though :)

[15:19:05.0000] <othermaciej>
I will try to get some art made and set up a Cafe Press store or something

[15:20:00.0000] <Lachy_>
/me will buy one :-)

[15:20:01.0000] <jdandrea>
/me will buy two. I mean five. :)

[15:21:00.0000] <hober>
or goodstorm.com -- their shirts seem higher quality than cafepress

[15:21:01.0000] <Lachy>
5 &gt; 2.0

[15:21:02.0000] <jdandrea>
Lachy: Better yet!

[15:21:03.0000] <gsnedders>
Lachy: but > doesn't need to be escaped!

[15:21:04.0000] <gsnedders>
why waste the bandwidth!

[15:21:05.0000] <gsnedders>
:P

[15:21:06.0000] <othermaciej>
Lachy: that's a little *too* geeky

[15:21:07.0000] <jdandrea>
2 &lt; 5

[15:21:08.0000] <jdandrea>
hehe

[15:21:09.0000] <gsnedders>
s/bandwidth/chest-width

[15:21:10.0000] <gsnedders>
:)

[15:21:11.0000] <gsnedders>
but yeah, that's too geeky.

[15:21:12.0000] <othermaciej>
hmmm

[15:21:13.0000] <gsnedders>
it needs to be discreet.

[15:22:00.0000] <othermaciej>
if we really want high quality, there's at least one place that makes their shirts based on American Apparel t-shirts

[15:22:01.0000] <othermaciej>
but I forget the name

[15:22:02.0000] <zcorpan>
"5 < 2.0" is nice

[15:22:03.0000] <jdandrea>
5 is the new 2? Hmm ...

[15:22:04.0000] <zcorpan>
er

[15:22:05.0000] <zcorpan>
"5 > 2.0"

[15:22:06.0000] <gsnedders>
/me can't imagine what people would think at school if it were obviously geeky

[15:23:00.0000] <gsnedders>
plenty people know I can touch type, but don't know to what extent I have to do with the web

[15:24:00.0000] <gsnedders>
how about… mmm… "Web 5.0"

[15:24:01.0000] <gsnedders>
:P

[15:24:02.0000] <KevinMarks>
5 > 2.0 is fun especially if we have them for next week

[15:24:03.0000] <othermaciej>
I like the minimal "5 > 2"

[15:24:04.0000] <othermaciej>
it's vague enough to be a little mysterious, even if you get it

[15:25:00.0000] <othermaciej>
what font should I use?

[15:25:01.0000] <othermaciej>
I am thinking it should be a font that is widely used on the web

[15:25:02.0000] <othermaciej>
but Times New Roman is kinda assy

[15:25:03.0000] <deltab>
Verdana?

[15:25:04.0000] <Lachy>
courier new

[15:25:05.0000] <KevinMarks>
all the 'web' fonts are

[15:25:06.0000] <hasather>
Monaco?

[15:25:07.0000] <gsnedders>
Helvetica?

[15:25:08.0000] <KevinMarks>
'cos they were done by MS on the cheap

[15:25:09.0000] <hober>
helfuckingvetica :)

[15:25:10.0000] <gsnedders>
Lucida Grande?

[15:25:11.0000] <Lachy>
comic sans ms

[15:26:00.0000] <zcorpan>
sans-serif

[15:26:01.0000] <KevinMarks>
Zapfino

[15:26:02.0000] <othermaciej>
Helvetica is probably the prettiest choice there

[15:26:03.0000] <zcorpan>
/me hides

[15:26:04.0000] <jgraham>
/me sees much of the web in Bitstream Vera

[15:26:05.0000] <jdandrea>
Zapf Dingbats. Let 'em decode it. hehe

[15:26:06.0000] <gsnedders>
othermaciej: and its common enough

[15:26:07.0000] <annevk>
comic sans ms +1

[15:26:08.0000] <gsnedders>
but comic sans…

[15:26:09.0000] <annevk>
although maybe we should have a pretty font for a pretty t-shirt...

[15:26:10.0000] <KevinMarks>
I set safari to Hoefler for serif and Gill Sans of sans

[15:26:11.0000] <gsnedders>
I'm not sure I could cope wearing comic sans…

[15:27:00.0000] <zcorpan>
georgia

[15:27:01.0000] <jdandrea>
Gill Sans is nice.

[15:27:02.0000] <gsnedders>
KevinMarks: how do you set serif and sans-serif seperatly?

[15:27:03.0000] <annevk>
Arial

[15:28:00.0000] <annevk>
I'm with Helvetica I think

[15:28:01.0000] <KevinMarks>
hm, maciej, did you hide the pref?

[15:28:02.0000] <KevinMarks>
I think it's still in the plist

[15:28:03.0000] <gsnedders>
com.apple.saf?

[15:29:00.0000] <gsnedders>
all I see is WebKitFixedFont and WebKitStandardFont

[15:29:01.0000] <othermaciej>
KevinMarks: it's only had prefs for standard and fixed-width font for some time

[15:29:02.0000] <KevinMarks>
dang, its gone

[15:32:00.0000] <gsnedders>
othermaciej: what's the default for sans-serif? Lucida Grande?

[15:32:01.0000] <jdandrea>
Helvetica's tried and true too. I could go for that. http://www.fonts.com/findfonts/mondosearchresults.htm?st=12&kid=Helvetica

[15:32:02.0000] <othermaciej>
gsnedders: probably Helvetica

[15:33:00.0000] <gsnedders>
othermaciej: one of the system fonts, I assume?

[15:34:00.0000] <KevinMarks>
Hoefler has non-lining numbers, which always looks cool imo

[15:34:01.0000] <othermaciej>
a lot of the commenters on <http://ajaxian.com/archives/proposal-for-the-w3c-to-adopt-html-5> must *clearly* be trolling

[15:37:00.0000] <Lachy>
/me wonders if the comment from Bill Gates is really him

[15:37:01.0000] <Hixie>
yeah i'm ignoring the comments after the last one i posted

[15:37:02.0000] <Hixie>
they're all just trolling or not interested in being informed

[15:38:00.0000] <Lachy>
it's clearly a troll, based on what he said

[15:38:01.0000] <jdandrea>
/me reads Hixie's comment (good response)

[15:38:02.0000] <gsnedders>
"Mozilla, Opera, and Apple should decrease the tolerance for invalid markup in their browsers" - is that even sane? :P

[15:39:00.0000] <annevk>
those people have no idea what they're talking about

[15:40:00.0000] <annevk>
most of them anyway

[15:40:01.0000] <Lachy>
good morning Hixie

[15:40:02.0000] <Hixie>
hi

[15:40:03.0000] <gsnedders>
http://codingforums.com/showpost.php?p=556210&postcount=26 – that's the forum post I was referring to earlier

[15:41:00.0000] <gsnedders>
maybe I should point him at liorean's article, as he is actually a supermoderator on that forum :P

[15:41:01.0000] <gsnedders>
it _used_ to be a very good intelligent forum

[15:42:00.0000] <Philip`>
Browsers won't want to do anything that makes the experience worse for users just because they're visiting invalid pages - but that shouldn't stop the browsers from searching the page for a webmaster email address and then sending anonymous hate mail to the author telling them to fix their markup. Users wouldn't know, so they'd be just as happy as before

[15:42:01.0000] <gsnedders>
but I've recently many times cited specifications and got argued that I'm wrong

[15:42:02.0000] <gsnedders>
:\

[15:43:00.0000] <gsnedders>
I'm a dumb teenager who thinks he knows it all, kthxbai.

[15:44:00.0000] <gsnedders>
oh, and please do say if you do get T-Shirts :)

[15:45:00.0000] <othermaciej>
I will let everyone know when/if there are t-shirts, need to do some other stuff for now but I shall try to get it set up tonight

[15:50:00.0000] <KevinMarks>
Having them for web 2.0 expo next week would be fun

[15:50:01.0000] <KevinMarks>
(starts on Sunday)

[15:51:00.0000] <annevk>
I'd love one for a presentation I do on HTML5 next Tuesday

[15:52:00.0000] <annevk>
but I suspect that might be difficult

[15:52:01.0000] <KevinMarks>
I'm doing one on Sunday

[15:52:02.0000] <annevk>
cool

[15:53:00.0000] <annevk>
Lachy is creating a wiki page where people can probably point to slides and such (if you intend to publish them)

[15:53:01.0000] <KevinMarks>
well, mine is on microformats, btu I can get an HTML5 reference in

[15:55:00.0000] <annevk>
ah ok

[15:55:01.0000] <annevk>
/me misunderstood

[16:02:00.0000] <Lachy>
I've added my presentation and Hixie's to the wiki page.  Are there any other publicly available presentations I should add?

[16:02:01.0000] <Lachy>
annevk, should I list yours in w3c-archive, even though it's member only?

[16:03:00.0000] <annevk>
I could make it public

[16:03:01.0000] <Lachy>
that would be good

[16:03:02.0000] <annevk>
it's not really about HTML5 though

[16:04:00.0000] <Lachy>
it's related to it a little, but I don't need to add it if you don't want

[16:07:00.0000] <Lachy>
http://wiki.whatwg.org/wiki/HTML5_Presentations

[16:09:00.0000] <zcorpan_>
/me will probably hold a presentation next month, in swedish

[16:11:00.0000] <zcorpan_>
at geekmeet

[16:13:00.0000] <annevk>
Lachy, http://www.w3.org/mid/op.tqm4d0o764w2qv@id-c0020

[16:13:01.0000] <annevk>
Lachy, http://lists.w3.org/Archives/Public/www-archive/2007Apr/att-0024/slides.xml has the slides

[16:16:00.0000] <Philip`>
"Not YAML" - that confused me until I realised it wasn't talking about the actual YAML (as in http://www.yaml.org/)

[16:17:00.0000] <annevk>
That's the great thing about using <abbr> in slides - nobody can ever tell what you meant until they look at the source code

[16:20:00.0000] <zcorpan_>
annevk: what's with xml:id?

[16:20:01.0000] <annevk>
did you notice the date?

[16:21:00.0000] <annevk>
March 2006

[16:21:01.0000] <zcorpan_>
ah

[16:21:02.0000] <zcorpan_>
ok

[16:21:03.0000] <annevk>
I was 19 back then and xml:id was new, cool and supported in Opera nightlies

[16:21:04.0000] <zcorpan_>
yup

[16:26:00.0000] <Hixie>
ok now i have to spec the playback rate stuff

[16:29:00.0000] <annevk>
Hixie, did you see the stuff Philip` made?

[16:29:01.0000] <annevk>
Hixie, would be nice if you could integrate that somehow in the spec build script

[16:30:00.0000] <Hixie>
ian⊙hc

[16:31:00.0000] <annevk>
Philip`, will you e-mail it?

[16:34:00.0000] <Hixie>
othermaciej: my concern with having a .defaultPlaybackRate that basically does nothing (but will be used by native ui), and having a .playbackRate for the current rate, and requiring .playbackRate = .defaultPlaybackRate to reset, is that nobody will use it

[16:34:01.0000] <Hixie>
they'll just have their own internal state

[16:34:02.0000] <othermaciej>
Hixie: I'll have to read what you wrote up

[16:34:03.0000] <othermaciej>
it's on the todo list

[16:34:04.0000] <Hixie>
and so if you set the speed to 2x in the ui, and then hit ffwd in the non-native ui, and then hit play, it'll go back to 1x

[16:34:05.0000] <Hixie>
i haven't written anything yet

[16:35:00.0000] <Hixie>
i'm trying to work out what to write

[16:35:01.0000] <othermaciej>
ah

[16:35:02.0000] <Hixie>
maybe we can force it by saying that sound will only play if defaultPlaybackRate = playbackRate

[16:35:03.0000] <Hixie>
then people will have to set both

[16:35:04.0000] <othermaciej>
so the idea is that defaultPlaybackRate is what you use when you hit play()

[16:36:00.0000] <othermaciej>
er

[16:36:01.0000] <othermaciej>
when you call play()

[16:36:02.0000] <Hixie>
yeah we can have play reset the rate as well, someone yesterday was against it though

[16:37:00.0000] <Hixie>
though they didn't give reasons

[16:37:01.0000] <annevk>
play(rate)

[16:37:02.0000] <othermaciej>
I think play() should set playbackRate to defaultPlaybackRate, this makes sense w/ the way controls work

[16:37:03.0000] <Hixie>
k

[16:37:04.0000] <Hixie>
i agree

[16:37:05.0000] <Hixie>
ok

[16:37:06.0000] <Hixie>
and also mute when the rates don't agree?

[16:38:00.0000] <Hixie>
or only mute when they don't agree for video?

[16:38:01.0000] <othermaciej>
FF of video might want to mute, but fast forward of audio likely wouldn't

[16:38:02.0000] <Lachy>
Hixie, no

[16:38:03.0000] <Hixie>
othermaciej: yeah

[16:38:04.0000] <Hixie>
maybe we should require ui to hit mute themselves

[16:38:05.0000] <Hixie>
though that would confuse other controllers maybe

[16:38:06.0000] <Hixie>
as all but one controller would be displaying the mute ui

[16:38:07.0000] <Lachy>
it depends on the reason they're using playback rate

[16:39:00.0000] <deltab>
other controllers?

[16:39:01.0000] <Hixie>
lachy: well there are two rates -- the default rate, which is the normal playback rate the user selected, and the current (temporary) rate, which, if different, is the rate used to ffwd or rew.

[16:39:02.0000] <Lachy>
the user may just want to speed it up to, e.g. watch a 20 min presention in just 10, and in that case audio is needed

[16:39:03.0000] <Hixie>
then they'd set the default rate to 2

[16:39:04.0000] <Lachy>
oh, ok.

[16:39:05.0000] <Hixie>
and the default rate would equal the rate

[16:40:00.0000] <Lachy>
I should go read the spec before I comment

[16:40:01.0000] <Hixie>
there is no spec

[16:40:02.0000] <Hixie>
yet

[16:40:03.0000] <Hixie>
trying to work out what it shouldbe

[16:40:04.0000] <Lachy>
ah, well, there's been heaps of stuff added to <video> since I last read it anyway

[16:40:05.0000] <Hixie>
true

[16:40:06.0000] <Hixie>
i wonder if we need a "mute while ffwding" flag

[16:41:00.0000] <Hixie>
/me goes to attend to biological needs like food while thinking about this

[16:41:01.0000] <deltab>
also whether it should pitch-shift

[16:41:02.0000] <ianloic>
hmm

[16:42:00.0000] <ianloic>
is it a good idea to put that much logic into <audio> & <video>?

[16:42:01.0000] <ianloic>
what if I want to use them to do more fancy multimedia things?

[16:43:00.0000] <KevinMarks>
pitch shifting is a useragent thing

[16:43:01.0000] <ianloic>
I should be able to build simple loop-based composition tools with just a little bit of html5, but tying the playback rate to play() would likely break that

[16:43:02.0000] <KevinMarks>
QT didn't used to do it, but it was added recently, and it helps a lot when you are doing 1.25x audio

[16:44:00.0000] <KevinMarks>
etc

[16:44:01.0000] <ianloic>
and there are so many different ways to pitch shift

[16:44:02.0000] <KevinMarks>
but requiring it is a bit harsh if you don't have the audio processing juice to do it

[16:44:03.0000] <ianloic>
if you don't do decent formant shifting a lot of content will sound really weird

[16:45:00.0000] <deltab>
KevinMarks: yeah, that was my concern

[16:45:01.0000] <ianloic>
so, what about user agents that can only play back one thing at once?

[16:46:00.0000] <ianloic>
does it make sense to track some kind of "media focus"?

[16:46:01.0000] <ianloic>
if I want to play each of the pieces of media on a page in order?

[16:46:02.0000] <KevinMarks>
you cna do it time domain on speech and have it work well, but music will sound weird

[16:47:00.0000] <KevinMarks>
that brings us back to sequencing, which I keep saying we need

[16:53:00.0000] <Lachy>
ha! Spammers are posting filthy jokes to the blog :-)

[16:54:00.0000] <othermaciej>
presenting audio above normal rate has lots of options

[16:54:01.0000] <Dashiva>
Lachy: You mean they talk about XHTML2?

[16:54:02.0000] <othermaciej>
you can do a pitch-distorting speedup, you can try to do a pitch-preserving speedup, you can stutter like many consumer CD players do, you can mute

[16:55:00.0000] <othermaciej>
not sure the spec needs to say which happens

[16:55:01.0000] <Lachy>
Dashiva, no, but that would be funny also

[16:56:00.0000] <Philip`>
annevk, Hixie: Yep - I've emailed it now

[16:57:00.0000] <deltab>
othermaciej: if your game, music composer, etc. relies on the pitch changing, the others aren't acceptable

[16:58:00.0000] <ianloic>
I would suggest that exposing fast-forward functionality independent of playback rate would be valuable

[16:58:01.0000] <othermaciej>
deltab: I'm not sure it is reasonable to expect user agents to support all 4 strategies

[16:59:00.0000] <deltab>
me neither, but the issue should be considered, imho

[16:59:01.0000] <zcorpan_>
Lachy: is the spam from bots or from humans?


2007-04-12
[17:00:00.0000] <Lachy>
not sure

[17:00:01.0000] <Lachy>
there was no URL, so I'm not sure what the purpose of it was

[17:00:02.0000] <ianloic>
are the jokes funny?

[17:01:00.0000] <ianloic>
the purpose could be to make you laugh

[17:01:01.0000] <Lachy>
I can post it here if you like

[17:01:02.0000] <Lachy>
it's not too rude

[17:02:00.0000] <Lachy>
Hello blog.whatwg.org! 

[17:02:01.0000] <Lachy>
A woman went to her doctor and complained that her husband was 300% impotent. 

[17:02:02.0000] <Lachy>
The doctor replied, "I'm not sure I understand what you mean." 

[17:02:03.0000] <Lachy>
She answered, "Well, the first 100% you can imagine. In addition, he burned his tongue and broke his finger!" 

[17:02:04.0000] <ianloic>
hah

[17:02:05.0000] <ianloic>
no buy viagra link though?

[17:02:06.0000] <Lachy>
nope

[17:03:00.0000] <othermaciej>
so would 19 USD be too much for a 5 > 2 t-shirt?

[17:03:01.0000] <zcorpan_>
not really

[17:04:00.0000] <othermaciej>
maybe I can get it cheaper by eliminating profit for me (which I don't want)

[17:04:01.0000] <Lachy>
that's about $AU23 + postage, which is reasonable

[17:04:02.0000] <othermaciej>
should I be optimistic and have a women's tee design as well?

[17:04:03.0000] <zcorpan_>
molly might want one :)

[17:05:00.0000] <Philip`>
(Do you mean you don't to eliminate profit for you, or that you don't want profit for you?)

[17:05:01.0000] <othermaciej>
I don't want profit for me

[17:05:02.0000] <zcorpan_>
how much cheaper would it be?

[17:06:00.0000] <Lachy>
does the seller make any profit from cafepress sales?

[17:06:01.0000] <Philip`>
Okay, I assumed that was most likely :-)

[17:07:00.0000] <Philip`>
You could always use any profit to buy an extra T-shirt and send it to the XHTML2 guys

[17:07:01.0000] <Dashiva>
I haven't heard much good about cafepress (but I've never used them myself)

[17:08:00.0000] <Lachy>
or just donate the profit to a worthwhile charity

[17:08:01.0000] <Lachy>
like the FSF or Mozilla Foundation

[17:10:00.0000] <Dashiva>
Does the whatwg have an economy, or is everything needed (like hosting) provided by the members?

[17:10:01.0000] <Lachy>
Hixie pays for the hosting

[17:10:02.0000] <ianloic>
he's trading hosting for stock though

[17:10:03.0000] <ianloic>
when WHATWG get's acquired he'll be rich

[17:11:00.0000] <ianloic>
(where did that apostrophe come from?)

[17:11:01.0000] <Dashiva>
From xhtml2

[17:11:02.0000] <Dashiva>
It's all their fault

[17:12:00.0000] <KevinMarks>
definitely do womens shirts

[17:12:01.0000] <hasather>
Dashiva: Yea, I've suspected that they've been tampering with this channel for a long time now

[17:12:02.0000] <KevinMarks>
the microformats strategy of equal numbers of male and female shirts definitely paid off

[17:12:03.0000] <Lachy>
the XHTML2 WG have their own IRC channel now

[17:13:00.0000] <Dashiva>
copycats

[17:13:01.0000] <Lachy>
irc://irc.w3.org:6665/xhtml

[17:13:02.0000] <Lachy>
/me joined in earlier

[17:13:03.0000] <Hixie>
is it logged?

[17:13:04.0000] <Lachy>
not yet

[17:13:05.0000] <Hixie>
Philip`: yt?

[17:13:06.0000] <Lachy>
but it's not as if there's heaps of discussion going on there

[17:13:07.0000] <hasather>
One would think that they can keep their apostrophes in their own channel

[17:14:00.0000] <KevinMarks>
I didn't think xhtml had &apos;

[17:14:01.0000] <Dashiva>
Maybe they're making a subtle comment about html allowing the absense of quoting in non-spaced attributes

[17:14:02.0000] <hasather>
KevinMarks: it has, HTML hasn't

[17:14:03.0000] <Lachy>
krijnh should join the XHTML2 channel and put logs up on his site too

[17:15:00.0000] <Philip`>
Hixie: Huh?

[17:15:01.0000] <KevinMarks>
aha

[17:15:02.0000] <zcorpan_>
hasather: html5 has

[17:15:03.0000] <hasather>
ahh, yea

[17:15:04.0000] <Hixie>
Philip`: (yt = are You There)

[17:15:05.0000] <Hixie>
Philip`: these scripts are awesome

[17:15:06.0000] <Philip`>
Oh, right

[17:15:07.0000] <Dashiva>
I usually go for 'syn' or 'ping' rather than 'yt'

[17:15:08.0000] <Philip`>
Yes, I'm here :-)

[17:16:00.0000] <Hixie>
Philip`: do you mind licensing them under the Apache License 2.0 so that i can check them into http://code.google.com/p/html5/ ?

[17:16:01.0000] <Lachy>
what's 'syn'?

[17:16:02.0000] <hasather>
synchronize

[17:16:03.0000] <Hixie>
Lachy: tcp connection packet

[17:16:04.0000] <Hixie>
response would be syn/ack

[17:16:05.0000] <hasather>
and then ack

[17:16:06.0000] <Hixie>
but if someone else says "ack" or "syn/ack" first, the connection breaks

[17:16:07.0000] <Hixie>
so i prefer yt :-P

[17:16:08.0000] <Lachy>
Hixie, why does that use the apache licence instead of MIT?

[17:17:00.0000] <Philip`>
Hixie: I'd be happy with that

[17:17:01.0000] <Hixie>
Lachy: google prefers the apache lincense

[17:17:02.0000] <Hixie>
Philip`: cool

[17:17:03.0000] <Hixie>
Philip`: do you have a gmail account?

[17:17:04.0000] <Philip`>
Hixie: I do

[17:17:05.0000] <Lachy>
oh, I wonder why.  It's GPL incompatible

[17:18:00.0000] <Hixie>
probably because the people who made the decision is on the apache board

[17:18:01.0000] <Hixie>
but that's another story :-)

[17:19:00.0000] <Hixie>
Philip`: added you to the members of the group

[17:19:01.0000] <Lachy>
/me wonders when he got added to that project. 

[17:19:02.0000] <Hixie>
Philip`: and checked the code in

[17:20:00.0000] <Hixie>
Lachy: i added everyone whose gmail address i knew who i thought could even remotely be helpful

[17:20:01.0000] <Hixie>
:-)

[17:20:02.0000] <Hixie>
Philip`: how do i run the script? (where do i have to put html5lib?)

[17:20:03.0000] <Lachy>
when was the project started?

[17:21:00.0000] <Hixie>
a while back

[17:21:01.0000] <Hixie>
how do you make python use the arguments passed to it?

[17:21:02.0000] <Lachy>
would have been nice if I were notified about it :-)

[17:22:00.0000] <Philip`>
Hixie: Okay, thanks!

[17:23:00.0000] <Hixie>
hm i'm gonna have to hack this to work on the site

[17:23:01.0000] <Philip`>
Hmm, I think you could rename the html5lib project's "src" directory to "html5lib" and stick it in the same place as the scripts that use it (if you can't just install it normally)

[17:23:02.0000] <Hixie>
since i don't need things like the non-relative uris

[17:23:03.0000] <Hixie>
Philip`: k

[17:23:04.0000] <KevinMarks>
what are you hosting the python with?

[17:24:00.0000] <Philip`>
Hopefully you've got Python 2.5, else some bits will probably need to be rewritten

[17:24:01.0000] <hasather>
Hixie: arguments is in a list: sys.argv

[17:24:02.0000] <hasather>
sys.argv[0] is the name of the program

[17:25:00.0000] <Hixie>
cool

[17:25:01.0000] <hasather>
and you have to import sys of course: import sys

[17:25:02.0000] <Hixie>
i have 22.2, 2.3, and .4 it seems

[17:26:00.0000] <Philip`>
Oh - if you can extrapolate those versions, it'll work fine

[17:26:01.0000] <Hixie>
er, 2.2, 2.3, and 2.4

[17:26:02.0000] <Hixie>
not sure why that two ended up in the wrong place

[17:27:00.0000] <Philip`>
Otherwise it'll probably complain obviously about "x for y in z" type stuff

[17:27:01.0000] <Hixie>
ah

[17:27:02.0000] <Hixie>
well we'll see

[17:29:00.0000] <deltab>
someone's asking if it's necessary to be an invited expert to get on the HTML WG mailing list

[17:30:00.0000] <deltab>
I'm not sure of the answer

[17:30:01.0000] <Philip`>
You can read the archives and post without being a member at all

[17:30:02.0000] <deltab>
thanks

[17:30:03.0000] <zcorpan_>
you can't subscribe though, aiui

[17:31:00.0000] <Philip`>
(Responding from the archives messes up the threading a bit, though)

[17:31:01.0000] <Hixie>
you can become an invited expert really really easily

[17:31:02.0000] <Hixie>
just by following the steps to join the list basically

[17:31:03.0000] <Hixie>
you don't have to be "invited" nor an "expect"

[17:31:04.0000] <Hixie>
expert

[17:31:05.0000] <Lachy>
Philip`, the reply link in the archive has the right fields in it to keep threading working

[17:32:00.0000] <Lachy>
you just need to be using a decent mail client

[17:32:01.0000] <Lachy>
I think it's really weird that you can post without agreeing to the patent policy, but can't subscribe

[17:33:00.0000] <Philip`>
I think it nearly worked, except the outgoing message had "References: %3Cp06240609c242b8eadd32@[192.168.0.101]%3E" and I think the %3C/%3E shouldn't be there

[17:33:01.0000] <Lachy>
oh, they should have been unescaped

[17:34:00.0000] <Philip`>
The mailto: link in the archives seems to double-escape them - it has "mailto:...%253C..."

[17:34:01.0000] <Lachy>
oh, then the archive is broken and should be patched

[17:35:00.0000] <othermaciej>
http://five-gt-two.spreadshirt.com/

[17:35:01.0000] <othermaciej>
I managed to turn off passing myself

[17:35:02.0000] <othermaciej>
does the design seem ok?

[17:35:03.0000] <zcorpan_>
arial?

[17:35:04.0000] <zcorpan_>
or is it helvetica?

[17:36:00.0000] <othermaciej>
that's arial, since they had support for that built-in and saved me the need to make my own EPS artwork with helvetica

[17:36:01.0000] <zcorpan_>
is it on both the front and the back?

[17:37:00.0000] <Lachy>
zcorpan_, there's preview photos that show all sides

[17:37:01.0000] <othermaciej>
it's only on the front I think

[17:38:00.0000] <zcorpan_>
it is

[17:39:00.0000] <zcorpan_>
looks good

[17:39:01.0000] <zcorpan_>
is it the whatwg green?

[17:39:02.0000] <Lachy>
doesn't look like it is

[17:40:00.0000] <zcorpan_>
would be nice if it was

[17:40:01.0000] <Lachy>
oh, maybe it should have been "5 > 2?"

[17:40:02.0000] <othermaciej>
there's a limited number of colors available

[17:41:00.0000] <zcorpan_>
Lachy: why the "?"?

[17:41:01.0000] <Lachy>
the WHATWG logo

[17:41:02.0000] <Lachy>
I don't there theres a ? in a circle available

[17:41:03.0000] <deltab>
are Unicode combining characters supported?

[17:42:00.0000] <Lachy>
?⃝

[17:42:01.0000] <Lachy>
they are in XChat

[17:42:02.0000] <Philip`>
That site looks quite like it's got unescaped user input in its title element. I think some should make a T-shirt called </title><style>*{text-decoration:blink}</style>

[17:42:03.0000] <Philip`>
*someone

[17:43:00.0000] <deltab>
there's one with  <body>  on the front and  </body>  on the back

[17:44:00.0000] <Lachy>
yeah, I've been thinking about getting that one from Remo with my voucher that I won

[17:45:00.0000] <Lachy>
http://remo.com.au/online/tshirts_item.cfm?PLU=1104

[17:48:00.0000] <othermaciej>
I think I will leave off the whatwg logo because (a) I am lazy and (b) it's more subtle this way

[17:48:01.0000] <Philip`>
Hixie: Sorry, Freenode doesn't like me sending private messages while unregistered - I'll try to fix that, but in the meantime: That's http://code.google.com/p/html5lib/issues/detail?id=33

[17:48:02.0000] <Hixie>
aah

[17:48:03.0000] <Hixie>
cood

[17:48:04.0000] <Hixie>
cool even

[17:48:05.0000] <Hixie>
ok who can i blame who works on html5lib :-)

[17:49:00.0000] <Philip`>
I believe jgraham said he didn't get that error, but I'm not sure what the difference is

[17:49:01.0000] <Lachy>
/me is wondering if he should explain 5>2 to people asking on twitter

[17:51:00.0000] <Hixie>
Philip`: http://mail.python.org/pipermail/xml-sig/2004-May/010252.html seems relevant

[17:53:00.0000] <Hixie>
wow, html5lib takes a long time to parse the spec

[17:54:00.0000] <Lachy>
I wonder how it could be optomised?

[17:54:01.0000] <deltab>
even Czech C code takes a couple of seconds to handle that much

[17:55:00.0000] <Lachy>
how long did it take?

[17:56:00.0000] <Lachy>
It's funny how, in response to the lack of support for data: URIs in IE, people propose new ideas that are equally unsupported

[17:56:01.0000] <Hixie>
(about 19s on this machine)

[17:59:00.0000] <Hixie>
and yes, this is a recurring pattern

[17:59:01.0000] <Hixie>
it baffles me

[17:59:02.0000] <Philip`>
I think this works around that html5lib problem:

[17:59:03.0000] <Philip`>
    import xml.dom.minidom

[17:59:04.0000] <Philip`>
    import html5lib

[17:59:05.0000] <Philip`>
    parser = html5lib.html5parser.HTMLParser(tree=html5lib.treebuilders.dom.TreeBuilder)

[17:59:06.0000] <Philip`>
    doc = parser.parse(open('current-work'), encoding='utf-8')

[17:59:07.0000] <Philip`>
    doc.removeChild(doc.firstChild) # remove the doctype, else html5lib's doc.cloneNode dies

[17:59:08.0000] <Philip`>
    head = doc.documentElement.getElementsByTagName('head')[0]

[17:59:09.0000] <Philip`>
    head.insertBefore(doc.createElement('meta'), head.firstChild).setAttribute('charset', 'UTF-8')

[17:59:10.0000] <Philip`>
...as a replacement for the top of the file

[17:59:11.0000] <Philip`>
Er, then you can get rid of that xml.dom.minidom line too

[17:59:12.0000] <Hixie>
oh the problem is the doctype?

[17:59:13.0000] <othermaciej>
Lachy: asking on twitter where?

[17:59:14.0000] <Philip`>
I think so - the problem seems to be triggered when you clone a node which has a doctype thingy as a child

[17:59:15.0000] <Lachy>
othermaciej, http://twitter.com/DmitryBaranovskiy

[17:59:16.0000] <Lachy>
I decided to respond

[17:59:17.0000] <Philip`>
(You can clone the xml.dom.minidom.DocumentType itself, but not a xml.dom.minidom.Document which contains it)

[17:59:18.0000] <Lachy>
http://twitter.com/Lachy

[18:00:00.0000] <Hixie>
Philip`: must be a bug in minidom, the TreeBuilder code looks correct to me

[18:00:01.0000] <Hixie>
anyway removing the doctype is fine

[18:01:00.0000] <othermaciej>
dammit, do I have to be on Twitter now to be cool?

[18:01:01.0000] <Hixie>
no

[18:02:00.0000] <Philip`>
The doctype gets added back in later, via string concatenation in html5Serializer because I couldn't be bothered to figure out the proper way of doing it...

[18:02:01.0000] <Hixie>
though if you are, make sure to add "WHATWG" as your friend

[18:02:02.0000] <Hixie>
the whatwg spec checkin script automatically twitters the checkin message

[18:02:03.0000] <Hixie>
Philip`: yeah

[18:02:04.0000] <Lachy>
othermaciej, being on twitter won't make you cool, but it's where all the cool kids hang out these days :-)

[18:05:00.0000] <ianloic>
twitter was cool a year ago

[18:05:01.0000] <ianloic>
when it was twttr

[18:05:02.0000] <Lachy>
/me only found out about twitter a few weeks ago

[18:05:03.0000] <ianloic>
its definitely one of those technologies that falls apart when too many people start using it

[18:06:00.0000] <ianloic>
it jumped the shark when it got a short-code

[18:06:01.0000] <othermaciej>
ianloic: so you liked their early stuff?

[18:07:00.0000] <ianloic>
othermaciej, I liked it when it was just me and a few friends :)

[18:07:01.0000] <KevinMarks>
so L is 42" chest?

[18:08:00.0000] <Hixie>
well that was weird

[18:08:01.0000] <zcorpan_>
/me ordered a white one

[18:08:02.0000] <Hixie>
i saw that twitter had spoken, went to have a look, and it was my script twittering

[18:08:03.0000] <Hixie>
how confusing

[18:09:00.0000] <Lachy>
Hixie, I thought it was supposed to post the description from the SVN check in, just just a generic "updated the spec... " message

[18:09:01.0000] <Hixie>
two scripts

[18:09:02.0000] <othermaciej>
KevinMarks: they are American Apparel sizes, might be size info on the AA site

[18:09:03.0000] <Hixie>
./.update.sh says the message you just got

[18:09:04.0000] <Hixie>
./.commit.pl says the checkin message

[18:10:00.0000] <Lachy>
does update.sh generate the spec from source?

[18:10:01.0000] <othermaciej>
KevinMarks: L is 42-44" chest

[18:10:02.0000] <Philip`>
The link to the multiple-page version is wrong

[18:10:03.0000] <Hixie>
yes

[18:10:04.0000] <Philip`>
("extract" probably isn't a great name - I'm not sure why I ever called it that)

[18:10:05.0000] <Hixie>
what should we call it?

[18:10:06.0000] <Philip`>
Also the annotations are missing, since it'll be looking for annotate-data.xml in that directory

[18:11:00.0000] <Hixie>
yeah just fixed that

[18:11:01.0000] <Hixie>
(not updated yet)

[18:11:02.0000] <zcorpan_>
M2 stopped working when i upgraded to opera 9.20. the error console keeps poping up saying "Error from server: -ERR bad command g1pf2579637muf. Did you change your password?"

[18:11:03.0000] <Philip`>
Maybe something like "multi"?

[18:12:00.0000] <Hixie>
i just independently called it multipage

[18:12:01.0000] <Philip`>
Could the Twitter message link to the multipage version, so unsuspecting link-followers don't have their browser explode?

[18:13:00.0000] <Hixie>
sure

[18:16:00.0000] <Hixie>
oops

[18:16:01.0000] <Hixie>
misfire

[18:16:02.0000] <Hixie>
let's try again

[18:16:03.0000] <Lachy>
hmm. So far there have been no other nominees for spec editor in the HTMLWG, which seems a bit unfair if there is to be an election

[18:17:00.0000] <Hixie>
we're having an election?

[18:17:01.0000] <Lachy>
don't know

[18:17:02.0000] <Hixie>
as far as i can tell me being editor was part of the package of getting the whatwg spec

[18:17:03.0000] <Lachy>
that's up to the chairs, I guess

[18:19:00.0000] <Lachy>
just the way you keep saying things like "If the HTML WG adopts the draft and me as editor" makes it seem like there's a remote possibility for someone else

[18:19:01.0000] <Hixie>
well i don't want to presume

[18:20:00.0000] <Lachy>
I know, but I think it's a fairly safe assumption

[18:24:00.0000] <Hixie>
Philip`, your script is awesometastic

[18:24:01.0000] <Philip`>
Whoops, I think the single-page spec got Opera stuck in an infinite loop

[18:24:02.0000] <othermaciej>
so far, I don't think anyone else has been nominated as editor

[18:25:00.0000] <othermaciej>
or volunteered

[18:25:01.0000] <deltab>
could the twitter message include the log comment?

[18:25:02.0000] <Hixie>
it does, when it's a checkin

[18:25:03.0000] <Hixie>
the messages you got weren't checkins

[18:25:04.0000] <deltab>
ah

[18:26:00.0000] <Philip`>
(It's reading the length of cache4\opr11VRW.htm twenty five thousand times per second, for the past few minutes, which I think is not intentional behaviour...)

[18:27:00.0000] <deltab>
it doesn't believe how big the file is :-)

[18:28:00.0000] <Philip`>
Hixie: Looks good to me now :-)

[18:28:01.0000] <Hixie>
:-)

[18:29:00.0000] <Hixie>
feel free to post to the html list telling people your script has been incorporated into the whatwg pipeline

[18:29:01.0000] <Lachy>
post it to both whatwg and public-html so everyone knows

[18:32:00.0000] <Philip`>
Looks like the tabs/spaces got a bit mangled in the version in SVN - I really should try to remember how to set my editor to do spaces all the time...

[18:32:01.0000] <Hixie>
i changed a bunch to spaces

[18:32:02.0000] <Hixie>
maybe not all

[18:32:03.0000] <Hixie>
so that was probably my fault

[18:33:00.0000] <Philip`>
Only the bits at the top got changed - the rest are still tabs, so Python might get a bit confused

[18:36:00.0000] <Hixie>
seems to work ok :-)

[18:50:00.0000] <Philip`>
Hixie: Do you delete the old multipage/*.html files before adding the new ones, so that stale links will become 404s instead of giving out-of-date content?

[18:50:01.0000] <Hixie>
yes

[18:53:00.0000] <Philip`>
Ah, good

[18:54:00.0000] <Philip`>
(That's the step that I always forget, and it always confuses me about why I'm not seeing any changes I've made...)

[18:54:01.0000] <Hixie>
heh

[19:03:00.0000] <Lachy>
why would you need to delete old versions?  Wouldn't the updated versions just replace them, keeping the same file names?\

[19:04:00.0000] <Philip`>
Sometimes the section IDs will change, so the filenames will change

[19:04:01.0000] <Lachy>
ah, but changing IDs and filenames breaks links

[19:06:00.0000] <Philip`>
Yep, and it's better not to leave dead link targets around when they're not being updated by the script any more

[19:07:00.0000] <Lachy>
yeah, ideally links shouldn't change, though that's probably unavoidable with the spec being edited so much

[19:09:00.0000] <Philip`>
It's using the IDs of the <hx> elements, which I believe are generated automatically from the textual content of the headings, so just changing a heading will break the links

[19:18:00.0000] <othermaciej>
HTML5 proposal made it to daringfireball: http://daringfireball.net/linked/2007/april#wed-11-html5

[19:23:00.0000] <Dashiva>
- The correct answer to this offer is “Yes.”

[19:24:00.0000] <Dashiva>
That's cheating!

[19:24:01.0000] <Hixie>
i love how even AFTER philip wrote the script to split the spec and i merged it with the pipeline, people are still asking for the spec to be split

[19:24:02.0000] <Hixie>
maybe, just maybe, if they did it themselves when they wanted it like philip did, we wouldn't have had to wait so long to have a split spec

[19:25:00.0000] <Dashiva>
The w3c doesn't stain its hands with implementations ;)

[19:27:00.0000] <Hixie>
that particular guy was complaining about this in the whatwg list a few months ago

[19:28:00.0000] <Lachy>
Hixie, Mike Schinkel?

[19:28:01.0000] <Hixie>
yeah

[19:28:02.0000] <Lachy>
/me is ignoring him

[19:28:03.0000] <Philip`>
It's not like it takes that much longer to implement than to complain about - it's just using someone else's code to do the hard stuff like parsing HTML, then twiddling bits of DOM

[19:29:00.0000] <Hixie>
right, dinner

[19:29:01.0000] <Hixie>
bbl

[19:42:00.0000] <Philip`>
(It's still failing to load annotate-data.xml - it'd probably be nice to either make that file available, or remove the script that tries to include it on every page (via the two page_body.appendChild(script.cloneNode(True)) lines))

[19:56:00.0000] <othermaciej>
I love that people are now taking the design principles document as having some kind of substantive force when we just kinda made it up

[19:56:01.0000] <Philip`>
(Just noticed one spec change that will break the page links: s/Auxillary/Auxiliary/)

[20:05:00.0000] <Hixie>
i thought annotate-data.xml was available

[20:05:01.0000] <Hixie>
what did i break

[20:05:02.0000] <Hixie>
oh i know

[20:05:03.0000] <Hixie>
there, should work now

[22:36:00.0000] <Hixie>
vote: addCuePoint() or addCueMark()?

[22:36:01.0000] <Hixie>
or something else altogether

[22:37:00.0000] <othermaciej>
brain not functional enough to have opinion

[22:38:00.0000] <Hixie>
also should it take an EventListener, a TimeoutHandler, or a Function (where Function is something I'd have to define)

[22:39:00.0000] <othermaciej>
probably not an EventListener

[22:39:01.0000] <othermaciej>
how is TimeoutHandler defined right now?

[22:40:00.0000] <Hixie>
http://www.whatwg.org/specs/web-apps/current-work/multipage/section-timers.html#timeouthandler

[22:40:01.0000] <othermaciej>
probably a no-arg function is fine, since you can check currentTime

[22:40:02.0000] <Hixie>
k

[22:40:03.0000] <othermaciej>
TimeoutHandler would be ok I guess, though the extra args thing that setTimeout does is a little weird

[22:43:00.0000] <Hixie>
or addCuePoint() could just cause an event to be fired

[22:43:01.0000] <Hixie>
but that seems silly

[22:43:02.0000] <Hixie>
from an authoring standpoint

[22:44:00.0000] <othermaciej>
yeah

[23:15:00.0000] <Hixie>
hmm

[23:15:01.0000] <Hixie>
are the update twitters useful?

[23:15:02.0000] <Hixie>
I'm thinking maybe not

[23:15:03.0000] <Hixie>
the checkin twitters presumably are

[23:15:04.0000] <Lachy>
Hixie, no, only the checkins

[23:15:05.0000] <Hixie>
ok, i'll dump them then

[23:16:00.0000] <Hixie>
i've also made the checkin script not bother twittering for purely editorial stuff

[23:17:00.0000] <mpt>
Hixie, http://googlefight.com/index.php?lang=en_GB&word1=%22cue+point%22&word2=%22cue+mark%22

[23:19:00.0000] <Hixie>
good good

[23:21:00.0000] <Hixie>
http://www.whatwg.org/specs/web-apps/current-work/multipage/section-video.html#cue-points

[23:30:00.0000] <Lachy>
Hixie, thanks for setting up whatwg.org/html5 - that's much easier to type!

[23:30:01.0000] <Hixie>
:-)

[23:31:00.0000] <Lachy>
 /wf2 might be useful too for web forms

[23:31:01.0000] <Hixie>
added, but i don't expect to do anything to wf2 until we merge the two

[23:32:00.0000] <Lachy>
yeah, but just incase I need to look somthing up in it

[23:50:00.0000] <Hixie>
wow, i'm down to only two things to do for media

[23:50:01.0000] <Hixie>
namely, try to work out whether video.error should be an object, and adding a security section.

[23:53:00.0000] <othermaciej>
as soon as you are done, I'll likely have to re-review comparing to the Apple version

[23:53:01.0000] <othermaciej>
and bring up some of the issues from our leftovers section

[23:53:02.0000] <othermaciej>
like (whee) metadata

[23:53:03.0000] <Hixie>
you can start now if you want

[23:54:00.0000] <Hixie>
the two things i have left are working out what to do with error objects (do you want an object or is the code enough?) and the security section, which i'll do last

[23:54:01.0000] <Hixie>
since i need to make sure i've taken care of security issues before i can write about how i've taken care of them

[23:54:02.0000] <Hixie>
and i can't take care of them before i've finished adding features

[00:02:00.0000] <KevinMarks>
QT can give you a nice long list of error codes...

[00:03:00.0000] <othermaciej>
I think an error object is a better idea

[00:03:01.0000] <othermaciej>
more reasonable to extend, and there is little value to keeping the error representation lightweight

[00:22:00.0000] <Hixie>
ok, it's an object now.

[00:25:00.0000] <Hixie>
ok, also added a dummy section for security

[00:25:01.0000] <Hixie>
so that's it as far as i'm concerned for v2 of the media object

[00:28:00.0000] <othermaciej>
ok, I'm planning time for review

[00:28:01.0000] <othermaciej>
though I also need to squeeze some coding time in

[00:29:00.0000] <Hixie>
:-)

[00:30:00.0000] <Hixie>
if you look at the <!----> comments in http://www.whatwg.org/specs/web-apps/current-work/multipage/section-video.html#media (search for 'v3') you'll see what _isn't_ included in this version yet

[00:30:01.0000] <othermaciej>
I need to hassle the Apple folks who were supposed write up accessibility

[00:30:02.0000] <othermaciej>
and I have a speclet bit to write up myself

[00:30:03.0000] <Hixie>
(which includes captions and metadata)

[00:35:00.0000] <KevinMarks>
chapters?

[00:35:01.0000] <Hixie>
that's part of metadata

[00:41:00.0000] <KevinMarks>
jolly good

[00:47:00.0000] <Hixie>
/me kills the Audio() API and replaces it with HTMLAudioElement

[00:56:00.0000] <Hixie>
removed an entire section.

[00:57:00.0000] <Hixie>
yum.

[01:22:00.0000] <hsivonen>
Lachy: Re: Apace vs. MIT: the Apache License 2.0 has an explicit patent grant. The MIT license does not. This is both the reason to prefer the Apache license and, due to the GPLv2 compatibility issue, prefer the MIT license

[01:32:00.0000] <Lachy>
hsivonen, is the Apache licence still incompatible with GPLv3?

[01:34:00.0000] <othermaciej>
how is the GPL incompatible w/ a license including a patent grant?

[01:34:01.0000] <othermaciej>
Hixie: no 'new Audio' API for compatibitlity?

[01:34:02.0000] <othermaciej>
er

[01:34:03.0000] <othermaciej>
for convenience?

[01:34:04.0000] <Lachy>
othermaciej, http://www.gnu.org/philosophy/license-list.html lists it as incompatible

[01:34:05.0000] <Hixie>
othermaciej: yeah, it's there

[01:35:00.0000] <Lachy>
I think it's because GPL requires no other restrictions to be imposed beyond those in GPL

[01:44:00.0000] <hsivonen>
othermaciej: GPLv2 isn't compatible with licenses that have an explicit patent grant

[01:45:00.0000] <hsivonen>
othermaciej: oh. I misread you. What Lachy said.

[01:46:00.0000] <hsivonen>
Lachy: my understanding is that it is a goal to draft GPLv3 so that Apache License 2.0 is compatible. I don't know the current status of that compat effort.

[01:46:01.0000] <Lachy>
ok

[02:31:00.0000] <Hixie>
i love "[discussion of how public WGs will and should operate]"

[02:31:01.0000] <Hixie>
there's some irony to the fact that the discussion of how a public group should operate wasn't minuted.

[02:32:00.0000] <Lachy>
:-)

[02:37:00.0000] <a-ja>
hi all.    <aside> is supposed to allow a <header>, isn't it?

[02:38:00.0000] <hsivonen>
a-ja: it isn't

[02:38:01.0000] <hsivonen>
doh

[02:38:02.0000] <hsivonen>
yes. header can be a descendant of aside

[02:38:03.0000] <Lachy>
wow, "By the way, the next XHTML 1.* issue will be number 10,000!".  With that many issues, you'd think they'd have fixed something ;-/

[02:39:00.0000] <hsivonen>
but a aside cannot be an descendant of header

[02:39:01.0000] <hsivonen>
/me should wake up some more before misreading more questions

[02:39:02.0000] <a-ja>
hsivonen: yep...that's what i'd expect

[02:40:00.0000] <a-ja>
heh...that too :)

[02:42:00.0000] <a-ja>
/me should get some shuteye before asking any more....but what the heck

[02:43:00.0000] <a-ja>
hsivonen: you'll fixup the conformance checker to allow header descendant of aside?

[02:47:00.0000] <hsivonen>
a-ja: it doesn't now?

[02:48:00.0000] <hsivonen>
a-ja: did you mix inline and <header>?

[02:48:01.0000] <a-ja>
nope....style descendant seems to be ok though

[02:49:00.0000] <hsivonen>
  aside.inner =

[02:49:01.0000] <hsivonen>
    (  style.elem*

[02:49:02.0000] <hsivonen>
    ,  common.inner.bimorphic 

[02:49:03.0000] <hsivonen>
    )

[02:49:04.0000] <hsivonen>
a-ja: test case?

[02:49:05.0000] <a-ja>
i have a <p> after the header

[02:50:00.0000] <a-ja>
http://a-ja.net/html5-sections.html

[02:51:00.0000] <annevk>
you are overdoing the <header> thing though

[02:51:01.0000] <annevk>
o sorry, it's a test :)

[02:51:02.0000] <annevk>
/me should wake up too

[02:52:00.0000] <hsivonen>
a-ja: you wrote heading instead of header

[02:53:00.0000] <a-ja>
doh!

[02:54:00.0000] <hsivonen>
http://hsivonen.iki.fi/validator/html5/?doc=http://hsivonen.iki.fi/test/header-in-aside.html

[02:56:00.0000] <a-ja>
it was bad enough before when i tried using a <h> elements in CSS

[02:57:00.0000] <a-ja>
/me knows that's heresy!

[02:58:00.0000] <a-ja>
on that night.....off for some zzzzzzzzzzzzz's

[03:03:00.0000] <annevk>
http://twitter.com/WHATWG/statuses/25578381

[03:06:00.0000] <Lachy>
I wonder how many of those shirts have sold so far.  othermaciej, are those stats available?

[03:21:00.0000] <annevk>
Lachy, how fast do they deliver to AU?

[03:22:00.0000] <Lachy>
don't know, but probably not fast enough

[03:22:01.0000] <annevk>
k

[03:22:02.0000] <zcorpan_>
hm. today i my inbox was filled with emails that i have sent to others. but no other emails. :S

[03:22:03.0000] <Lachy>
it'd probably take at least a week if they're shipping from the US

[03:22:04.0000] <annevk>
well, I'll have to wait a few weeks then until I have my credit cards back

[03:22:05.0000] <annevk>
zcorpan_, you spammer

[03:23:00.0000] <zcorpan_>
i don't get it

[03:23:01.0000] <Lachy>
zcorpan_, consider yourself lucky

[03:23:02.0000] <zcorpan_>
lol

[03:23:03.0000] <Lachy>
but you should have at least got some from public-html

[03:23:04.0000] <zcorpan_>
yeah

[03:23:05.0000] <Lachy>
and whatwg

[03:24:00.0000] <Lachy>
woah, 34 messages on whatwg since midnight!

[03:24:01.0000] <zcorpan_>
i have them on gmail.com. but opera says no new messages

[03:25:00.0000] <zcorpan_>
should i reinstall opera?

[03:25:01.0000] <annevk>
neh

[03:25:02.0000] <annevk>
POP? IMAP?

[03:25:03.0000] <zcorpan_>
pop

[03:25:04.0000] <annevk>
dunno about that

[03:25:05.0000] <Lachy>
and 68 on public-html!

[03:26:00.0000] <zcorpan_>
Lachy: midnight UTC?

[03:26:01.0000] <Lachy>
no, midnight my time

[03:26:02.0000] <zcorpan_>
ok

[03:26:03.0000] <Lachy>
UTC+10

[03:26:04.0000] <annevk>
hit check/send a few times...

[03:26:05.0000] <zcorpan_>
annevk: done that

[03:26:06.0000] <zcorpan_>
"no messages received or sent"

[03:26:07.0000] <annevk>
does it connect to the server?

[03:26:08.0000] <zcorpan_>
yes

[03:27:00.0000] <zcorpan_>
"connected"

[03:27:01.0000] <zcorpan_>
i can send emails

[03:27:02.0000] <Lachy>
try recreating your profile or something.  That's what I do when FF and TB screw up badly

[03:27:03.0000] <zcorpan_>
ok

[03:27:04.0000] <annevk>
sometimes this works

[03:27:05.0000] <zcorpan_>
won't my emails and their labels go away then though?

[03:28:00.0000] <annevk>
tools -> mail and chat accounts -> edit your account -> servers -> add a space after the server name and try again

[03:28:01.0000] <annevk>
I don't really encounter that problem with Opera 9.2 anymore but maybe it helps

[03:28:02.0000] <zcorpan_>
same

[03:28:03.0000] <Lachy>
ooh, 9.2's been released?

[03:28:04.0000] <Lachy>
/me goes to upgrde

[03:29:00.0000] <annevk>
hah, and I'm joking. I'm using Opera 9.1

[03:29:01.0000] <annevk>
but 9.2 has indeed been released

[03:29:02.0000] <zcorpan_>
this started when i installed 9.2 i think

[03:32:00.0000] <annevk>
/me goes back to reading the HTML5 changelog

[03:34:00.0000] <Lachy>
the new Opera speed dial looks interesting

[03:47:00.0000] <SuitCase>
oh no! html5 is going to take over the world now?

[03:48:00.0000] <SuitCase>
i am reading this interesting proposal to the w3c. it kinda makes me feel bad for the people who have seemingly spent so long on something that may end up irrelevant.

[03:51:00.0000] <Hixie>
which ones? whatwg? rdf? xhtml2?

[03:52:00.0000] <SuitCase>
xhtml2. i'm referring to this: http://lists.w3.org/Archives/Public/public-html/2007Apr/0429.html

[03:53:00.0000] <Hixie>
right

[03:54:00.0000] <annevk>
/me notes that people worked hard and long on HTML5 too

[03:54:01.0000] <annevk>
in fact, if you compare XHTML2 and HTML5... but lets not go there

[03:55:00.0000] <SuitCase>
oh, sure, but html5 people are the cool kids that get all the recognition anyway.

[03:55:01.0000] <annevk>
heh, you should read some comments on articles

[03:56:00.0000] <SuitCase>
oh? this is just my perception.

[03:57:00.0000] <annevk>
See http://ajaxian.com/archives/proposal-for-the-w3c-to-adopt-html-5 for instance

[03:58:00.0000] <annevk>
It has quite a big group of supporters too though, it varies per community

[04:00:00.0000] <annevk>
(People opposing are also mostly flamebaits, which is encouraging.)

[04:01:00.0000] <SuitCase>
that's interesting. thanks for the link.

[04:04:00.0000] <SuitCase>
does it look like it will be successful? there seems to be little resistance from the w3c people there.

[04:04:01.0000] <annevk>
it already is successful

[04:04:02.0000] <annevk>
see <canvas> for instance

[04:07:00.0000] <SuitCase>
well, i mean the specific proposal to have the w3c adopt the html5 spec in favour of xhtml.

[04:07:01.0000] <SuitCase>
err.. instead of xhtml.

[04:08:00.0000] <annevk>
HTML5 has an XML and HTML serialization

[04:09:00.0000] <hsivonen>
SuitCase: well, since the W3C already established the new HTML WG, it is pretty clear that XHTML 2.0 is not the answer to the question that the new HTML WG is expected to answer

[04:09:01.0000] <annevk>
XHTML1 also uses the semantics from HTML4

[04:09:02.0000] <annevk>
I suppose many people don't really understand that

[04:12:00.0000] <annevk>
The answer is pretty clear, we still don't get the question though :)

[04:16:00.0000] <annevk>
SimonW, how did your presentation go?

[04:16:01.0000] <SimonW>
reasonably well

[04:17:00.0000] <SimonW>
cramming the entire history of HTML / WHATWG in to 5 minutes was, in hindsight, a bit ambitious

[04:17:01.0000] <SimonW>
but I think it got people interested / thinking, which is good

[04:17:02.0000] <SimonW>
could have happily done 45 minutes

[04:18:00.0000] <SimonW>
slides : http://simonwillison.net/2007/talks/oxford-html5/

[04:19:00.0000] <SuitCase>
hsivonen: oh! i was not aware of this. guess i need to learn more about the spec process.

[04:19:01.0000] <Hixie>
oh hey, SimonW is simon willison, didn't realise that

[04:20:00.0000] <Hixie>
hey simon

[04:20:01.0000] <SimonW>
hi Hixie 

[04:21:00.0000] <SimonW>
are you at XTech again this year?

[04:24:00.0000] <Hixie>
nah

[04:24:01.0000] <Hixie>
nice slides for your talk, btw

[04:27:00.0000] <SimonW>
thanks

[04:28:00.0000] <Hixie>
othermaciej, dude, you have to bring out your patent bat on the list again

[04:29:00.0000] <SimonW>
I was going to use your diagram of IE building something that isn't a tree from a malformed DOM, but I had to drop it due to time constraints

[04:36:00.0000] <annevk>
slides are nice

[04:39:00.0000] <annevk>
btw, when I did performance testing long ago on Web Apps 1.0 it took 6 seconds or so... not 19

[04:39:01.0000] <annevk>
either someone regressed it heavily or Web Apps became a lot larger :)

[04:43:00.0000] <Hixie>
or the machine i'm using isn't as good as yours...

[04:43:01.0000] <Hixie>
makes no sense to compare performance metrics on different machines

[04:43:02.0000] <Hixie>
even on the same machine many factors can affec it

[04:45:00.0000] <annevk>
yeah, fair enough

[04:45:01.0000] <annevk>
some day we'll port the thing to C

[04:46:00.0000] <zcorpan>
"Fetching messages (17/685)"

[04:46:01.0000] <zcorpan>
after reinstall

[04:47:00.0000] <annevk>
fancy

[04:47:01.0000] <zcorpan>
i still have my bookmarks, that's nice

[04:47:02.0000] <zcorpan>
not anything else though :|

[04:49:00.0000] <zcorpan>
my feeds! sigh, how should i know which feeds i were subscribed to?

[04:50:00.0000] <zcorpan>
hm, seems like my old profile is in another folder

[05:14:00.0000] <gsnedders>
/me now really wants 5 > 2 T-shirts

[05:14:01.0000] <gsnedders>
:P

[05:30:00.0000] <zcorpan>
SimonW: i'll add your presentation to http://wiki.whatwg.org/wiki/HTML5_Presentations

[05:31:00.0000] <SimonW>
thanks zcorpan 

[05:31:01.0000] <SimonW>
should be video of it up online in the next few days

[05:32:00.0000] <zcorpan>
SimonW: you have a whatwg wiki account?

[05:33:00.0000] <SimonW>
not yet

[05:33:01.0000] <zcorpan>
was it presented today?

[05:34:00.0000] <SimonW>
yesterday

[05:34:01.0000] <zcorpan>
ok

[05:34:02.0000] <SimonW>
(you should install the MediaWiki OpenID plugin)

[05:34:03.0000] <SimonW>
creating an acconut...

[05:36:00.0000] <annevk>
SimonW, pointer for that plugin? if it's easy to set up I'm sure Hixie will integrate it

[05:36:01.0000] <SimonW>
I'm SimonW

[05:36:02.0000] <annevk>
I believe DanC was willing to contribute as well if we had an OpenID login mechanism

[05:36:03.0000] <Hixie>
surely openid would just make it easier for spammers

[05:37:00.0000] <Hixie>
since they'd just have to set up an open authentication openid

[05:37:01.0000] <SimonW>
http://www.openidenabled.com/software/mediawiki

[05:37:02.0000] <SimonW>
yup, OpenID is great for spammers

[05:37:03.0000] <SimonW>
but right now there's no captcha on the wiki account creation (is there?)

[05:37:04.0000] <SimonW>
so shouldn't make a difference in practise

[05:38:00.0000] <Hixie>
there's no captcha but there's some tricks going on behind the scenes to make it harder for them to log in and create an account

[05:38:01.0000] <Hixie>
actually since we implemented those hacks i don't think we've had any spam at all (we used to have quite a lot)

[05:38:02.0000] <Hixie>
or were the hacks we added just for the forums

[05:38:03.0000] <Hixie>
i forget

[05:38:04.0000] <zcorpan>
is "[[User:SimonW|Simon Willison]]" correct?

[05:38:05.0000] <SimonW>
yup

[05:38:06.0000] <SimonW>
OpenID users still have to create a wiki account when they use OpenID

[05:39:00.0000] <SimonW>
the only difference is that they don't have to set Yet Another Password

[05:39:01.0000] <SimonW>
so it might be compatible with the measures you have at the moment

[05:39:02.0000] <zcorpan>
http://wiki.whatwg.org/wiki/HTML5_Presentations

[05:39:03.0000] <annevk>
Hixie, there was some spam, I wasn't sure how to undo it and such

[05:39:04.0000] <Hixie>
i undo it when i see it

[05:39:05.0000] <SimonW>
http://wikitravel.org/en/Main_Page has the OpenID thing installed (they developed it)

[05:39:06.0000] <Hixie>
(i get mailed for every change to the wiki)

[05:40:00.0000] <annevk>
http://wiki.whatwg.org/index.php?title=Category:Implementations&curid=1645&diff=2188&oldid=2096

[05:40:01.0000] <annevk>
http://wiki.whatwg.org/index.php?title=User:Matthew_Raymond:dataentry_element&curid=734&diff=2186&oldid=786

[05:40:02.0000] <annevk>
not sure they qualify as spam

[05:40:03.0000] <annevk>
but they're not doing useful stuff

[05:41:00.0000] <annevk>
is Jarvklo online here?

[05:42:00.0000] <SimonW>
http://www.mediawiki.org/wiki/Extension:OpenID looks like a better bet than that patch

[05:42:01.0000] <Hixie>
annevk: weird

[05:42:02.0000] <Hixie>
didn't get mailed for those either

[05:42:03.0000] <Hixie>
wonder why not

[05:44:00.0000] <zcorpan>
annevk: why?

[05:44:01.0000] <annevk>
re: Talk:Changes from HTML4

[05:46:00.0000] <zcorpan>
i could ask him to get in here

[05:47:00.0000] <annevk>
i replied in the wiki

[05:47:01.0000] <annevk>
Hixie, not going to bed today?

[05:48:00.0000] <Hixie>
meh

[05:48:01.0000] <gsnedders>
/me realises the time in Hixie-land

[05:48:02.0000] <gsnedders>
/me used to live on 4 hours sleep without much issue

[05:54:00.0000] <zcorpan>
wonder if i should make html5-elements link to the multipage version or not

[05:54:01.0000] <annevk>
depends on how much work you want in mainting the thing

[05:54:02.0000] <annevk>
maintaining*

[05:55:00.0000] <zcorpan>
i'll leave it as is for now

[05:56:00.0000] <annevk>
the multipage version is not yet ideal

[05:56:01.0000] <Philip`>
zcorpan: If you do, it should be relatively straightforward to use the existing spec-splitter script to get the same ID -> filename mapping

[05:57:00.0000] <annevk>
oh right, that makes sense

[05:57:01.0000] <zcorpan>
Philip`: ok

[05:58:00.0000] <Hixie>
annevk: if you want it improved, feel free to commit patches

[05:58:01.0000] <annevk>
yeah, I was just thinking that I should shut up :)

[05:58:02.0000] <Hixie>
:-D

[05:58:03.0000] <Hixie>
i didn't mean that sarcasticly though

[05:58:04.0000] <Hixie>
i mean, you've checked in patches before :-)

[05:59:00.0000] <Hixie>
right well i really should sleep

[05:59:01.0000] <Hixie>
nn

[06:00:00.0000] <zcorpan>
g'night Hixie

[06:00:01.0000] <annevk>
g'night

[06:01:00.0000] <virtuelv>
ah, editing

[06:01:01.0000] <virtuelv>
I'm on a slow track to make any spec editing 100x more useful from within my editor

[06:09:00.0000] <MikeSmith>
virtuelv - hei

[06:10:00.0000] <virtuelv>
MikeSmith: hey

[06:10:01.0000] <virtuelv>
MikeSmith: you going to xtech?

[06:10:02.0000] <MikeSmith>
yah

[06:10:03.0000] <annevk>
MikeSmith, your former former employer joined the HTML WG

[06:11:00.0000] <MikeSmith>
annevk - yep. Because I made them an offer they couldn't refuse :)

[06:12:00.0000] <MikeSmith>
Christian Sejersen is a friend

[06:12:01.0000] <MikeSmith>
he's the engineering manager for their client product dev group (which includes their browser)

[06:13:00.0000] <MikeSmith>
an implementor

[06:14:00.0000] <MikeSmith>
I don't know how much coding his does now, be he did a lot of hands-on work on their browser over the years 

[06:14:01.0000] <annevk>
i see

[06:15:00.0000] <MikeSmith>
virtuelv - we are doing this Gelocation BoF

[06:16:00.0000] <MikeSmith>
http://2007.xtech.org/public/schedule/detail/222

[06:17:00.0000] <MikeSmith>
about developing a standard scripting interface to location-sensing (GPS, etc.) capabilities on devices

[06:18:00.0000] <MikeSmith>
you and Geir should please be there

[06:20:00.0000] <annevk>
Geir is planning, it seems

[06:21:00.0000] <MikeSmith>
annevk - great. Will be good to see him

[06:22:00.0000] <MikeSmith>
annevk, virtuelv - other than you guys and Geir and Haakon, anybody else from Opera planning to be there?

[06:22:01.0000] <MikeSmith>
gorm maybe?

[06:23:00.0000] <virtuelv>
MikeSmith: dunno, perhaps p01

[06:23:01.0000] <MikeSmith>
whoah

[06:23:02.0000] <MikeSmith>
that'll be cool if he's there

[06:23:03.0000] <MikeSmith>
/me misses #staff very very much

[06:29:00.0000] <annevk>
do you need to register btw?

[06:29:01.0000] <annevk>
/me hasn't been busy much with XTech stuff yet

[06:34:00.0000] <MikeSmith>
annevk - yeah, you need to register

[06:36:00.0000] <annevk>
i believe last time I just went there and bought a ticket at the door

[06:37:00.0000] <MikeSmith>
ah, yeah, I'm sure they'll be happy to do that

[06:37:01.0000] <MikeSmith>
btw, no Reboot this year?

[06:37:02.0000] <annevk>
can't come

[06:37:03.0000] <annevk>
there will be one though

[06:45:00.0000] <Philip`>
annevk: If you have ideas for improving the multipage version, it could be useful to write them down somewhere in case someone is bored enough to work on it again in the future :-)

[06:46:00.0000] <Philip`>
(I'd at least like to merge some of the empty h2-level pages)

[06:46:01.0000] <Philip`>
((but I'm away for the next five days so I can't actually do anything now))

[06:54:00.0000] <annevk>
1. Empty pages. 2. Unstable links. 3. Ugly links.

[06:54:01.0000] <annevk>
I suppose only 1 can be solved

[06:54:02.0000] <annevk>
The others can probably be solved by getting stable header IDs

[07:04:00.0000] <zcorpan>
all emails read! phew

[07:44:00.0000] <MikeSmith>
http://www.reboot.dk/article-184-en.html

[08:08:00.0000] <annevk>
from www-svg: "

[08:08:01.0000] <annevk>
That said, unless the interfaces for working with HTML improve in its next version, I'm going to prefer to work with pure SVG, which has a much cleaner and more consistent API."

[08:09:00.0000] <annevk>
(signed by Doug Schepers)

[08:09:01.0000] <annevk>
/me isn't entirely sure what it means or implies

[08:11:00.0000] <hsivonen>
annevk: which clients implement the pure svg apis?

[08:12:00.0000] <annevk>
Opera and Adobe

[08:12:01.0000] <annevk>
dunno how much Firefox does

[08:12:02.0000] <annevk>
I'm not sure what definition of clean and consistent he's using though

[08:13:00.0000] <hsivonen>
annevk: how does Opera reconcile legacy hhtml api impurities ? is svg contradictory ?

[08:18:00.0000] <annevk>
no, it's just different

[08:19:00.0000] <annevk>
i haven't paid much attention to it lately though

[09:06:00.0000] <zcorpan>
/me has to subscribe to hsivonen's rss feed as opposed to his atom feed, because opera doesn't show anything for the atom feed :(

[09:07:00.0000] <annevk>
feel free to make a bug report :)

[09:08:00.0000] <zcorpan>
you don't have one already?

[09:09:00.0000] <annevk>
dunno

[09:09:01.0000] <annevk>
I'm not up to date with our feed stuff

[09:15:00.0000] <zcorpan>
#260205, used http://hsivonen.iki.fi/feed/atom/ as the test case

[09:16:00.0000] <annevk>
cc'ed qa contact for those bugs

[09:19:00.0000] <Dashiva>
Hmm... namespacing atom as non-default

[10:49:00.0000] <Dashiva>
not very encouraging reading, that essay Chris wrote

[10:57:00.0000] <zcorpan_>
i still don't understand why we can't make the spec say whatever MS needs it to say in order to be compatible (e.g. <object classid>)

[10:57:01.0000] <zcorpan_>
and <!doctype html5> triggers quirks mode in gecko and webkit, so can't be used

[10:58:00.0000] <Dashiva>
The second word is supposed to be the root element, after all

[10:58:01.0000] <zcorpan_>
e.g. the spec could define how classid works, but make activex optional or something

[10:58:02.0000] <zcorpan_>
Dashiva: that's not the relevant part, although it is the reason why it triggers quirks mode

[10:58:03.0000] <Dashiva>
All that aside, he didn't make a case for versioning the spec at all

[10:58:04.0000] <hsivonen>
zcorpan_: are you goint to say that the doctype he proposed doesn't work?

[10:59:00.0000] <hsivonen>
going

[10:59:01.0000] <zcorpan_>
dunno

[10:59:02.0000] <hsivonen>
(on the list, that is)

[10:59:03.0000] <zcorpan_>
should i?

[10:59:04.0000] <Philip`>
(Hmm, is it intentional that Firefox stays in standards mode even if you put up to 1024 minus sizeof(doctype) bytes of text in front of it?)

[10:59:05.0000] <hsivonen>
zcorpan_: someone has to

[10:59:06.0000] <Dashiva>
He wants IE versioning, not spec versioning. No matter the doctype, IE6 -> IE7 was on the same pages

[10:59:07.0000] <zcorpan_>
/me responds then

[11:00:00.0000] <Dashiva>
In conclusion, we want a <meta ie-support-version="5,6,7">

[11:01:00.0000] <Dashiva>
Or rather, he wants it

[11:15:00.0000] <Philip`>
It seems he calls it "opt[ing] in to standards compliance", but it sounds like the proposal is more 'opting in to the behaviour of the oldest version of IE that I want my site to work correctly in'

[11:16:00.0000] <Philip`>
Somebody needs to beat his "I think I win for longest post to this list so far" by replying with "+1" (or "-1") followed by quoting his entire message

[11:18:00.0000] <annevk>
<!doctype html> should be the new IE opt in and we'll make damn sure (by means of thousands of tests) that they get it right

[11:19:00.0000] <annevk>
but the browser sniffing is indeed a real pain

[11:20:00.0000] <annevk>
hey othermaciej!

[11:20:01.0000] <Philip`>
Would the IE team be able to pass all the tests that they're given, rather than having to ship a half-finished IE$n by some fixed deadline? (I have no idea how flexible their release schedule is)

[11:21:00.0000] <annevk>
if they ship frequently hpoefully authors don't rely on broken behavior too much

[11:21:01.0000] <annevk>
that's how other browsers are able to fix their standards support anyway

[11:22:00.0000] <annevk>
of course, our baseline is slightly better

[11:23:00.0000] <Philip`>
Is that browser sniffing mostly on the client side, with the <!--[if gt IE 6]> type stuff? If so, they could circumvent that sniffing and get the same content as other browsers by just removing support for those conditional comments

[11:25:00.0000] <othermaciej>
hi annevk 

[11:33:00.0000] <hsivonen>
zcorpan_: well, I pointed out that the doctype won't work

[11:36:00.0000] <zcorpan_>
hsivonen: ok

[11:42:00.0000] <hsivonen>
hmm. I think the generic "solution" to the IE problem is <html bugs-as-of='2007-04-12'>

[11:42:01.0000] <Philip`>
If HTML5 sticks with <!DOCTYPE html>, presumably HTML6 wouldn't be able to use a nice doctype and it'd have to do something ugly like <!DOCTYPE html PUBLIC "6"> or <!DOCTYPE html><html version="6">, which I suppose is one problem with not looking ahead to support whatever versioning HTML6 might need

[11:43:00.0000] <hsivonen>
Philip`: how would it help if HTML5 was as ugly?

[11:43:01.0000] <Philip`>
The HTML5 spec could say UAs must handle <!DOCTYPE html(\d+)> identically to <!DOCTYPE html>, though it'd still be non-conforming for authors

[11:44:00.0000] <Philip`>
then at some point in the future, when all non-negligible UAs support HTML5, HTML7 can happily use <!DOCTYPE html7>

[11:44:01.0000] <zcorpan_>
Philip`: <!doctype htmlx> still triggers quirks mode in gecko and webkit, so it will never be an option

[11:44:02.0000] <hsivonen>
Philip`: and what would you do on the XML side?

[11:45:00.0000] <Philip`>
zcorpan_: <!doctype html> still triggers quirks-like-mode in IE4 - but nobody cares, since it's too old and few people use it

[11:46:00.0000] <zcorpan_>
Philip`: ie4 only has quirks mode

[11:46:01.0000] <zcorpan_>
there's nothing we can do about ie4

[11:47:00.0000] <hsivonen>
/me notes that the HTML5 doctype triggers the quirks mode in legacy Geckos that have rickg's sniffing instead of dbaron's

[11:47:01.0000] <zcorpan_>
true

[11:48:00.0000] <zcorpan_>
ok, i get your point

[11:48:01.0000] <zcorpan_>
but i don't see why it would be useful

[11:48:02.0000] <zcorpan_>
we shouldn't encourage future revisions of html to introduce versioning

[11:49:00.0000] <Philip`>
Hmm, I suppose the difference is that there's no way at all to get standards mode in IE4 (and IE5?), even with ugly version identifiers - whereas HTML7 could still get standards mode from FF2, if it kept using <!DOCTYPE html>, and so it's worthwhile even if approximately nobody uses FF2 that far in the future

[11:51:00.0000] <Philip`>
(and so it's not worth making HTML5 require support for <!DOCTYPE html7>, because HTML7 should still use <!DOCTYPE html> and if it really needs a version number then it can do something else)

[11:52:00.0000] <Philip`>
(*something else which won't trigger quirks mode in ancient browsers like FF2)

[11:52:01.0000] <zcorpan_>
indeed

[11:59:00.0000] <zcorpan_>
is it possible for other browser vendors to ship a browser with activex support?

[12:00:00.0000] <annevk>
i'd think so

[12:06:00.0000] <hsivonen>
zcorpan_: only on Windows

[12:06:01.0000] <zcorpan_>
ok

[12:12:00.0000] <Philip`>
http://plugindoc.mozdev.org/windows-all.html#ActiveX

[12:24:00.0000] <zcorpan_>
http://simon.html5.org/temp/Versioning%20and%20html[5].txt -- comments before i send it?

[12:26:00.0000] <Philip`>
http://developers.slashdot.org/article.pl?sid=07/04/12/152245 - in case anyone really likes reading comments sections [and hasn't found this already]

[12:31:00.0000] <hasather>
zcorpan_: in the first paragraph, you have "doens't"

[12:32:00.0000] <zcorpan_>
hasather: fixed

[12:40:00.0000] <zcorpan_>
no other comments?

[12:40:01.0000] <hsivonen>
zcorpan_: typo: In my optinion

[12:41:00.0000] <zcorpan_>
thanks

[12:44:00.0000] <Philip`>
With conditional comments, is it reasonable to suggest that if IE fixed enough bugs then they could remove support for conditional comments and they would no longer suffer because of authors' expectations of IE-brokenness? (I don't know if that's realistic or not)

[12:44:01.0000] <zcorpan_>
on the long term, yes

[12:45:00.0000] <zcorpan_>
on the short term they won't be able to fix enough bugs to be able to remove CCs

[12:46:00.0000] <zcorpan_>
they are useful to fix bugs without relying on other bugs or expecting the next version of ie to have the same bug

[12:46:01.0000] <zcorpan_>
they are too powerful though, it should have only been possible to say "if lte IE X"

[12:47:00.0000] <zcorpan_>
because too many authors just say "if IE" and thus apply the bug fixes to all future versions of ie that still implements CCs

[12:48:00.0000] <zcorpan_>
(i guess they could drop make "if IE" return false, though, but that would be ugly)

[12:48:01.0000] <zcorpan_>
s/drop//

[12:48:02.0000] <Philip`>
Then people might say "if lte IE 999" if they want to apply to all version of IE forevermore

[12:48:03.0000] <Philip`>
...but at least then it'd be obviously ugly and nasty, so maybe they wouldn't do that as often as they do "if IE"

[12:48:04.0000] <zcorpan_>
yes, but that would be an active choise

[12:49:00.0000] <zcorpan_>
when they use "if IE" today they only intended to fix IE6

[12:49:01.0000] <Philip`>
Ah, makes sense

[12:50:00.0000] <Philip`>
It would be nice if IE shipped with a list of all its bugs, so you could do conditional comments based on which ones they've fixed yet...

[12:50:01.0000] <Philip`>
(I suppose impossible solutions aren't much good, though)

[12:52:00.0000] <zcorpan_>
given only typos have been commented on i take it as it isn't fundamentally bogus, so i'll send it now

[12:53:00.0000] <Philip`>
I didn't see any fundamental bogosity, though I don't think I'm very sensitive at picking up things like that

[12:54:00.0000] <zcorpan_>
sent

[12:56:00.0000] <Philip`>
Hmm, should the multipage version of the HTML5 spec be written in HTML 4.01 (like the single-page one is)?

[12:56:01.0000] <zcorpan_>
naw

[12:57:00.0000] <Philip`>
If it has the same normative value as the single-page one (which it appears it does), it would make more sense to not write it in itself

[12:57:01.0000] <zcorpan_>
i'd expect the single-page one to be html5 too when the tools used to produce it are updated

[12:58:00.0000] <zcorpan_>
the html4 spec is written in html4

[12:58:01.0000] <zcorpan_>
having it in html5 would be a proof of concept that it is indeed compatible with html4 UAs :)

[12:59:00.0000] <Philip`>
Ah, okay then

[12:59:01.0000] <Philip`>
http://www.w3.org/TR/WD-html40-970708/cover.html seems to be the latest HTML4 that wasn't written in HTML4

[12:59:02.0000] <zcorpan_>
ok

[12:59:03.0000] <Philip`>
HTML5 is nicer to generate than HTML4, since it makes it obvious how you're meant to do it, so I'm happy with sticking to that :-)

[13:05:00.0000] <zcorpan_>
hsivonen: what was the conclusion of http://www.effi.org/sananvapaus/videotiedostomuoto.html , briefly?

[13:11:00.0000] <zcorpan_>
wey! opera 9.2 has new style sheets!

[13:12:00.0000] <zcorpan_>
nice

[13:12:01.0000] <zcorpan_>
now i don't need my "reveal IDs" bookmarklet anymore

[13:15:00.0000] <zcorpan_>
what is "Inline structure"?

[13:17:00.0000] <zcorpan_>
"Table of contents" would be more useful to have in the sidebar, i think

[13:17:01.0000] <Philip`>
Would it be feasible to implement <video> for IE6/7 with JavaScript+Flash? (It'd only support .flv files but at least it could use the same syntax and API as the alternate non-FLV video sources)

[13:18:00.0000] <zcorpan_>
why flash? i thought you could use html+time to render video in ie

[13:19:00.0000] <Philip`>
Oh, I forgot about that

[13:19:01.0000] <zcorpan_>
don't know about the feasability though

[13:19:02.0000] <zcorpan_>
would be cool if it worked

[13:20:00.0000] <zcorpan_>
(given what dean edwards has implemented with js in ie, i would be surprised if it didn't work)

[13:43:00.0000] <annevk>
jgraham, your suggestion is pretty evil

[13:43:01.0000] <annevk>
having said that, I need to go

[13:43:02.0000] <jgraham>
annevk: I know :)

[13:44:00.0000] <Hixie>
we hit /.

[13:44:01.0000] <kingryan>
http://developers.slashdot.org/developers/07/04/12/152245.shtml ?

[13:45:00.0000] <Hixie>
yah

[13:45:01.0000] <Hixie>
sorry if it's old news, i'm going through my new referrers :-)

[13:45:02.0000] <zcorpan_>
authors can use CCs today without making assumptions about future IEs. i don't see why that isn't good enough

[13:45:03.0000] <kingryan>
looks like it was posted today

[13:45:04.0000] <annevk>
(was in this channel before though)

[13:46:00.0000] <zcorpan_>
(i.e. <!--[if lte IE 7]> to fix bugs in current IEs)

[13:47:00.0000] <hsivonen>
zcorpan_: "EFFI recommends that videos distributed on the Internet be offered at least as HTTP-downloadable H.264 (Main or Baseline profile)/Low Complexity AAC/MP4 or Theora/Vorbis/Ogg format video files that are directly linked to from an HTML page and that have no kind of Digital Restrictions Management (DRM) applied to them. In particular, it is noted that Windows Media Video or Flash video should not be offered as the only choice."

[13:48:00.0000] <zcorpan_>
hsivonen: thanks

[13:49:00.0000] <hsivonen>
zcorpan_: the document was written in response to the national public broadcaster going with a WMV9-only solution that doesn't even work properly with Flip4Mac

[13:50:00.0000] <hsivonen>
(this happened after they got a new CEO who previously worked from Microsoft Finland)

[13:50:01.0000] <zcorpan_>
ok

[13:51:00.0000] <hsivonen>
s/from/for/

[14:20:00.0000] <zcorpan_>
Hixie: to be fair to ms, the classid and codebase attributes (and indeed the entire <object> element) might have been invented by ms before it was included in html4

[14:21:00.0000] <Hixie>
what's in html4 doesn't match what they do

[14:22:00.0000] <zcorpan_>
i don't understand the question

[14:25:00.0000] <Hixie>
it's not a question

[14:26:00.0000] <Hixie>
ie 'codebase' != html4 'codebase'

[14:26:01.0000] <zcorpan_>
oh

[14:26:02.0000] <zcorpan_>
indeed

[14:26:03.0000] <Hixie>
so ie's attribute wasn't included in the html4 spec

[14:26:04.0000] <Hixie>
even if the name was used by ie before the spec came out

[14:27:00.0000] <zcorpan_>
true, but if ms used the attribute name first it isn't their fault that it is in conflict with html4

[14:28:00.0000] <zcorpan_>
that's what i meant

[14:29:00.0000] <zcorpan_>
...unless ms were in the html wg back then defining it

[14:30:00.0000] <Philip`>
Chris said "the WHATWG HTML5 removes a few key attributes from the <object> element that were in HTML 4.01 - namely, classid and codebase - that are heavily used by ActiveX in IE", which is why it's relevant to say that IE isn't actually using those HTML 4.01 attributes

[14:30:01.0000] <Philip`>
(independent of whose fault it is that the spec and the implementation disagree)

[14:31:00.0000] <zcorpan_>
yes

[14:33:00.0000] <othermaciej>
hi everyone

[14:34:00.0000] <zcorpan_>
hi othermaciej

[14:43:00.0000] <zcorpan_>
"inline structure" reveals at least <code> and <abbr> it seems

[15:06:00.0000] <zcorpan_>
is display:table-cell on its own handled correctly in latest webkit? (i.e. is http://simon.html5.org/temp/w3c-home-in-html5.html rendered in three columns or one column?)

[15:09:00.0000] <othermaciej>
I'm surprised the HTML5 proposal made it to slashdot but not digg or reddit (yet)

[15:16:00.0000] <zcorpan_>
apparently it isn't. i can't find a bug about it in bugzilla either.

[15:16:01.0000] <zcorpan_>
/me thought it was a known bug

[15:17:00.0000] <othermaciej>
zcorpan_: please file a bug if there isn't one

[15:17:01.0000] <zcorpan_>
oh wait, http://bugs.webkit.org/show_bug.cgi?id=3235

[15:17:02.0000] <webben>
Hixie, re http://lists.w3.org/Archives/Public/public-html/2007Apr/0108.html

[15:18:00.0000] <webben>
where could i read your proposal for CSS replacement of "

[15:18:01.0000] <Hixie>
don't think i ever wrote it down

[15:19:00.0000] <Hixie>
basically it's ::text-before(/regexp/) or something

[15:19:01.0000] <webben>
Hixie, Interesting. Do you know how easy or hard that would be to implement?

[15:20:00.0000] <Hixie>
probably quite hard, pseudo-elements in general are a bitch

[15:21:00.0000] <Hixie>
i can't believe chris honestly doesn't see how what he's doing is exactly what microsoft keeps doing to sustain their monopolies

[15:21:01.0000] <Hixie>
sometimes people laugh at "don't be evil", but it's exactly this kind of thing that google's culture prevents

[15:22:00.0000] <webben>
/me wonders what MS have  done now...

[15:22:01.0000] <Hixie>
the versioning thing in the html list

[15:27:00.0000] <othermaciej>
we do try to build the anonymous rendering strctures needed for CSS tables but there could certainly be bugs

[15:28:00.0000] <othermaciej>
Hixie: I was going to point out that his citation of Windows and Office does little to dispel fears of acting in bad faith

[15:28:01.0000] <othermaciej>
but I have not finished reading his essay yet

[15:28:02.0000] <Hixie>
i will let you finish

[15:29:00.0000] <othermaciej>
don't let me stop you from ranting -)

[15:29:01.0000] <othermaciej>
:-)

[15:30:00.0000] <Hixie>
i'm curious in your opinion unbiased by my rants :-)

[15:30:01.0000] <ianloic>
it's an interesting read

[15:30:02.0000] <Hixie>
i already recorded my opinion in replies to his mail

[15:31:00.0000] <zcorpan_>
oops, sometimes threaded view in opera doesn't reveal the entire three

[15:31:01.0000] <zcorpan_>
/me responded to something that was already responded to

[15:35:00.0000] <hasather>
zcorpan_: yea, that bothers me too, the only indication of new mail is that the parent mail becomes bold, but the new mail shouldn't be hidden IMO

[15:53:00.0000] <zcorpan_>
should i respond to henrik's last email?

[15:54:00.0000] <zcorpan_>
too late

[15:54:01.0000] <zcorpan_>
although i could be more specific

[15:54:02.0000] <hsivonen>
zcorpan_: kingryan didn't mention the charter

[15:54:03.0000] <zcorpan_>
indeed

[15:55:00.0000] <kingryan>
/me is lost

[15:55:01.0000] <kingryan>
are you talking about my email re: implementors vs. authors?

[15:56:00.0000] <zcorpan_>
no

[15:56:01.0000] <zcorpan_>
version information

[15:56:02.0000] <kingryan>
ah, yes

[15:56:03.0000] <kingryan>
I should have mentioned that too


2007-04-13
[20:16:00.0000] <Hixie>
so...

[20:16:01.0000] <Hixie>
there's a problem with the spec splitter script

[20:16:02.0000] <Hixie>
it causes me to hit my CPU limit.

[20:16:03.0000] <Hixie>
I've tried 'nice'ing it but that doesn't help

[20:39:00.0000] <mpt_>
Hixie, that's easy to solve, just shorten the spec

[20:40:00.0000] <mpt_>
/me ducks

[20:54:00.0000] <Lachy>
are there any comments in that slashdot article that actually respond to the issue?  So far, all comments I've read are mostly off topic talk about stylesheet issues

[20:57:00.0000] <mpt_>
Not that I could see (but then, I browse at ≥3, so maybe the moderators weren't interested in on-topic discussion)

[20:59:00.0000] <Lachy>
I leave all comments visible.  I find it annoying reading comments that respond to others that aren't shown

[20:59:01.0000] <Lachy>
and I don't particularly trust people I don't know to be able to choose what I should and shouldn't read

[20:59:02.0000] <mpt_>
It's great that you have that much time to read :-)

[21:00:00.0000] <Lachy>
ah, the benefits of being unemployed :-)

[21:01:00.0000] <mpt_>
You are, largely, trusting people you don't know to choose what you should and shouldn't read every time you do a Web search and don't go beyond the first page of results

[21:02:00.0000] <Lachy>
I often go beyond the first results page if I don't find what I want

[22:22:00.0000] <Hixie>
mpt_: not clear what I should drop that would make it shorter enough. The script that's failing is the one that MAKES it shorter :-P

[00:17:00.0000] <annevk>
Everyone should agree on this by now. Versioning debates are awesome!

[00:19:00.0000] <othermaciej>
I'm going to have to make my own entry in the versioning debate, I fear

[00:21:00.0000] <annevk>
The browsing sniffing from authors is a major pain though.

[00:22:00.0000] <othermaciej>
I do think what cwilso is asking for is not only a bad idea but also kind of insulting to the working group

[00:22:01.0000] <othermaciej>
"we won't ever conform to the spec because at some point we will freeze bugs, please change the spec to make it easier for us to not conform"

[00:22:02.0000] <annevk>
If a site uses features browser A and B equally support they can still render it differently because the author is uninformed

[00:23:00.0000] <Lachy>
Chris' fallicious arguments are starting to get annoying

[00:24:00.0000] <annevk>
(Because the site uses feature detection with a feature outside of the compatible rangebetween A and B.)

[00:24:01.0000] <annevk>
othermaciej, yeah, that's not very clear to me either. If they completely implement HTML5, there's no need for versioning in HTML6...

[00:25:00.0000] <othermaciej>
Chris has pretty much said they will never support it: "- support <!DOCTYPE html> as always-the-latest-version,

[00:25:01.0000] <othermaciej>
Not once it's widely used, as we will break back-compat."

[00:28:00.0000] <annevk>
Yeah, I didn't get that statement

[00:28:01.0000] <Lachy>
I wish Chris would join us in here or #html-wg

[00:31:00.0000] <annevk>
did David P. try to beat cwilso in the longest e-mail match?

[00:31:01.0000] <annevk>
/me quit reading it

[00:33:00.0000] <othermaciej>
I find it funny that he says no changes are acceptable for current content, but any IE changes to HTML5 support will be acceptable so long as less than 0.5% of the web uses that doctype

[00:33:01.0000] <othermaciej>
so which is it, absolutely 0, or up to 0.5%?

[00:33:02.0000] <othermaciej>
because I think "breaks less than 0.5% of sites" would be a fine threshold for anything in HTML5

[00:33:03.0000] <othermaciej>
perhaps even too lenient

[00:34:00.0000] <annevk>
if done incrementally we could fix the web :)

[01:03:00.0000] <annevk>
That people still believed HTML5 was based on SGML

[01:03:01.0000] <annevk>
How about reading what type of effort you join...

[01:05:00.0000] <hsivonen>
annevk: I guess having Web Apps 1.0 and Patent Law for Dummies as required reading is too ambitious if people don't read the Charter, which is required reading that everyone had to affirmatively claim to have read

[01:05:01.0000] <othermaciej>
We need a FAQ

[01:05:02.0000] <othermaciej>
people won't read that either but at least you can point them to it

[01:08:00.0000] <othermaciej>
I may have to write an essay of my own to properly reply to Chris

[01:10:00.0000] <Lachy>
othermaciej, what questions would you like me to add to the WHATWG FAQ to deal with these issues?

[01:10:01.0000] <KevinMarks>
he basically syas people serve specific things to fucked up ie versions

[01:10:02.0000] <annevk>
Q: Am I an idiot?

[01:10:03.0000] <annevk>
A: Most likely.

[01:11:00.0000] <othermaciej>
Lachy: I guess one about the format not being based on SGML, and maybe one about the reasoning behind the doctype

[01:12:00.0000] <KevinMarks>
so, how about an ie version tag with a range

[01:12:01.0000] <KevinMarks>
and he can commit to which version he'll fix it in

[01:12:02.0000] <Lachy>
othermaciej, we already have those questions

[01:13:00.0000] <Lachy>
but they could be improved

[01:14:00.0000] <othermaciej>
Lachy: then I just need to read the FAQ so I know when to point people to it :-)

[01:14:01.0000] <Lachy>
http://blog.whatwg.org/faq/#doctype

[02:39:00.0000] <krijnhoetmer>
Lachy: http://krijnhoetmer.nl/irc-logs/whatwg/latest/

[02:40:00.0000] <hasather>
krijnhoetmer: great, thanks for that

[02:40:01.0000] <krijnhoetmer>
hasather: np

[02:44:00.0000] <krijnhoetmer>
More ideas?

[02:45:00.0000] <met_>
http://krijnhoetmer.nl/irc-logs/whatwg/ sould redirect to latest ?

[02:45:01.0000] <met_>
and not to the index

[02:45:02.0000] <krijnhoetmer>
Doh

[02:45:03.0000] <krijnhoetmer>
:)

[02:47:00.0000] <krijnhoetmer>
met_: Done

[02:47:01.0000] <met_>
thx

[02:47:02.0000] <met_>
and same for http://krijnhoetmer.nl/irc-logs/html-wg/ ? 

[02:48:00.0000] <met_>
8-)

[02:48:01.0000] <krijnhoetmer>
I did

[02:48:02.0000] <met_>
not workfor me

[02:48:03.0000] <met_>
http://krijnhoetmer.nl/irc-logs/html-wg/ still redirect to index

[02:48:04.0000] <krijnhoetmer>
Lies!

[02:48:05.0000] <krijnhoetmer>
:(

[02:49:00.0000] <met_>
sorry, but till 8-(

[02:49:01.0000] <met_>
now, the change

[02:49:02.0000] <met_>
ok mabe it was some cache

[02:50:00.0000] <met_>
ff cache, in IE it redirect thx and sorry for confusion 8-)

[02:50:01.0000] <krijnhoetmer>
Yeah, Fx sucks

[02:50:02.0000] <krijnhoetmer>
*runs*

[02:50:03.0000] <krijnhoetmer>
/me is going to erase that line from the logs :P

[02:51:00.0000] <met_>
/me used to do ff propaganda in czech republic

[02:51:01.0000] <krijnhoetmer>
I hope you used 'Fx' for that propaganda then? ;)

[02:51:02.0000] <met_>
/me sometime agree, but only sometime

[02:51:03.0000] <krijnhoetmer>
http://www.mozilla.org/support/firefox/faq.html#spell-abbreviate

[02:53:00.0000] <met_>
/me will prefers ff for the end of his life

[04:13:00.0000] <Lachy>
krijnhoetmer, that FAQ entry is poor.  FF is the most commonly used abbreviation, although FX does get used in SFX for SpreadFirefox

[04:14:00.0000] <Lachy>
although it's correcrt about not calling it FireFox - that just annoys me everytime I see it

[04:15:00.0000] <html6>
don't read, test sentance

[04:16:00.0000] <Lachy>
oops, sorry, I read the sentence and now it's too late :-(

[04:17:00.0000] <zcorpan_>
Lachy: "fx" or "Fx" is the preferred abbreviation according to Mozilla's FAQ

[04:18:00.0000] <Lachy>
yeah, like I said, that's wrong

[04:18:01.0000] <Lachy>
it's always been FF for as long as I've known it

[04:21:00.0000] <Toolskyn>
Lachy, as far as I know Mozilla made that FAQ entry as soon as the name Firefox was introduced, so how can it be wrong? ;)

[04:22:00.0000] <zcorpan_>
yeah, people naturally abbreviate it to FF, i don't know why the moz guys don't take a step back and say that it is ok

[05:54:00.0000] <krijnhoetmer>
Lachy: Ah, okay :)

[06:32:00.0000] <Ian_>
may i ask why you so choose HTML4 rather than XHTML1.1 to base your new idea on?

[06:34:00.0000] <Lachy>
XHTML 1.1 is a useless spec

[06:35:00.0000] <virtuelv>
worse than useless, in fact

[06:35:01.0000] <Lachy>
HTML4 is also useless, but still much closer to reality

[06:36:00.0000] <Ian_>
see, i think XHTML is quite good. For one thing it encourages that the webpage should centralise around content, and the CSS style sheet around styling.Rather useful for low-capacity low-bandwidth devices such as mobiles to just get the data.

[06:37:00.0000] <Lachy>
that's the zeldman myth, it's not true

[06:37:01.0000] <Lachy>
XHTML1 was just a reformulation of HTML4 as XML

[06:37:02.0000] <Lachy>
the semantics are idenitical

[06:37:03.0000] <Ian_>
well - i certainly think style stuff should not be in HTML tags but in style sheets

[06:38:00.0000] <virtuelv>
Ian_: there's _nothing_ preventing you from doing that to HTML4

[06:38:01.0000] <Lachy>
it does not encourage the separation of content and presentation any more or less than HTML does.

[06:38:02.0000] <Ian_>
and what is so bad with XML?

[06:38:03.0000] <virtuelv>
or WA1.0

[06:38:04.0000] <Lachy>
it's just a different syntax for the same vocabulary

[06:39:00.0000] <Ian_>
well - will this HTML5 idea remove things such as td height and width and leave that to style sheets

[06:39:01.0000] <virtuelv>
Ian_: what's so bad about sending user content as XML? read http://annevankesteren.nl/2005/11/draconian

[06:39:02.0000] <Lachy>
Ian_, HTML5 defines 2 serialisations: One HTML and one XML

[06:39:03.0000] <Lachy>
so there is both HTML5 and XHTML5

[06:39:04.0000] <Lachy>
both are the same langauage, just different syntaxes

[06:40:00.0000] <Ian_>
okay

[06:41:00.0000] <Ian_>
it is just the styling element which concerns me

[06:41:01.0000] <annevk>
hey, a non-Hixie Ian

[06:41:02.0000] <jdandrea>
:)

[06:42:00.0000] <Ian_>
*)

[06:42:01.0000] <Ian_>
* 8)

[06:42:02.0000] <annevk>
FWIW: HTML5 is also based on XHTML 1.1

[06:42:03.0000] <Ian_>
okay

[06:42:04.0000] <Ian_>
i should really look into some of the height, width and align etc tags to see if they remain - that would just be annoying :p

[06:42:05.0000] <annevk>
Ian_, http://simon.html5.org/html5-elements should give you an idea of what elements and attributes HTML5 will have

[06:43:00.0000] <Ian_>
thanks

[06:43:01.0000] <Ian_>
may i ask, i dont see it there, but will he style="" tag remain?

[06:44:00.0000] <Lachy>
hopefully not

[06:44:01.0000] <zcorpan_>
Ian_: that's not resolved yet. currently only <font style> is allowed by wysiwyg

[06:44:02.0000] <Lachy>
though, it will be defined, I hope it will remain non-conforming

[06:44:03.0000] <annevk>
Ian_, it's likely that the attribute will be added back, yes

[06:45:00.0000] <Lachy>
though, I think it's better than the alternative of <font style>

[06:45:01.0000] <hasather>
Lachy: agreed

[06:45:02.0000] <annevk>
Ian_, http://wiki.whatwg.org/wiki/Changes_from_HTML4 lists some changes from HTML4

[06:45:03.0000] <annevk>
Ian_, most of them, I think

[06:45:04.0000] <Ian_>
<font> is evil

[06:45:05.0000] <annevk>
<font> is the only thing that makes sense for WYSIWYG editors

[06:46:00.0000] <Ian_>
guess you should tell WYSIWYG editors to update themselves if they want to use HTML5

[06:46:01.0000] <Ian_>
oh - scraping <br/> for the plain <br>

[06:47:00.0000] <annevk>
Ian_, <br/> is allowed

[06:47:01.0000] <Lachy>
does anyone know if javascript: URIs have ever been formally defined in any spec?

[06:47:02.0000] <Ian_>
hmm - i dont like the idea of a double standard

[06:47:03.0000] <annevk>
/me doesn't think syntax is worthy of much discussion though

[06:48:00.0000] <Ian_>
i think i may be much more constructive if i actuctly contributed to discussions :P

[06:48:01.0000] <zcorpan_>
Lachy: i think there's an rfc or i-d for it

[06:48:02.0000] <Lachy>
HTML5 has a reference to [JSURI], but there's no references in the spec yet

[06:48:03.0000] <annevk>
Ian_, specifically, the trailing slash is allowed as a symbol of faith

[06:48:04.0000] <annevk>
Ian_, or something in that direction, for people who believe in that sort of thing :)

[06:48:05.0000] <Ian_>
yey - all hail />

[06:48:06.0000] <Lachy>
it's not listed here http://www.iana.org/assignments/uri-schemes.html

[06:49:00.0000] <zcorpan_>
Lachy: http://www.ietf.org/internet-drafts/draft-hoehrmann-javascript-scheme-00.txt

[06:50:00.0000] <Lachy>
awesome!

[06:50:01.0000] <jdandrea>
And on my birthday no less. :)

[06:51:00.0000] <jdandrea>
This post appears to have predated the Hoehrmann draft too: Lachy: See also - http://www1.ietf.org/mail-archive/web/discuss/current/msg00330.html

[06:51:01.0000] <Ian_>
http://www.w3.org/Help/Account/Request/Public - hmm , i never give my phone number online- no need to give it

[06:52:00.0000] <annevk>
i don't think it will be shown to non members

[06:52:01.0000] <annevk>
then again...

[06:52:02.0000] <Lachy>
jdandrea, I already found that email thread when I searched :-)

[06:52:03.0000] <jdandrea>
Lachy: Ahh, got it.

[06:53:00.0000] <Ian_>
ok - i will poke around on the site a bit

[06:53:01.0000] <Ian_>
cya

[07:09:00.0000] <Lachy>
html6, I have no idea what you were trying to say in this email http://lists.w3.org/Archives/Public/public-html/2007Apr/0694.html

[07:10:00.0000] <Lachy>
is anyone else able to comprehend it?

[07:11:00.0000] <jdandrea>
Lachy: I think it refers to wildcarding.

[07:11:01.0000] <Lachy>
wild carding in the browsers address bar?  What for?

[07:11:02.0000] <jdandrea>
:)

[07:12:00.0000] <jdandrea>
To get a list of matching, previously visited links. (As Laurens replies, not something for an HTML spec.)

[07:13:00.0000] <annevk>
I think it's a feature request

[07:13:01.0000] <annevk>
for the UI of web browsers

[07:13:02.0000] <Lachy>
I realise that from reading the replies to it, but I don't understand what the feature request actually is

[07:13:03.0000] <jdandrea>
/me nods

[07:13:04.0000] <Lachy>
oh, like a histroy search feature

[07:14:00.0000] <jdandrea>
/me nods again

[07:27:00.0000] <Lachy>
oh my gosh, does anyone know what a "tag PACK" is? - another one of Dmitry Turin's emails.

[07:32:00.0000] <jdandrea>
Lachy: I'm reading the msg it supposedly replies to ... and - this is just a wild guess - I'm wondering if it refers to a related set of elements (table, thead, tfoot, et. al.) ???

[07:33:00.0000] <Lachy>
I'll reply and ask him.

[07:34:00.0000] <jdandrea>
Lachy: A ha! Don't know if you've seen this yet, but: http://lists.w3.org/Archives/Public/public-html/2007Apr/0520.html

[07:34:01.0000] <jdandrea>
See <pack host="id1"> ...

[07:34:02.0000] <jdandrea>
It's literally a pack element.

[07:36:00.0000] <Lachy>
Do you think it would be rude if I suggested that he seek assistence from someone who is more fluent in english than he is?

[07:37:00.0000] <Lachy>
I still have no idea what a pack element is supposed to be

[07:38:00.0000] <jdandrea>
Neither do I. host .. ? Hmm.

[07:39:00.0000] <annevk>
i believe it's declaritive drawing with HTML

[07:39:01.0000] <annevk>
similar to the <line> proposal

[07:39:02.0000] <Lachy>
oh well, I'm going to add him to my growing list of people to ignore

[07:40:00.0000] <Lachy>
currently up to 3 people, I think

[07:40:01.0000] <annevk>
the problem with that is that some ideas might be good

[07:42:00.0000] <Lachy>
I'll skim the messages.  If I can comprehend anything that's worthwhile, then I'll see it, but I'm not going to spend time trying to interpret any more

[07:43:00.0000] <annevk>
/me has 95 draft messages from started replies that never have been sent

[07:43:01.0000] <Lachy>
though, I've seen the ideas on his website.  I doubt there will be any that are worthwhile

[07:43:02.0000] <Lachy>
wow!

[07:43:03.0000] <Lachy>
how long have those drafts been building up?

[07:43:04.0000] <annevk>
accumulated over some time

[07:43:05.0000] <annevk>
i won't be sending them btw

[07:44:00.0000] <annevk>
more things I can simply answer an then realize that someone else will probably do that

[07:44:01.0000] <annevk>
so I won't bother

[07:45:00.0000] <Lachy>
I have 6 in my drafts.  1 for whatwg, 1 for www-tag, 1 for Mike Schinkel (probably won't send) and one I'm writing now for public-html, and one personal

[07:45:01.0000] <Lachy>
s/6/5/

[08:02:00.0000] <annevk>
see, jgraham just answered the e-mail I had a draft for

[08:02:01.0000] <annevk>
and better than my one liner

[08:03:00.0000] <annevk>
/me had something like "Take it up with your browser vendor. No need for interoperability on UI features."

[08:04:00.0000] <Lachy>
I was drafting a reply to that one too :-)

[08:05:00.0000] <annevk>
anyway, time to move places

[08:05:01.0000] <annevk>
s/places/somewhere else/

[08:05:02.0000] <Lachy>
http://webstandardsgroup.org/features/lachlan-hunt.cfm is published!

[08:06:00.0000] <jdandrea>
Congrats! Reading.

[08:23:00.0000] <krijnh>
"I�ve heard rumours that you were once fired from a job for refusing to top post." - priceless :p

[08:31:00.0000] <Lachy>
yeah, it seems companies just can't stop thinking up pathetic reasons to fire me

[08:32:00.0000] <krijnh>
I got fired once, because I didn't want to wear a tie

[08:32:01.0000] <krijnh>
(and because I made fun of my boss, that as well)

[08:33:00.0000] <Lachy>
I got fired for: 1. Refusing to to replace all &amp; with & in an XHTML document (even though it was text/html) and then publicly call my boss a moron

[08:33:01.0000] <Lachy>
2. Refusig to top post

[08:34:00.0000] <Lachy>
3. Not having any skills with Flash

[08:34:01.0000] <jdandrea>
Lachy: They wanted you to REPLACE &amp; with &?

[08:34:02.0000] <Lachy>
yes!

[08:34:03.0000] <jdandrea>
/me shakes head

[08:34:04.0000] <krijnh>
Why would they want that?

[08:35:00.0000] <Lachy>
I had to remove the blog entry about it cause they threatened legal action and I couldn't be bothered fight it on free speech grounds

[08:36:00.0000] <Lachy>
but the reason was that on my boss' computer, an image was failing to load and he assumed that the cause was the use of &amp; in the query string

[08:36:01.0000] <Lachy>
despite the fact that it worked on every other comptuer I tested

[08:39:00.0000] <krijnh>
Silly boss

[12:27:00.0000] <hendry>
anyone seen the OLPC? http://flickr.com/photos/hendry/457911926/

[15:52:00.0000] <othermaciej>
jdandrea: "oh my" where?

[16:01:00.0000] <Philip`>
Hixie: Is the slowness in the spec splitter just the parsing phase? It would be much better if the spec was written in XHTML, so we could use existing fast XML parsing libraries instead of having to write and optimise an HTML5 one instead ;-)

[16:02:00.0000] <Philip`>
(Alternatively, I suppose it's a use case that provides some motivation to rewrite html5lib in C...)

[16:14:00.0000] <Hixie>
Philip`: no, it's the output phase as far as i can tell

[16:14:01.0000] <Hixie>
it's not slowness that's the problem

[16:15:00.0000] <Hixie>
it's cpu spiking

[16:15:01.0000] <Hixie>
i could probably just stick some sleep statements in there

[16:33:00.0000] <gsnedders>
othermaciej: manage to do anything about the t-shirts?

[16:37:00.0000] <othermaciej>
gsnedders: yes

[16:37:01.0000] <othermaciej>
http://five-gt-two.spreadshirt.com/

[16:37:02.0000] <othermaciej>
should I send it to the whatwg list or would that be too inflammatory?

[16:38:00.0000] <gsnedders>
*giggles at the URL*

[16:38:01.0000] <hasather>
othermaciej: how many have you sold?

[16:38:02.0000] <gsnedders>
othermaciej: am I right seeing nothing more than 5 > 2 on it?

[16:39:00.0000] <zcorpan_>
othermaciej: you could blog about it on blog.whatwg.org

[16:40:00.0000] <hasather>
gsnedders: your eyes are just fine

[16:41:00.0000] <gsnedders>
yay. I needn't get my glasses.

[16:42:00.0000] <gsnedders>
othermaciej: is it possible to get any other colours?

[16:46:00.0000] <othermaciej>
gsnedders: yes

[16:47:00.0000] <othermaciej>
gsnedders: I tried to use the higher-quality t-shirts for it, which are available in a limited number of colors - do you have a specific request?

[16:48:00.0000] <othermaciej>
3 orders so far btw

[16:48:01.0000] <zcorpan_>
#3c790a

[16:48:02.0000] <othermaciej>
if anyone wants to blog it, feel free, I don't want to promote it too much myself (I set it up but I'm not getting a cut of the sales or anything)

[16:48:03.0000] <zcorpan_>
...is the whatwg green, i think

[16:49:00.0000] <othermaciej>
2 white and one black so far

[16:49:01.0000] <zcorpan_>
/me ordered a white one

[16:49:02.0000] <othermaciej>
zcorpan_: the set of available colors for printing is fixed - I tried to get the closest match I could find

[16:49:03.0000] <zcorpan_>
ok

[16:51:00.0000] <zcorpan_>
what should the blog post say?

[16:54:00.0000] <Hixie>
i love the t-shirts, but if you blog about it, remember that it's not in our best interests to piss off the xhtml2wg more than we already do...

[16:54:01.0000] <othermaciej>
if you blogged it, I would make it vague

[16:54:02.0000] <zcorpan_>
should i blog it at all, then?

[16:55:00.0000] <othermaciej>
like "T-Shirts for Fans of Obvious Mathematical Statements"


2007-04-14
[17:03:00.0000] <zcorpan_>
er, i thought TIME predated SMIL, no?

[17:03:01.0000] <Hixie>
HTML+TIME is a variant of SMIL for HTML. Not sure which came first.

[17:03:02.0000] <zcorpan_>
ok

[17:15:00.0000] <zcorpan_>
seems i've lost my password on blog.whatwg.org. i haven't received any email with my password, despite using the "lost your password?" thing twice

[17:15:01.0000] <Hixie>
weird

[17:17:00.0000] <zcorpan_>
ah, found it

[17:17:01.0000] <zcorpan_>
better change it to something i can remember

[17:28:00.0000] <zcorpan_>
blog post: "5 > 2. So now you know." should i say anything more? :)

[17:29:00.0000] <zcorpan_>
(title being the one om_beer suggested)

[17:29:01.0000] <zcorpan_>
(with "5 > 2" being a link)

[17:39:00.0000] <othermaciej>
zcorpan_: that's enough for people who know

[17:49:00.0000] <deltab>
http://www.whatwg.org/specs/web-apps/current-work/multipage/section-dynamic.html  is empty and everything after it is missing

[17:50:00.0000] <Hixie>
yeah working on it

[17:50:01.0000] <Hixie>
the script is taking too much cpu on my box so i'm farming out the actual splitting to another box in the UK

[17:50:02.0000] <deltab>
why does it take too much cpu?

[17:51:00.0000] <zcorpan_>
http://blog.whatwg.org/t-shirts

[17:51:01.0000] <Hixie>
deltab: it's a big spec :-)

[17:52:00.0000] <deltab>
what's involved in splitting it? indexing and rewriting all the links, I guess

[17:53:00.0000] <Hixie>
http://html5.googlecode.com/svn/trunk/spec-splitter/spec-splitter.py

[17:53:01.0000] <Hixie>
it should be fixed now, btw

[17:56:00.0000] <Philip`>
deltab: It parses it into a DOM, then extracts the common header/etc bits, then it keeps moving bits from the original single page into the new output page, and it starts a new output page when it reaches certain points (mostly <h2>/<h3>), and when it's finished it goes through every page and finds every ID, then it goes through every page and updates the links, and then it serialises everything out to HTML, and that's about

[17:56:01.0000] <Philip`>
... it.

[17:58:00.0000] <deltab>
ah, DOM + 1.5 MB source

[17:59:00.0000] <Philip`>
+ Python :-)

[17:59:01.0000] <Hixie>
my update script now calls a web service on w3.org to create the big index page, then calls a web service on a server i have access to in the UK to split the spec up, and that web service calls a web service on whatwg.org to fetch the resulting tarball and unzip it, and then the original script goes ahead and does a svn diff.

[17:59:02.0000] <Hixie>
talk about a contraption.

[18:00:00.0000] <jdandrea>
Now THAT's quite a process. Hey, if it works, it works.

[18:01:00.0000] <Hixie>
it would be great parallel processing if it wasn't for the way that each script waits for the previous one to finish before doing anything

[18:03:00.0000] <Hixie>
by far the slowest part of this process now is the generation of the cross-references in the first place

[18:05:00.0000] <Lachy>
good morning

[18:05:01.0000] <Hixie>
might be time soon for me (or someone else) to write the script that takes the source file, and numbers the headers, generates the toc, and generates the cross-references

[18:05:02.0000] <Hixie>
hey lachy

[18:06:00.0000] <Philip``>
Hixie: I have a surprisingly similar (but totally unrelated) system - user commits to SVN on server A, A notifies server B, B updates SVN from A, B notifies C, C updates SVN from A, B spends ~20 minutes compiling the C++ application from SVN, C creates HTML/Atom of SVN history and SCPs it to server D, and after an hour B commits the compiled program back into SVN...

[18:07:00.0000] <Hixie>
hah

[18:07:01.0000] <Philip``>
The really surprising thing is that it actually works at all

[18:07:02.0000] <Hixie>
yeah

[18:07:03.0000] <deltab>
compiled programs in SVN?

[18:07:04.0000] <Philip``>
It's kind of grown over the years

[18:08:00.0000] <Philip``>
(Oh, I forgot that A also notifies Trac so it can automatically close tickets and email the relevant people)

[18:08:01.0000] <deltab>
oh, I suppose you don't want to spend time recompiling just to run an old version

[18:09:00.0000] <Philip``>
deltab: It's for a game where there are people (e.g. artists) who can't compile it themselves, so the best solution seems to be to put the executable file into SVN - they can just do a simple checkout and run it

[18:10:00.0000] <Hixie>
ok the multipage stuff now runs in the background so i no longer have to wait for that to finish to actually check in

[18:11:00.0000] <Hixie>
anyone know what the status of the annotate-data.xml file is? how to update it, etc?

[18:13:00.0000] <zcorpan_>
i haven't looked at it yet

[18:13:01.0000] <Hixie>
k

[18:13:02.0000] <Hixie>
i guess i should do an update to it, it's getting a bit stale

[18:14:00.0000] <zcorpan_>
i guess i should look at making it easier to update

[18:14:01.0000] <zcorpan_>
:)

[18:14:02.0000] <Hixie>
heh

[18:14:03.0000] <othermaciej>
I'm still snickering at Lachy's #xhtml log

[18:14:04.0000] <Hixie>
dude that was some disturbing stuff

[18:15:00.0000] <othermaciej>
at least they are safe in their padded cell (mostly)

[18:16:00.0000] <Hixie>
would be interesting to see shane speak to chris

[18:16:01.0000] <Hixie>
given shane wants to break content that isn't valid (over 93% of the web) and christ wants to not fix any bugs

[18:16:02.0000] <Hixie>
er, chris, not christ.

[18:17:00.0000] <othermaciej>
I'm not really sure what to say in response to Chris's stuff

[18:18:00.0000] <othermaciej>
but I'll have to write something

[18:18:01.0000] <Hixie>
you realise we're going to end up having to implement all these modes, right

[18:18:02.0000] <Hixie>
that's the most frightening part of this

[18:19:00.0000] <Hixie>
by "we" i mean mozilla/opera/apple in terms of browsers and me in terms of the spec

[18:20:00.0000] <othermaciej>
well, it depends on how quickly browsers fix other compat issues such that the modes are the most important compat isssue vs. changing balance of market share

[18:20:01.0000] <othermaciej>
it is somewhat true that due to browser targetting there are some bugs that are compat issues *only* for IE

[18:20:02.0000] <Hixie>
yeah the only way to get out of this mess would be for them to lose 50% market share, at which point they'd be even more screwed by their modes than we would

[18:21:00.0000] <othermaciej>
if we emulated some IE bugs in Safari, more content would break than stort working

[18:21:01.0000] <Hixie>
since we wouldn't have a mode, but they couldn't render the new stuff in their default mode

[18:21:02.0000] <Hixie>
which would be hilarious

[18:21:03.0000] <Hixie>
i really don't know what should happen, yeah

[18:21:04.0000] <Dashiva>
No hope of them committing to just fixing bugs before they become relied upon, I suppose

[18:22:00.0000] <Lachy>
I once thought that if FF, O and S implemented the spec and proved it was compat with the web, MS would follow, but it's now clear that sadly not the case

[18:22:01.0000] <Dashiva>
The reason people rely on e.g. peekaboo is because it wasn't fixed, not because they want to

[18:22:02.0000] <zcorpan_>
Lachy: indeed

[18:23:00.0000] <jdandrea>
I was just going to say that. "a bunch of people did expect that we didn't implement child selectors in CSS." No, speaking for myself, I _didn't_ expect it. It just became the norm for IE ... eventually.

[18:23:01.0000] <Hixie>
we're not going to convince microsoft not to do versionning

[18:23:02.0000] <Hixie>
we can definitely stand firm against spec-condoned opt-ins

[18:24:00.0000] <othermaciej>
well, there's one aspect of this that maybe cwilso hasn't thought about clearly

[18:24:01.0000] <Hixie>
and we can try to make the spec even more compatible than we have so far

[18:24:02.0000] <Hixie>
/me bites his tongue

[18:24:03.0000] <othermaciej>
which is that the likely content-breaking points of incompatibility are mostly in CSS and core DOM, not in HTML

[18:24:04.0000] <zcorpan_>
let's make the html4 strict and xhtml1 strict doctypes conforming html5 doctypes :)

[18:24:05.0000] <Hixie>
othermaciej: there's certainly some incompatibilities in html too, but yes

[18:24:06.0000] <othermaciej>
none of the examples he raised were actually HTML examples

[18:24:07.0000] <Hixie>
<button type=> default, parsing, those are html 

[18:25:00.0000] <othermaciej>
(except for the <object> example which was fallacious)

[18:25:01.0000] <jdandrea>
othermaciej: Interesting point there ... !

[18:25:02.0000] <Hixie>
yeah i don't know if he really believes his <object> argument or what

[18:25:03.0000] <Lachy>
zcorpan_, that may not be such a bad idea from an authoring tool perspective, since so many already insert such doctypes by default

[18:25:04.0000] <othermaciej>
amusingly, there is an Apple intranet site I know of that won't work in IE because of their differing button type default)

[18:26:00.0000] <Hixie>
hah

[18:26:01.0000] <zcorpan_>
i filed bugs about <button type> default and <button> using innerText instead of .value, but they were dismissed as "by design"

[18:26:02.0000] <Hixie>
!

[18:27:00.0000] <zcorpan_>
i shit you not

[18:27:01.0000] <Dashiva>
"by design" means "by not being fixed long enough for sites to rely on it" :/

[18:27:02.0000] <Lachy>
as I wrote in an off-list maill to Hixie last night, I think it would be good if <button> defaulted to type=button, but <button type=unknown> defaulted to type=submit

[18:28:00.0000] <othermaciej>
I wonder if it is true that MS got sued for fixing IE bugs

[18:28:01.0000] <zcorpan_>
<button> defaulting to type=button makes it harder to make repetition templates fallback server side

[18:28:02.0000] <Lachy>
ask him for evidence that it would happen

[18:28:03.0000] <zcorpan_>
you have to use client side script

[18:28:04.0000] <Hixie>
zcorpan_: well that's already hard

[18:28:05.0000] <Hixie>
zcorpan_: because of IE

[18:28:06.0000] <zcorpan_>
yes

[18:28:07.0000] <Hixie>
zcorpan_: the point is Lachy's suggestion doesn't break that

[18:28:08.0000] <jdandrea>
Didn't someone mention the EULA sort of limits their liability??

[18:29:00.0000] <zcorpan_>
true

[18:29:01.0000] <Hixie>
zcorpan_: since those have <button type=unknown>

[18:29:02.0000] <Dashiva>
Even if the EULA says stuff, it can't override local law

[18:29:03.0000] <zcorpan_>
i would be fine with making <button> default to type=button then, given that ms don't intend to fix it

[18:30:00.0000] <Lachy>
he seems to be going from: Microsoft gets sued a lot to Microsoft will get sued for breathing

[18:30:01.0000] <Dashiva>
I would also enjoy default type button, since styling regular buttons is my main use for button

[18:30:02.0000] <Hixie>
zcorpan_: send a mail and i'll add it to the wf2 pile

[18:30:03.0000] <zcorpan_>
ok

[18:31:00.0000] <Hixie>
thanks

[18:31:01.0000] <Hixie>
ok i've reorganised the browsing contexts section to have a saner order, this should make it easier to write the spec

[18:31:02.0000] <Hixie>
and read it

[18:31:03.0000] <Dashiva>
So gentlemen, how do we get IE's market share down to 50% so they'll listen to reason?

[18:31:04.0000] <Hixie>
make ff3 suck less would be one start

[18:32:00.0000] <Dashiva>
That's replacing one impossibility with another

[18:32:01.0000] <othermaciej>
does FF3 suck?

[18:33:00.0000] <othermaciej>
as in, more than IE7?

[18:33:01.0000] <Hixie>
the ff3 trunk is bad

[18:33:02.0000] <Hixie>
crashy, slow, bloated, bigger download than ff2

[18:33:03.0000] <Hixie>
not to mention flaky

[18:33:04.0000] <Hixie>
it's rather sad

[18:34:00.0000] <Lachy>
I had to uninstall teh FF trunk cause it kept crashing whenever I used it

[18:34:01.0000] <zcorpan_>
print preview in my fx trunk is broken

[18:35:00.0000] <Dashiva>
It's not like pre-beta builds are all that stable in general, of course

[18:35:01.0000] <Hixie>
this is worse than usual

[18:36:00.0000] <Hixie>
i've been using pre-alpha browsers for literally 9 or more years and this is the worst i've seen since the netscape 6 days

[18:37:00.0000] <zcorpan_>
what about .innerText vs .value on <button>?

[18:37:01.0000] <Hixie>
haven't done anything with .innerText yet

[18:37:02.0000] <Hixie>
what is it?

[18:37:03.0000] <Hixie>
i mean, what is it in the context of buttons

[18:38:00.0000] <Hixie>
i know what the attribute is in general

[18:38:01.0000] <zcorpan_>
ie submits .innerText to the server instead of .value

[18:38:02.0000] <zcorpan_>
<button type=submit value=pass>fail</button>

[18:38:03.0000] <Hixie>
ah

[18:38:04.0000] <Dashiva>
We can't spec that, it would make them even more useless than currently

[18:38:05.0000] <Hixie>
do other browsers do the right thing?

[18:38:06.0000] <Dashiva>
Might as well deprecate type=submit

[18:38:07.0000] <zcorpan_>
Hixie: yes

[18:39:00.0000] <Hixie>
let's leave it then, clearly it's not THAT important

[18:39:01.0000] <zcorpan_>
ok

[18:41:00.0000] <Dashiva>
If anyone ever made a page depending on that, I want to stab them :D

[18:41:01.0000] <Hixie>
they might not do it on purpose

[18:42:00.0000] <Dashiva>
It would require not specifying value at all, and still expecting one

[18:42:01.0000] <Hixie>
yeah

[18:42:02.0000] <Hixie>
easy mistake

[18:42:03.0000] <Hixie>
you know a lot of people write code by seeing what they get and coding to it

[18:42:04.0000] <Hixie>
not by thinking about what they're doing

[18:42:05.0000] <Hixie>
even you, i'm sure, do that

[18:42:06.0000] <zcorpan_>
logical way of working

[18:43:00.0000] <zcorpan_>
reading specs is more work

[18:43:01.0000] <Dashiva>
But the behavior when value in unspecified is unspecified, so other browsers could conform to that

[18:43:02.0000] <Dashiva>
Would be like <option> with no value

[18:44:00.0000] <zcorpan_>
yeah, that could work

[18:45:00.0000] <Dashiva>
Who wants to bring it up with MS? :)

[18:46:00.0000] <zcorpan_>
on public-html?

[18:46:01.0000] <Hixie>
just send a mail to the whatwg list and i'll update the spec in due course (make it explicit since i won't look at this for several months probably)

[18:46:02.0000] <Hixie>
btw if the multipage page gets out of sync let me know, the system is a bit brittle and i'm not sure why it's not always working

[18:47:00.0000] <zcorpan_>
Hixie: the mail i just sent, was it too brief?

[18:47:01.0000] <Hixie>
no that was fine

[18:47:02.0000] <Hixie>
cos it was a simple issue :-)

[18:47:03.0000] <zcorpan_>
ok

[18:47:04.0000] <Dashiva>
Should've mentioned defaulting to submit for unknown

[18:47:05.0000] <Hixie>
this issue is a bit more complex

[18:47:06.0000] <zcorpan_>
/me drafts what we have discussed so far

[18:48:00.0000] <Dashiva>
"initial value of the button element is the value attribute, or the contents of the element if it is missing"

[18:52:00.0000] <Lachy>
Hixie, how widely used is <button type=submit>?

[18:52:01.0000] <Lachy>
do you have stats on that?

[18:53:00.0000] <Hixie>
looks like i don't

[18:53:01.0000] <Dashiva>
Anyone with Safari care to report the submitted values on http://folk.ntnu.no/magnusrk/test/buttonvalue.html ?

[18:53:02.0000] <Lachy>
I know I've used it, but none of my sites depend on IE's non-standard behaviour.  They were all forced to work around it by not using a value

[18:53:03.0000] <Lachy>
1. novalue=

[18:54:00.0000] <Lachy>
2. emptyvalue=

[18:54:01.0000] <Lachy>
3, value=non-empty+value

[18:54:02.0000] <Dashiva>
Same as FF and Opera then

[18:59:00.0000] <Dashiva>
Can anyone imagine a site specifying value but relying on innerText being submitted?

[19:00:00.0000] <Lachy>
it seems unlikely, <input type=submit> would be used far more often for that purpose

[19:00:01.0000] <zcorpan_>
well, some might use it as a way to identify ie on the server, for whatever reason

[19:01:00.0000] <Lachy>
but then if IE is was as standards compliant as other browsers, then it wouldn't really matter that much

[19:01:01.0000] <zcorpan_>
indeed

[19:01:02.0000] <Lachy>
unless the stupid site was using it to lock out other browsers

[19:01:03.0000] <Lachy>
but that seems unlikely to use that method

[19:02:00.0000] <zcorpan_>
from what i've heard the reason to process .innerText was to work around the bug. so fixing it wouldn't break those sites

[19:02:01.0000] <zcorpan_>
...on the server that is

[19:02:02.0000] <zcorpan_>
or with js

[19:03:00.0000] <Dashiva>
It would only break if a site specifies value, but relies on only innerText, which only makes sense from a viewpoint of intentional breakage, not from laziness or ignorance

[19:03:01.0000] <zcorpan_>
Dashiva: what's the difference between the first and second button?

[19:03:02.0000] <Dashiva>
Empty attribute and missing attribute

[19:04:00.0000] <zcorpan_>
/me doesn't see the empty attribute

[19:04:01.0000] <Dashiva>
... right

[19:04:02.0000] <Dashiva>
/me fix

[19:05:00.0000] <Dashiva>
It doesn't really matter now, since it would only matter in browsers that implement the option-like behavior

[19:05:01.0000] <Hixie>
can someone rephrase this to fix the obvious problem without removing the precision and conformance criteria?: "When invoked as a constructor, this constructor must construct a"

[19:06:00.0000] <othermaciej>
the problem being word repetition?

[19:06:01.0000] <Dashiva>
How do you invoke a constructor as a non-constructor? Leaving out new?

[19:06:02.0000] <othermaciej>
many constructors can also be called as functions

[19:07:00.0000] <Hixie>
yes and yes

[19:07:01.0000] <Hixie>
i can replace the last "construct" with "replace a new"

[19:07:02.0000] <Hixie>
euh

[19:07:03.0000] <Hixie>
"return a new"

[19:07:04.0000] <othermaciej>
"return a new", yes

[19:08:00.0000] <Dashiva>
Could move the when clause to the end, but that would be easier to miss

[19:08:01.0000] <othermaciej>
and you could replace "this constructor" with the referent (which I assume is "the Audio constructor")

[19:08:02.0000] <Hixie>
Audio, Image, and Option at the moment

[19:08:03.0000] <Dashiva>
e.g. This constructor must return a new (...) when invoked as a constructor.

[19:09:00.0000] <Hixie>
the (...) is too long for that

[19:10:00.0000] <Dashiva>
"The constructor, when invoked as a constructor, must return a new" sounds silly

[19:10:01.0000] <Hixie>
ok i've used "When invoked as constructors, these must return a new"

[19:10:02.0000] <Hixie>
since they have multiple variants in each case

[19:13:00.0000] <othermaciej>
Hixie: if you really want to get technical, per the ES3 spec what happens during "new" is that the "[[Construct]]" internal property is called

[19:14:00.0000] <othermaciej>
but that's probably moredetail than you want and may be incompatible with future ES specs

[19:14:01.0000] <Hixie>
i'm also supposed to be pretending to attempt to stay language-neutral

[19:14:02.0000] <Hixie>
but yeah

[19:14:03.0000] <Hixie>
maybe i should just say that

[19:15:00.0000] <Dashiva>
Is whatwg involved in es4 in any way?

[19:15:01.0000] <zcorpan_>
*now* i got my password from the blog. very helpful :)

[19:15:02.0000] <Hixie>
i want a spec to define the whole HTMLFooElement and .prototype stuff too

[19:15:03.0000] <othermaciej>
other than vague overlap of participants, not really

[19:16:00.0000] <othermaciej>
Hixie: it would indeed be nice to have a spec for that, I wish it had existed before we reverse-engineered Firefox to add it to Safari

[19:16:01.0000] <othermaciej>
hmm, Chris Wilson's vacation mail says he will be at web 2.0 expo, I wonder if he'd be interested in meeting up for dinner or something

[19:17:00.0000] <Hixie>
you going to web 2.0 expo?

[19:18:00.0000] <othermaciej>
no, but it's in SF apparently

[19:19:00.0000] <Hixie>
ah

[19:19:01.0000] <Hixie>
well if he does want to meet up, let me know

[19:20:00.0000] <othermaciej>
roger that

[19:21:00.0000] <Hixie>
is this accurate?:

[19:21:01.0000] <Hixie>
Each Document has a scripting context, which provides a structure in which scripts can execute. Scripting contexts are associated with a domain, which defines the security context of that scripting context. Each scripting context also has a global scope, which is represented by the object representing the Document's default view, the Window object.

[19:21:02.0000] <othermaciej>
I'm not sure what "scripting context" means, but I think it is more properly associated with the browsing context than with the document

[19:22:00.0000] <Hixie>
oh right Window elements survive page loads

[19:25:00.0000] <Hixie>
so by the web's model, you couldn't make a browser that actively rendered and let you interact with more than one Document per browsing context, right?

[19:25:01.0000] <Hixie>
like, you couldn't have a live view of all the documents in history, one of them would have to be the "live" one

[19:27:00.0000] <Dashiva>
Hixie: How do iframes figure into that?

[19:28:00.0000] <Hixie>
oh, ignore those

[19:28:01.0000] <Hixie>
they introduce new browsing contexts

[19:29:00.0000] <Dashiva>
So browsing context and window are 1:1?

[19:29:01.0000] <Hixie>
yes

[19:30:00.0000] <Dashiva>
XMLHttpRequest lets you access a Document without a new window

[19:31:00.0000] <Dashiva>
Just creating empty Documents aside, I imagine there are other ways to instantiate new ones (LSParser, etc)

[19:31:01.0000] <Hixie>
the Document without a window isn't in a browsing context

[19:31:02.0000] <Hixie>
that's just a Document

[19:31:03.0000] <Hixie>
it can't be "live"

[19:33:00.0000] <Dashiva>
Well, if neither internal (above) nor external documents (frames) satisfy it, I don't see how it could happen

[19:33:01.0000] <Hixie>
imagine a UI where you can interact with all the documents in your session history at once

[19:33:02.0000] <Hixie>
instead of the current UI where you can only see one at a time

[19:34:00.0000] <Dashiva>
How would that be different from a tabbed browser with one static tab per page in history?

[19:34:01.0000] <Hixie>
with the tabbed ui, only one of the pages is "active"

[19:35:00.0000] <Hixie>
script isn't running in all of them at once, presumably

[19:35:01.0000] <Hixie>
or if it is, then it's what i'm describing, and my question is, does the model that html uses support that at all

[19:35:02.0000] <Dashiva>
Running a CPU-intensive script in one tab and focusing another should show scripts still run in all tabs

[19:36:00.0000] <Hixie>
current browsers don't let you put all the documents in your history in different tabs.

[19:37:00.0000] <Dashiva>
Caching details aside, could you not get the behavior by always using 'open in new tab'?

[19:38:00.0000] <Hixie>
no, because then they would be in different browsing contexts with different session histories.

[19:38:01.0000] <Hixie>
i'm talking about one session history

[19:38:02.0000] <Dashiva>
ah

[19:41:00.0000] <Dashiva>
In the JS circles I frequent, it's a truism that neither scripting context nor window survives a page load

[19:41:01.0000] <Hixie>
Window survives page load

[19:42:00.0000] <Hixie>
if you have an iframe and take its Window object, then navigate to another page in the iframe and take that Window object, and compare them, they're the same.

[19:42:01.0000] <Hixie>
with === equality

[19:42:02.0000] <Dashiva>
But a large amount of its properties are changed/reset

[19:42:03.0000] <Hixie>
same object though

[19:43:00.0000] <Dashiva>
Probably for implementation reasons, it doesn't offer any beneficial behavior that I can think of

[19:44:00.0000] <Dashiva>
To say, nothing would change if browsers started making new objects

[19:45:00.0000] <Hixie>
oh lots of sites would break

[19:45:01.0000] <Hixie>
lots of sites will grab the window object once and manipulate it across page loads

[19:45:02.0000] <Hixie>
(of a child iframe, e.g.)

[19:45:03.0000] <Hixie>
e.g.  var x = window.open(...); /* time passes */ x.close();

[19:46:00.0000] <Dashiva>
Ah, like that

[19:46:01.0000] <Hixie>
where while time passed, the window navigated to other pages

[19:47:00.0000] <Dashiva>
Yeah, that would break. A consequence of the mixing of window object and the frame/tab/window it's contained in

[19:48:00.0000] <Dashiva>
But Opera does manage to preserve the window properties of each page when going back and forward, somehow

[19:48:01.0000] <Hixie>
sure, all browsers do

[19:48:02.0000] <Hixie>
the html5 spec even says to do it

[19:48:03.0000] <Dashiva>
/me envisions time-share of the window object to allow concurrent use of a window object with different documents

[19:49:00.0000] <Dashiva>
But other than references to the window from other windows, is window object persistence otherwise used?

[19:50:00.0000] <Hixie>
not sure how else it could be used

[19:53:00.0000] <Dashiva>
Then I'll agree the current model doesn't allow concurrent live pages in the same session history, without some crazy time-sharing, but it could potentially be changed by making window references point to the containing frame/tab/etc and its current window object, rather than a specific one

[20:00:00.0000] <zcorpan_>
/me updates notes at http://simon.html5.org/test/ie7b2-bugs/

[20:05:00.0000] <zcorpan_>
perhaps the html parser should ignore </li> tags (like ie does)

[20:07:00.0000] <Hixie>
test case?

[20:07:01.0000] <Hixie>
i'm pretty sure i found it didn't ignore </li>

[20:12:00.0000] <zcorpan_>
http://simon.html5.org/temp/li.html

[20:12:01.0000] <Hixie>
interesting

[20:12:02.0000] <Hixie>
send mail

[20:12:03.0000] <Hixie>
i've gotta go

[20:13:00.0000] <Hixie>
thanks for the help

[20:13:01.0000] <zcorpan_>
ok

[20:13:02.0000] <zcorpan_>
np

[03:47:00.0000] <met_>
why html5 is not sgml? (just red http://annevankesteren.nl/2007/04/html-red-pill ) i am not sgml-fun, just thought almost any markup can be described by sgml, what is the difference

[03:48:00.0000] <zcorpan_>
sgml doesn't define error handling

[03:48:01.0000] <Lachy>
sgml isn't compatible with the web

[03:49:00.0000] <met_>
 if you define what should happen for tag soup <x><y></x></y> it stops being sgml?

[03:50:00.0000] <zcorpan_>
Lachy: well, an application of sgml could be made compatible with the web (if we also define error handling), but assuming that the sgml declaration for html4 is in place then no. (and defining error handling for sgml is not appropriate for a higher-level spec to do)

[03:50:01.0000] <Lachy>
SGML rules would force <y> to close immediate before the </x>

[03:50:02.0000] <Lachy>
I don't think it's possible to write an SGML declaration that could handle all conforming documents

[03:51:00.0000] <zcorpan_>
and in any case, it's simpler to define the whole thing as a separate language than to continue to pretent that it is sgml

[03:51:01.0000] <Lachy>
you could get close, but not perfect

[03:51:02.0000] <zcorpan_>
ok

[03:51:03.0000] <met_>
thx

[03:51:04.0000] <zcorpan_>
and saying that it is sgml doesn't help implementors

[04:00:00.0000] <zcorpan_>
i should have an email filter that filters out all +1 and -1 emails

[04:01:00.0000] <hendry>
Maulkin: 

[04:02:00.0000] <met_>
zcorpan_, just sum +1 and -1, this gives you zero and zero is not neccessary to read

[04:02:01.0000] <zcorpan_>
the problem is i can't figure out which emails are +1 and -1s until i read them

[04:02:02.0000] <zcorpan_>
and i don't want to read them

[04:03:00.0000] <zcorpan_>
they shouldn't enter my inbox

[04:03:01.0000] <zcorpan_>
obviously it would be nice if people just didn't send +1 or -1 emails

[04:04:00.0000] <met_>
they are used tom from digg and reddit this time

[04:04:01.0000] <met_>
just post mail to reddit and they can vote there 8-)

[04:04:02.0000] <zcorpan_>
i don't read digg or reddit

[05:37:00.0000] <mpt_>
/me is amused that the XHTML2 WG inherited the venerable http://www.w3.org/MarkUp/ URL

[05:39:00.0000] <zcorpan>
should i add http://diveintomark.org/public/2005/12/whitelights/ to the html5 presentations page in the wiki? (it's about other stuff too)

[05:40:00.0000] <zcorpan>
(great presentation)

[05:40:01.0000] <krijnh>
Yes

[05:40:02.0000] <zcorpan>
(or slides, i didn't see the presentation)

[05:40:03.0000] <zcorpan>
ok

[05:41:00.0000] <mpt_>
it's an interesting presentation, but it mentions HTML5 only tangentially

[05:41:01.0000] <krijnh>
It's also from December 2005

[05:41:02.0000] <zcorpan>
indeed

[05:41:03.0000] <krijnh>
Perhaps that's why it mentions HTML5 only tangentially

[05:41:04.0000] <krijnh>
(Learned a new word, yay)

[05:42:00.0000] <zcorpan>
perhaps he didn't want to talk about html5 exclusively

[05:42:01.0000] <mpt_>
It's mainly about writing

[05:42:02.0000] <zcorpan>
so should i add it or not?

[05:44:00.0000] <mpt_>
If this is a vote, mine's -1

[05:44:01.0000] <krijnh>
Hmm

[05:44:02.0000] <krijnh>
What's the wiki URI?

[05:45:00.0000] <Lachy>
wiki.whatwg.org

[05:45:01.0000] <zcorpan>
mpt_: ok

[05:47:00.0000] <krijnh>
Ah, it only has 3 links, nah, then don't add it :)

[09:02:00.0000] <Lachy>
I started drafting up that blog entry to explain the situation - http://lachy.id.au/temp/mistakes

[13:29:00.0000] <gsnedders>
om_brunch: light blue would be nice (re: t-shirt colours)

[14:50:00.0000] <zcorpan>
othermaciej: http://blog.whatwg.org/t-shirts#comments

[14:52:00.0000] <othermaciej>
zcorpan: I'm not sure fantasai hating it is a bad thing (although she might be one of few potential customers for the girly version)

[14:53:00.0000] <othermaciej>
I think she missed the point that the obscurity is completely intentional

[14:53:01.0000] <othermaciej>
it's an inside joke, not an advocacy statement

[14:53:02.0000] <othermaciej>
as for doing it in a better font, that is probably being my personal skill level

[15:28:00.0000] <zcorpan>
just read through http://krijnhoetmer.nl/irc-logs/xhtml/20070413 . amusing.

[15:30:00.0000] <othermaciej>
/me cries

[15:35:00.0000] <jdandrea>
/me is reading ... "oh my ..."

[15:45:00.0000] <jdandrea>
"opera and apple ... are also-rans." ??? Ouch.

[15:46:00.0000] <zcorpan>
as if the "run" was over

[15:47:00.0000] <jdandrea>
exactly.

[15:47:01.0000] <hsivonen>
/me wonders how XHTML 2.0 fared :-)

[15:47:02.0000] <jdandrea>
Sounds like a cautionary tale to me.

[15:48:00.0000] <zcorpan>
sorry guys, ie won and firefox came second, no need to improve safari or opera anymore. let's pack our bags and go home.

[15:48:01.0000] <jdandrea>
heh

[15:52:00.0000] <jdandrea>
othermaciej: "oh my" the entire transcript @ http://krijnhoetmer.nl/irc-logs/xhtml/20070413

[15:52:01.0000] <jdandrea>
Just finished it. Wow. Interesting times ...

[15:52:02.0000] <othermaciej>
oh, right

[15:52:03.0000] <othermaciej>
yeah

[15:53:00.0000] <othermaciej>
well, all Gecko-based browsers total used to have less market share than Safari does now

[15:53:01.0000] <othermaciej>
and Safari's is growing

[15:53:02.0000] <othermaciej>
anyway, that guy is a tool

[15:58:00.0000] <zcorpan>
interesting view though. "i don't care about the existing content on the web, if it stops working then who cares? it was invalid anyway. let's make up a new perfect language with no defined error handling, which isn't a problem because there will be tools that produce perfect code and UAs won't accept non-compliant code."

[15:59:00.0000] <zcorpan>
"and this is handled with a new doctype."

[15:59:01.0000] <zcorpan>
(that's how i interpreted it, anyway)

[16:00:00.0000] <jdandrea>
I'm a (relatively) late arrival to the WHAT-WG myself, but I've been reading the docs backwards and forwards and taking it all in. I like what I see!

[16:00:01.0000] <hsivonen>
what's the business interest of ApTest in XHTML 2.0?

[16:00:02.0000] <jdandrea>
I wonder what the ratio of real-world HTML-to-CSS concerns is for MS. Even a ballpark figure.

[16:01:00.0000] <othermaciej>
I think IE7 has made them scared to change anything ever

[16:01:01.0000] <jdandrea>
hmm

[16:01:02.0000] <jdandrea>
Someone warn Molly.

[16:02:00.0000] <othermaciej>
I wonder how she would feel about Chris's stance on versioning

[16:02:01.0000] <jdandrea>
Bingo.

[16:02:02.0000] <Dashiva>
Frustrated?

[16:03:00.0000] <zcorpan>
perhaps cwilso is afraid of losing his job

[16:05:00.0000] <zcorpan>
(just speculating)

[16:06:00.0000] <jdandrea>
Has anyone read his "Albatross!" blog posts?

[16:06:01.0000] <zcorpan>
pointer?

[16:06:02.0000] <jdandrea>
"I should be really clear here – I don’t really want to be the chair of the HTML Working Group." http://blogs.msdn.com/cwilso/archive/2007/01/10/you-me-and-the-w3c-aka-reinventing-html.aspx

[16:06:03.0000] <Dashiva>
zcorpan: He did say he thought he was close after IE7

[16:06:04.0000] <jdandrea>
Though I may be quoting out of context. There's more to it ...

[16:07:00.0000] <zcorpan>
Dashiva: yeah, i meant that it might be the reason he wants versioning in ie

[16:07:01.0000] <othermaciej>
it's getting increasingly clear that he is not doing any actual chairing

[16:07:02.0000] <zcorpan>
jdandrea: yes i read it before

[16:07:03.0000] <jdandrea>
zcorpan: ahh, ok

[16:08:00.0000] <jdandrea>
/me answers the clarion call of din-din - bbl ...

[16:08:01.0000] <Dashiva>
He still hasn't explained how versioning is supposed to help when browser releases don't coincide with standard changes

[16:10:00.0000] <othermaciej>
if you think XHTML2 WG is crazy: http://news.yahoo.com/s/ap/20070413/ap_on_hi_te/rebuilding_the_internet

[16:11:00.0000] <zcorpan>
ha

[16:11:01.0000] <othermaciej>
forget replacing the web, how about replacing the whole internet?

[16:12:00.0000] <zcorpan>
why don't we move to another planet altogether?

[16:12:01.0000] <othermaciej>
a lot of Earth does not use proper semantic geography

[16:12:02.0000] <othermaciej>
so you have a good point there

[16:12:03.0000] <othermaciej>
we need a solar system that refuses to process planets which are not well-formed

[16:12:04.0000] <zcorpan>
i mean, the earth sucks, the only way to address security and mobility is to start from a clean slate

[16:13:00.0000] <zcorpan>
exactly


2007-04-15
[17:16:00.0000] <zcorpan>
could we say that class names in html are case insensitive? (so that they match case insensitively in css) would it break any documents? (it is required for quirks mode, i'd presume)

[17:17:00.0000] <zcorpan>
or is that just considered a css quirk not in html's domain to fix?

[17:21:00.0000] <othermaciej>
zcorpan: they are case insensitive in quirks mode, but making them CI in quirks mode would likely break things

[17:21:01.0000] <othermaciej>
zcorpan: I don't know what non-CSS use of classes would be relevant here though

[17:22:00.0000] <zcorpan>
getElementsByClassName, getElementsBySelector

[17:22:01.0000] <othermaciej>
neither of those would be used in old content

[17:22:02.0000] <othermaciej>
so there's no need for compatibility workarounds

[17:23:00.0000] <zcorpan>
indeed

[17:23:01.0000] <zcorpan>
you said "but making them CI in quirks mode would likely break things" -- i don't follow, they already are CI in quirks mode

[17:23:02.0000] <othermaciej>
er

[17:23:03.0000] <othermaciej>
in strict mode I meant

[17:23:04.0000] <zcorpan>
ah

[17:24:00.0000] <zcorpan>
ok

[17:24:01.0000] <zcorpan>
so it's considered a css quirk then

[17:24:02.0000] <othermaciej>
you could make those APIs switch on quirks vs. strict mode, but I think that would be unhelpful

[17:25:00.0000] <zcorpan>
yeah

[17:26:00.0000] <zcorpan>
i'm just interested in having quirks mode specced...

[17:26:01.0000] <zcorpan>
most of the web rely on quirks

[17:27:00.0000] <zcorpan>
not speccing them means new vendors would have to reverse engineer others anyway

[17:28:00.0000] <zcorpan>
(implementing html5 and css21 would only work with standards mode content)

[17:30:00.0000] <othermaciej>
yes, although most quirks mode issues are CSS issues, not HTML

[17:30:01.0000] <othermaciej>
there are a few exceptions, mostly parsing issues

[17:30:02.0000] <othermaciej>
but I think many parsing quirks could safely be put in both modes

[17:32:00.0000] <zcorpan>
if not all, though that would perhaps also include <p><table> (could content rely on it being a child? how?)

[17:33:00.0000] <zcorpan>
why can't <p><table> parsing be dropped from quirks mode? is there an example page that relies on it?

[17:34:00.0000] <zcorpan>
(either way, i'd like it to be parsed the same in both modes...)

[18:33:00.0000] <zcorpan>
so ms don't actually want to implement the spec. just use it as a guide and implement something near it. how depressing. and they are introducing more rendering modes that all other browsers will be forced to implement.

[18:35:00.0000] <zcorpan>
...unless they become a minority vendor

[18:36:00.0000] <zcorpan>
in which case it's just a problem for them

[18:36:01.0000] <jdandrea>
... and we instigate that by ... :)

[18:36:02.0000] <jdandrea>
I hear ya.

[18:37:00.0000] <Lachy>
I wonder if Chris would be willing to compromise on the versioning switch...

[18:37:01.0000] <Lachy>
Let the authors decide if they want the page to render according to IE[n] bugs when they write the page, or if they want to always render in the latest standards mode

[18:38:00.0000] <zcorpan>
that could work

[18:38:01.0000] <Lachy>
e.g. so <!DOCTYPE html> would always trigger the latest standards mode, but if the author adds some version switch, they get that versions

[18:38:02.0000] <Dashiva>
He would agree to that, I think. However, he argues that most/many/some authors would opt in without meaning to

[18:38:03.0000] <Lachy>
it's the lesser of two evils

[18:39:00.0000] <Dashiva>
After all, standards mode was supposed to be for standards, but now it's a frozen bug state

[18:40:00.0000] <zcorpan>
we should remember what standards are for... to foster interop

[18:42:00.0000] <zcorpan>
or a shortcut to get interop, if you will. without standards vendors could still reverse engineer each other to get interoperable

[18:42:01.0000] <Lachy>
ha! "Huh?  Wrong about what, exactly?" -- Chris Wilson :-)

[01:53:00.0000] <hsivonen>
Lachy: do we have a mechanism for blacklisting commenter URLs on the WHATWG blog?

[01:53:01.0000] <hsivonen>
/me deals with yet another einemillioneurohomepage spam

[01:56:00.0000] <hsivonen>
/me finds the Akismet blacklist

[01:58:00.0000] <hsivonen>
hmm. I guess it is called Spam Karma 2

[01:58:01.0000] <hsivonen>
Lachy: how do Akismet and Spam Karma 2 interact?

[02:12:00.0000] <gavins>
Lachy: yt?

[02:14:00.0000] <gavins>
nm :)

[02:35:00.0000] <Lachy>
hey

[02:37:00.0000] <Lachy>
hsivonen, those plugins seem to work well together.  I'm not sure of the details, but if one doesn't catch the spam, the other generally does

[02:37:01.0000] <Lachy>
gavins, yes?

[02:37:02.0000] <gavin>
Lachy: I was going to ask where I could find "Lachy's #xhtml log", but I found it

[02:42:00.0000] <Hixie>
hsivonen: ask lachy

[03:37:00.0000] <hsivonen>
Hixie: ask Lachy about spam?

[03:46:00.0000] <Hixie>
about the blog, yeah

[03:46:01.0000] <Hixie>
i don't admin it

[09:58:00.0000] <gsnedders>
what would be a good way to mark up a list of aims, and justifications behind those aims? a <dl>?

[10:01:00.0000] <hsivonen>
gsnedders: depends on the presentation you want

[10:01:01.0000] <hsivonen>
/me hides

[10:03:00.0000] <gsnedders>
a numbered list of aims (so <ol>), with the justifications set off in a box out of line from the content (an <i>?)

[10:06:00.0000] <gsnedders>
I think really what I'm unsure about is how to mark up the justification… just another paragraph?

[10:11:00.0000] <gsnedders>
/me settles on nested ol elements, combined with headers for each aim, with a paragraph(s) for the justification

[12:26:00.0000] <hsivonen>
http://seo.has.no.com/2007/04/seo-tip-for-html5/

[14:41:00.0000] <bewest>
/me is wondering where to find the appropriate venue for feedback regarding http://www.w3.org/TR/2006/WD-web-forms-2-20060821/#form-submission

[14:42:00.0000] <zcorpan>
/me points out that bewest isn't reading the most resent version of the spec

[14:43:00.0000] <zcorpan>
but, since the html-wg haven't decided yet about adopting wf2 (might be mistaken here, still have 45 emails to read), -> whatwg list

[14:44:00.0000] <bewest>
that seems to be the most recent version

[14:44:01.0000] <bewest>
erm

[14:44:02.0000] <zcorpan>
it's the most resent WD

[14:44:03.0000] <bewest>
oh

[14:44:04.0000] <bewest>
should I be looking somewhere else?

[14:45:00.0000] <zcorpan>
http://whatwg.org/wf2

[14:45:01.0000] <bewest>
great

[14:46:00.0000] <bewest>
so whatwg mailing list is good?

[14:46:01.0000] <hsivonen>
would be nice to get the decision in the HTML WG...

[14:46:02.0000] <hsivonen>
bewest: yes

[14:46:03.0000] <zcorpan>
bewest: yes

[14:47:00.0000] <bewest>
hmm I should probably do some homework before posting

[14:48:00.0000] <bewest>
I'll do some searching as well, but can anyone recall any discussion around using some protocol to allow the application to communicate it's input expectations to the user agent, instead of doing strict type checking of input on the client side?

[14:49:00.0000] <bewest>
AFAICT the application will have to perform the validation routines again, anyway, since I don't see any mechanism for ensuring that incoming input has already been checked

[14:50:00.0000] <hsivonen>
bewest: yes, the server has to do it again, because it would be dangerous to trust stuff arriving over HTTP

[14:50:01.0000] <bewest>
right

[14:51:00.0000] <bewest>
and in addition, the format/type of input is probably some subset of requirements for validity for many applications, I would suspect

[16:18:00.0000] <othermaciej>
hsivonen: I will leave it to the chairs to decide when to call the question on adopting the WHATWG specs

[16:19:00.0000] <othermaciej>
so far there has been no actual objection to the proposal

[16:19:01.0000] <othermaciej>
I'm not sure if there are even any really serious amendments


2007-04-16
[17:25:00.0000] <Dashiva>
othermaciej: Your mail about canvas suggestions seemed to suffer from candlejacking

[17:31:00.0000] <othermaciej>
Dashiva: what's "candlejacking"?

[17:32:00.0000] <Philip`>
The ever-reliable Urban Dictionary seems to suggest that term is used when sentences get cut off near the end

[17:33:00.0000] <othermaciej>
oops

[17:33:01.0000] <othermaciej>
I do that sometimes b/c I reply to things out of order and don't proofread

[17:38:00.0000] <jdandrea>
I can't find candlejacking in urbandictionary.com ... link?

[17:41:00.0000] <Philip`>
It has the terms "candlejack" and "candle jack"

[22:46:00.0000] <mpt>
"Google at present ever so slightly favours <b> over <strong> for weighing keywords and content on a page"? weird

[00:27:00.0000] <annevk>
http://mbro.belajar.net/2007/04/15/standar-web-idealis-vs-realistis/ might give feedback on HTML5

[00:27:01.0000] <annevk>
I can't read it though

[00:28:00.0000] <annevk>
lang="id"

[00:38:00.0000] <annevk>
"Stoneship uses HTML5, to be precise. HTML5 is not a standard, it is not widely adopted, and there’s virtually no browser support yet. It is still usable, though, because HTML5, unlike XHTML2, is backward compatible: a HTML4 document can be a HTML5 one with very little changes (changing the doctype should suffice in most cases)."

[00:38:01.0000] <annevk>
from http://stoneship.org/journal/marked-up/

[00:43:00.0000] <krijnh>
<meta charset="utf-8"> is supported by current browsers?

[00:43:01.0000] <annevk>
yes

[00:44:00.0000] <annevk>
because lots of people write this: <meta http-equiv=content-type value=text/html; charset=utf-8>

[00:44:01.0000] <krijnh>
Ah, makes sense

[00:44:02.0000] <annevk>
otherwise it wouldn't be in ;) :)

[00:44:03.0000] <krijnh>
I know

[00:44:04.0000] <krijnh>
:)

[00:48:00.0000] <krijnh>
Fatal Error: & not followed by # or name start.

[00:48:01.0000] <krijnh>
:S

[00:48:02.0000] <annevk>
Fatal Error: You're using XML

[00:48:03.0000] <krijnh>
No I'm not

[00:48:04.0000] <annevk>
oh, is this html5lib?

[00:49:00.0000] <annevk>
oh wait, the validator?

[00:49:01.0000] <krijnh>
No, the conformance checker ;)

[00:49:02.0000] <krijnh>
Ow, and scope isn't allowed on <th> elements?

[00:50:00.0000] <krijnh>
Damn, that would make my CMS non conformant then :(

[00:52:00.0000] <annevk>
scope is iirc

[00:52:01.0000] <krijnh>
Error: Attribute scope not allowed at this point; ignored.

[00:52:02.0000] <krijnh>
Perhaps it's not in the checker yet

[00:52:03.0000] <annevk>
are you using it correctly?

[00:52:04.0000] <annevk>
that could be true as well

[00:52:05.0000] <annevk>
(don't trust the validator!)

[00:52:06.0000] <krijnh>
<th scope="row">

[00:53:00.0000] <krijnh>
The & problem was my fault btw ^

[00:55:00.0000] <krijnh>
But an <a> isn't allowed directly inside a <div>?

[00:56:00.0000] <annevk>
either only block level or only inline

[00:56:01.0000] <annevk>
if you mix, you're wrong

[00:56:02.0000] <annevk>
(for the moment anyway)

[00:56:03.0000] <krijnh>
Ah, makes sense as well :)

[00:56:04.0000] <krijnh>
The document conforms to the machine-checkable conformance requirements for HTML5 (subject to the utter previewness of this service).

[00:56:05.0000] <krijnh>
\o/

[00:56:06.0000] <annevk>
apparently it's hard to generate better error messages for that

[00:57:00.0000] <krijnh>
http://www.bvalmere.nl/ is HTML5 :)

[00:58:00.0000] <krijnh>
<input type=url> is slow in Opera btw

[00:58:01.0000] <krijnh>
(depends on your history settings I think)

[00:59:00.0000] <krijnh>
accept-charset for form is dropped, cause nobody uses it?

[00:59:01.0000] <annevk>
is it dropped?

[00:59:02.0000] <annevk>
why?

[00:59:03.0000] <krijnh>
Don't know

[00:59:04.0000] <annevk>
i don't think it is

[00:59:05.0000] <krijnh>
K

[00:59:06.0000] <annevk>
Web Forms 2 is a superset of HTML4

[00:59:07.0000] <krijnh>
summary for <table> ?

[00:59:08.0000] <annevk>
it makes a few changes, such as deprecating size=

[01:00:00.0000] <annevk>
krijnh, yeah, that's dropped

[01:00:01.0000] <krijnh>
Why?

[01:00:02.0000] <krijnh>
My CMS auto generates those :/

[01:00:03.0000] <krijnh>
"table has x columns and y rows"

[01:00:04.0000] <annevk>
it was thought of to be pretty useless

[01:00:05.0000] <annevk>
that would be a pretty useless summary

[01:00:06.0000] <annevk>
an example of a*

[01:00:07.0000] <krijnh>
Yeah, auto generated fluff most of the times doesn't make sense

[01:01:00.0000] <krijnh>
I think a caption is enough summary already

[01:01:01.0000] <krijnh>
Some sites say the summary attribute should contain a summary of the complexity of the table

[01:01:02.0000] <krijnh>
*could

[01:02:00.0000] <annevk>
some people are misinformed though

[01:02:01.0000] <krijnh>
Jup

[01:03:00.0000] <krijnh>
I thought it was Russ Weakley who said that

[01:03:01.0000] <annevk>
well, I once promoted XHTML2

[01:03:02.0000] <krijnh>
Hehe

[01:04:00.0000] <virtuelv>
(w|m)ildly offtopic: Does anyone know of a blog-type CMS which doesn't suck, and actually allows me to output HTML

[01:05:00.0000] <krijnh>
I don't

[01:05:01.0000] <krijnh>
Mine doesn't allow HTML :)

[01:07:00.0000] <krijnh>
annevk: What's your opinion on http://webrichtlijnen.overheid.nl/handleiding/ontwikkeling/productie/tabellen/layout/ ?

[01:07:01.0000] <virtuelv>
krijnh: more to the point, all of the CMSes I've seen try to f*ck around with strings

[01:08:00.0000] <krijnh>
Content is a string most of the times

[01:08:01.0000] <krijnh>
So it probably makes sense CMSes try to mess around with it

[01:09:00.0000] <krijnh>
http://www.usability.com.au/resources/tables.cfm#summary btw

[01:09:01.0000] <krijnh>
Brb

[01:09:02.0000] <virtuelv>
krijnh: no, it doesn't really make sense to treat markup as strings

[01:10:00.0000] <virtuelv>
the content of any given text node or an attribute value, perhaps, but not the markup

[01:37:00.0000] <annevk>
the serialization of markup is a string

[01:41:00.0000] <met_>
whats the font on '5>2' t-shirts? no delivery to czech republic, so i try to make it myself

[01:41:01.0000] <annevk>
Arial

[01:41:02.0000] <annevk>
but that's not that relevant I suppose

[01:41:03.0000] <annevk>
if you can make it prettier, go for it

[02:21:00.0000] <krijnh>
Courier!

[02:23:00.0000] <met_>
it's a guess?

[02:23:01.0000] <annevk>
i think that's a bad suggestion for a prettier font (meant as a joke)

[02:24:00.0000] <annevk>
(it really is Arial)

[02:35:00.0000] <Hixie>
annevk?

[02:35:01.0000] <annevk>
yeah

[02:35:02.0000] <Hixie>
annevk: where's the CSSOM you're writing?

[02:35:03.0000] <Hixie>
/me can't find it on dev.w3.org

[02:35:04.0000] <annevk>
http://dev.w3.org/cvsweb/csswg/cssom/

[02:35:05.0000] <Hixie>
but the w3c date system makes everything impossible to find anyway, so

[02:35:06.0000] <Hixie>
cool

[02:35:07.0000] <Hixie>
thanks

[02:35:08.0000] <annevk>
I took your html5 example

[02:35:09.0000] <Hixie>
heh, there was already a CSS directory there :-P

[02:36:00.0000] <annevk>
oh :)

[02:36:01.0000] <annevk>
got to go

[03:45:00.0000] <hsivonen>
the versioning debate is hard, because the line of argument one might make depends on whether the Microsoft position is taken as a thing that cannot be changed

[03:46:00.0000] <hsivonen>
that is, whether to argue for minimizing the damage assuming that the damage is inevitable

[03:46:01.0000] <hsivonen>
or whether to argue for not inflicting the damage in the first place

[03:46:02.0000] <ROBOd>
hsivonen: correct, and i agree with Microsoft's position: they cannot "break the web"

[03:47:00.0000] <ROBOd>
that would be really Evil™

[03:47:01.0000] <virtuelv>
I just don't understand that argument

[03:47:02.0000] <ROBOd>
virtuelv: why not?

[03:47:03.0000] <virtuelv>
'Breaking the web' can happen in one of two ways

[03:47:04.0000] <virtuelv>
1) Scripting interfaces break

[03:47:05.0000] <virtuelv>
authors will make use of whatever scripting interfaces they like, no matter what a DTD says

[03:48:00.0000] <virtuelv>
2) CSS breaks

[03:48:01.0000] <virtuelv>
Microsoft can make a clean break[sic] starting with HTML5

[03:48:02.0000] <hsivonen>
virtuelv: and they want to

[03:49:00.0000] <hsivonen>
virtuelv: the problem is that if IE8 makes a clean break with HTML5, Microsoft believes they can no longer change anything in IE9

[03:49:01.0000] <hsivonen>
I understand why they think that way

[03:49:02.0000] <ROBOd>
hsivonen: that's not accurate

[03:50:00.0000] <hsivonen>
ROBOd: without a new mode

[03:50:01.0000] <ROBOd>
they can no longer change anything in IE 9, only *if* there are *many* pages relying on bugs in IE8 implementation of HTML5

[03:50:02.0000] <ROBOd>
(many, according to their definition, of course)

[03:50:03.0000] <hsivonen>
ROBOd: right

[03:50:04.0000] <virtuelv>
hsivonen: IMHO, they're wrong

[03:51:00.0000] <ROBOd>
so... that will *only* happen if IE 9 comes after 5 years

[03:51:01.0000] <virtuelv>
the major thing here being if they actually bother to fix CSS

[03:51:02.0000] <hsivonen>
ROBOd: and they don't know how to measure the breakage, so they'd err on the side of assuming that any change breaks something

[03:51:03.0000] <ROBOd>
(or a long enough period)

[03:51:04.0000] <virtuelv>
if they ship IE8 being broken enough, then yes, they're going to be 'needing' versioning

[03:51:05.0000] <ROBOd>
hsivonen: i believe they know how to measure breakage

[03:51:06.0000] <virtuelv>
however, I think versioning is downright harmful

[03:52:00.0000] <virtuelv>
the problem with versioning is that other browser vendors will no longer be able to work on progressive enhancment

[03:52:01.0000] <hsivonen>
virtuelv: I think versioning is harmful, too

[03:53:00.0000] <hsivonen>
virtuelv: but if MS does it, I'd rather they did it the way I suggest ;-)

[03:53:01.0000] <hsivonen>
ROBOd: I thought Chris said they don't

[03:53:02.0000] <ROBOd>
hsivonen: hmm... maybe... i didn't read *all* his emails

[03:54:00.0000] <virtuelv>
hsivonen: since I'm not actively following any lists these days, mind giving me a two-line summary?

[03:54:01.0000] <ROBOd>
but they should be able to measure breakage, otherwise, it's pretty bad

[03:54:02.0000] <hsivonen>
virtuelv: of what I'm suggesting as the not-as-bad way?

[03:54:03.0000] <virtuelv>
hsivonen: yes

[03:55:00.0000] <hsivonen>
virtuelv: an attribute on the root element that declares the date on which the page designer tested the page against then-current browsers

[03:55:01.0000] <ROBOd>
(almost) the same as versioning, imho

[03:56:00.0000] <virtuelv>
hsivonen: and how is that going to be maintainable?

[03:56:01.0000] <hsivonen>
virtuelv: it isn't going to be less maintainable than having a flag that binds to IE versions rather than dates

[03:57:00.0000] <hsivonen>
virtuelv: if only IE uses the flag and documents their release dates, the alternatives are equally bad

[03:57:01.0000] <hsivonen>
virtuelv: however, a date levels the playing field a bit in case another vendor wants to play the MS game

[03:58:00.0000] <hsivonen>
virtuelv: and having it in an attribute minimizes the damage to intermediate tool vendors for whom doctypes and conditional comments suck big time

[03:59:00.0000] <hsivonen>
virtuelv: but regardless of what the flag is, having as many modes as IE versions will be bad for competition and for maintainability

[03:59:01.0000] <virtuelv>
hsivonen: I think you're asking us to choose between the black plague and cholera here

[04:00:00.0000] <Hixie>
it's exactly the same mistake they've made again and again with Word

[04:00:01.0000] <hsivonen>
virtuelv, Hixie: agreed on both counts

[04:00:02.0000] <Hixie>
and given the advantage it's given them there, and how they're losing the web, i don't see why microsoft _wouldn't_ explicitly and intentionally go down this route

[04:00:03.0000] <Hixie>
for HTML

[04:01:00.0000] <Hixie>
it's possible chris doesn't realise he's doing it, though

[04:01:01.0000] <Hixie>
and that he honestly believes it's the best thing for his users

[04:01:02.0000] <hsivonen>
Hixie: which is why I'm in the damage minimization mode as opposed to the let's-not-even-think-about-it mode

[04:01:03.0000] <zcorpan>
/me considers advocating authors to use quirks mode

[04:01:04.0000] <Hixie>
hsivonen: yeah, you and lachy both

[04:02:00.0000] <Hixie>
/me doesn't even know how to do damage limitation at this stage

[04:02:01.0000] <Hixie>
i'm more concerned with saving our short doctype

[04:02:02.0000] <virtuelv>
let's go for damage maximization, then :)

[04:02:03.0000] <hsivonen>
Hixie: out of curiosity, do you think Lachy and I are arguing in favor of MS?

[04:02:04.0000] <Hixie>
what does "in favour of MS" mean?

[04:03:00.0000] <hsivonen>
Hixie: I'm concerned about that, too. (hence, attribute)

[04:03:01.0000] <Hixie>
yeah

[04:03:02.0000] <Hixie>
i think the attribute with a date is the better plan

[04:03:03.0000] <Hixie>
theoretically at least

[04:03:04.0000] <Hixie>
but if we do that i'm sure that people will just say tested-on="9999-12-31"

[04:04:00.0000] <virtuelv>
Hixie: I quite agree

[04:04:01.0000] <Hixie>
and we'll be back to square 1

[04:04:02.0000] <Hixie>
basically any system where you can predict the future values will be gamed

[04:04:03.0000] <zcorpan>
Hixie: that should be equivalent to not specifying it at all

[04:04:04.0000] <Hixie>
zcorpan: and trigger legacy mode? ("quirks v2"?)

[04:05:00.0000] <virtuelv>
zcorpan: I think that's rather the point. The attribute won't matter or be respected by application authors anyway

[04:05:01.0000] <virtuelv>
and you'll have the same augmentation you have now, but you'll have one more useless attribute to stick in the document

[04:05:02.0000] <hsivonen>
whee! arms race time! browsers embarassing authors who specify a date ahead of wall-clock

[04:05:03.0000] <zcorpan>
Hixie: no, latest mode. (i haven't followed the whole discussion here so i may be missing something)

[04:06:00.0000] <Hixie>
hsivonen: can't do that either, computer clocks are notoriously unreliable

[04:06:01.0000] <virtuelv>
/me would just rather see Microsoft actually adhering to both HTML5 and CSS2.1

[04:06:02.0000] <Hixie>
zcorpan: right. anything that triggers a moving target will end up having to be frozen.

[04:06:03.0000] <hsivonen>
virtuelv: don't we all?

[04:07:00.0000] <Hixie>
zcorpan: what today is called "standards mode" is exactly that -- "use the latest standards, don't do any quirks, break me if you want to fix a bug"

[04:07:01.0000] <virtuelv>
we do, but I'll say that budging to microsoft on this one is going to cost the web big-time in the years to come

[04:07:02.0000] <Hixie>
zcorpan: but now enough people use it that Microsoft feel they can't change rendering in that mode, and hence it now becomes quirks v2

[04:07:03.0000] <virtuelv>
because we'll be no better off tomorrow than we were three years ago

[04:07:04.0000] <zcorpan>
Hixie: yeah :(

[04:07:05.0000] <virtuelv>
Hixie: wouldn't it then be prudent for them to have a 'real standards mode'?

[04:08:00.0000] <Hixie>
zcorpan: if we have a system that can be gamed, then authors will ask for "latest mode", gaming the system, other authors will copy that, and we'll end up with yet another quirks mdoe trigger, this time quirks v3

[04:08:01.0000] <hsivonen>
virtuelv: the problem is that the WG does not have a mechanism to enforce compliance on the part of Microsoft

[04:08:02.0000] <Hixie>
virtuelv: it would be prudent of any browser vendor to not have rendering modes at all and just have a spec that can be followed to render all the web

[04:08:03.0000] <Hixie>
virtuelv: but the working assumption here is that microsoft disagree with what would be prudent

[04:09:00.0000] <virtuelv>
Hixie: not that I disagree, but noone's done it right. Ever.

[04:09:01.0000] <virtuelv>
hsivonen: I know. It's depressing

[04:09:02.0000] <Hixie>
virtuelv: well there was a time where we wanted to fix the browsers instead of hte specs, and that's how we ended up with quirks vs standards

[04:09:03.0000] <Hixie>
virtuelv: but now, notwithstanding that disaster, all the browsers are doing what i describe

[04:10:00.0000] <Hixie>
virtuelv: only microsoft is suggesting doing something else in future.

[04:11:00.0000] <zcorpan>
not so long ago, i thought html had a bright future. i'm not so sure anymore :(

[04:11:01.0000] <Hixie>
it has a fine future, so long as microsoft keep losing market share

[04:11:02.0000] <zcorpan>
yeah, that's out only hope

[04:12:00.0000] <zcorpan>
s/out/our

[04:12:01.0000] <virtuelv>
either keep losing market share, or get their act together

[04:14:00.0000] <krijnh>
Or both

[04:16:00.0000] <virtuelv>
in Utopia, no browser has more than ~30%

[04:24:00.0000] <hsivonen>
view source at http://diveintomark.org/archives/2007/04/15/ejookashun and look for <dialog></p>

[04:25:00.0000] <mpt>
virtuelv, I've long said that to anyone who will listen (which is hardly anyone:-)

[04:25:01.0000] <krijnh>
hsivonen: why's that?

[04:25:02.0000] <hsivonen>
krijnh: trying to find out if its WP or Gecko

[04:26:00.0000] <krijnh>
I see it 4 times in Opera

[04:26:01.0000] <hsivonen>
krijnh: seems to be in the real source

[04:26:02.0000] <krijnh>
Jep

[04:26:03.0000] <mpt>
<p><dialog></p>

[04:26:04.0000] <krijnh>
And <p></dialog></p>

[04:26:05.0000] <krijnh>
Two times

[04:26:06.0000] <krijnh>
Funny stuff :)

[04:27:00.0000] <mpt>
It's probably doubling as a testcase for Mark to fix that

[04:27:01.0000] <krijnh>
It's probably WordPress putting <p> </p> around it

[04:27:02.0000] <mpt>
yes

[04:27:03.0000] <krijnh>
Doh

[04:28:00.0000] <krijnh>
The <p><i>One minute later&#8230;</i></p> could be put inside the <dialog> right?

[04:28:01.0000] <Hixie>
it's been suggested

[04:28:02.0000] <krijnh>
With an <li> or something? Or isn't that allowed yet?

[04:28:03.0000] <Hixie>
haven't dealt with that feedback yet

[04:29:00.0000] <Hixie>
i have two main things to do before i start replying to feedback

[04:29:01.0000] <Hixie>
one is finish off the first draft of the browsing context section

[04:29:02.0000] <Hixie>
and the other is figuring out some alternative to the <switch> proposl

[04:29:03.0000] <Hixie>
one that actually works

[04:29:04.0000] <Hixie>
some mechanic that conveys the semantic of "only one of these options is active"

[04:30:00.0000] <Hixie>
so that, e.g., you can have one document with a login form and a logged in form, and have a UA never show both

[04:30:01.0000] <Hixie>
or so you can have two <video> elements and an <img> element, and have them all right there active in the DOM, but only one rendering at any one time, ever

[04:31:00.0000] <Hixie>
right now there's no way to do this

[04:31:01.0000] <Hixie>
no way to say "this is simply not active, not relevant, don't show it"

[04:31:02.0000] <krijnh>
Only with CSS

[04:32:00.0000] <Hixie>
CSS is about presentation

[04:32:01.0000] <krijnh>
How would that be backwards compatible?

[04:32:02.0000] <Hixie>
i'm talking about the actual semantic

[04:32:03.0000] <Hixie>
dunno

[04:32:04.0000] <Hixie>
haven't found a solution yet

[04:32:05.0000] <krijnh>
Hmm

[04:32:06.0000] <krijnh>
But you do think that should be possible in HTML?

[04:32:07.0000] <Hixie>
i think it's a feature that is missing and is very important

[04:33:00.0000] <Hixie>
i see people working around it all the time

[04:33:01.0000] <krijnh>
I mostly work around it server side

[04:33:02.0000] <hsivonen>
Hixie: why wouldn't a login form state be tightly coupled with server-side page round tripping or tightly coupled with XHR and scripted DOM manipulation?

[04:33:03.0000] <mpt>
krijnh, Mark's dialog also evokes that unusual idea someone posted to the whatwg list a month or so ago, about replacing <dl> with <table>

[04:33:04.0000] <Hixie>
hsivonen: why should it be?

[04:34:00.0000] <mpt>
dt -> th, dd -> td

[04:34:01.0000] <krijnh>
Yeah, why not :)

[04:34:02.0000] <hsivonen>
Hixie: to make sure the state and login processing are tied together

[04:35:00.0000] <Hixie>
hsivonen: but why should the state manipulation be so complicated as to require a complete change to the dom?

[04:35:01.0000] <krijnh>
mpt: I like <dl> better though

[04:35:02.0000] <Hixie>
hsivonen: shouldn't, once you've logged in, the change simply be an attribute change somewhere? (or equivalent)

[04:35:03.0000] <mpt>
I'd like <dl> a whole lot better if <dl compact> was still possible

[04:35:04.0000] <krijnh>
Yeah, me too

[04:35:05.0000] <Hixie>
hsivonen: the current system leads to people putting all the content on the page but with display:none

[04:35:06.0000] <Hixie>
display:none doesn't mean anything

[04:36:00.0000] <Hixie>
it doesn't hide the contents except in the default case of css being enabled and honoured

[04:36:01.0000] <mpt>
But then it demonstrates the table-ness again -- you have terms and you have definitions, but where do you put the pronunciation, and the etymology, and the part of speech...

[04:36:02.0000] <hsivonen>
Hixie: I see your point. 

[04:36:03.0000] <krijnh>
I don't do that, if it shouldn't be there, it shouldn't be there

[04:36:04.0000] <hsivonen>
Hixie: otoh, any new solution has the same problem as far as legacy uas go

[04:37:00.0000] <hsivonen>
krijnh: how do you load DOM templates?

[04:37:01.0000] <krijnh>
Most of the times the ugly way

[04:37:02.0000] <krijnh>
With some server side language along with it

[04:37:03.0000] <Hixie>
hsivonen: yeah, sometimes the solutions have poor legacy answers (like input type=datetime), but we still need them for the future, and script can handle most legacy cases

[04:37:04.0000] <hsivonen>
krijnh: what's the ugly way?

[04:38:00.0000] <krijnh>
I'm a bozo most of the times, if that's what you mean :)

[04:38:01.0000] <hsivonen>
Hixie: ok

[04:38:02.0000] <hsivonen>
krijnh: I mean: you have an app with repeating stuff, don't you put a display:none; copy of the repeating part somewhere so that you can clone it easily?

[04:39:00.0000] <krijnh>
No

[04:39:01.0000] <krijnh>
I try to avoid having to use display: none

[04:39:02.0000] <zcorpan>
couldn't <menu type=popup> be abused for that purpose? (not saying it should)

[04:39:03.0000] <krijnh>
hsivonen: Perhaps I'm misunderstanding you

[04:41:00.0000] <zcorpan>
the idea is something like <switch><section>A</section><section>B</section></switch>, right?

[04:42:00.0000] <krijnh>
And how do you choose the 'active' one?

[04:42:01.0000] <zcorpan>
perhaps with normal links

[04:42:02.0000] <krijnh>
I only see a use for tabs or something

[04:42:03.0000] <Hixie>
yeah e.g.:

[04:43:00.0000] <zcorpan>
<ul><li><a href="#a">A</a><li><a href="#b">B</a></ul>

[04:43:01.0000] <Hixie>
<switch> <section> login page </section> <section> game </section> <section> game over screen </section> </switch>

[04:43:02.0000] <Hixie>
with script switching between them

[04:43:03.0000] <Hixie>
tabs would specifically _not_ be a good use of this

[04:43:04.0000] <Hixie>
because the fact that tabs are displayed one after the other is an artifact of presentation

[04:43:05.0000] <krijnh>
This could be abused for it then

[04:43:06.0000] <Hixie>
nothing would stop a ui from showing all the tabs at once

[04:43:07.0000] <Hixie>
yes

[04:43:08.0000] <zcorpan>
oh

[04:45:00.0000] <zcorpan>
so a link pointing to an inactive section wouldn't make the section be relevant when the link is followed?

[04:45:01.0000] <hsivonen>
hmm. regarding a date-based switch and gaming it with future dates: a conformance checker could be trusted to know what today's date is with 24h accuracy. therefore, future dates could be non-conforming

[04:46:00.0000] <krijnh>
And make UAs not render documents with future dates?

[04:47:00.0000] <Hixie>
zcorpan: like i said, i haven't got a solution for this yet. so i don't know :-)

[04:47:01.0000] <hsivonen>
krijnh: Hixie said that wouldn't work because end user clocks are inaccurate

[04:47:02.0000] <Hixie>
zcorpan: do you think it should? what's the use case?

[04:47:03.0000] <krijnh>
"This page is optimized for 1024x768 and the year 2064" :)

[04:48:00.0000] <Hixie>
hsivonen: over 93% of pages are non-conforming

[04:48:01.0000] <krijnh>
How would a UA know which section is active? An attribute on <switch> or on <section> ?

[04:48:02.0000] <krijnh>
And shouldn't that attribute be set with JS or server side code anyway?

[04:49:00.0000] <hsivonen>
Hixie: yes, but the kind of people who think they can outsmart the switch are the kind of people who want a conformance checker to say that they conform

[04:49:01.0000] <Hixie>
krijnh: i have no idea, that's why i need to look at this to get ideas

[04:49:02.0000] <zcorpan>
Hixie: well. tabs. :)

[04:49:03.0000] <Hixie>
hsivonen: no, they're people like you, me and lachy, e.g. writing test case, who want to always have the page use standards mode

[04:50:00.0000] <Hixie>
hsivonen: without having to change the date in the test every week

[04:50:01.0000] <Hixie>
zcorpan: quite

[04:51:00.0000] <zcorpan>
Hixie: if it did, it would be easy to implement tabs that fallback to css or js for legacy

[04:51:01.0000] <Hixie>
zcorpan: tabs are a presentation idiom, they don't belong in html

[04:52:00.0000] <zcorpan>
Hixie: ditto a table of contents?

[04:52:01.0000] <Hixie>
zcorpan: whether a series of control groupings (fieldsets) get rendered as tabs or as side by side boxes is completely presentational

[04:52:02.0000] <Hixie>
zcorpan: table of contents?

[04:52:03.0000] <zcorpan>
i see a tab bar pretty similar to a list of toc

[04:53:00.0000] <hsivonen>
Hixie: well, yeah, but you, me and lachy (or the people on this channel basically) are perhaps not the optimization target when the alternative is everyone else. :-(

[04:53:01.0000] <zcorpan>
you don't need the toc list, it is implied by the document headings (the ua can generate one itself)

[04:53:02.0000] <Hixie>
zcorpan: yeah, it's a fair comparison. and with a toc the sections are always visible.

[04:53:03.0000] <zcorpan>
Hixie: indeed

[04:53:04.0000] <Hixie>
hsivonen: my point is that making it not conform wouldn't stop us, therefore it wouldn't solve the problem

[04:54:00.0000] <Hixie>
hsivonen: since we're exactly the people who copy paste us

[04:54:01.0000] <Hixie>
er

[04:54:02.0000] <Hixie>
who people copy and paste, rather

[04:54:03.0000] <zcorpan>
people would use script to insert today's date anyway

[04:54:04.0000] <zcorpan>
server side

[04:55:00.0000] <hsivonen>
zcorpan: that would solve the cargo cultist problem for cargo cultists who don't

[04:56:00.0000] <zcorpan>
hsivonen: sorry i don't follow

[04:57:00.0000] <hsivonen>
zcorpan: if we had server-side scripts updating test cases, cargo cultists who copy and paste our stuff would not have their pasted date auto-increment

[04:58:00.0000] <zcorpan>
ok

[05:21:00.0000] <zcorpan>
oh well, a filter couldn't wipe out +1s anyway, given we now have +100,000s

[11:42:00.0000] <jdandrea>
Am I reading the "Picture-perfect rendering" msg timeline correctly (over on public-html⊙wo)?

[11:42:01.0000] <jdandrea>
CWilson sez there's some vagueness within HTML5. Hixie asks for concrete examples. CWilson offers up <canvas> - no rendering rules to get pixel-perfect rendering. Hixie asks for more specifics.

[11:43:00.0000] <jdandrea>
TV Raman notes pixel-perfect rendering was never HTML's goal and should not become its goal ... and then CWilson agrees.

[11:50:00.0000] <zcorpan>
jdandrea: yes

[11:53:00.0000] <Philip`>
Then people think "HTML is not about pixel-perfect rendering" entails "HTML should be contain <canvas>", as opposed to concluding "<canvas> should not define pixel-perfect rendering" (which is the current situation)

[11:53:01.0000] <Philip`>
s/should be/should not/

[11:54:00.0000] <zcorpan>
indeed

[11:56:00.0000] <Philip`>
(I think my response (of the (non-pixel-perfect) specifics that I believe are missing) fell out of the threading - does anyone happen to know of a way to write from Gmail via mailto: links while preserving References?)

[11:57:00.0000] <zcorpan>
include the relevant headers in the mailto: link, perhaps?

[11:58:00.0000] <jdandrea>
Philip `: Exactly.

[11:58:01.0000] <jdandrea>
(wrt the conclusion/perception)

[11:58:02.0000] <zcorpan>
not sure if the gmail web interface preserves them, though. if it doesn't then enable pop and use an email client that does


2007-04-17
[21:07:00.0000] <Hixie>
wow, i just specced popup blocking without using the word "popup" or "blocking".

[21:07:01.0000] <Hixie>
didn't even realise it until i was done

[21:28:00.0000] <jdandrea>
Hixie: How'd you pull that one off? Nice.

[21:29:00.0000] <Hixie>
last bit of http://www.whatwg.org/specs/web-apps/current-work/multipage/section-windows.html

[21:29:01.0000] <jdandrea>
Thx

[21:30:00.0000] <jdandrea>
I find that choosing good words/names/etc. is one of the hardest (and most important) things you can do, hence my interest. Reading ...

[21:31:00.0000] <Hixie>
no guarentee the words are well chosen!

[21:31:01.0000] <jdandrea>
's OK. It's a start.

[21:31:02.0000] <Hixie>
it's about iteration 50 of this particular section

[21:31:03.0000] <Hixie>
i keep rewriting it to make it better

[21:32:00.0000] <Hixie>
still not overly happy with it

[21:32:01.0000] <Hixie>
but it'll do for now

[21:52:00.0000] <mpt>
/me is unfortunately reminded of the checkbox label "Let Web pages ... [/] Open new windows by themselves"

[21:53:00.0000] <mpt>
Quite possibly the least understandable interface design I ever perpetrated

[21:54:00.0000] <mpt>
Hixie, "auxillary" -> "auxiliary"

[21:55:00.0000] <jdandrea>
In 4.1.2 - ah.

[21:55:01.0000] <jdandrea>
I'm just about to get to the good stuff. :)

[22:03:00.0000] <jdandrea>
Hixie: Nice. As for the wording, it's consistent with the rest of 4.1.3 rule 5. I like the use of auxiliary as well.

[22:19:00.0000] <annevk>
/me wonders if the rest of the world is asleep right now

[22:25:00.0000] <Hixie>
mpt: thanks, will try to remember to fix that

[22:50:00.0000] <annevk>
/me wonders if it's ok to give the same presentation, albeit modified, twice

[22:51:00.0000] <othermaciej>
if in different venues, sure

[22:52:00.0000] <annevk>
I'd like to make the one I just gave public

[22:52:01.0000] <annevk>
oh well, I'll just do it

[23:00:00.0000] <annevk>
FYI: http://www.w3.org/mid/op.tqwwnwan64w2qv@id-c0020 (presentation on HTML5 I gave today at QUT)

[00:02:00.0000] <othermaciej>
every time I've spoken in person to one or more Microsoft employees I ended up with a headache

[00:02:01.0000] <othermaciej>
I wonder if there is a pattern there

[00:02:02.0000] <Hixie>
really?

[00:03:00.0000] <othermaciej>
well, it's only a sample space of 3

[00:03:01.0000] <Hixie>
do you get headaches often unrelatd to meeting microsoft employees?

[00:04:00.0000] <othermaciej>
sometimes, but not that often

[00:06:00.0000] <Hixie>
my recommendation: get more data

[00:06:01.0000] <othermaciej>
I'm not sure its worth it for the information value alone

[00:07:00.0000] <othermaciej>
*it's

[00:14:00.0000] <annevk>
I never had a headache before in relation to Microsoft employees

[00:15:00.0000] <othermaciej>
I think I might be allergic to Microsoft

[02:29:00.0000] <KevinMarks>
yay, my t-shirt came

[02:29:01.0000] <KevinMarks>
http://www.spreadshirt.com/shop.php?sid=108788

[04:12:00.0000] <nickshanks>
"5 > 2" ?

[04:22:00.0000] <hsivonen>
wow. a SEO spammer or SEO spammers have picked up liorean's future of the web article and Google is happily informing me about all the bogus blogs that contain a copy of the article

[04:35:00.0000] <Hixie>
hsivonen: which query?

[04:54:00.0000] <hsivonen>
Hixie: http://blogsearch.google.com/blogsearch_feeds?hl=en&q=link:blog.whatwg.org&filter=0&ie=utf-8&num=10&output=atom

[04:54:01.0000] <Hixie>
thanks

[07:41:00.0000] <krijnh>
Anybody in here know a way to style active links in Opera?

[07:42:00.0000] <krijnh>
Active in the way you navigate to them with Shift + Arrows

[07:43:00.0000] <krijnh>
:focus only works when you hover an active anchor

[07:44:00.0000] <citoyen>
Theoretically :focus, though the styling is hidden by the selected text

[07:44:01.0000] <zcorpan>
/me got his t-shirt today

[07:44:02.0000] <krijnh>
citoyen: It works with an outline, but only once you hover the selected element

[07:45:00.0000] <citoyen>
Yeah, known

[07:45:01.0000] <krijnh>
K

[07:50:00.0000] <krijnh>
zcorpan: congrats btw ;)

[08:30:00.0000] <Dashiva>
Hixie: What's the reasoning for the void Table.delete* methods, rather than (e.g.) Table.remove* which return the removed element?

[08:35:00.0000] <zcorpan_>
krijnh: thanks

[10:50:00.0000] <zcorpan_>
http://shadow2531.com/opera/testcases/mht/000.mht the background doesn't cover the canvas in opera

[10:51:00.0000] <zcorpan_>
should it?

[11:00:00.0000] <hasather>
zcorpan_: yea, that looks like a bug

[11:01:00.0000] <zcorpan_>
/me won't report it, though

[11:03:00.0000] <zcorpan_>
what's with people wanting to change element names?

[11:03:01.0000] <zcorpan_>
/me hates naming

[11:10:00.0000] <Philip`>
The background appears to cover the canvas of the document which it's the background of - but that document is in an <object> inside the page that's actually being displayed, and the object doesn't fill that page

[11:12:00.0000] <hasather>
Philip`: ahh, you're right, didn't check that

[11:13:00.0000] <zcorpan_>
so that you can have multiple html pages after each other?

[11:15:00.0000] <zcorpan_>
makes sense

[15:14:00.0000] <zcorpan_>
/me goes to add a subforum

[15:49:00.0000] <zcorpan_>
http://forums.whatwg.org/viewforum.php?f=3


2007-04-18
[17:05:00.0000] <Hixie>
Since we are "paving the cow paths" it seems to me that we must also pave

[17:05:01.0000] <Hixie>
foreseeable future cow paths as well.  To that end, add a version

[17:05:02.0000] <Hixie>
attribute and just be done with it.  Everyone else can ignore it.

[17:05:03.0000] <jdandrea>
Heh.

[17:05:04.0000] <jdandrea>
(from the list, aye)

[17:05:05.0000] <Hixie>
/me wonders if the author understands the cowpath principle

[17:05:06.0000] <Dashiva>
Hixie: What's the reasoning for the void Table.delete* methods, rather than (e.g.) Table.remove* which return the removed element?

[17:06:00.0000] <Hixie>
the reasoning is probably "that's what DOM2 HTML has"

[17:08:00.0000] <Dashiva>
And related, create* now explicitly only returns new elements, while DOM2 HTML states they may return an existing one

[17:10:00.0000] <Hixie>
probably based on what browsers did, i forget. however if you find an incompatibility, feel free to mail the list and i'll look at it

[17:11:00.0000] <Dashiva>
Personally I consider the whole bunch redundant and low-utility, so no worries :)

[17:12:00.0000] <Philip`>
http://www.ipfw.edu/chem/104/kimble/cowpath.htm - that seems to advocate against paving cow paths, because they're often far from the best route and it's just that nobody has thought to look around and notice there's a problem. But I think this is probably taking the metaphor a bit too far...

[17:13:00.0000] <Dashiva>
No, it's a good lesson

[17:14:00.0000] <Dashiva>
The cowpaths to pave are those with a sensible foundation, not mindless drones enforcing it

[17:14:01.0000] <jdandrea>
/me moos

[17:15:00.0000] <jdandrea>
/me keeps thinking the versioning request is meant (on the whole) more for CSS/DOM, not so much HTML.

[17:15:01.0000] <Dashiva>
Prototyping Object is a cowpath that should be blocked off on penalty of death :)

[17:15:02.0000] <jdandrea>
lol

[17:20:00.0000] <Philip`>
(Hmm, not sure why that page lacks the proper attribution - http://www.mitcharf.com/mitcharf/art/poems/calfpath.html seems a better resource)

[18:01:00.0000] <Lachy>
good morning

[18:02:00.0000] <zcorpan_>
Lachy: morning

[18:02:01.0000] <Philip`>
(http://www.archive.org/details/whiffsfromwildme00fossiala (p. 77) - aha, found the original! and hooray for copyright expiration)

[18:10:00.0000] <jdandrea>
Philip`: "The Calf-Path" - thanks for finding this!

[18:12:00.0000] <jdandrea>
/me thinks it would make good starter lyrics for a folk-rock song

[18:20:00.0000] <zcorpan_>
isn't name="" on <iframe> missing? (opera supports id="", but firefox and ie don't)

[18:21:00.0000] <Hixie>
possibly

[18:23:00.0000] <zcorpan_>
if we want <a target> and <form taget> to work with iframes then we need <iframe name>, i think

[18:25:00.0000] <zcorpan_>
should i mail the list?

[18:26:00.0000] <Hixie>
zcorpan_: sure

[18:26:01.0000] <Hixie>
and thanks

[18:26:02.0000] <annevk>
does it need to be conforming? prolly...

[18:27:00.0000] <zcorpan_>
annevk: <a target> is conforming, what use is it if you can't use it together with iframes?

[18:27:01.0000] <annevk>
i meant <iframe name>

[18:27:02.0000] <annevk>
much like <map name>

[18:28:00.0000] <zcorpan_>
<map> works with id. <iframe> doesn't

[18:28:01.0000] <annevk>
the former doesn't in Firefox

[18:29:00.0000] <zcorpan_>
but it does in ie. the latter doesn't in both firefox and ie.

[18:30:00.0000] <annevk>
Firefox has like 10% market share...

[18:30:01.0000] <annevk>
If working is an argument, <map name> should certainly be conforming

[18:30:02.0000] <zcorpan_>
not against that

[18:31:00.0000] <annevk>
they should be handled in a consistent way imo

[18:31:01.0000] <annevk>
same for <object id/name> etc.

[18:32:00.0000] <om_coffee>
I think it's preferrable to have only id be conforming if there aren't any major UAs where name works but id doesn't for some construct

[18:33:00.0000] <zcorpan_>
othermaciej: <iframe id> doesn't work in ie nor firefox (don't know about safari)

[18:34:00.0000] <othermaciej>
zcorpan_: yeah, I'm just making a broad generalization based on what annevk said

[18:34:01.0000] <othermaciej>
I don't know what the situation is with object id/name

[18:38:00.0000] <zcorpan_>
well, i guess i need to do more testing here first then

[18:38:01.0000] <zcorpan_>
not today though

[18:38:02.0000] <Hixie>
:-)

[18:46:00.0000] <annevk>
zcorpan_, what does http://xhtml.se/2007/04/17/x-and-html-sitting-in-a-tree/ say?

[18:49:00.0000] <Dashiva>
It's basically saying "Now that HTML and XHTML will be the same in v5, it won't matter which one you use"

[18:49:01.0000] <zcorpan_>
annevk: it basically says that html5 is the future, and that it doesn't matter if you use xhtml1 or html4 today, and it is pointless to switch from xhtml1 to html4, and pointless to argue about why xhtml is bad/shouldn't be used

[18:50:00.0000] <zcorpan_>
yeah

[18:50:01.0000] <annevk>
blah

[18:50:02.0000] <Dashiva>
It fails to address the non-HTML aspects

[18:50:03.0000] <annevk>
if he means XHTML as text/html though, fair enough

[18:50:04.0000] <Dashiva>
Like mime-types, CSS background, you know the list better than me

[18:50:05.0000] <annevk>
is that made explicit?

[18:51:00.0000] <zcorpan_>
annevk: he does, although he also likes content negotiation and mobile stuff

[18:51:01.0000] <Dashiva>
He doesn't mention it in the article itself, though

[18:52:00.0000] <zcorpan_>
he will hold a presentation about xhtml at geekmeet together with my html5 presentation next month

[18:52:01.0000] <Dashiva>
Seems to be a defense for people using xhtml, rather than saying xhtml is good :)

[18:52:02.0000] <zcorpan_>
Dashiva: indeed, i just know because i've talked with him about it a lot

[18:52:03.0000] <annevk>
"Look, I'm right after all!"

[18:54:00.0000] <zcorpan_>
annevk: i don't think it's benefitical to argue against these guys. he's not alone. making html5 accept xhtmlish syntax makes them happy. :)

[18:55:00.0000] <annevk>
I'm not arguing, I'm just making fun of them :)

[18:55:01.0000] <zcorpan_>
he's not ignorant at all, i just think he sticks to the xhtml idea because he bought the domain name ;)

[18:55:02.0000] <zcorpan_>
ok

[18:55:03.0000] <annevk>
lol

[18:56:00.0000] <Dashiva>
Wow

[18:56:01.0000] <Dashiva>
Now it's "why not a namespace on the html element"

[18:58:00.0000] <zcorpan_>
cwilso also thought it would be appropriate to use the namespace as a versioning switch, didn't he?

[18:58:01.0000] <annevk>
ouch

[19:02:00.0000] <zcorpan_>
http://www.w3.org/mid/5C276AFCCD083E4F94BD5C2DA883F05A27D7192F39⊙twwnmc

[19:03:00.0000] <annevk>
crazy

[19:03:01.0000] <annevk>
same as the nonsense people supporting that idea

[19:06:00.0000] <othermaciej>
I think cwilso doesn't want a version attribute because he (probably correctly) thinks people will just leave it off

[19:07:00.0000] <Dashiva>
That's counter to his other line of arguments, that people copy the header cargo cultishly

[19:07:01.0000] <Hixie>
no, he pointed out 50% of sites leave off the <!DOCTYPE>

[19:09:00.0000] <Dashiva>
Yes, but he also argues that many sites who opt for standards mode don't really want standards mode, so we could never have a real "give me the latest and best" switch

[19:11:00.0000] <Lachy>
As a web developer, I find Chris' implication that those of us who use Strict DOCTYPEs to get standards mode now aren't competent enough to want "always standards mode" offensive

[19:11:01.0000] <Dashiva>
XHTML has taught us a lot of people will say it without meaning it, though :/

[19:11:02.0000] <othermaciej>
it seems to me that Microsoft having such major bugs that were left around for years is a major contributing factor to the breakage level

[19:12:00.0000] <Lachy>
the incompetence of others shouldn't reflect on the whole

[19:13:00.0000] <Lachy>
MS actually needs to do some real work and research each change on a case by case basis, rather than ruling them all out at once

[19:15:00.0000] <Dashiva>
And not fix the "solution bugs" before the problems that depend on it 

[19:15:01.0000] <zcorpan_>
Lachy: they will only do that if they become a minority vendor and the web will ignore their versioning switches and work better on the competitor's browsers

[19:15:02.0000] <zcorpan_>
i think

[19:15:03.0000] <zcorpan_>
competitors' even

[19:16:00.0000] <Lachy>
zcorpan_: yes, it's quite sad

[19:17:00.0000] <zcorpan_>
so that scenario is what we should aim for then :)

[19:19:00.0000] <Philip`>
Perhaps the HTML5 spec should add an author conformance requirement that all pages must end in <a href="http://getfirefox.com">Get Firefox!</a>

[19:20:00.0000] <annevk>
right

[19:21:00.0000] <annevk>
what about opera.com

[19:21:01.0000] <annevk>
we actually implemented WF2!

[19:21:02.0000] <Dashiva>
It should rotate

[19:23:00.0000] <Philip`>
opera.com doesn't have enough swooshy blue and orange bits :-(

[19:38:00.0000] <annevk>
you know, some people appreciate that :)

[22:53:00.0000] <Hixie>
hey did anyone ever write a faq entry on why versioning is bad?

[22:55:00.0000] <annevk>
don't think so

[22:55:01.0000] <annevk>
we need more wiki entries

[22:55:02.0000] <annevk>
on role=, href=, versioning, etc.

[22:55:03.0000] <annevk>
of course, this is easy to say and hard to do :)

[22:56:00.0000] <annevk>
"hard"

[22:59:00.0000] <hober>
@role makes my head hurt

[23:00:00.0000] <hober>
I honestly can't figure out how it is any different from @class

[23:01:00.0000] <hober>
besides in some kind of "@class has been abused, therefore let us introduce a new attr that'll also be abused" sense

[23:04:00.0000] <Lachy>
if someone sends me a Q&A for versioning, I can add it to the FAQ.  If not, I'll get around to it later this week.

[23:05:00.0000] <Lachy>
it'll probably form part of that blog entry I'm writing, and will finish after I get back from the WAF F2F this weekend

[23:05:01.0000] <Hixie>
cool

[23:19:00.0000] <annevk>
What's the idea for "wizards" and "model dialogs"?

[23:20:00.0000] <Hixie>
modal, you mean?

[23:20:01.0000] <Hixie>
both are pretty bad ui

[23:20:02.0000] <Hixie>
but window.open() can handle them

[23:21:00.0000] <annevk>
I quite like the "step 1 of 4" kind of UI

[23:21:01.0000] <annevk>
not sure about modal dialogs

[23:21:02.0000] <annevk>
window.open() doesn't make it modal though

[23:22:00.0000] <Hixie>
wizard ui does very badly in usability studies, as i understand it

[23:22:01.0000] <Hixie>
you can easily make things modal, just ignore events in the other browsing context

[23:22:02.0000] <Hixie>
but basically the <switch> thing i've been talking about is the real solution to this

[23:23:00.0000] <annevk>
is it going in?

[23:23:01.0000] <annevk>
or HTML6?

[23:25:00.0000] <Hixie>
if we can find a working decent solution, it's the last thing i personally want to put in

[23:25:01.0000] <Hixie>
i.e. other than everyone else's requests

[23:25:02.0000] <annevk>
it's not used that often

[23:26:00.0000] <Hixie>
any time someone tries to do ui using display:none, they're asking for it

[23:26:01.0000] <annevk>
hmm

[23:58:00.0000] <annevk>
So you have HTML tokenization which does some sort of case mangling. You have the HTML DOM APIs which uppercase stuff and there's CSS matching...

[23:58:01.0000] <annevk>
/me played around with the <pÉ> element

[23:58:02.0000] <annevk>
and also <pé>

[00:01:00.0000] <annevk>
It seems that some browsers actually uppercase that through the DOM

[00:01:01.0000] <Lachy>
there are some interop problems with case sensitivity of selectors that matching non-ASCII chars

[00:01:02.0000] <annevk>
Internet Explorer and Firefox

[00:01:03.0000] <annevk>
yes

[00:06:00.0000] <Lachy>
it would probably make sense to try and define that UAs must use the Unicode Case Mappings, but we should actually try and do some tests to figure out how closely UAs follow them

[00:07:00.0000] <Lachy>
I hope Unicode is more interop than HTML4, cause I don't want to write Unicode5, but somehow I doubt it.

[00:07:01.0000] <annevk>
Unicode 5 is already here

[00:08:00.0000] <annevk>
/me rather has ASCII case-insensitive when possible

[00:08:01.0000] <Lachy>
that's Unicode 5.0, I meant like a WHATWG version

[00:08:02.0000] <Lachy>
that's one possibility, but isn't quite what UAs do at the moment

[00:11:00.0000] <Lachy>
I don't think we'll need to change parsing reqs to address this issue, but we'll need to more thouroughly define the DOM APIs upper-casing rules and selector matching

[00:11:01.0000] <Lachy>
since Selectors says case sensitivity depends on the markup language, we should be able to define explicity case sensitivity rules

[00:12:00.0000] <annevk>
selectors should just be fixed...

[00:12:01.0000] <Lachy>
could we define that for XHTML5, selectors are case insensitive?

[00:13:00.0000] <Lachy>
it should be the language that matters, not the serialisation

[00:13:01.0000] <annevk>
i would think so

[00:24:00.0000] <Hixie>
are you also suggesting that <INPUT> and createElementNS(... 'INPUT') create HTML <input> elements?

[00:24:01.0000] <annevk>
createElement should

[00:25:00.0000] <annevk>
createElementNS prolly not

[00:25:01.0000] <Hixie>
then the selectors shouldn't match either, imho

[02:06:00.0000] <Hixie>
hmm

[02:07:00.0000] <Hixie>
i wonder if window.location and window.history are per-Document or per-Window

[02:23:00.0000] <aroben>
Hixie: I should think window.history would have to be per-Window

[02:24:00.0000] <Hixie>
turns out it's actually per-Document

[02:24:01.0000] <aroben>
Hixie: what would a single-document history mean?

[02:24:02.0000] <aroben>
Hixie: ??

[02:24:03.0000] <Hixie>
a new object is created for each Document

[02:24:04.0000] <Hixie>
they all represent the same underlying data, naturally

[02:24:05.0000] <aroben>
Hixie: ah, I see

[02:49:00.0000] <mpt>
Ahhh, that reminds me of Netscape 2

[02:50:00.0000] <mpt>
where the Back and Forward buttons took you back and forward entire framesets, and frames had separate "Back in This Frame" and "Forward in This Frame" shortcut menu items

[02:50:01.0000] <mpt>
madness

[02:53:00.0000] <Hixie>
not that the more modern frame ui is much more usable

[02:55:00.0000] <Hixie>
the public-html and public-xhtml2 are amusing to compare

[02:56:00.0000] <Hixie>
both groups are doing basically nothing

[02:56:01.0000] <Hixie>
but the former is doing it with significantly more acticity

[02:56:02.0000] <Hixie>
activity

[03:47:00.0000] <zcorpan_>
whohoo!!

[03:47:01.0000] <zcorpan_>
i got the job!

[03:47:02.0000] <zcorpan_>
:D :D :D

[03:48:00.0000] <Hixie>
congrats!

[03:48:01.0000] <Hixie>
which job? :-)

[03:48:02.0000] <zcorpan_>
at opera

[03:48:03.0000] <Hixie>
nice

[03:48:04.0000] <zcorpan_>
for the summer

[03:49:00.0000] <Hixie>
do you know what you'll be doing?

[03:49:01.0000] <zcorpan_>
writing a test suite for html5

[03:49:02.0000] <Hixie>
sweet

[03:49:03.0000] <zcorpan_>
yup

[03:49:04.0000] <Hixie>
guess i'd better finish this spec off, then :-P

[03:49:05.0000] <zcorpan_>
heh

[03:50:00.0000] <hasather>
zcorpan_: congrats :)

[03:50:01.0000] <zcorpan_>
thanks

[03:53:00.0000] <hsivonen>
zcorpan_: congrats

[03:53:01.0000] <zcorpan_>
thanks :)

[03:54:00.0000] <hsivonen>
zcorpan_: it would be superb if you could include a categorization of expected syntactic correctness

[03:55:00.0000] <zcorpan_>
hsivonen: you mean test cases intended for conformance checkers?

[03:55:01.0000] <hsivonen>
anne's wf2.0 browser tests were very useful for conformance checkin, but the suite lacked this data

[03:56:00.0000] <hsivonen>
zcorpan_: i mean using browser test as conformance checker tests as a byproduct

[03:56:01.0000] <zcorpan_>
ah, yeah that would be simple to arrange

[04:40:00.0000] <Dashiva>
I have this feeling everyone on the list is ignoring the html60 guy

[04:47:00.0000] <Hixie>
i didn't ignore him

[04:47:01.0000] <Hixie>
i just didn't understand what he was proposing

[04:47:02.0000] <Hixie>
so i didn't feel qualified to comment

[04:47:03.0000] <Hixie>
anyway, well beyond bed time

[04:47:04.0000] <Hixie>
nn

[04:48:00.0000] <hasather>
good night Hixie

[08:42:00.0000] <zcorpan_>
/me downloads swift_0.2.msi

[08:44:00.0000] <sgillies>
i just found out about the time element from Sam Ruby's blog. is anyone considering geo-spatial markup? i'm asking this because Google is proposing that KML (the Google Earth language) be the markup for what the GIS industry likes to call the "Geo-Web". i'd rather there weren't two webs where there could be one 

[08:55:00.0000] <hsivonen>
sgillies: what's the use case for geospatial markup in HTML?

[08:59:00.0000] <sgillies>
find pages, or elements of pages, that have a spatial context. much the same as the use case for time, really. i admit that it is not as generally useful

[08:59:01.0000] <sgillies>
substitute "where" for "when" in the time use case

[09:01:00.0000] <hsivonen>
hmm. looking at http://microformats.org/wiki/geo and comparing the situation with <time>, perhaps Hixie should bless Tantek's general <abbr> design pattern somehow

[09:02:00.0000] <hsivonen>
the main problem I see with the <abbr> pattern is that it is designed to be backwards compatible with HTML 4.01 validation instead of being designed to be backwards-compatible with the browser rendering of <abbr> and title="".

[09:03:00.0000] <sgillies>
hsivonen: http://georss.org is probably a better example than the microformat. all the major web mapping applications can consume spatially referenced RSS.

[10:16:00.0000] <zcorpan_>
/me is playing around with im.css in opera... can someone mention my name?

[10:16:01.0000] <gavin_>
zcorpan_ 

[10:17:00.0000] <zcorpan_>
works :)

[10:17:01.0000] <zcorpan_>
/me now has a counter at the right top of the window of how many times my name has been mentioned

[10:18:00.0000] <Dashiva>
zcorpan_ zcorpan_ zcorpan_ zcorpan_ zcorpan_ zcorpan_ zcorpan_ zcorpan_ mushroom mushroom

[10:18:01.0000] <zcorpan_>
lol

[10:19:00.0000] <zcorpan_>
snake! snake!

[10:34:00.0000] <hendry>
I can't access this URL with my W3 username http://www.w3.org/Member/Intro

[10:34:01.0000] <hendry>
can you guys?

[10:35:00.0000] <h3h>
that's because you're probably not a Member

[10:35:01.0000] <h3h>
you're a member

[10:35:02.0000] <h3h>
there's a difference :)

[10:36:00.0000] <hendry>
so 'invited expert' isn't a member, eh?

[10:36:01.0000] <h3h>
invited expert is a member

[10:36:02.0000] <h3h>
invited expert is not a Member

[10:37:00.0000] <h3h>
Members: http://www.w3.org/Consortium/Member/List

[10:37:01.0000] <hendry>
so a member must be a company, not an individual?

[10:37:02.0000] <hendry>
and I guess a member must pay fees

[10:40:00.0000] <h3h>
heh. is your shift key broken?

[11:00:00.0000] <zcorpan_>
hendry: h3h's point is that there's a difference between members and Members. you're a member but not a Member.

[11:01:00.0000] <hendry>
zcorpan_: oh really?

[11:01:01.0000] <hendry>
h3h: that's really confusing

[11:01:02.0000] <h3h>
tell that to them

[11:02:00.0000] <hendry>
I must say I fealt a little smug with the "invited expert" title at first and didn't ask too many questions

[11:02:01.0000] <hendry>
now I feel like a victim ;)

[12:41:00.0000] <kingryan>
hmm, I can't seem to find where in the current spec it says that "void elements" must not have an end tag

[12:42:00.0000] <kingryan>
it's also not clear that contentmodel == empty is equivalent to "void element"

[12:44:00.0000] <billmason>
kingryan: http://www.whatwg.org/specs/web-apps/current-work/#elements0 "Void elements only have a start tag; end tags must not be specified for void elements."

[12:45:00.0000] <kingryan>
billmason: ah, thanks

[12:46:00.0000] <kingryan>
so, starting at http://www.whatwg.org/specs/web-apps/current-work/#param with the question "why does the validator say 'Fatal Error: End tag param seen even though the element is an empty element.'?" I couldn't find the answer

[13:08:00.0000] <jdandrea>
From Browser Wars IV: A New Hope - http://slashdot.org/comments.pl?sid=231321&cid=18784983

[16:25:00.0000] <zcorpan_>
krijnh: perhaps the logs should state the timezone?


2007-04-19
[04:15:00.0000] <Dashiva>
So Apple doesn't want theora internally because of patent risk, while MS got the eolas patent case for having support externally. Joy.

[06:26:00.0000] <zcorpan_>
svg in 'background'?? whoh! this will be interesting to write ad-hoc testcases against

[06:26:01.0000] <zcorpan_>
re http://annevankesteren.nl/2007/04/html5-talk

[06:41:00.0000] <hsivonen>
I wonder if I should offer to talk about how the WHATWG has done stuff in the orientation part of the telecon assuming that Hixie and Anne aren't participating...

[06:46:00.0000] <met_>
probably spam http://blog.whatwg.org/feed-autodiscovery#comment-3484

[06:47:00.0000] <hsivonen>
met_: do you still see it if you reload? I think I already removed it.

[06:47:01.0000] <met_>
its still there, even after hardreload

[06:48:00.0000] <met_>
hasather, probably second spam?

[06:49:00.0000] <Dashiva>
I don't see it

[06:49:01.0000] <Dashiva>
3468, on the other hand

[06:49:02.0000] <hsivonen>
I'm going to blacklist the Rechtsanwalt

[06:49:03.0000] <hsivonen>
who would want an attorney who spams?

[06:50:00.0000] <met_>
'Rechtsanwalt Strafrecht  - Thanks for it. I searched for it and now i find it here. Thnaks' ?

[06:50:01.0000] <met_>
oh this 3468, wrong anchod

[06:51:00.0000] <Dashiva>
2960 and 2990 are also suspicious

[06:51:01.0000] <met_>
/me published czech translation of xhtml.com - Hixie interview http://www.root.cz/clanky/ian-hickson-o-nove-verzi-html/

[06:57:00.0000] <hsivonen>
met_, Dashiva: bad comments removed. thanks

[07:00:00.0000] <zcorpan_>
was annevk's talk held yesterday?

[07:00:01.0000] <hsivonen>
zcorpan_: I think so

[07:01:00.0000] <zcorpan_>
/me updated http://wiki.whatwg.org/wiki/HTML5_Presentations

[07:10:00.0000] <Lachy>
good evening

[07:15:00.0000] <zcorpan_>
Lachy: evening

[07:16:00.0000] <Dashiva>
Lachy is trying to impose his personal timezone on us against our will (also, hi)

[07:16:01.0000] <zcorpan_>
lol

[07:17:00.0000] <Lachy>
ha!

[07:18:00.0000] <Dashiva>
I was thinking some more about the whole bugmode thing. IE6 isn't compatible with IE7 in many cases, so to write a page now you might actually have to write three pages (IE6,IE7,rest)

[07:18:01.0000] <Dashiva>
And now they're saying IE8 (probably) won't be compatible with IE7. So that's four pages. 

[07:19:00.0000] <Dashiva>
Will this lead to people deprecating older IE versions, or to not taking advantage of new ones?

[07:19:01.0000] <Dashiva>
people as in authors

[07:20:00.0000] <met_>
if you have ie6 - qurk mode, ie7,ff,opera,safari - standard mode, you can write one page 8-) for basic things

[07:20:01.0000] <zcorpan_>
or you write one page that works in opera, mozilla, safari, works less well in all ies, and let ie's market share contine to drop until they decide to implement html5 without versioning :)

[07:20:02.0000] <met_>
for me it just works

[07:20:03.0000] <Lachy>
unfortunately, authors will need to continue supporting IE6 for as long as it retains a significant market share, but we should still be working towards more standards compliance, not holding it back with forced bug modes

[07:21:00.0000] <Lachy>
met_, why would you put IE6 into quirks mode?

[07:21:01.0000] <Lachy>
or did I misunderstand you?

[07:22:00.0000] <met_>
because it leads to less problems

[07:22:01.0000] <Lachy>
really?

[07:22:02.0000] <Dashiva>
I would say pages that only worked in the latest IE would be an incentive to upgrade, but then users on olders windows don't even get the choice...

[07:22:03.0000] <met_>
for css posiotioning yes

[07:22:04.0000] <Lachy>
quirks mode is far too unpredictable

[07:22:05.0000] <zcorpan_>
met_: do you also want to support ie5?

[07:22:06.0000] <met_>
zcorpan_: e-e 

[07:22:07.0000] <met_>
just ie6+7

[07:23:00.0000] <Lachy>
unfortunately, pages that don't work in a users browser aren't an incentive to upgrade, but rather to go elsewhere

[07:23:01.0000] <Dashiva>
They are an incentive to downgrade, though... heh

[07:23:02.0000] <zcorpan_>
Lachy: i didn't mean that it should break entirely, just work less well or not be as pretty

[07:24:00.0000] <Lachy>
ah, degrade gracefully. sure!

[07:24:01.0000] <Lachy>
many personal sites of standards activists do that already

[07:24:02.0000] <met_>
Lachy: my technique is make page for FF.Opr etc, put ie6 in quirk mode and fix 3-4 found bugs, put ie7 in standard mode loom for bugs (there is often none) and finish, it just my technigue, someone can have differrent

[07:25:00.0000] <zcorpan_>
my client's pages do that already too, to some degree

[07:25:01.0000] <Lachy>
met_, I just don't understand what advantages IE6 quirks offers over IE6 standards

[07:25:02.0000] <zcorpan_>
to the degree that is acceptable by the client :)

[07:26:00.0000] <met_>
Lachy: for me it works better, maybe I know better bugs in ie-quirk than ie-standard, but everytime I tried ie-6 in standard mode it gives me more problems

[07:26:01.0000] <Lachy>
zcorpan_, I try to do that with some pages too.  It doesn't always work, though

[07:26:02.0000] <Lachy>
ok, fair enough

[07:27:00.0000] <Dashiva>
How do you get IE6 into quirks, but not IE7? Sending different content?

[07:27:01.0000] <zcorpan_>
xml decl

[07:28:00.0000] <Lachy>
a comment before the DOCTYPE works too

[07:28:01.0000] <met_>
e.g this http://www.vscht.cz/15esfc/

[07:28:02.0000] <zcorpan_>
Lachy: no, that will force ie7 into quirks

[07:28:03.0000] <Lachy>
really?  I thought they fixed that

[07:28:04.0000] <zcorpan_>
nope

[07:28:05.0000] <Lachy>
of, crap!

[07:28:06.0000] <Lachy>
s/of/oh/

[07:28:07.0000] <zcorpan_>
xml declaration on two lines will also trigger quirks in ie7

[07:29:00.0000] <zcorpan_>
i.e. <?xml\nversion="1.0"?>

[07:29:01.0000] <Lachy>
damn, MS are hopeless

[07:29:02.0000] <met_>
Lachy: look at table on http://www.lupa.cz/clanky/navrat-modreho-e/

[07:29:03.0000] <met_>
just scoll to 1st table there

[07:29:04.0000] <zcorpan_>
/me sent bug reports about that during the "beta" period

[07:29:05.0000] <virtuelv>
zcorpan_: what about <?xml\tversion="1.0"?>

[07:30:00.0000] <zcorpan_>
virtuelv: haven't tested that

[07:30:01.0000] <Lachy>
my latest off-list reply from Chris Wilson seems to indicate that he's not willing to negotiate or compromise in any way whatsoever :-(

[07:30:02.0000] <met_>
xml trigger quirk for ie6

[07:31:00.0000] <zcorpan_>
virtuelv: yep, tab triggers quirks too

[07:31:01.0000] <zcorpan_>
or wait

[07:32:00.0000] <met_>
<?xml anything ?> triggers is what i know

[07:32:01.0000] <virtuelv>
what about <?xml\t \tversion="1.0"?>

[07:32:02.0000] <virtuelv>
met_: in IE6, yes

[07:32:03.0000] <virtuelv>
we're discussing IE7

[07:32:04.0000] <met_>
yes ie6

[07:32:05.0000] <met_>
in ie7 it is standard mode for me

[07:32:06.0000] <zcorpan_>
yup, tab triggers quirks in ie7

[07:34:00.0000] <zcorpan_>
seems they look for "<?xml "

[08:11:00.0000] <zcorpan_>
/me observes that most canvas demos on the web are in quirks mode.

[08:12:00.0000] <zcorpan_>
so. ms will have to add new features to quirks mode

[08:12:01.0000] <Philip`>
My canvas demo is valid XHTML, since I wanted to find as many ways as possible to be incompatible with IE ;-)

[08:12:02.0000] <zcorpan_>
pointer?

[08:14:00.0000] <Philip`>
http://canvex.lazyilluminati.com/83/play.xhtml

[08:16:00.0000] <Philip`>
(The DTD stuff is rather evil and I don't know if it's correct, but at least it makes the document valid according to the W3C validator...)

[08:17:00.0000] <csarven>
Philip` im impressed =)

[08:18:00.0000] <zcorpan_>
Philip`: i thought you'd know by now that dtds are irrelevant? :)

[08:18:01.0000] <zcorpan_>
why not drop the doctype altogether?

[08:18:02.0000] <met_>
Philip`: Whow!

[08:19:00.0000] <zcorpan_>
Philip`: very nice demo btw

[08:20:00.0000] <zcorpan_>
Philip`: you should figure out a way to port all doom maps to canvas programatically :)

[08:20:01.0000] <Philip`>
zcorpan_: Because the DTD gives an opportunity to add pointless complexity to the project :-)

[08:21:00.0000] <Philip`>
(which is the same reason why it uses SVG for the HUD graphics)

[08:21:01.0000] <hsivonen>
Philip`: that's all canvas 2D, right?

[08:21:02.0000] <zcorpan_>
svg for the hud is not pointless, it adds to the quality afaict

[08:21:03.0000] <Philip`>
(and why it's written in JavaScript and <canvas>, instead of something sane like C++)

[08:21:04.0000] <hsivonen>
Philip`: cool either way!

[08:22:00.0000] <Philip`>
zcorpan_: http://canvex.lazyilluminati.com/84/play.xhtml might look familiar (though I didn't get around to adding any textures or sprites)

[08:24:00.0000] <Philip`>
hsivonen: Yep, it's just 2D (mainly using drawImage quite a lot of times)

[08:27:00.0000] <zcorpan_>
Philip`: did you port the map? write it from scratch?

[08:28:00.0000] <Philip`>
It's a mostly automatic conversion

[08:28:01.0000] <zcorpan_>
cool

[08:29:00.0000] <Philip`>
(I think I used some other tool to get a human-readable version of the map file, then parsed/processed/converted it)

[08:31:00.0000] <Philip`>
((The conversion is a bit buggy - it's missing the lake outside the window and the square thingy in the large room beyond the first door. I think that would just need more complex conversion into the map structure that my engine understands, but I gave up / ran out of time before then))

[08:32:00.0000] <zcorpan_>
i'd presume that it would perform better with canvas 3d

[08:35:00.0000] <Philip`>
A hardware-acclerated 2D canvas would probably perform much better than the current one - but real 3D would be even better, since I could remove all the perspective hacks (drawing lots of thin columns of decreasing sizes) and just draw real 3D walls

[08:35:01.0000] <Philip`>
*accelerated

[08:38:00.0000] <Philip`>
http://canvex.lazyilluminati.com/misc/photos.html - that uses the moz-gles11 canvas, which seems to work pretty nicely (if you've compiled the extension for it)

[08:40:00.0000] <Philip`>
(http://canvex.lazyilluminati.com/misc/giraffes.png - output (except it rotates and changes the picture each time it turns))

[08:43:00.0000] <zcorpan_>
nice

[08:47:00.0000] <csarven>
Philip` got a source to moz-gles11 canvas extension?

[08:49:00.0000] <Philip`>
http://lxr.mozilla.org/mozilla/source/extensions/canvas3d/

[08:50:00.0000] <Philip`>
(You can just add an extra flag when compiling the browser, then wait n hours until it's finished, and then it should work (though I think it's Windows-only right now))

[08:51:00.0000] <csarven>
oh

[08:58:00.0000] <haustein>
hey so I could use glFontTextOut() and finally draw text in canvas? :)

[08:59:00.0000] <Philip`>
OpenGL doesn't do text :-)

[09:00:00.0000] <Philip`>
The usual methods (at least for games) seem to be rendering onto a bitmap using FreeType/GDI/etc then using that as a texture, or pre-rendering a texture with all the characters you're going to use and then drawing lots of little textured rectangles

[09:00:01.0000] <haustein>
ok i guess should read the stuff before copy pasting some random google result.... :)

[09:01:00.0000] <Philip`>
There are quite a few libraries that try to handle text in GL, but none are standard (and I think none are ubiquitous)

[09:02:00.0000] <haustein>
I think I'll just continue waiting for drawString in 2d....

[09:05:00.0000] <Dashiva>
Make a bitmap font :)

[09:08:00.0000] <Philip`>
http://zaynar.demon.co.uk/misc2/font4.png - bitmap fonts are not excellent when you want to handle non-English characters, but they're not that bad if you pack them properly and if you only need a single font/size/weight, and they're pretty easy to use

[09:12:00.0000] <haustein>
i think the font image is about 10 times the size of my application.... and you still need the meta information and the code 


2007-04-20
[17:18:00.0000] <Hixie>
anyone have an installation of IE that handles flash plugins correctly?

[17:18:01.0000] <zcorpan_>
i do (i think)

[17:19:00.0000] <Hixie>
can you go to: http://software.hixie.ch/utilities/js/live-dom-viewer/

[17:19:01.0000] <Hixie>
and click "download"

[17:20:00.0000] <zcorpan_>
sec

[17:20:01.0000] <Hixie>
and see if it says "pass" in the rendered view?

[17:20:02.0000] <Hixie>
(i think that's your flash test file, in fact)

[17:21:00.0000] <zcorpan_>
heh

[17:21:01.0000] <zcorpan_>
it says pass

[17:21:02.0000] <Hixie>
ok hold on

[17:23:00.0000] <zcorpan_>
not in ie6 though, it seems...

[17:23:01.0000] <Hixie>
download again and see what the log says (bottom box)

[17:24:00.0000] <zcorpan_>
rendering mode: CSS1Compat

[17:24:01.0000] <zcorpan_>
document has no title

[17:24:02.0000] <zcorpan_>
log: [object]

[17:24:03.0000] <zcorpan_>
log: [object]

[17:24:04.0000] <zcorpan_>
log: undefined

[17:24:05.0000] <zcorpan_>
but there's more...

[17:25:00.0000] <Hixie>
interesting

[17:25:01.0000] <zcorpan_>
there's an infobar warning me about a blocked download

[17:25:02.0000] <Hixie>
so basically IE doesn't let you access the Document of a full-plugin pyage

[17:25:03.0000] <Hixie>
page

[17:26:00.0000] <zcorpan_>
if i click the infobar and click "download", the page reloads and the log is:

[17:26:01.0000] <zcorpan_>
rendering mode: CSS1Compat

[17:26:02.0000] <zcorpan_>
document has no title

[17:27:00.0000] <zcorpan_>
plus there's a second box below that says "This script puts a function w(s) into the global scope of the test page, "

[17:27:01.0000] <Hixie>
yeah i just added that box just now

[17:27:02.0000] <Hixie>
ignore that :-)

[17:27:03.0000] <zcorpan_>
ok

[17:28:00.0000] <Hixie>
wait if you reload and click "download" again it doesn't show the flash file at all?

[17:28:01.0000] <zcorpan_>
now it does

[17:29:00.0000] <zcorpan_>
rendering mode: CSS1Compat

[17:29:01.0000] <zcorpan_>
document has no title

[17:29:02.0000] <zcorpan_>
log: [object]

[17:29:03.0000] <zcorpan_>
log: [object]

[17:29:04.0000] <Hixie>
then an exception is raised?

[17:29:05.0000] <Hixie>
(should show little yellow warning icon in the status bar if so)

[17:29:06.0000] <zcorpan_>
yeah

[17:29:07.0000] <zcorpan_>
"Permission denied" or something

[17:30:00.0000] <Hixie>
k

[17:30:01.0000] <zcorpan_>
ie's error console isn't very helpful :(

[17:31:00.0000] <Hixie>
heh

[18:22:00.0000] <zcorpan_>
as a fulltime job, how many testcases can one create in one day, in avarage?

[18:23:00.0000] <zcorpan_>
given you know the spec you're writing test cases against and you know how to write test cases?

[18:25:00.0000] <zcorpan_>
/me is trying to estimate how many test cases he will be able to write this summer

[18:25:01.0000] <Lachy>
zcorpan_, given roughly 5 minutes per test case, you could get close to 100 per day

[18:25:02.0000] <Lachy>
that 5 min includes writing and testing in several browsers, and fixing any mistakes you find in your test

[18:26:00.0000] <Hixie>
zcorpan_: depends

[18:26:01.0000] <Hixie>
zcorpan_: some test cases take upwards of 2 hours to make

[18:26:02.0000] <Hixie>
zcorpan_: and then you can write a script that spits out 100 with just 30 minutes of work

[18:26:03.0000] <zcorpan_>
Hixie: yeah, indeed

[18:27:00.0000] <Hixie>
zcorpan_: i'd estimate 1 hour per test per person on average over several months

[18:27:01.0000] <Hixie>
since you'll also have to comment on the spec, learn the spec, etc

[18:27:02.0000] <Hixie>
so 40 tests a week

[18:27:03.0000] <Hixie>
these being high-quality tests

[18:27:04.0000] <Lachy>
wow, I seriously over estimated!

[18:27:05.0000] <Hixie>
if you're working fulltime and are doing simple tests, you can get more done

[18:27:06.0000] <Hixie>
it varies

[18:28:00.0000] <zcorpan_>
6 weeks, that's 240 test cases

[18:28:01.0000] <Hixie>
if you're doing testing exclusively you'll get more done than that, probably

[18:28:02.0000] <Hixie>
depends on the tests

[18:29:00.0000] <zcorpan_>
i'll probably be working on the implemented parts of html5 that don't already have tests

[18:29:01.0000] <zcorpan_>
as a start

[18:29:02.0000] <Hixie>
like History and Location?

[18:29:03.0000] <zcorpan_>
yeah

[18:30:00.0000] <Hixie>
for those 90% of the work will be reverse engineering browsers and trying to work out what parts of the spec are wrong

[18:30:01.0000] <Hixie>
so 1 hour per test is quite reasonable for those sections :-)

[18:30:02.0000] <zcorpan_>
probably also things that are new but have experimental implementations, like video

[18:34:00.0000] <zcorpan_>
if we want to have 20,000 tests by 2010, then we'd need at least 3 people working full-time writing test cases

[18:35:00.0000] <zcorpan_>
given i didn't miscalculate

[18:36:00.0000] <karlUshi>
5 days * 8 hours a day = 40 test cases a week 

[18:37:00.0000] <karlUshi>
for the corner cases that Ian was mentionning

[18:37:01.0000] <karlUshi>
another thing to take into consideration is the format to report the results

[18:38:00.0000] <karlUshi>
if we have a common format, it will be a lot easier to produce an interop report

[18:39:00.0000] <Hixie>
i love how every time i commit something and the script twitters the commit message, WHATWG gets new twitter friends

[18:39:01.0000] <Hixie>
that's hilarious

[18:40:00.0000] <zcorpan_>
automatically or because people add whatwg as their twitter friend?

[18:41:00.0000] <Hixie>
i assume people see WHATWG on the public timeline and go "oo, i should add that"

[18:41:01.0000] <zcorpan_>
ah

[18:49:00.0000] <zcorpan_>
calculating again from the other end, if one person would work full-time writing test cases, it would take 11.6 years to get 20,000 tests

[18:50:00.0000] <Hixie>
yeah but if one person worked full time on tests for 11 years, by the end of it they'd be going way faster than one test every hour

[18:50:01.0000] <Hixie>
:_)

[18:51:00.0000] <zcorpan_>
true

[18:52:00.0000] <Hixie>
ok i am not looking forward to the next section i have to do

[18:52:01.0000] <Hixie>
/me shudders

[18:52:02.0000] <Hixie>
("scripting")

[18:52:03.0000] <zcorpan_>
why not?

[18:52:04.0000] <Hixie>
mostly i'm not sure what it should say

[18:54:00.0000] <karlUshi>
Hixie: define only one element ;) no scripting :p

[18:55:00.0000] <karlUshi>
/me wonders if Hixie has a template when he's writing sections. As in "having a set of questions, I should go through to write"

[19:04:00.0000] <Hixie>
i wonder what happens if you invoke a method associated with a document that isn't the active document in a browsing context

[20:30:00.0000] <Hixie>
ok i'm basically down to the scripting security model section

[22:20:00.0000] <paulproteus>
Hixie, That is hilarious re: twitter friends. (-:

[22:28:00.0000] <othermaciej>
/me looks for an antacid and an aspirin before reading cwilso's mail

[23:09:00.0000] <annevk>
is Simon around?

[23:13:00.0000] <Lachy>
annevk, Simon is asleep

[23:15:00.0000] <annevk>
He reported a bug on Opera that <foo><dd></foo> should have closed the <dd> while in fact the </foo> is ignored

[23:15:01.0000] <annevk>
However, this is exactly what html5lib is doing too

[23:16:00.0000] <Lachy>
is that what the spec says to do?

[23:16:01.0000] <annevk>
and Firefox

[23:16:02.0000] <annevk>
maybe I misread the bug report

[23:17:00.0000] <annevk>
oh, he uses http://simon.html5.org/test/html/parsing/tree-construction/the-main/how-to/in-body/003.htm as a testcase

[23:18:00.0000] <annevk>
never mind, that's not what Firefox does

[23:19:00.0000] <annevk>
it's what html5lib does

[23:19:01.0000] <Lachy>
IE does the same as FF

[23:20:00.0000] <annevk>
IE can't be relied upon with unknown elements

[23:20:01.0000] <Lachy>
oh, right

[23:20:02.0000] <tantek>
that sounds like an overgeneralization

[23:21:00.0000] <annevk>
Well, unless you involve namespaces you can't

[23:23:00.0000] <othermaciej>
can't be relied upon?

[23:23:01.0000] <othermaciej>
does it do something weird with unknown elements?

[23:23:02.0000] <annevk>
</foo> becomes an element /foo

[23:23:03.0000] <othermaciej>
!

[23:24:00.0000] <annevk>
all unknown tags basically become void elements

[23:24:01.0000] <annevk>
same for some "known" elements if they are outside their usual context (td, etc.)

[23:24:02.0000] <othermaciej>
that is pretty weird

[23:25:00.0000] <othermaciej>
I see now why they are scared to change anything in parsing

[23:27:00.0000] <annevk>
whoa, I'm quoted in the news

[23:27:01.0000] <annevk>
well, some tech site

[23:29:00.0000] <Lachy>
annevk, pointer to the article?

[23:30:00.0000] <annevk>
I can't find the original

[23:30:01.0000] <annevk>
http://risal.wordpress.com/2007/04/19/is-the-web-ready-for-html-5/

[23:30:02.0000] <annevk>
has the quotes though

[23:30:03.0000] <annevk>
I did actually say that, but I didn't think I'd get the attribution

[23:31:00.0000] <annevk>
“HTML5 is about preserving the information people have accumulated over the years,” Opera spokesperson Anne van Kesteren told internetnews.com. “By remaining backward and forwards compatible, we hope to ensure that people will be able to interpret HTML for decades if not centuries to come.”

[23:37:00.0000] <annevk>
http://www.internetnews.com/dev-news/article.php/3672011 is the original

[23:38:00.0000] <annevk>
found it by using http://www.google.com/search?q=%22HTML5+is+about+preserving+the+information+people+have+accumulated+over+the+years%22

[23:39:00.0000] <Lachy>
heh, IE actually allows .createElement("/foo") to work

[23:39:01.0000] <Lachy>
it looks like it's the same article

[23:40:00.0000] <annevk>
I suppose people are just making copies

[23:42:00.0000] <Lachy>
do you know where you actually said what they quoted?  Did they just copy it from some post to whatwg or public-html?

[23:43:00.0000] <annevk>
yes, no

[23:51:00.0000] <annevk>
/me wonders how it helps if cwilso became chair given his argument about competitors

[23:51:01.0000] <annevk>
then again, he said that point wasn't as important as his other

[23:52:00.0000] <Lachy>
the one about Hixie working for Google?

[23:52:01.0000] <annevk>
yes

[23:53:00.0000] <Lachy>
I was surprised about that too, especially since Google isn't competing with IE, they're competing with MS in other areas

[00:05:00.0000] <annevk>
Google has a Firefox dev team

[00:06:00.0000] <Lachy>
so?

[00:06:01.0000] <Lachy>
They also build stuff for IE too, like the google toolbar for instance

[00:07:00.0000] <annevk>
/me would count both as competing

[00:16:00.0000] <krijnh>
Morning people

[00:24:00.0000] <annevk>
good afternoon

[00:24:01.0000] <annevk>
apparently JavaScript isn't as interoperable as claimed: http://my.opera.com/hallvors/blog/2007/04/11/new-adventures-in-date-parsing

[00:25:00.0000] <annevk>
I suppose it's time for another X5 remark

[00:25:01.0000] <othermaciej>
ECMAScript5?

[00:27:00.0000] <annevk>
Yeah, I wonder if they are solving these type of issues too

[00:28:00.0000] <othermaciej>
they plan to write the spec via a reference implementation in SML

[00:28:01.0000] <othermaciej>
I think spec by reference implementation is a really bad idea

[00:30:00.0000] <annevk>
the "Web browsers" section of HTML5 is complicated

[00:30:01.0000] <annevk>
though browsers are like that, I guess :)

[00:31:00.0000] <Hixie>
ECMAScript is one of the better-written specs

[00:32:00.0000] <othermaciej>
it suffers mainly from lack of maintenance, but yes, it is written better than HTML4 certainly

[00:34:00.0000] <annevk>
"The rules for chosing a browsing context given a browsing context name" don't take into account what happens when there's no parent browsing context

[00:34:01.0000] <annevk>
for instance

[00:42:00.0000] <annevk>
So it seems that <foo><dd></foo> should indeed close <dd>

[00:42:01.0000] <annevk>
as in <foo><dd></foo>x "x" should be a sibling of foo

[00:48:00.0000] <annevk>
Actually, no...

[00:49:00.0000] <annevk>
when you hit </foo> <dd> is the bottommost node of the spec

[00:49:01.0000] <annevk>
s/spec/stack/

[00:49:02.0000] <annevk>
<dd> is in the special category and therefore </foo> has to be ignored

[01:22:00.0000] <othermaciej>
I'm really amused by Murray asking for Dan to explain all the technical details to him on public-html

[01:22:01.0000] <othermaciej>
even with his 30 years of experience with markup languages

[01:22:02.0000] <annevk>
apparently he misses all the +1 messages to dbaron his e-mail

[01:23:00.0000] <annevk>
I was wondering whether I should reply that people who agree with dbaron don't do +1

[01:24:00.0000] <othermaciej>
I am amazed at how many times you can make arguments against versioning in the doctype or the namespace URI and people keep bringing it up

[01:24:01.0000] <othermaciej>
not even refuting the arguments, just acting like they didn't happen

[01:25:00.0000] <annevk>
yeah, arguments -> wiki

[01:26:00.0000] <annevk>
everyone just ignored John Boyer :)

[01:27:00.0000] <othermaciej>
well, he didn't say anything relevant to the group

[01:27:01.0000] <othermaciej>
I wanted to point out that the Forms task force hadn't started yet, but thought better of it

[01:29:00.0000] <annevk>
I like what marcos said about XForms Transitional

[01:29:01.0000] <annevk>
"Seems illogical to me to have transitional technology to something that will be incompatible with the web (XForms)."

[01:31:00.0000] <Hixie>
the xforms wg thinks the forms task force has started, fwiw

[01:31:01.0000] <Hixie>
see their minutes

[01:32:00.0000] <othermaciej>
who do they think is in the task force?

[01:32:01.0000] <othermaciej>
just them?

[01:32:02.0000] <Hixie>
no idea

[01:32:03.0000] <Hixie>
didn't investigate

[02:04:00.0000] <annevk>
bugmode= ...

[02:04:01.0000] <annevk>
what if all our testcases don't include that attribute?

[02:04:02.0000] <annevk>
IE will forever fail?

[02:05:00.0000] <Lachy>
indeed

[02:05:01.0000] <annevk>
wfm

[02:05:02.0000] <annevk>
i don't think we should worry much about it

[02:05:03.0000] <othermaciej>
it would be good to report its non-conformance accfurately

[02:06:00.0000] <Lachy>
annevk, as I just wrote in my last email, we should worry about it because all authors would be unconditionally required to use it

[02:06:01.0000] <annevk>
until MS stops seeing the spec as a guide there's not much point in arguing anyway

[02:07:00.0000] <annevk>
they seem to think fundamentally different about standards

[02:07:01.0000] <annevk>
unless cwilso responded to some of Hixie's points about quirks mode in other browsers, specs as guide, etc. already and denied that

[02:07:02.0000] <Lachy>
yes, and given their monopoly position, that is a problem

[02:08:00.0000] <annevk>
not much we can do about it on the mailing list it seems

[02:08:01.0000] <Lachy>
not sure what else we can do that would be even remotely successful

[02:08:02.0000] <annevk>
/me was wondering whether a F2F or telcon would be good for this

[02:09:00.0000] <Lachy>
it should definitely be discussed in the telcon if we have one

[02:10:00.0000] <Lachy>
I'll even stay awake for the telcon if it's going to be discussed

[02:10:01.0000] <annevk>
oh right, a telcon was proposed

[02:11:00.0000] <annevk>
I don't think I'll make it

[02:22:00.0000] <othermaciej>
/me wonders who Terje Bless is

[02:54:00.0000] <virtuelv>
othermaciej: think he used to work for w3c, but I can't remember as what

[02:55:00.0000] <hsivonen>
othermaciej: Terje Bless is from the W3C Validator team

[02:55:01.0000] <othermaciej>
hsivonen: I'm surprised at his views on implementation conformance

[02:56:00.0000] <hsivonen>
othermaciej: do you mean on IRC yesterday?

[02:56:01.0000] <othermaciej>
no, in email today

[02:56:02.0000] <othermaciej>
his apparent view that an implementation which handles content in a nonconformant way can still be conformant

[02:57:00.0000] <hsivonen>
so it seems

[02:58:00.0000] <met_>
/me wonders if case location=url1; location=url2 is specified on html5, cannot found it

[02:58:01.0000] <hsivonen>
I feel pretty strongly that if MS is going to have a switch no matter what, it should be done with an attribute that the spec makes conforming even if we all hold our noses while doing so

[03:00:00.0000] <othermaciej>
the spec shouldn't make triggering nonconforming behavior based on that attribute conforming, however

[03:00:01.0000] <othermaciej>
if there is one

[03:01:00.0000] <met_>
now is behaviour different among browsers, this is why i care

[03:02:00.0000] <met_>
 location=url1; alert(1); location=url2; alert(2); location=url3 etc.

[03:02:01.0000] <othermaciej>
met_: I specified it in the Window spec I think; I believe the last location within a single batch of script execution wins

[03:02:02.0000] <met_>
shouldn't be specified recommended behaviour?

[03:03:00.0000] <othermaciej>
<script>location = l1; location = l2;</script> --> you end up at l2

[03:03:01.0000] <met_>
/me is looking on window spec

[03:03:02.0000] <othermaciej>
<script>location = l1;</script><script>location = l2;</script>

[03:03:03.0000] <othermaciej>
that ends up at l1

[03:05:00.0000] <hsivonen>
othermaciej: if a conformance checker doesn't pass an IE bug mode attribute as conforming, MS will use a switch that is cloaked from conformance checkers

[03:06:00.0000] <hsivonen>
othermaciej: the net result would be no change to the verdict given by a conformance checker, but there'd be more pain for intermediate tool vendors who build on XML tooling

[03:08:00.0000] <hsivonen>
othermaciej: I would be OK, though, with making the conformance checker only treat the latest IE bug mode opt-in as conforming

[03:08:01.0000] <othermaciej>
hsivonen: I'm not opposed to such an attribute, I am just saying it should not be conformant for implementations to actually introduce bugs based on the value or absence of a bugmode attribute

[03:09:00.0000] <hsivonen>
othermaciej: oh, you were talking about UA conformance. I was talking about document conformance.

[03:11:00.0000] <othermaciej>
hsivonen: I understood what you meant; sorry for not being sufficiently clear myself

[06:18:00.0000] <virtuelv>
what does it take to convince the editor that the WA1.0 spec needs to be split into multiple documents

[06:19:00.0000] <krijnh>
zcorpan_: it probably should, yeah

[06:19:01.0000] <krijnh>
virtuelv: http://www.whatwg.org/specs/web-apps/current-work/multipage/ ?

[06:19:02.0000] <virtuelv>
$ lynx -nolist -dump http://whatwg.org/specs/web-apps/current-work/ | wc -w

[06:19:03.0000] <virtuelv>
150202

[06:21:00.0000] <virtuelv>
krijnh: I'd like something in between

[06:21:01.0000] <virtuelv>
3. on one page, then 3.1. on another probably isn't the best thing either

[06:21:02.0000] <virtuelv>
but thanks anyway, it's much better to work with those slightly smaller docs

[06:22:00.0000] <Lachy>
virtuelv, yeah, splitting on just the higher level headings, rather than the sub headings would be sensible

[06:22:01.0000] <krijnh>
Philip` can probably fix that

[06:23:00.0000] <Lachy>
I don't like the split version though, it makes searching through it annoying

[06:23:01.0000] <krijnh>
Afaik he made the current multipage stuff

[06:23:02.0000] <Lachy>
I wish whatwg.org/html5 redirected to the full page version instead

[06:43:00.0000] <Philip`>
Lachy: I think splitting on just the higher levels (1., 2., 3., etc) wouldn't be much help, since section 3 is about half of the whole document so it'd still be huge

[06:44:00.0000] <Philip`>
(so you might as well use the single-page version, if you have a computer/browser that doesn't fall to pieces when trying to render it)

[06:45:00.0000] <Philip`>
((I don't know why people actually have the problems they've reported - for me (in Opera and Firefox) it just takes ten seconds to load and then it's perfectly fine...))

[06:45:01.0000] <Lachy>
doesn't everyone have a dual core, hyperthreaded Pentium 4 with 2GB of RAM that can handle such documents with ease? ;-)

[06:46:00.0000] <Philip`>
Mine's only single core and the hyperthreading is disabled ;-)

[06:46:01.0000] <Lachy>
actually, I got the one without hypertheading, but it's dual core

[06:47:00.0000] <Lachy>
would it be much effort to special case section 3 then, so that it gets split on the sub headings, but the smaller sections were split on the higher level?

[06:47:01.0000] <Lachy>
or would that just be illogical

[06:48:00.0000] <Philip`>
virtuelv: The nearly-empty pages for e.g. "3." should be considered a bug - it's designed to work that way, but the design should be changed to merge nearly-empty pages with more contentful pages (but I don't have time to work on that now/soon)

[06:49:00.0000] <Philip`>
Lachy: Section 3 is actually special-cased already, to split on some third-level headings (so that the big areas like video and canvas are separated)

[06:50:00.0000] <Lachy>
oh, ok

[06:50:01.0000] <Philip`>
If I remember correctly, I was trying to keep each of the pages under about 100KB

[06:52:00.0000] <Philip`>
It should be easy to do the splitting in a different way, so the individual sections are much larger - I guess it'd just be a bit confusing to have three official versions of the same document, and I don't know which variations would be the best to keep

[06:55:00.0000] <Philip`>
(http://html5.googlecode.com/svn/trunk/spec-splitter/spec-splitter.py in the place where it has 'h3' is how it decides what to split on, and that could just be changed to something different)

[06:58:00.0000] <Lachy>
I think I finally figured out Chris' insane logic for wanting for wanting version numbers in the DOCTYPE and how that relates to the bug mode switch!

[06:58:01.0000] <Lachy>
:-)

[06:59:00.0000] <Lachy>
I think the logic is this:...

[07:00:00.0000] <Lachy>
1. <!DOCTYPE html PUBLIC "html5"> triggers HTML5 mode, which will eventually become a frozen bug state

[07:00:01.0000] <Lachy>
2. After it becomes a frozen bug state, IE introduces another form of explicit opt-in. e.g. <html ie-version="9">

[07:01:00.0000] <Lachy>
3. IE continues updating the bug mode switch with each new release for several years, each time improving their compliance

[07:02:00.0000] <Lachy>
4. HTML6 is released a while after that with a new DOCTYPE. <!DOCTYPE html PUBLIC "html6">

[07:02:01.0000] <Lachy>
5. That provides a brand new opt-in, and by that time IE will be compliant enough to not need a specific bug mode opt-in

[07:07:00.0000] <Philip`>
I'm not sure about step 5 - I've thought it is "5. That provides a brand new opt-in, which can replace the proprietary ie-version="13", but they still don't quite comply with HTML6, so they'll add a proprietary ie-version="14" back in for the next release"

[07:07:01.0000] <virtuelv>
Lachy: that is going to be a nightmare for every other browser vendor

[07:07:02.0000] <jdandrea>
/me is catching up on public-html emails ... !

[07:07:03.0000] <Lachy>
Philip`, yes, there's always the possibility that MS will keep repeating the cycle

[07:08:00.0000] <Lachy>
virtuelv, I know.  As I said, it's *insane* logic

[07:08:01.0000] <Philip`>
and then "6. After enough iterations (hopefully only one, but maybe that's optimistic and they want to plan ahead), the rate of new spec versions will be equal to the rate at which new IE versions break sites, so they will converge and ie-version will never be necessary again"

[07:08:02.0000] <Lachy>
/me is also catching up.  I have about 242 to read

[07:08:03.0000] <virtuelv>
Lachy: if then Opera, Mozilla and Safari added their own <html [vendor]-version="n" and multiple incompatible attributes were selected, which one should trump

[07:09:00.0000] <Lachy>
virtuelv, I have no idea.  I hope no other browser but IE introduces a bug mode switch

[07:09:01.0000] <jdandrea>
Lachy: Ow! OK, I thought I was in worse shape but I'm current as of Thursday 23:00 -0400 (just got past Chris' official response and Hixie's response to that ...)

[07:09:02.0000] <Lachy>
I wish IE wouldn't either

[07:10:00.0000] <Dashiva>
IE8 should automatically reform pages to fix its own bugs, sort of like opera's browserjs :)

[07:10:01.0000] <Lachy>
some of them I already read, but since I was away for the WAF WG meeting, using a crappy email client, I was unable to process them all well

[07:12:00.0000] <Philip`>
(Assuming this is inevitable, the job of the HTML WG should perhaps be to make step 6 happen as quickly as possible, by making the first IE-with-HTML5 as compliant as possible, presumably by following IE7's behaviour when sane and by providing enough test cases for the IE developers to know what they're doing wrong before release)

[07:12:01.0000] <Lachy>
right, so what IE is really after is another Transitional DOCTYPE with which they can abuse for their opt-ins

[07:12:02.0000] <jdandrea>
I was just speaking with a former editor for ANSI C++ about the whole versioning issue (briefly). His $0.02: Detecting multiple versions of any language/markup/etc. are a nightmare.  Old versions stay around much longer than anyone imagines.  (Then, with no prodding from me, he cited Visual C++ v6 as an example.) :)

[07:14:00.0000] <Lachy>
jdandrea, yes.  Once IE introduces a frozen bug state, it will be up to them to maintain that state forever.

[07:14:01.0000] <jdandrea>
Ugh.

[07:14:02.0000] <jdandrea>
He then mentioned that Python does some interesting things WRT versions. I responded: "I don't see anything like a version switch." He replied: "I'm thinking about 'from _future_ import ...'" !

[07:15:00.0000] <Lachy>
I don't know what _future_ is in python

[07:15:01.0000] <jdandrea>
Future versions that aren't known or haven't been released, IIRC.

[07:15:02.0000] <Philip`>
jdandrea: I don't think one can complain (in the sense of blaming MS) about VC++6 being non-standard-compliant, since it was released before the standard, nor about VC++2003 since that did a good job of being standard-compliant; but it's fair to complain that it took them five years between those releases

[07:16:00.0000] <jdandrea>
Philip ` - agreed, and I think that was his ultimate point, the length of time it took to make good.

[07:17:00.0000] <Philip`>
Lachy: Python 2.5 adds the keyword "with", but it's disabled by default (since it would break compatibility with existing code that uses variables/etc named "with")

[07:17:01.0000] <Philip`>
In Python 2.6, it will be enabled by default, so those programs have to be fixed before then

[07:17:02.0000] <Philip`>
In the meantime, Python 2.5 programs can write "from __future__ import with" to get the 2.6-like behaviour

[07:18:00.0000] <jdandrea>
There you go.

[07:18:01.0000] <Lachy>
unfortunately, that sort of method couldn't work with HTML though

[07:19:00.0000] <jdandrea>
Aye. I figured it was worth getting other POVs to see if there's any more precedent we can bring to the table, but I gather this horse is beaten to a pulp.

[07:19:01.0000] <zcorpan_>
quirks mode will be around forever...

[07:20:00.0000] <zcorpan_>
/me intends to figure out quirks mode and spec it, unless someone else does it

[07:20:01.0000] <Lachy>
zcorpan_, good luck getting IE to conform to your spec :-)

[07:20:02.0000] <zcorpan_>
yeah

[07:21:00.0000] <jdandrea>
LOL

[07:21:01.0000] <jdandrea>
Bizarro-HTML: http://lists.w3.org/Archives/Public/public-html/2007Apr/1244.html ... sigh ...

[07:21:02.0000] <Lachy>
it's clear that IE won't make any changes at all to quirks mode, except maybe adding new elements (I assume the new HTML5 elements will be supported in quirks mode)

[07:21:03.0000] <zcorpan_>
they'd only do that if they become a minority vendor

[07:21:04.0000] <zcorpan_>
Lachy: it's not clear to me they will add new features to quirks mode

[07:21:05.0000] <Philip`>
With Python programs, you'd fix them (preferably as soon as Python 2.5 is released and gives you warnings that your variable called "with" is going to break in the next release) or stick with an old version of Python - that doesn't work at all when those 'programs' are HTML pages that you can't change and are owned by somebody who isn't going to change them ever :-(

[07:22:00.0000] <zcorpan_>
but nevertheless <canvas> is already used in quirks mode

[07:22:01.0000] <zcorpan_>
for one

[07:22:02.0000] <Philip`>
Maybe they would decide the amount of <canvas> content is far less than the amount of quirks-mode content that would break if they started handling <canvas>

[07:23:00.0000] <Lachy>
zcorpan_, if they didn't allow them in quirks mode, it would be a way to force all HTML5 pages to use standards mode, but AFAIK, other browsers are going to make them work in quirks

[07:23:01.0000] <Lachy>
<canvas> already works in quirks, for instance

[07:23:02.0000] <Philip`>
(I suppose that's where you need real statistics of how many pages use <canvas> accidentally without wanting the HTML5 interpretation of it)

[07:23:03.0000] <zcorpan_>
Lachy: indeed

[07:23:04.0000] <zcorpan_>
Philip`: google for "canvas demo"

[07:23:05.0000] <zcorpan_>
and check the first few

[07:26:00.0000] <Philip`>
Yep, those are the real <canvas> content that would be fixed if IE handled it in quirks mode - the problem (maybe less so for canvas than for other elements) is old sites that used <canvas> when its meaning was undefined and they never noticed because every browser ignored it and just displayed the content, and those would be broken if IE started to handle it

[07:27:00.0000] <Philip`>
Presumably Microsoft needs to be convinced that the second group is nonexistent or negligible compared to the first group, otherwise they wouldn't want to risk handling <canvas> in case it breaks old sites

[07:28:00.0000] <virtuelv>
heh, Bizarro-HTML it truly is

[07:34:00.0000] <jdandrea>
/me is finally caught up (catches breath).

[07:35:00.0000] <krijnh>
I'm still 143 mails behind :/

[07:36:00.0000] <zcorpan_>
ah! html60 has a name

[07:36:01.0000] <krijnh>
Dmitry Turin ?

[07:36:02.0000] <zcorpan_>
yeah

[07:37:00.0000] <met_>
any way how subscribe for receiving html-wg mails and not be html-wg member?

[07:37:01.0000] <zcorpan_>
met_: don't think so

[07:37:02.0000] <krijnh>
Perhaps only via RSS

[07:38:00.0000] <met_>
krijnh, rss? which url?

[07:38:01.0000] <krijnh>
http://lists.w3.org/Archives/Public/public-html/ has a feed

[07:38:02.0000] <krijnh>
And use RssFwd to mail it to you :)

[07:38:03.0000] <Philip`>
That seems to not have the actual content of the messages, so you've still got to go to the archive to read them

[07:38:04.0000] <met_>
/me will try, thanks

[07:39:00.0000] <krijnh>
Ah

[07:39:01.0000] <krijnh>
Then it doesn't make sense to use it

[07:39:02.0000] <met_>
no problem with the archive i want to see whic message i read and which not 8-)

[07:41:00.0000] <Philip`>
The real problem with versioning is that it assumes there will be another version of HTML after HTML5 - but that cannot possibly happen, because Dmitry has already taken the name HTML6 and it would be silly to jump straight from HTML5 to HTML7 and so development of HTML will have to stop

[07:45:00.0000] <Lachy>
lol

[07:57:00.0000] <Philip`>
[Correction from some time ago: in Python 2.5 you say "from __future__ import with_statement" (not "with") to get the "with" keyword - http://docs.python.org/whatsnew/pep-343.html]

[07:57:01.0000] <jdandrea>
Philip `: Thx!

[07:59:00.0000] <Philip`>
[http://www.python.org/dev/peps/pep-0236/ for more details on __future__ in case anyone's overly interested]

[07:59:01.0000] <zcorpan_>
Hixie: perhaps the <dialog> example in the spec should have end tags... with the current parsing spec the last DD is left open

[08:01:00.0000] <zcorpan_>
or the parsing spec is wrong? all browsers except opera close the DD with "<foo><dd></foo>X"

[08:02:00.0000] <Lachy>
zcorpan_, anne and I discussed that issue in here earlier today.  check the logs to see what was said about it

[08:08:00.0000] <zcorpan_>
Lachy: ok, read it. conclusion was that the bug report is invalid given the current spec?

[08:09:00.0000] <Lachy>
I think so, but I think it might be a bug in the spec

[08:09:01.0000] <zcorpan_>
yeah

[08:10:00.0000] <zcorpan_>
/me mails the list

[08:11:00.0000] <Lachy>
Nooo!!! I've got enough mail to read already ;-)

[08:13:00.0000] <zcorpan_>
heh

[08:13:01.0000] <Lachy>
I got through about 100 today, only 202 left on public-html

[08:14:00.0000] <Lachy>
and 252 on whatwg

[08:15:00.0000] <jdandrea>
It's a bear. I have to skim them just to get oriented, then go back and really read each one carefully (esp. the long ones). It's not casual reading by any means.

[09:06:00.0000] <Dashiva>
zcorpan_: Did you forget a message in your latest mail?

[09:06:01.0000] <zcorpan_>
Dashiva: ?

[09:07:00.0000] <Dashiva>
The mail I got for "Parsing: should <foo><dd></foo> close the DD?" didn't have any content after headers. Wondering if it's my mail server acting up, or if it was empty

[09:08:00.0000] <zcorpan_>
http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-April/010968.html

[09:09:00.0000] <Dashiva>
ok, thanks

[10:45:00.0000] <zcorpan_>
http://xhtml.se/2007/04/20/im-not-arguing-im-just-making-fun-of-them/

[10:53:00.0000] <krijnh>
Nice huh? Public logs :)

[10:53:01.0000] <zcorpan_>
yeah

[10:56:00.0000] <Philip`>
Public logs with links that show up when people check Referers, too :-)

[10:57:00.0000] <krijnh>
Indeed

[10:57:01.0000] <krijnh>
Now if only they'd link back

[10:57:02.0000] <krijnh>
That'd be good for my pagerank :p

[11:00:00.0000] <krijnh>
Probably fair to remove rel="nofollow" on my links then

[11:02:00.0000] <Philip`>
Hmm, SYSTRAN provides interesting translations for parts of the original article - '"X" wild duck "HTML" sitting in a tree', '[...]HTML5 seems [...] to create the the entire wife solving situation [...]', etc

[11:03:00.0000] <hasather>
Philip`: no, no, that's correct. It actually said that in Swedish ;)

[11:05:00.0000] <Philip`>
Swedish seems to have been slightly neglected by automatic translation services, compared to things like French and German :-(

[11:05:01.0000] <krijnh>
Anybody here using Opera 9.2 ?

[11:05:02.0000] <hasather>
krijnh: yeah

[11:06:00.0000] <Philip`>
krijnh: Me too

[11:06:01.0000] <krijnh>
Why isn't "double click to flag/unflag as important" working anymore..

[11:06:02.0000] <krijnh>
It did in 9.1

[11:06:03.0000] <Philip`>
It works if I double-click on the line, but not if I double click on the textual part of it

[11:06:04.0000] <krijnh>
It even works in IE7 :(

[11:07:00.0000] <Philip`>
(because that just selects the text)

[11:07:01.0000] <krijnh>
Ah

[11:07:02.0000] <hasather>
krijnh: also, I get a JS error on the log pages

[11:07:03.0000] <hasather>
might be related

[11:07:04.0000] <krijnh>
A JS error?

[11:07:05.0000] <krijnh>
Or :target ?

[11:08:00.0000] <gsnedders>
hasather: what is the entire wife solving solution, then? :P

[11:08:01.0000] <hasather>
krijnh: "message: Statement on line 2634: Could not convert undefined or null to object"

[11:08:02.0000] <krijnh>
Interesting

[11:08:03.0000] <krijnh>
Line 2634 does not seem to exist though

[11:09:00.0000] <hasather>
gsnedders: no idea

[11:10:00.0000] <hasather>
krijnh: ehum, sorry about that. It was in my of my UserJS files :D

[11:11:00.0000] <krijnh>
If I needed 2634 lines to make this work I'd suck even more than I do already ;)

[11:11:01.0000] <hasather>
hehe

[11:12:00.0000] <krijnh>
Ow well, nobody's using that feature anyway

[11:14:00.0000] <krijnh>
http://www.google.co.uk/search?hl=en&q=stream+read+error+oneview&btnG=Google+Search&meta= :P

[11:15:00.0000] <hasather>
krijnh: wow, congrats man, top result for that

[11:15:01.0000] <krijnh>
Tee hee

[11:15:02.0000] <krijnh>
One hooray for social software

[11:16:00.0000] <krijnh>
Err, make that user generated content

[11:16:01.0000] <zcorpan_>
html5 spec down, two more to go

[12:17:00.0000] <jdandrea>
Reality check Q (read: "I could be reading this incorrectly.") In section 3.2.3.6, is the "(e.g. no space characters)" at odds with major steps 4 and 6?

[12:18:00.0000] <jdandrea>
Put another way, is "1, 2, 3, 4" a valid list of integers or must it be "1,2,3,4"?

[12:19:00.0000] <Philip`>
As I read it, that's the difference between author conformance and UA conformance

[12:20:00.0000] <Philip`>
Authors must write that list as "1,2,3,4", but UAs must parse the invalid string "1, 2, 3, 4" in the same way

[12:21:00.0000] <jdandrea>
Ahh, I see - author it the former way, but forgive the spaces (or a leading comma) when parsing.

[12:25:00.0000] <Philip`>
Excepting any mistakes I've made, ",,,01,2.5--,3 , 4!" would be parsed the same way too

[12:27:00.0000] <jdandrea>
nice

[12:27:01.0000] <jdandrea>
/me is learning a good deal by fine-toothed-combing through this

[12:29:00.0000] <Philip`>
This is the kind of place where it'd be nice to have an explicit guide for authors about what they must / must not do - otherwise you have to put on a spec-reading hat and read very carefully, which most people won't do

[12:30:00.0000] <kingryan>
Philip`: we'll have one eventually.

[12:31:00.0000] <jdandrea>
True. Hmm ... perhaps I'll be able to contribute to that end.

[12:31:01.0000] <Philip`>
That will be good - I guess the main problem is just finding someone to write it

[12:31:02.0000] <jdandrea>
:)

[12:32:00.0000] <Philip`>
jdandrea: That would be good :-)

[12:32:01.0000] <jdandrea>
I want to work through it a few more times. Heh. Though I will say this spec is more, how shall I say, literate (?) than others I've attempted to read!

[12:32:02.0000] <jdandrea>
So - an author guide that describes the must and must nots.

[12:33:00.0000] <jdandrea>
"So You Want To Write Valid X/HTML 5"

[12:35:00.0000] <Philip`>
I think it would be useful to have something written like a spec or a reference manual (rather than a tutorial), perhaps based on the same structure as the full spec, but only including the bits that are relevant for conforming documents

[12:35:01.0000] <zcorpan_>
people asked for a view of the spec that only had the information that applied to authors, i think

[12:35:02.0000] <Philip`>
Could we add a new HTML element for it? <relevance who="authors">

[12:35:03.0000] <zcorpan_>
it would be nice but it's hard to do programmatically as it looks now

[12:35:04.0000] <zcorpan_>
a class would be sufficient

[12:35:05.0000] <Philip`>
That's no fun :-(

[12:35:06.0000] <krijnh>
How about that <switch> element? :)

[12:36:00.0000] <kingryan>
Philip`: such a reference would probably be most useful if it focused on just one of the serializations (just html5, or just xhtml5)

[12:36:01.0000] <zcorpan_>
kingryan: why?

[12:36:02.0000] <zcorpan_>
i mean, the difference would be that the #writing section could be skipped in case you write xhtml5

[12:37:00.0000] <zcorpan_>
the rest applies to both

[12:37:01.0000] <kingryan>
zcorpan_: http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-April/010957.html

[12:38:00.0000] <kingryan>
zcorpan_:  because it's hard to track down the specific things that authors need

[12:38:01.0000] <zcorpan_>
ok

[12:38:02.0000] <kingryan>
b/c part of the contstraints are in the language and part are in the serialization

[12:38:03.0000] <kingryan>
if you collapse the 2 for a particular serialization, it'd help authors alot

[12:39:00.0000] <Philip`>
Hmm, are there any preferences on whether to use/promote <!DOCTYPE HTML> or <!DOCTYPE html> or <!doctype html> etc? Whenever I'm writing a page, I always waste time worrying about which one I should use...

[12:39:01.0000] <zcorpan_>
Philip`: don't worry about it

[12:39:02.0000] <zcorpan_>
it's case insensitive, just like tags

[12:39:03.0000] <jdandrea>
I tend to just go with the latter. YMMV.

[12:40:00.0000] <zcorpan_>
/me doesn't have a caps lock key :)

[12:40:01.0000] <Philip`>
zcorpan_: But I can't make myself not worry about it :-)

[12:40:02.0000] <zcorpan_>
Philip`: then <!doctype html> ;)

[12:40:03.0000] <jdandrea>
lol

[12:40:04.0000] <jdandrea>
<!Doctype Html>

[12:41:00.0000] <Philip`>
I don't worry about tags/quotes because I always do <lowercase thing="quoted"> and anything else looks kind of wrong, but I haven't had time to settle on a spelling for the doctype

[12:41:01.0000] <zcorpan_>
<!dOcTyPe HtMl>

[12:42:00.0000] <Philip`>
(I worry about single vs double quotes every time I write JavaScript too)

[12:42:01.0000] <Philip`>
(Too many choices :-( )

[12:42:02.0000] <zcorpan_>
in existing documents, if i am to convert the doctype to the html5 doctype, i usually preserve the previous case

[12:42:03.0000] <zcorpan_>
Philip`: you must like python

[12:43:00.0000] <krijnh>
Philip`: Me too :|

[12:44:00.0000] <krijnh>
But perhaps that's because it does matter in php

[12:44:01.0000] <zcorpan_>
well. it's like /> vs >. don't waste your time thinking about it :)

[12:44:02.0000] <krijnh>
With /> it saves a space and a slash

[12:45:00.0000] <krijnh>
In php it saves some CPU

[12:45:01.0000] <zcorpan_>
yeah, but if you have a template that has />, would you spend time converting it to >?

[12:45:02.0000] <Dashiva>
Philip`: Always use ' unless the string contains ' and no ". Simple.

[12:45:03.0000] <krijnh>
zcorpan_: I probably would, yes

[12:45:04.0000] <zcorpan_>
i used to as well. now i wouldn't

[12:46:00.0000] <zcorpan_>
you have to learn to ignore things that are irrelevant :)

[12:46:01.0000] <krijnh>
Especially if it's a template for my cms

[12:46:02.0000] <krijnh>
Which breaks on xhtml ;p

[12:46:03.0000] <jdandrea>
"Old habits ..."

[12:46:04.0000] <zcorpan_>
krijnh: then you have a bug in your cms... :P

[12:47:00.0000] <krijnh>
Yeah, irrelevant ones

[12:47:01.0000] <Dashiva>
Old habits die hard, quirks mode lives forever

[12:47:02.0000] <jdandrea>
:)

[12:47:03.0000] <Philip`>
/> vs > is easy - the spec already has a "should" conformance criteria for the latter case

[12:47:04.0000] <zcorpan_>
Philip`: only for atheists :)

[12:48:00.0000] <Dashiva>
I prefer 'godless heathen' myself

[13:26:00.0000] <zcorpan_>
mmm... tinned tuna

[13:30:00.0000] <zcorpan_>
/me has to get used to eating 6 times a day again

[13:40:00.0000] <krijnh>
zcorpan_: Working that hard?

[13:41:00.0000] <zcorpan_>
at the gym, yes

[13:41:01.0000] <krijnh>
Ah, right

[13:41:02.0000] <Hixie>
is spartanicus here?

[13:48:00.0000] <zcorpan_>
wow, firefox does the same wrt <iframe id/name> as it does with <map id/name>

[13:55:00.0000] <zcorpan_>
@_@ ... opera's behaviour with <a target> freaked me out there

[13:57:00.0000] <zcorpan_>
if i had <object name=foo data=data:text/plain,x></object> then <a href=foo.txt target=foo>test</a> opened in the object (no surprise). then i removed the data attribute and suddenly the link opened in another object in another tab, that was opened independently from the one where the link was found!

[13:58:00.0000] <zcorpan_>
if i closed that tab, the link would open in a new window

[13:58:01.0000] <zcorpan_>
(or well, tab)

[14:01:00.0000] <zcorpan_>
nevertheless <a target> only works in <object> in opera, it seems, so a more pragmatic way to interop would be to say that <a target> doesn't work with <object> at all (like in ie, firefox)

[14:23:00.0000] <zcorpan_>
hm. <map> is an inline element in html4

[14:24:00.0000] <zcorpan_>
block in html5

[14:37:00.0000] <zcorpan_>
/me uploads test cases without pass conditions... i.e. demos

[14:40:00.0000] <zcorpan>
...or not, can't connect with the server :|

[14:45:00.0000] <zcorpan>
conclusion of my tests anyway is that <iframe name> and <map name> should be conforming (name perhaps required), but UAs must support both id and name for both. <object> shouldn't work with <a target> at all

[14:45:01.0000] <zcorpan>
i'll send this to the list as soon as i can upload my tests

[14:52:00.0000] <met_>
/me examined source of http://www.microsoft.com/silverlight/asp/default.aspx it's html but inside it is <?xml ?> declaration, awful

[14:56:00.0000] <krijnh>
I like how they presented IE7 with a nice stylesheet with hacks

[14:57:00.0000] <krijnh>
IE6 that is

[15:05:00.0000] <Philip`>
I like how the "cross-browser, cross-platform plug-in" doesn't support either the browser or the platform that I'm currently using

[15:14:00.0000] <Philip`>
(Actually, it looks like it wouldn't support either even if I rebooted into Windows (2000), though I'm unsure whether that's support in the "we will provide official support" sense or the "we will not refuse to install" sense (though my experience is that the latter is most often the case for Microsoft software...))

[15:42:00.0000] <zcorpan>
test

[15:58:00.0000] <ajnewbold>
it worked


2007-04-21
[02:02:00.0000] <Hixie>
maybe the solution to <switch> is simply to have a "hidden" attribute

[02:02:01.0000] <Hixie>
and some API to toggle one element off and another one on

[02:03:00.0000] <Hixie>
or just require people to do  loginform.hidden = true; game.hidden = false;

[02:03:01.0000] <Hixie>
certainly would be simple

[02:05:00.0000] <krijnh>
Hixie: A global hidden attribute?

[02:05:01.0000] <Hixie>
yeah

[02:05:02.0000] <hsivonen>
Hixie: would it in terms of CSS toggle display: none; instead of visibility: hidden;?

[02:05:03.0000] <Hixie>
i'm just trying to work out the semantic part, i haven't really thought about the rendering

[02:06:00.0000] <Hixie>
i guess from a rendering perspective you'd have [hidden] { display: none; }

[02:06:01.0000] <krijnh>
<foo hidden="false">

[02:06:02.0000] <hsivonen>
I wonder if the name confuses authors who know about CSS "hidden"

[02:07:00.0000] <krijnh>
Prototype JS uses hidden() as well

[02:07:01.0000] <krijnh>
foo.hidden() => boolean

[02:07:02.0000] <aroben>
Hixie: might be easier to understand .visible instead of .hidden, since it's positive

[02:08:00.0000] <Hixie>
we need something that defaults to visible

[02:08:01.0000] <krijnh>
With .visible I'd think about visibility: hidden|visible

[02:08:02.0000] <Hixie>
so the attribute, if it's an attribute, has to hide it

[02:08:03.0000] <hsivonen>
invisible

[02:08:04.0000] <krijnh>
I still don't get why that should be in HTML..

[02:08:05.0000] <Hixie>
invisible could work, though i'd like a name that more obviously means the content is irrelevant... maybe irrelevant=""

[02:09:00.0000] <Hixie>
krijnh: imagine you're doing a web based game

[02:09:01.0000] <Hixie>
krijnh: and you have a login page, then a game page, then a game over page with the high score

[02:09:02.0000] <Hixie>
krijnh: and these are all just <section>s in a single HTML page

[02:09:03.0000] <Hixie>
krijnh: the game page is semantically irrelevant until the login page has been used to log in

[02:10:00.0000] <Hixie>
krijnh: and then the login page is semantically irrelevant once the game has started

[02:10:01.0000] <krijnh>
Yeah, I get that part

[02:10:02.0000] <Hixie>
krijnh: even without CSS, you'd still want those parts of the page to not be shown

[02:10:03.0000] <Hixie>
krijnh: that is, their presence is not stylistic

[02:10:04.0000] <Hixie>
krijnh: it's semantic

[02:11:00.0000] <krijnh>
For backwards compatibility it should be three separate pages anyway,

[02:11:01.0000] <Hixie>
well for now, sure, but eventually once html5 has a large install base, that won't be a problem

[02:12:00.0000] <Hixie>
and in the meantime you can use it if you put something in your stylesheet or script

[02:12:01.0000] <krijnh>
So you could have a website in one big file, and swap hidden sections while you navigate through it

[02:13:00.0000] <krijnh>
Or is it only for a page where you have to 'start' at one section

[02:14:00.0000] <Hixie>
krijnh: it's for things where it would make no sense to show everything at once

[02:15:00.0000] <Hixie>
krijnh: where the other thing would actually be completely irrelevant, unusable, and nonsensical if shown

[02:15:01.0000] <Hixie>
krijnh: if it would make sense to have the two things opened in two tabs or two windows at once, then this feature wouldn't apply

[02:16:00.0000] <Hixie>
e.g. it wouldn't be appropriate to use this for tabs, or to split a big page into multiple panels, etc

[02:16:01.0000] <krijnh>
So what happens when you make multiple sections visible at once?

[02:17:00.0000] <krijnh>
If it's really a switch thing, only one can be 'visible', right?

[02:17:01.0000] <Hixie>
how do you mean?

[02:17:02.0000] <krijnh>
Well, say you have 3 sections, for that game

[02:17:03.0000] <krijnh>
Only one is relevant

[02:18:00.0000] <krijnh>
Now you switch to a different section

[02:18:01.0000] <krijnh>
foo1.hidden = true; foo2.hidden = false

[02:18:02.0000] <krijnh>
Or something

[02:18:03.0000] <krijnh>
What if you don't hide foo1

[02:18:04.0000] <krijnh>
Then it's no switch

[02:19:00.0000] <peepo>
Is there an accessibility bod contributing to this thread?

[02:19:01.0000] <krijnh>
So those two statements should be just one, relevantRightNow = foo2

[02:19:02.0000] <Hixie>
krijnh: well, you could imagine cases where something might become relevant but otherwise might not be, independent of other things, e.g. a notification "window"

[02:19:03.0000] <krijnh>
Hixie: If you could make all sections visible at once, I don't get why they couldn't be separate pages

[02:20:00.0000] <peepo>
http://ua-games.gr/game-over

[02:20:01.0000] <peepo>
We have just released "Game Over!", the world's first universally  

[02:20:02.0000] <peepo>
inaccessible game!

[02:20:03.0000] <peepo>
Game Over! aims to provide game developers with a first-hand  

[02:21:00.0000] <peepo>
(frustrating) experience of how it feels interacting with a game that  

[02:21:01.0000] <peepo>
is not accessible due to the fact that important accessibility design  

[02:21:02.0000] <peepo>
rules were not considered or applied.

[02:21:03.0000] <Hixie>
krijnh: imagine your page is a status dashboard for some network service, and there are three servers, and each one can sent a status message at any time

[02:21:04.0000] <peepo>
have to run..later

[02:21:05.0000] <Hixie>
krijnh: the status window for a server is irrelevant/nonsensical/unusable unless the server is sending a message

[02:22:00.0000] <Hixie>
krijnh: but all three servers might send a message at once, in which case all three windows are relevant/usable

[02:22:01.0000] <Hixie>
krijnh: but the messages aren't "pages" as far as the user is concerned

[02:22:02.0000] <Hixie>
(though i guess they might well be "pages" in the REST sense)

[02:23:00.0000] <krijnh>
Hmm

[02:34:00.0000] <krijnh>
Hixie: Sorry, I was thinking about a <switch><case>..</case><case>..</case></switch> kinda thing :)

[02:34:01.0000] <Hixie>
yeah that was my original thought too

[02:34:02.0000] <Hixie>
i think an attribute is more flexible

[02:35:00.0000] <Hixie>
but the problem is people will use it for tabs, and for multi-page pages that really should just be a big page with some styles to show/hide sections

[02:35:01.0000] <Hixie>
not sure how to design the feature so it's not misused

[02:35:02.0000] <krijnh>
I think I would..

[02:36:00.0000] <krijnh>
Tabs you don't want to see aren't relevant, right? :)

[02:46:00.0000] <krijnh>
Hixie: Would children of irrelevant elements show up in the DOM?

[02:47:00.0000] <Hixie>
tabs are just a way of re-rendering a group of <fieldset>s so that only one of them is rendered -- the other tabs are just as relevant, they're just not shown

[02:47:01.0000] <Hixie>
yes, the DOM would be unaffected

[03:58:00.0000] <Hixie>
/me reads the xhtml2 logs and flags some fun lines

[04:01:00.0000] <krijnh>
From yesterday?

[04:01:01.0000] <Hixie>
the latest logs, whatever that is

[04:39:00.0000] <krijnh>
zcorpan: I included a current timestamp on the logs

[04:40:00.0000] <zcorpan>
krijnh: great

[04:55:00.0000] <zcorpan>
i have serious trouble telling relatives and friends about what i am to work with at opera :(

[04:56:00.0000] <zcorpan>
they don't know what test cases are, they don't know what html5 is, some don't even know what a browser is

[04:56:01.0000] <zcorpan>
if i go "i'll write test cases for html5" they usually go @_@

[05:01:00.0000] <krijnh>
What's wrong with people going @_@ ? :)

[05:01:01.0000] <zcorpan>
they won't stop bugging me until they understand

[05:02:00.0000] <krijnh>
I'm sure you're able to explain it :)

[05:02:01.0000] <zcorpan>
yeah, but it takes effort... it's not like "i work at the post office"

[05:06:00.0000] <zcorpan>
it goes like "you know when you browse the web, you see web pages?", "oh yep", "the browser receives html code so that it can render the page", "oh yep", "for that to work the browser vendor has to implement html", "what's implement?", "programming the browser in a way so that it can handle the code", "oh yep", "so *i* will create test cases for html so that the programmers get things right", "right", "comprendo?", "no"

[05:07:00.0000] <krijnh>
"Okay, in that case, I work at the post office"

[05:07:01.0000] <zcorpan>
tempting

[05:08:00.0000] <krijnh>
"Ah, so you're a web designer?"

[05:08:01.0000] <krijnh>
That's what I get most of the time

[05:08:02.0000] <zcorpan>
yeah i got that too

[05:08:03.0000] <zcorpan>
although i am a web designer, it's not what i'll do at opera

[05:08:04.0000] <krijnh>
"No, I try to make sites for blind people, although I don't know anybody who's blind"

[05:09:00.0000] <krijnh>
Btw

[05:09:01.0000] <krijnh>
"The a element must not be empty." <-- does that mean <a name="foo"></a> is non conforming?

[05:09:02.0000] <zcorpan>
or well, i don't design, just usnig the term

[05:09:03.0000] <zcorpan>
krijnh: correct

[05:09:04.0000] <krijnh>
Hmm

[05:09:05.0000] <zcorpan>
<a name> isn't either

[05:09:06.0000] <zcorpan>
in the first place

[05:10:00.0000] <krijnh>
But it's used a lot

[05:10:01.0000] <zcorpan>
so?

[05:10:02.0000] <zcorpan>
this has nothing to do with ua conformance

[05:11:00.0000] <webben>
um... why has <a name="foo"></a> been made non-conforming?\

[05:11:01.0000] <webben>
that's a perfectly good way of making anchors that work cross-browser

[05:12:00.0000] <zcorpan>
id="" has worked too for the past 10 years afaict

[05:12:01.0000] <zcorpan>
only not in nn4

[05:12:02.0000] <krijnh>
It didn't in Opera a while ago

[05:12:03.0000] <zcorpan>
what was that, opera4?

[05:12:04.0000] <webben>
zcorpan: also, not in JAWS earlier than 4.51

[05:12:05.0000] <krijnh>
zcorpan: No, 7

[05:13:00.0000] <zcorpan>
krijnh: you're kidding, right?

[05:13:01.0000] <krijnh>
No, wait

[05:13:02.0000] <krijnh>
http://krijnhoetmer.nl/stuff/html/jump-inside-div/ didn't work in Opera once, and I'm not using Opera since version 4

[05:14:00.0000] <krijnh>
But that probably something different :)

[05:14:01.0000] <krijnh>
*certainly

[05:14:02.0000] <zcorpan>
webben: right... you can still use html4 if you need to support legacy uas :)

[05:16:00.0000] <zcorpan>
not that i'm opposed to <a name> though

[05:16:01.0000] <zcorpan>
if you think it should be allowed then bug the list :)

[05:16:02.0000] <krijnh>
I don't think it should

[05:16:03.0000] <webben>
i'm just wondering why it was suddenly disallowed

[05:16:04.0000] <krijnh>
It's probably written somewhere

[05:17:00.0000] <webben>
it just declares perfectly good existing content non-conformant AFAICT

[05:17:01.0000] <krijnh>
That some documents use <a name=foo></a>

[05:17:02.0000] <zcorpan>
webben: it's disallowed in xhtml1.1 and deprecated in xhtml1.0, probably part of the reason

[05:17:03.0000] <krijnh>
And what to do with those

[05:17:04.0000] <krijnh>
webben: So did removing the table summary attribute, for instance

[05:21:00.0000] <zcorpan>
wow, table summary probably shouldn't have been removed

[05:22:00.0000] <krijnh>
zcorpan: http://krijnhoetmer.nl/irc-logs/whatwg/20070416#l-118

[05:23:00.0000] <hsivonen>
zcorpan: is there AT that makes useful use of the table summary? are sites using it?

[05:23:01.0000] <krijnh>
Are sites using it?

[05:23:02.0000] <krijnh>
Yes

[05:23:03.0000] <zcorpan>
hsivonen: at least jaws uses it

[05:24:00.0000] <krijnh>
What is AT?

[05:24:01.0000] <zcorpan>
don't have data about how common it is in the wild but i have seen it being used and i've used it myself

[05:24:02.0000] <krijnh>
Assistive Technology ?

[05:24:03.0000] <zcorpan>
krijnh: yes

[05:24:04.0000] <krijnh>
K

[05:24:05.0000] <hsivonen>
I wonder if Hixie had a good reason for removing it or whether it is something that hasn't been added yet

[05:24:06.0000] <krijnh>
Could be AnyThing :]

[05:25:00.0000] <krijnh>
I'm also using it

[05:26:00.0000] <zcorpan>
getting information from a table in non-visual media is not a simple task, so getting a summary is useful so you can skip the table but still get the essentials

[05:26:01.0000] <zcorpan>
/me used a summary for http://simon.html5.org/articles/mobile-results

[05:28:00.0000] <zcorpan>
though, if summary is included in the spec, it should have guidelines about what it's for... "this table has x columns and y rows" is not a useful summary because that information is implied

[05:29:00.0000] <krijnh>
Shh

[05:29:01.0000] <krijnh>
;)

[05:29:02.0000] <zcorpan>
"this is a layout table" also isn't useful

[05:29:03.0000] <zcorpan>
:)

[05:30:00.0000] <krijnh>
I have 5 customers! I cannot and will not break pages for them by removing the auto generated summary ;]

[05:30:01.0000] <krijnh>
So, I need versioning.

[05:30:02.0000] <zcorpan>
lol

[05:31:00.0000] <krijnh>
Hmm, even WCAG (1.0!) says a summary of the complexity of the table can be handy..

[05:32:00.0000] <zcorpan>
krijnh: yeah, also see html4 spec

[05:32:01.0000] <krijnh>
Yep

[05:32:02.0000] <krijnh>
Summary of purpose and structure

[05:32:03.0000] <krijnh>
Indeed x cols and y rows is implied

[05:32:04.0000] <krijnh>
But that wouldn't count as a complex table

[05:33:00.0000] <krijnh>
Hmm, is it implied btw?

[05:33:01.0000] <krijnh>
hsivonen: your table integrity checker, can that tell the number of cols and rows in a table, without walking through the entire table?

[05:33:02.0000] <zcorpan>
if you have a more complex table, then that's even more reason to give a useful summary so that blind people don't have to walk through the entire table to get an idea about it

[05:34:00.0000] <zcorpan>
krijnh: jaws can state x rows and y columns for tables

[05:34:01.0000] <krijnh>
Grmbl

[05:35:00.0000] <krijnh>
zcorpan: I think the summary for your mobile test results is handy for visual UAs as well btw

[05:35:01.0000] <zcorpan>
krijnh: fair enough, i'll move it out of the table then :)

[05:35:02.0000] <krijnh>
And keep it in the summary attribute?

[05:36:00.0000] <zcorpan>
<p>summary<table> (no summary="")

[05:36:01.0000] <krijnh>
Okay, so now we know why it's dropped ;p

[05:36:02.0000] <krijnh>
<p>summary<details><table> ;)

[05:37:00.0000] <krijnh>
<details><legend>summary</legend><table>.. even

[05:37:01.0000] <zcorpan>
that's not what details is for

[05:37:02.0000] <krijnh>
I know :)

[05:38:00.0000] <zcorpan>
summary="" is information that would be redudant for visual media

[05:39:00.0000] <zcorpan>
hm. i can't connect to any of my servers via ftp anymore. wonder why

[05:39:01.0000] <krijnh>
I think that a summary is only redundant if it's a very simple table

[05:39:02.0000] <krijnh>
If it's complex, a summary could be useful for everybody

[05:40:00.0000] <zcorpan>
ok

[05:40:01.0000] <krijnh>
I think

[05:40:02.0000] <krijnh>
Stuff describing the table could be in a caption

[05:41:00.0000] <zcorpan>
still, a summary for "simple tables" could be useful for blind people (because even a simple table is non-trivial to walk through with a screen reader)

[06:15:00.0000] <webben_>
hsivonen and zcorpan: Window-Eyes also supports summary and has done for some time: http://www.w3.org/WAI/UA/implementation/eval_win_wineyes411.html

[06:19:00.0000] <webben_>
It's also worth checking whether tools to distinguish table layouts from data tables use summary as a hint.

[06:19:01.0000] <webben_>
My guess is they would, but it rather depends on the prevalence of summary abuse.

[06:20:00.0000] <zcorpan>
webben_: yeah, they do, question is how

[06:20:01.0000] <zcorpan>
and do we need to spec it

[06:20:02.0000] <webben_>
zcorpan: I thought HTML5 doesn't spec UA behaviour.

[06:20:03.0000] <webben_>
we could suggest some behaviours for it

[06:21:00.0000] <webben_>
I think WCAG probably has a testcase for it

[06:21:01.0000] <webben_>
one could use that as a base.

[06:21:02.0000] <zcorpan>
i didn't suggest we spec behaviour

[06:21:03.0000] <zcorpan>
only an algorithm to determinate whether a table is a layout table or not

[06:22:00.0000] <webben_>
Ah ... interesting.

[06:22:01.0000] <webben_>
best ask Aaron Leventhal, as he's looking into integrating such algorithms into Mozilla.

[06:23:00.0000] <zcorpan>
not sure if it's needed, but it would aid competition in the screen reader space i'd presume, given the web uses layout tables and a screen reader that interprets them all as data tables isn't very useful

[06:23:01.0000] <webben_>
zcorpan: True.

[06:24:00.0000] <zcorpan>
also, conformance checkers could use that algorithm to flag layout tables!

[06:25:00.0000] <zcorpan>
not sure if that's a good thing

[06:25:01.0000] <zcorpan>
would probably only lead to authors working around the errors

[06:25:02.0000] <zcorpan>
making the algorithm useless

[06:25:03.0000] <webben_>
Why would an author bother to do that?

[06:26:00.0000] <zcorpan>
are you serious? i see *lots* of authors going through hoops to make their pages validate, even if it involves the most redicilous workarounds

[06:26:01.0000] <zcorpan>
and makes the pages worse

[06:26:02.0000] <SpookyET>
Hello.

[06:27:00.0000] <webben_>
hmm ... that would seem to militate against any sort of conformance checking equally

[06:27:01.0000] <webben_>
not specific to distinguishing layout/data tables.

[06:27:02.0000] <webben_>
also, /if/ we actually can distinguish layout from non-layout tables, would the use of tables for layout matter half as much

[06:27:03.0000] <SpookyET>
I've been playing with HTML5. It seems that Opera have implemented quite a bit of it.

[06:27:04.0000] <zcorpan>
webben_: only where working around the error is simpler than doing the right thing

[06:28:00.0000] <zcorpan>
SpookyET: yup

[06:28:01.0000] <webben_>
zcorpan: I suspect the key there is to give people an easier way of laying out pages than misusing tables.

[06:29:00.0000] <webben_>
maybe (worst-case scenario) giving them an attribute like <table type="layout">

[06:29:01.0000] <ajnewbold>
the only reason tables were ever used for layout was poor css support in the early browsers, wasn't it?

[06:29:02.0000] <SpookyET>
I'm still do not understand what is going on at W3C. Do they care more about their other members than the actual browser makers?

[06:29:03.0000] <webben_>
ajnewbold: Probably tables were being misused before CSS existed.

[06:30:00.0000] <ajnewbold>
webben_: heh, well, before CSS even existed, the layout options were severely limited :P

[06:30:01.0000] <webben_>
SpookyET: Why shouldn't they?

[06:30:02.0000] <zcorpan>
SpookyET: who are "they"?

[06:31:00.0000] <webben_>
something I've been wondering is whether you could create a sort of layout markup language to which elements in an HTML file could be mapped

[06:32:00.0000] <webben_>
perhaps with simple selectors like footer, .photo, #main

[06:32:01.0000] <webben_>
so that you could get all the ease of laying out with markup

[06:32:02.0000] <webben_>
but still separate content from presentation

[06:33:00.0000] <webben_>
in theory XSLT could be used to map between them, but in practice XSLT seems prohibitively hard to use

[06:33:01.0000] <webben_>
possibly XPath might be some use, but I think element, identifier and class selectors would be easiest

[06:35:00.0000] <webben_>
and with JS one could actually implement it in existing browsers, much as with WF2.

[06:37:00.0000] <webben_>
e.g. <view><box halign="center"><row><col select="#main" width="30%"><arrange type="grid" select=".photo"></col><col select="#sidebar"></row><row select="footer"></row></box></view>

[06:38:00.0000] <webben_>
obviously web designers might well prefer CSS, but I wonder if that would help ordinary authors.

[06:40:00.0000] <webben_>
<link rel="stylelayout" type="text/slml" media="screen"> could import it

[06:40:01.0000] <webben_>
that way mobiles could be given different layouts for instance

[06:40:02.0000] <webben_>
and old browsers wouldn't have to deal with it at all

[06:40:03.0000] <Lachy>
webben_, I think you're looking for XBL

[06:41:00.0000] <webben_>
Lachy: I thought XBL used CSS selection?

[06:41:01.0000] <Lachy>
it does

[06:41:02.0000] <Lachy>
so?

[06:41:03.0000] <webben_>
Lachy: So many authors seem to prefer markup.

[06:42:00.0000] <webben_>
It may just be that nobody has come up with sufficiently sophisticated/intelligent CSS properties

[06:42:01.0000] <Lachy>
I don't understand, let me finish reading all your messages first

[06:42:02.0000] <webben_>
e.g. .photo {auto-arrange: grid}

[06:43:00.0000] <webben_>
even that though, you still have a list of rules not a markup layout

[06:46:00.0000] <Lachy>
ok, finished.  it seems that either css3-layout or XBL <template> could solve the problems you describe

[06:47:00.0000] <webben_>
ah XBL <template> sounds interesting

[06:48:00.0000] <Lachy>
you can include whatever markup you like inside it, and style it however you want, and include content from the document in various places

[06:48:01.0000] <webben_>
that's this thing is it? http://www.w3.org/TR/sXBL/

[06:48:02.0000] <Lachy>
no, that's sXBL

[06:48:03.0000] <Lachy>
try /TR/xbl/

[06:49:00.0000] <Lachy>
sXBL is like a predecessor to XBL2

[06:52:00.0000] <webben_>
ah okay

[06:58:00.0000] <webben_>
Lachy: Looks very interesting, especially if it would be possible to concretize some of that into something ordinary authors could use, perhaps in conjunction with a JS lib. Do you know of any implementations or pseudo-implementations?

[07:03:00.0000] <hsivonen>
krijnh: to tell the number of row and columns in an HTML table, you have to examine the entire table

[07:04:00.0000] <hsivonen>
krijnh: so no, my table integrity checker cannot tell the number of rows and columns until the entire table has been seen

[07:39:00.0000] <SpookyET>
<!Doctype html> Since it does not have a version, does it assume that HTML will be backwards compatible forever?

[07:40:00.0000] <Dashiva>
That's the idea

[07:42:00.0000] <SpookyET>
It makes sense. You can clean up HTML like XHTML2, but keep the old stuff in the spec in the vintage section. That way, 50 years from now, HTML 3.2 pages can still be displayed assuming that layout engines are well designed and won't be rewritten from scratch.

[07:42:01.0000] <SpookyET>
Except, IE, of coarse.

[07:43:00.0000] <Philip`>
It also assumes that even if HTML isn't backwards compatible forever, it's up to that future HTML to add versioning - <!doctype html> doesn't prevent HTML7 from using <!doctype html public "7"> if it turns out that we're wrong now

[07:43:01.0000] <Philip`>
(Er, maybe that has to be "system" instead?)

[07:43:02.0000] <SpookyET>
what's with the public and system words?

[07:44:00.0000] <zcorpan>
Philip`: only if you want to introduce doctype sniffing to xml...

[07:44:01.0000] <SpookyET>
what's wrong with <!doctype html 7> or the xml style attribute based prolog

[07:44:02.0000] <Philip`>
Oops, I keep forgetting about XML

[07:44:03.0000] <zcorpan>
SpookyET: it triggers quirks mode in gecko and webkit

[07:44:04.0000] <zcorpan>
SpookyET: plus it's not well-formed xml

[07:45:00.0000] <zcorpan>
(should you care about sniffing in xml)

[07:45:01.0000] <Philip`>
There's still nothing to stop (X)HTML7 from doing <html version="7"> if it needs to, but people believe we shouldn't encourage that path by doing version="5" now

[07:45:02.0000] <Philip`>
(I think)

[07:46:00.0000] <SpookyET>
if you don't have DTDs, the <!doctype syntax> is a relic. <html version=5> makes more sense.

[07:46:01.0000] <Philip`>
(With public vs system: if I remember correctly, IE treated one as quirks and the other as standards, but I could be wrong about that)

[07:48:00.0000] <zcorpan>
Philip`: <!doctype html public "7"> and <!doctype html system "7"> both trigger standards mode in ie7

[07:48:01.0000] <Philip`>
Hmm, maybe I'm wrong

[07:48:02.0000] <Philip`>
or maybe it was only IE6, but I didn't think they'd changed that at all

[07:49:00.0000] <zcorpan>
same in ie6

[07:49:01.0000] <Philip`>
Okay, I'm wrong :-)

[07:54:00.0000] <zcorpan>
it's funny how the css validator warns about all various things. then students have to pass validation without warnings. so they have to add redudant and bogus rules to their style sheets

[07:55:00.0000] <zcorpan>
like background:inherit when the parent is transparent

[07:56:00.0000] <zcorpan>
(it doesn't do anything, but it makes the warnings go away)

[07:57:00.0000] <SpookyET>
I despise those warnings.

[07:58:00.0000] <SpookyET>
Sometimes, adding background or foreground colours to fix those warnings breaks the style of your page.

[07:58:01.0000] <SpookyET>
You can't always fix them. transparent or inherit don't always work

[07:58:02.0000] <zcorpan>
transparent is the default

[07:58:03.0000] <zcorpan>
if the parent is transparent, then inherit == transparent

[07:59:00.0000] <SpookyET>
hmm, Sir Bray's page is invalid :-(

[08:00:00.0000] <zcorpan>
who?

[08:00:01.0000] <zcorpan>
tim?

[08:00:02.0000] <SpookyET>
yep

[08:01:00.0000] <zcorpan>
so?

[08:01:01.0000] <zcorpan>
:)

[08:02:00.0000] <SpookyET>
XHTML 1.1 as text/html with 2 validation errors

[08:03:00.0000] <SpookyET>
It proves that WHATWG are doing is important. If Sir. Bray can't have his own page coded/served according to the spec, how can the rest of us? :-)

[08:12:00.0000] <mpt>
SpookyET, when does {background: inherit} not work?

[08:13:00.0000] <mpt>
oh, wait, silly question

[08:15:00.0000] <mpt>
{background: inherit} has undesirable results when the parent has an image background

[08:16:00.0000] <zcorpan>
or possibly also a background color

[08:16:01.0000] <zcorpan>
(if the element is positioned outside its parent or overflows or something)

[08:16:02.0000] <mpt>
A background color would be fine, the repetition would be a no-op

[08:17:00.0000] <mpt>
ohhh

[08:17:01.0000] <mpt>
yik

[08:34:00.0000] <SpookyET>
mpt: When my menu lists have images in them, and it wants a background-color for the anchor element

[08:36:00.0000] <zcorpan>
SpookyET: some would add <span>s in between just to get rid of the warning

[08:51:00.0000] <SpookyET>
zcorpan: No, thank you.

[08:51:01.0000] <SpookyET>
I write clean code -- http://www.icrhealthcare.com for example

[08:52:00.0000] <zcorpan>
SpookyET: oh, i didn't suggest you should

[08:52:01.0000] <zcorpan>
personally i don't find the css validator a useful QA tool at all

[08:52:02.0000] <zcorpan>
so i don't use it

[08:53:00.0000] <zcorpan>
i'm just observing that some people *are* going through hoops just to make the warnings go away

[08:54:00.0000] <SpookyET>
I tried that once or twice. It's not worth it.

[08:54:01.0000] <SpookyET>
I'm more interested in writing boxed-in css

[08:54:02.0000] <hsivonen>
zcorpan: I'd like to have a CSS3 Paged Media plus Prince profile for the Oxygen local copy of the CSS validator

[08:55:00.0000] <SpookyET>
http://www.ksuicehockey.com/css/global-screen.css

[08:55:01.0000] <zcorpan>
hsivonen: ok

[08:56:00.0000] <hsivonen>
zcorpan: but with conservatively locked-down profiles, the tool isn't that useful to me

[08:57:00.0000] <hsivonen>
so I think it is a useful tool in principle, but it needs to grow as fast as CSS

[08:57:01.0000] <zcorpan>
indeed

[08:58:00.0000] <SpookyET>
Writing CSS that way doesn't always provide extreme redability, but it makes it very easy to embed applications in your site and referencing your css instead of fully restyling them to match your sites style

[08:59:00.0000] <zcorpan>
i find the error consoles in browsers more useful to find errors in my style sheets

[09:00:00.0000] <SpookyET>
firebug

[09:01:00.0000] <hsivonen>
zcorpan: the validating CSS editor in Oxygen is a really cool idea, but the problem is that when you have enough prince-* properties with squiggly red underline, you stop caring

[09:01:01.0000] <zcorpan>
yup

[09:04:00.0000] <SpookyET>
Oxygen has more icons in the toolbars than smallpox victim

[09:05:00.0000] <zcorpan>
does document.onmessage = function(e) {...} work? or should it be window.onmessage?

[09:06:00.0000] <SpookyET>
winodw i think, but it's better to use addEventListner

[09:06:01.0000] <zcorpan>
why?

[09:08:00.0000] <SpookyET>
You can have more than one listner, and it can't be overriden by document.onmessage = null.

[09:10:00.0000] <zcorpan>
right. so if i will only use one listener and don't care about it being overridden, window.onmessage would be fine

[09:11:00.0000] <Philip`>
For cross-document messaging? The spec says it's a bubbling event and it's dispatched on the Document object, so I assume that means document.onmessage would work fine, but I don't know if things bubble from document to window...

[09:12:00.0000] <zcorpan>
Philip`: events bubble from document to window

[09:14:00.0000] <SpookyET>
one thing that sucks is that they haven't added event.eventListnersList until DOM3

[09:14:01.0000] <SpookyET>
It was stupid to have add/remove with list in DOM2/1, whenever those 2 were added

[09:42:00.0000] <zcorpan>
/me tries to find a good swedish word for "form control"

[10:42:00.0000] <webben_>
zcorpan: What's wrong with background:transparent when the containing element has a background image? And isn't it the case that background:inherit would do something if a UA (or extension to a UA) applied a default style to (say) a microformat component?

[10:42:01.0000] <webben_>
it's not obvious to me that such values are pointless

[10:49:00.0000] <zcorpan>
webben_: don't disagree, i wasn't advocating

[10:50:00.0000] <webben_>
i thought you said such rules were redundant?

[10:51:00.0000] <webben_>
maybe I misunderstood

[10:51:01.0000] <zcorpan>
redundant at best

[10:51:02.0000] <zcorpan>
harmful at worst

[10:52:00.0000] <webben_>
that's what i mean, inherit is only redundant when there isn't a default style

[10:52:01.0000] <webben_>
but the set of default styles is unknown and expanding.

[10:52:02.0000] <zcorpan>
indeed

[14:41:00.0000] <met_>
/me found http://www.w3.org/TR/xhtml2/xhtml2-changes.html#a_changes very descriptive

[14:43:00.0000] <gsnedders>
a very informative appendix indeed :)

[14:44:00.0000] <SpookyET>
I think the W3C are smoking  their own gases. I do not know whom they represent, but I do know they do not represent web developers -- nor browser vendors for that matter.

[14:45:00.0000] <gsnedders>
the W3C has an idealistic view of the web in a perfect world.

[14:45:01.0000] <gsnedders>
this != perfect.

[14:45:02.0000] <Lachy>
that appendix will essentially just repeat the rest of the spec, since nothing is the same as XHTML 1.1

[14:46:00.0000] <met_>
/me rofl

[14:46:01.0000] <gsnedders>
apart from the namespace, possibly.

[14:47:00.0000] <zcorpan_>
xhtml2 using the xhtml1 namespace is a good way to get xhtml2 even more ignored, if that's even possible

[14:47:01.0000] <SpookyET>
You have to change your spec to reality, not reality to your spec.

[14:48:00.0000] <Lachy>
zcorpan_, they did plan to use the XHTML1 namespace.  I'm not sure if they've changed their mind about that yet

[14:48:01.0000] <zcorpan_>
/me has spent some time what reality looks like wrt target="" pointing to id="" or name=""

[14:49:00.0000] <zcorpan_>
Lachy: yeah, either way, i don't care :)

[14:50:00.0000] <zcorpan_>
not that i want xhtml2 to fail or anything, i simply don't care, i have other things to care about

[14:50:01.0000] <Lachy>
yeah, I don't care either.

[14:51:00.0000] <Lachy>
they rejected my suggestion to fix the XHTML1.1 MIME type issue too, but I'm not even going to bother objecting, since the spec is irrelevant anyway

[14:51:01.0000] <SpookyET>
Safari have said that they will NOT implement XHTML2.

[14:52:00.0000] <gsnedders>
almost everyone has :P

[14:52:01.0000] <SpookyET>
almost?

[14:53:00.0000] <Lachy>
someone will attempt to implement XHTML2, it just won't be anyone of any importance

[14:53:01.0000] <SpookyET>
KHTML?

[14:54:00.0000] <gsnedders>
nobody who is relevant whatsoever.

[14:54:01.0000] <SpookyET>
neah, KHTML is important for WebKit

[14:54:02.0000] <gsnedders>
(and calling KHTML is pushing it)

[14:55:00.0000] <zcorpan_>
i thought there was a project to port webkit to kde and merge the good stuff from both in the process

[14:55:01.0000] <gsnedders>
there was.

[14:55:02.0000] <gsnedders>
whether they replace KHTML with WebKit is undecided

[14:56:00.0000] <gsnedders>
http://dot.kde.org/1152645965/

[14:59:00.0000] <SpookyET>
calling KHTML is pushing it? that's the engine's name, isn't it? It stands for KDE HTML.

[15:00:00.0000] <gsnedders>
I mean, calling KHTML relevant

[15:00:01.0000] <gsnedders>
it has such a small marketshare. it supporting XHTML2 wouldn't make much difference (not that they plan to)

[15:00:02.0000] <SpookyET>
KHTML is relevant to WebKit.

[15:00:03.0000] <SpookyET>
WebKit is relevant to Safari.

[15:00:04.0000] <gsnedders>
not really at all.

[15:01:00.0000] <gsnedders>
WebKit is so different now

[15:01:01.0000] <gsnedders>
remember that WebKit was forked off many years ago

[15:02:00.0000] <SpookyET>
but they share patches

[15:02:01.0000] <gsnedders>
there are plenty of parts of the sourcebase which they don't at all.

[15:02:02.0000] <gsnedders>
as they are completely incompatible in plenty of parts

[15:03:00.0000] <SpookyET>
KDE 4.0 might change that.

[15:04:00.0000] <gsnedders>
might. depends on what happens with Unity.

[15:05:00.0000] <SpookyET>
I like the KDE interface. But, I hate its interface with a passion.

[15:06:00.0000] <SpookyET>
It's got a billion icons, tabs, and entries.

[15:34:00.0000] <zcorpan_>
i have 16 test cases and 2 support files to augment an email i've written that i'll send to the whatwg list. ftp doesn't seem to work for me today so i can't upload them. should i zip them and send as attachment?

[15:39:00.0000] <zcorpan_>
or perhaps someone else can upload them somewhere?

[15:39:01.0000] <hasather>
zcorpan_: sure

[15:41:00.0000] <zcorpan_>
hasather: cheers, i'll email them to you

[15:41:01.0000] <hasather>
ok

[15:43:00.0000] <zcorpan_>
sent

[15:46:00.0000] <hasather>
http://hasather.net/test/html/id-vs-name/

[15:46:01.0000] <hasather>
oh, and you can write in Swedish to me ;)

[15:47:00.0000] <zcorpan_>
@_@ right.

[15:47:01.0000] <hasather>
should I delete text.txt and empty.txt?

[15:47:02.0000] <zcorpan_>
no, put them in support/

[15:47:03.0000] <hasather>
ok

[15:48:00.0000] <hasather>
ahh, sorry, they were part of the tests of course

[15:48:01.0000] <zcorpan_>
:)

[15:53:00.0000] <zcorpan_>
there

[16:41:00.0000] <zcorpan_>
"Simple animation support: Flash makes creating animations extremely easy. Doing the same thing in anything Free (HTML, SVG, even Gtk) requires being a good code monkey. The example above required NO coding." -- http://www.advogato.org/person/company/diary.html?start=39

[16:42:00.0000] <zcorpan_>
thus, we need authoring tools for canvas or svg animations

[16:42:01.0000] <zcorpan_>
probably svg is more suitable for animations

[16:56:00.0000] <zcorpan>
http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-April/010975.html where did that leading space in the table come from?

[16:56:01.0000] <zcorpan>
/me ponders


2007-04-22
[17:12:00.0000] <Philip`>
zcorpan: "<map name> works in all except in Gecko in XHTML and WebKit in XHTML." - but the table's data for 007 seems to indicate it does work in WebKit-XHTML

[17:13:00.0000] <zcorpan>
Philip`: ah, yes, thanks

[17:23:00.0000] <zcorpan>
/me likes the "no alt = image is meaningful" idea

[17:38:00.0000] <jgraham>
http://wordsandpictures.dyndns.org/html5/parsetree.html (a HTML5 DOM viewer based on html5lib -- still very rough)

[17:39:00.0000] <jgraham>
/me is going to sleep now

[17:40:00.0000] <zcorpan>
jgraham: nice!

[17:40:01.0000] <zcorpan>
why are the elements uppercased?

[17:42:00.0000] <jgraham>
Ah, in innerHTML view they shouldn't be... I'll remove that CSS :)

[17:42:01.0000] <zcorpan>
it shouldn't be in the tree either, should it?

[17:44:00.0000] <jgraham>
I guess it makes more sense to have everything lowercase (it's also easier)

[17:44:01.0000] <zcorpan>
yeah

[17:49:00.0000] <zcorpan>
innerHTML view says "<!DOCTYPE: HTML>"

[17:52:00.0000] <zcorpan>
perhaps the textarea should be POST?

[17:53:00.0000] <zcorpan>
/me stops bugging jgraham now

[18:05:00.0000] <Lachy>
jgraham, #text nodes shouldn't appear outside of the root element.

[18:08:00.0000] <zcorpan>
Lachy: i think they should

[18:08:01.0000] <zcorpan>
(per spec that is)

[18:09:00.0000] <zcorpan>
(if the spec is wrong is another matter)

[18:09:01.0000] <Lachy>
I mean in the DOM view

[18:10:00.0000] <zcorpan>
yes?

[18:10:01.0000] <Lachy>
so the DOM spec applies

[18:10:02.0000] <Lachy>
and I'm fairly sure they shouldn't

[18:10:03.0000] <zcorpan>
the html5 parsing spec says to append text nodes to the Document object in some cases

[18:11:00.0000] <Lachy>
that's weird

[18:13:00.0000] <zcorpan>
what would you do with whitespace between the doctype and the first tag?

[18:13:01.0000] <Lachy>
drop it

[18:13:02.0000] <zcorpan>
why?

[18:13:03.0000] <Lachy>
outside of the document, it has no meaning and irrelevant

[18:14:00.0000] <Lachy>
s/document/root/

[18:14:01.0000] <Lachy>
that's what browsers do

[18:14:02.0000] <zcorpan>
i think at least firefox appends the whitespace to the HEAD instead

[18:15:00.0000] <zcorpan>
not sure about the others

[18:16:00.0000] <zcorpan>
and whitespace after </body> and </html> is appended to the body or whatever unclosed element was open in body... in firefox, and iirc

[18:19:00.0000] <zcorpan>
i don't feel strongly about it either way, but the way it's defined now preserves the "source code" after parsing and serializing (fwiw)

[18:20:00.0000] <Lachy>
firefox doesn't retain whitespace before the root

[18:23:00.0000] <zcorpan>
Lachy: compare data:text/html,<!doctype html>%0A<html> and data:text/html,<!doctype html><html> in DOM viewer (check the contents of HEAD)

[18:24:00.0000] <zcorpan>
the first has a text node in HEAD, the second doesn't

[18:25:00.0000] <Lachy>
which version of FF are you using?

[18:25:01.0000] <annevk>
morning

[18:25:02.0000] <Lachy>
hi annevk 

[18:25:03.0000] <zcorpan>
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a4pre) Gecko/20070417 Minefield/3.0a4pre

[18:25:04.0000] <Lachy>
I've only got FF2

[18:25:05.0000] <zcorpan>
got the same results last time i checked, which was probably a year ago or something

[18:25:06.0000] <Lachy>
and it doesn't show the whitespace

[18:25:07.0000] <zcorpan>
really?

[18:25:08.0000] <annevk>
testing parsing?

[18:26:00.0000] <zcorpan>
annevk: yeah

[18:26:01.0000] <Lachy>
testing how whitespace outside of the root should be handled

[18:26:02.0000] <zcorpan>
whitespace between doctype and root

[18:26:03.0000] <annevk>
it would make sense to contribute to the testsuite format already developed for parsing I think

[18:26:04.0000] <Lachy>
annevk,  see http://wordsandpictures.dyndns.org/html5/parsetree.html

[18:27:00.0000] <annevk>
fancy

[18:27:01.0000] <annevk>
what I meant was that there's already a testsuite input and output format

[18:28:00.0000] <annevk>
through some scripting that can be used for browser testsuites etc.

[18:28:01.0000] <annevk>
/me slowly fetches 208 e-mails

[18:29:00.0000] <zcorpan>
Lachy: tested in fx2 now and indeed it gets dropped

[18:30:00.0000] <Lachy>
annevk, where have you gone touring around so far? Been to the rainforest yet?

[18:31:00.0000] <Lachy>
or sand surfing?

[18:31:01.0000] <annevk>
heh, all public-uwa messages get marked as spam

[18:31:02.0000] <annevk>
Lachy, I've been doing the rainforest walk

[18:31:03.0000] <annevk>
yesterday

[18:31:04.0000] <annevk>
it was quite exhausting

[18:31:05.0000] <annevk>
we walked over 20km

[18:32:00.0000] <annevk>
after about 12km rainforest we found some water

[18:32:01.0000] <annevk>
after 20km we found a beer and some chips :)

[18:32:02.0000] <Lachy>
:-)

[18:32:03.0000] <Lachy>
who did you go with?

[18:32:04.0000] <annevk>
the plan was to take a short walk of about an hour

[18:33:00.0000] <annevk>
but we missed something underway...

[18:33:01.0000] <annevk>
with marcos

[18:34:00.0000] <zcorpan>
data:text/html,%3Cp%3Ex%3C%2Fbody%3E%20%3C%21--foo--%3E%20%3C%2Fhtml%3E%20%3C%21--bar--%3E%20x

[18:34:01.0000] <Lachy>
jgraham, test FF3 with XML to see if it keeps whitespace outside of the root

[18:35:00.0000] <annevk>
that's conforming btw

[18:35:01.0000] <annevk>
and should be done per XML

[18:35:02.0000] <annevk>
Opera should have some fix for that for the next major Opera release

[18:37:00.0000] <zcorpan>
Lachy: fx3 drops whitespace outside root in xml

[18:37:01.0000] <Lachy>
right

[18:37:02.0000] <annevk>
dropping?!

[18:38:00.0000] <annevk>
it's simply not added per the XML grammar

[18:38:01.0000] <zcorpan>
what's the difference?

[18:40:00.0000] <zcorpan>
annevk: did you see http://xhtml.se/2007/04/20/im-not-arguing-im-just-making-fun-of-them/ ?

[18:43:00.0000] <Dashiva>
I'm famous!

[18:43:01.0000] <zcorpan>
lol

[18:44:00.0000] <annevk>
no

[18:45:00.0000] <Dashiva>
How come I'm the only one with "" around my name?

[18:45:01.0000] <annevk>
lol

[18:46:00.0000] <Lachy>
probably because he thinks that's not your real name

[18:46:01.0000] <Dashiva>
Unlike zcorpan? :)

[18:46:02.0000] <zcorpan>
:D

[18:46:03.0000] <zcorpan>
it is! promise!

[18:46:04.0000] <Lachy>
:-)

[18:47:00.0000] <Dashiva>
What's the status of that 'conformant html5' image?

[18:47:01.0000] <zcorpan>
?

[18:48:00.0000] <Lachy>
do you mean the Valid HTML5 logo with Hixie's cat on it?

[18:48:01.0000] <Dashiva>
Yeah, that one

[18:48:02.0000] <zcorpan>
ah

[18:48:03.0000] <zcorpan>
http://simon.html5.org/temp/valid-html5.png

[18:48:04.0000] <annevk>
you should move that out of temp

[18:49:00.0000] <Dashiva>
Is valid the right word? Is using the cat legal? How many submarine patents are expected?

[18:49:01.0000] <Lachy>
ha!

[18:49:02.0000] <zcorpan>
ftp doesn't work for me today

[18:49:03.0000] <zcorpan>
Dashiva: cat used without permission :)

[18:50:00.0000] <zcorpan>
/me expects to get sued by Hixie

[18:50:01.0000] <annevk>
whoa

[18:50:02.0000] <Lachy>
permission from Hixie is implied, he hasn't complained about it

[18:50:03.0000] <annevk>
dhyatt has enough time to be co-editor

[18:50:04.0000] <Lachy>
probably just enough time to keep those who think we need 2 editors happy

[18:51:00.0000] <Dashiva>
Hixie can send him some diffs so Hyatt can commit them

[18:51:01.0000] <zcorpan>
heh

[18:52:00.0000] <Lachy>
why is http://simon.html5.org/temp/ forbidden?  I want to see what other cool stuff you have hidden in there

[18:52:01.0000] <annevk>
it's temp

[18:52:02.0000] <Lachy>
so?

[18:52:03.0000] <Dashiva>
temp stuff shouldn't be accessible, it leads to people linking it and expecting permanence

[18:52:04.0000] <zcorpan>
don't have cool stuff there

[18:54:00.0000] <zcorpan>
http://simon.html5.org/temp/program my gym program from last year, just started with it again now

[18:54:01.0000] <zcorpan>
in swedish

[18:56:00.0000] <zcorpan>
http://simon.html5.org/temp/browsers/ icons used in the web-apps-tracker

[18:56:01.0000] <annevk>
heh, collegue of dhyatt proposes dhyatt and then people wonder whether dhyatt actually agreed to it...

[18:56:02.0000] <annevk>
i suppose it's possible he didn't...

[18:57:00.0000] <zcorpan>
not when the mail proposing it says "he has said he's willing to do it"

[18:58:00.0000] <annevk>
oh right

[18:58:01.0000] <annevk>
guess that tells how closely I read some of my e-mail initially

[19:26:00.0000] <zcorpan>
Lachy: http://www.google.com/search?q=site%3Asimon.html5.org%2Ftemp should bring up anything "cool" i've hidden there :)

[19:26:01.0000] <Dashiva>
/me links to hsivonen's checker using valid-html5 image. Go precedent!

[19:26:02.0000] <zcorpan>
/me doesn't think that's a good idea

[19:26:03.0000] <Lachy>
why not?

[19:27:00.0000] <Dashiva>
"Some markup validation/checking services give out a small badge for advertising that a document passed. This service does not."

[19:28:00.0000] <zcorpan>
stickies linking to a checker generally gives the impression that the checker result is some sort of status indicator

[19:29:00.0000] <zcorpan>
the valid html5 image was mostly a joke on my part and i think hsivonen wants his tool to be a QA tool, not a status indicator

[19:29:01.0000] <annevk>
ooh, security

[19:30:00.0000] <Dashiva>
I fully agree, but that doesn't stop me from being subversive

[19:31:00.0000] <zcorpan>
fair enough :)

[19:32:00.0000] <zcorpan>
although i'd suggest that the link point to whatwg.org/html5 if anything

[19:32:01.0000] <annevk>
/me wonders when Web Forms 2 will be merged with HTML5

[19:34:00.0000] <zcorpan>
Dashiva: also see first paragraph in (4) in http://lists.w3.org/Archives/Public/public-html/2007JanMar/0433.html

[19:35:00.0000] <Dashiva>
I think the page is valid pretty much anything in its current state

[19:36:00.0000] <Dashiva>
Well, the <img> tag doesn't have the / needed for xhtml

[19:36:01.0000] <annevk>
<img> can also be <img></img> in XHTML

[19:36:02.0000] <annevk>
fwiw

[20:00:00.0000] <annevk>
http://www.w3.org/mid/op.tq008yhnsmjzpq⊙al

[20:02:00.0000] <annevk>
http://therealcrisp.xs4all.nl/blog/2007/04/22/html5-microsoft-and-the-opt-in-catch/

[20:03:00.0000] <annevk>
http://labs.google.com/goog411/

[20:18:00.0000] <annevk>
+  <h4>Origin</h4>

[20:18:01.0000] <annevk>
+  <!-- Hallowed are the Ori -->

[20:18:02.0000] <annevk>
heh

[20:25:00.0000] <Lachy>
annevk, which page had that ori comment in it?

[20:26:00.0000] <annevk>
HTML5?

[20:26:01.0000] <annevk>
Hixie defined origin for scripts

[20:33:00.0000] <Lachy>
That solution on crisp's blog won't work as-is, we tried it already

[20:34:00.0000] <Lachy>
Though, my latest response from Chris (off list) indicates that he'd listen to a solution that starts out as an opt-in and can transition to an opt-out, once the dependence on opt-ins isn't required

[20:35:00.0000] <zcorpan>
i do think making [if IE] return false in ie8 would help them not breaking things, but he didn't reply to that suggestion. perhaps i should bug him about that again

[20:35:01.0000] <annevk>
it requires too many fixes

[20:35:02.0000] <annevk>
in one go

[20:36:00.0000] <zcorpan>
then in ie9

[20:36:01.0000] <annevk>
if we got the proper testsuites ready by then...

[20:36:02.0000] <zcorpan>
/me will be working on that this summer

[20:37:00.0000] <Lachy>
I think the only way we're going to get Chris to compromise is to make HTML4/XHTML1 DOCTYPEs require an opt-in, and use the change in DOCTYPE in HTML to convert to an opt-out system

[20:38:00.0000] <annevk>
we shouldn't have to do much

[20:38:01.0000] <annevk>
this would only apply to IE

[20:38:02.0000] <annevk>
i hope

[20:38:03.0000] <Lachy>
that way, IE can use the time it takes to transition from HTML4/XHTML1 to HTML5 to improve its standards support and get to a point where it's close enough to interop with outhers, to not require an explicit opt-in

[20:38:04.0000] <Lachy>
yes, it would be an IE only opt-in/out

[20:39:00.0000] <annevk>
personally I want to keep fixing quirks mode, documenting quirks, etc.

[20:39:01.0000] <Lachy>
yeah, that would be good, but there's no chance IE is going to change it's quirks mode at all

[20:40:00.0000] <zcorpan>
in any case, it should be possible for authors to trigger a "latest standards mode", author against opera/firefox/safari, then work around bugs in current IEs without making assumptions about IE.next

[20:40:01.0000] <zcorpan>
annevk: yeah, me too

[20:45:00.0000] <annevk>
/me wonders if http://dev.w3.org/cvsweb/~checkout~/2006/waf/widgets/Overview.html?content-type=text/html;%20charset=utf-8#autodiscovery is good enough to define a new link type

[20:47:00.0000] <zcorpan>
/me thinks it is

[20:48:00.0000] <zcorpan>
or how is it handled today in opera?

[20:48:01.0000] <Lachy>
annevk, typo: "...to create an hyperlink" s/an/a/

[20:50:00.0000] <zcorpan>
ah, Content-Type: application/x-opera-widgets

[20:51:00.0000] <annevk>
I'm pretty sure it's always an before h...

[20:51:01.0000] <Lachy>
no

[20:51:02.0000] <zcorpan>
depends on how it's pronounced

[20:51:03.0000] <zcorpan>
an hour

[20:51:04.0000] <Lachy>
it doesn't depend on the letter, it depends on the sound

[20:53:00.0000] <annevk>
yeah nm me

[20:54:00.0000] <zcorpan>
annevk: if there is application/widget, why is there a need for rel="widget"?

[20:54:01.0000] <zcorpan>
surely the opera widget page works fine with only the mime type

[20:55:00.0000] <Lachy>
same reason there is rel=feed

[20:55:01.0000] <zcorpan>
for UI?

[20:55:02.0000] <Lachy>
the rel allows the UA to detect the widget before fetching it and offer a better UI

[20:56:00.0000] <zcorpan>
ok

[20:56:01.0000] <annevk>
right

[21:04:00.0000] <annevk>
fixed the typo and some other stuff fwiw

[21:06:00.0000] <annevk>
might be back later... food

[21:39:00.0000] <Hixie>
anyone got any pet bugs in browsers that they'd want fixed? any bugs. any browsers.

[21:39:01.0000] <Hixie>
ideally a DOM2 or HTML4 bug, maybe CSS2.1

[21:39:02.0000] <Hixie>
or JavaScript

[21:42:00.0000] <Lachy>
I want to see display: table-* properties fixed to work with less bugs in FF

[21:43:00.0000] <zcorpan>
/me points to http://simon.html5.org/test/ie7b2-bugs/

[21:43:01.0000] <Hixie>
/me takes notes

[21:44:00.0000] <Hixie>
Lachy: any specifics in mind?

[21:44:01.0000] <Hixie>
zcorpan: wow cool thanks

[21:44:02.0000] <zcorpan>
np

[21:45:00.0000] <Lachy>
see my site in FF.  Sometimes the columns don't work properly, but can usually be fixed by reloading

[21:46:00.0000] <zcorpan>
Lachy: oh yeah i get that too on some layouts

[21:46:01.0000] <zcorpan>
very unpredictable

[21:46:02.0000] <Hixie>
weird

[21:46:03.0000] <Hixie>
http://simon.html5.org/test/ie7b2-bugs/003.html <-- wtf is IE doing here

[21:47:00.0000] <Lachy>
also, if you create elements with JS and insert them into the DOM with display: table-cell; applied, it doesn't work well either

[21:47:01.0000] <zcorpan>
Hixie: no idea

[21:47:02.0000] <Lachy>
I'll create a test case

[21:49:00.0000] <zcorpan>
Hixie: also http://simon.html5.org/test/opera/

[21:49:01.0000] <Hixie>
cool, more tests

[21:52:00.0000] <Lachy>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%0A%3Cstyle%3E%0Adiv%20%7B%20display%3A%20table-cell%3B%20border%3A%201px%20solid%20blue%3B%20%7D%0A%3C/style%3E%0A%3Cp%3EThese%20columns%20are%20in%20the%20markup%3A%0A%3Cdiv%3Ecolumn1%3C/div%3E%0A%3Cdiv%3Ecolumn2%3C/div%3E%0A%3Cdiv%3Ecolumn3%3C/div%3E%0A%0A%3Cp%3EThese%20columns%20are%20created%20by%20a%20script%3A%0A%0A%3Cscript%3E%0Avar%20div%3B%0Afor%20%28va

[21:52:01.0000] <Lachy>
r%20i%20%3D%201%3B%20i%20%3C%3D%203%3B%20i++%29%20%7B%0A%20%20div%20%3D%20document.createElement%28%22div%22%29%3B%0A%20%20div.innerHTML%20%3D%20%22column%20%22%20+%20i%3B%0A%20%20document.body.appendChild%28div%29%3B%0A%7D%0A%3C/script%3E%0A

[21:52:02.0000] <Hixie>
hit the upload link

[21:52:03.0000] <Lachy>
I uploaded that to the DOM viewer clipboard, so you can just download it

[21:52:04.0000] <Hixie>
cool

[21:52:05.0000] <Hixie>
got it

[21:57:00.0000] <annevk>
/me wouldn't mind interop on feature "x"

[21:58:00.0000] <Hixie>
any specifics in mind?

[21:59:00.0000] <annevk>
innerHTML

[21:59:01.0000] <annevk>
HTML table model

[21:59:02.0000] <Hixie>
zcorpan: these are really good tests

[21:59:03.0000] <zcorpan>
:)

[22:00:00.0000] <annevk>
heh, why do you think we hired him?

[22:00:01.0000] <annevk>
/me doesn't have much tests for either of his suggestions

[22:03:00.0000] <annevk>
what are the feature requests for?

[22:04:00.0000] <annevk>
/me wonders why Opera doesn't do the dynamically generated <div> stuff correctly

[22:05:00.0000] <annevk>
ooh, Firefox has problems too

[22:05:01.0000] <Lachy>
annevk, are you talking about the test I made?  What does opera get wrong?

[22:05:02.0000] <annevk>
some build i have doesn't render the generated divs at all

[22:05:03.0000] <zcorpan>
same here

[22:06:00.0000] <annevk>
actually, all builds so far...

[22:06:01.0000] <Hixie>
i'm preparing for the writing of acid3

[22:06:02.0000] <Lachy>
that's weird.  It renders them as soon as I click download, but they disappear after the source is edited

[22:06:03.0000] <Hixie>
trying to get bugs lined up

[22:06:04.0000] <annevk>
(this would be covered by my "HTML table model" suggestion above which really should be named "HTML table layout model")

[22:06:05.0000] <Lachy>
oh, awesome!

[22:07:00.0000] <Hixie>
especially interested in DOM tests

[22:07:01.0000] <Hixie>
and JS

[22:07:02.0000] <Lachy>
when do you expect acid3 to be released?

[22:08:00.0000] <Hixie>
when it's ready

[22:08:01.0000] <Lachy>
fair enough

[22:08:02.0000] <Hixie>
http://hixie.ch/www/tests/evil/acid/003/ is where things will happen at first

[22:08:03.0000] <annevk>
http://tc.labs.opera.com/html/canvas/getContext/2d/ still shows quite some differences between browsers

[22:08:04.0000] <annevk>
although I suppose <canvas> may be too new

[22:09:00.0000] <Hixie>
yeah this is going to cover only things that were specced around 2003 or earlier

[22:09:01.0000] <annevk>
k

[22:09:02.0000] <annevk>
"PNG that contains some HTML"?!

[22:10:00.0000] <annevk>
IE still has a bunch of CSS parsing issues

[22:11:00.0000] <annevk>
http://annevankesteren.nl/2007/02/ie7-css-hacks contains a list of copy & paste testcases

[22:12:00.0000] <annevk>
oh, \n\r in document.write() and scripts in general might be good

[22:12:01.0000] <annevk>
or \r\n whatever it is

[22:13:00.0000] <Hixie>
has to be only things that can be seriously argued out of existing specs, unfortunately

[22:15:00.0000] <zcorpan>
having ie support the events model correctly would be nice, i think... and addEventListener, event.target, etc.

[22:15:01.0000] <Hixie>
yeah, gonna test that for sure

[22:15:02.0000] <annevk>
handling of <i><b></i> is to new too?

[22:15:03.0000] <annevk>
or is it too new too?

[22:16:00.0000] <Lachy>
I didn't realise FF3 passed acid2 already.  That's awesome! :-)

[22:16:01.0000] <Hixie>
parsing of html is too new if it's error handling

[22:16:02.0000] <Hixie>
Lachy: yeah, hence it being time for acid3

[22:16:03.0000] <zcorpan>
Lachy: better late than never :)

[22:18:00.0000] <zcorpan>
getAttribute("for") in IE

[22:18:01.0000] <Lachy>
Hixie, add this: <!--[if IE]>FAIL<![endif]--> :-)

[22:18:02.0000] <zcorpan>
(IE needs getAttribute("htmlFor"))

[22:19:00.0000] <Lachy>
same for "className"

[22:19:01.0000] <zcorpan>
yeah

[22:19:02.0000] <annevk>
<style type="">

[22:19:03.0000] <Lachy>
and there's something else too

[22:19:04.0000] <Lachy>
/me checking the DOM ref...

[22:19:05.0000] <annevk>
versus <style type="unknown"> versus <style> versus <style type="text/css">

[22:20:00.0000] <annevk>
maybe in the CSSOM? for float? which becomes .cssFloat as attribute

[22:20:01.0000] <Lachy>
I think getAttribute("http-equiv"); also has the problem

[22:20:02.0000] <zcorpan>
likely

[22:21:00.0000] <zcorpan>
using getAttribute on event handler attributes doesn't work right either iirc

[22:21:01.0000] <annevk>
events on body / document / window

[22:23:00.0000] <Lachy>
event object isn't passed to the handler in IE, needs window.event

[22:23:01.0000] <annevk>
already mentioned

[22:23:02.0000] <Lachy>
oh, sorry

[22:25:00.0000] <Lachy>
http://developer.mozilla.org/en/docs/Core_JavaScript_1.5_Reference:Global_Objects:Array#Methods  some of those aren't supported in IE

[22:25:01.0000] <annevk>
i suppose investigating in what Dean Edwards has to work around in his library might give some good suggestions on to what to include

[22:26:00.0000] <Lachy>
I assume you'll test the NS methods in the DOM too?

[22:27:00.0000] <Hixie>
already looked at dean's stuff

[22:27:01.0000] <Hixie>
nah, no point testing namespaces really

[22:27:02.0000] <Lachy>
ok

[22:27:03.0000] <Hixie>
since we don't particularly want anyone to use them

[22:27:04.0000] <Lachy>
alright

[22:28:00.0000] <annevk>
nasty SVG bits?

[22:31:00.0000] <Lachy>
Hixie, <link rel="icon"> - IE uses rel="shortcut icon"

[22:32:00.0000] <Lachy>
although, technically it wasn't specified till recently, icons have been widely supported for a while

[22:37:00.0000] <Hixie>
yeah but there's no spec that says it has to be "icon"

[22:38:00.0000] <Lachy>
I suppose you're not counting HTML5 then?

[22:38:01.0000] <annevk>
before 2003

[22:38:02.0000] <annevk>
HTML5 _started_ in 2004

[22:38:03.0000] <Lachy>
I know that, but it's been widely supported for much longer than that

[22:38:04.0000] <annevk>
the other requirement is for it to be in a spec

[22:39:00.0000] <Hixie>
unless i can point at a REC with clear conformance criteria, it doesn't count

[22:39:01.0000] <Lachy>
ok

[22:39:02.0000] <Hixie>
since you can argue that it isn't required for compliance

[22:39:03.0000] <annevk>
that leaves out testing HTML features :p

[22:44:00.0000] <Hixie>
mostly, yeah

[22:45:00.0000] <virtuelv_>
One thing that definetly would need to go into acid3 is how 'apply' works

[22:46:00.0000] <Hixie>
which spec?

[22:46:01.0000] <Lachy>
virtuelv_, in JavaScript?

[22:46:02.0000] <virtuelv_>
Lachy: yup

[22:46:03.0000] <virtuelv_>
I've run into cases where the behavior is rather exotic across browser

[22:47:00.0000] <virtuelv_>
+s

[22:47:01.0000] <Lachy>
is it defined in ECMA 262?

[22:49:00.0000] <virtuelv_>
iirc, yes

[22:49:01.0000] <virtuelv_>
lemme check, though

[22:49:02.0000] <Lachy>
yep, it's on page 87 (section 15.3.4.3

[22:50:00.0000] <Hixie>
if you can give me a testcase and explain how it tests behaviour described in the spec i can add it, sure

[22:52:00.0000] <Hixie>
anne, wouldn't rel=widget be better as a type="" value?

[22:52:01.0000] <annevk>
similar to rel=feed

[22:53:00.0000] <Lachy>
Hixie, why?

[22:53:01.0000] <virtuelv_>
Hixie: will do, bug me at the end of the coming week

[22:53:02.0000] <Hixie>
virtuelv_: k

[22:53:03.0000] <Lachy>
rel=widget is easier for authors to type and remember, compared with MIME types

[22:53:04.0000] <Hixie>
i guess

[22:54:00.0000] <virtuelv_>
inability to add stuff to the Element prototype is also, IMHO one thing browsers should get right in 2007

[22:55:00.0000] <virtuelv_>
s/in//

[22:55:01.0000] <Hixie>
unfortunately prototypes for elements aren't defined anywhere (yet)

[22:56:00.0000] <Hixie>
so that's not a standards compliance issues

[22:56:01.0000] <Hixie>
issue

[22:56:02.0000] <annevk>
It's more consistent with rel=stylesheet rel=feed etc. too. It means that authors have to learn one less trick (the trick that type= might lead to special handling too).

[22:56:03.0000] <Lachy>
but Elements are just Objects in ECMAScript, and Object is defined

[22:56:04.0000] <virtuelv_>
Hixie: that's rather unfortunate, really

[22:57:00.0000] <Hixie>
Lachy: if you can find a spec that says that, let me know :-)

[22:57:01.0000] <Lachy>
http://www.w3.org/TR/DOM-Level-2-Core/ecma-script-binding.html binds all the DOM objects to Object in ECMAScript

[22:58:00.0000] <Lachy>
ECMA 262 defines .prototype

[22:58:01.0000] <Lachy>
and it's how all other major browsers implement it, AFAIK

[22:59:00.0000] <Hixie>
but where does it say that the object has to have a prototype?

[23:00:00.0000] <virtuelv_>
Hixie: all objects that have constructors must have a prototype

[23:00:01.0000] <Hixie>
but Element doesn't have a constructor

[23:02:00.0000] <virtuelv_>
sigh. Neither does Node

[23:02:01.0000] <Lachy>
all Objects have constructors though.  It inherits from Object, and so it has a constructor.  It's just not available to scripts

[23:06:00.0000] <annevk>
but that Element is available somewhere in the global scope as attribute is also not defined

[23:12:00.0000] <Lachy>
border-collapse should be tested too (or was it tested in acid2 already?) http://lachy.id.au/dev/css/tests/bordercollapse/bordercollapse.html

[23:13:00.0000] <Lachy>
I think IE is the only one to fail those tests

[23:14:00.0000] <Hixie>
not really interested in css

[23:14:01.0000] <Lachy>
ok

[23:15:00.0000] <annevk>
location.hash when the page contains a single # is sometimes a compat issue iirc

[23:15:01.0000] <annevk>
oh, nm

[23:16:00.0000] <Hixie>
location was specced post 2003

[23:16:01.0000] <Hixie>
namely, last week

[23:16:02.0000] <annevk>
/me was under the illusion location was somehow part of old dom specs

[23:16:03.0000] <annevk>
did you and Maciej agree that the Window spec will be merged into HTML5?

[23:31:00.0000] <Lachy>
OMG WTF? "Long release cycles are a very good thing" on public-html

[23:32:00.0000] <annevk>
citing IE6 as an example

[23:32:01.0000] <Lachy>
Frequent releases of increasingly buggy browsers may be a bad thing, but the quicker we can get more standards compliant browsers into the market the better

[23:56:00.0000] <Hixie>
annevk: haven't asked

[23:57:00.0000] <annevk>
k

[23:57:01.0000] <Hixie>
annevk: but since Window was no longer being developped, and I needed to spec things that depended on it...

[23:57:02.0000] <annevk>
yeah, just wondering

[23:58:00.0000] <annevk>
now the chairs sort of "accepted" HTML5 will you merge in WF2? or let dhyatt do it? :p

[00:06:00.0000] <Hixie>
they have?

[00:08:00.0000] <Lachy>
Hixie, Chris responded fairly positively about the HTML5 spec

[00:09:00.0000] <Hixie>
uri?

[00:09:01.0000] <Lachy>
http://www.w3.org/mid/5C276AFCCD083E4F94BD5C2DA883F05A27D9CDDC38⊙twwnmc

[00:12:00.0000] <annevk>
whoa, schepers is asking for a third editor...

[00:12:01.0000] <Lachy>
annevk, I was just typing the same thing :-)

[00:13:00.0000] <Lachy>
2 editors is more than enough

[00:15:00.0000] <Hixie>
Lachy: it wasn't clear to me that he was accepting the proposal maciej put forward exactly. but yeah, that was positive regarding the spec itself.

[00:15:01.0000] <Hixie>
(my reply was http://lists.w3.org/Archives/Public/public-html/2007Apr/1241.html)

[00:16:00.0000] <Hixie>
i'm curious as to what people think "editor" means

[00:17:00.0000] <Hixie>
why does it matter what the opinions of the editor are?

[00:17:01.0000] <Lachy>
yeah, I don't understand why he wants the whatwg to stop either

[00:18:00.0000] <Hixie>
chris?

[00:18:01.0000] <Lachy>
yes

[00:22:00.0000] <Lachy>
I have no idea how I should respond to this about selectors API http://lists.w3.org/Archives/Public/public-webapi/2007Apr/0025.html

[00:22:01.0000] <Lachy>
I already said it's not wise to explicitly introduce undefined behaviour

[00:24:00.0000] <annevk>
Does anyone know what the last dot http://www.w3.org./ can possibly do?

[00:25:00.0000] <Hixie>
Lachy: well, what you have now doesn't guarentee interop, so what they're asking for doesn't really change that

[00:25:01.0000] <Hixie>
annevk: guarentees you're not going to do local dns lookup (with a suffix, e.g.)

[00:25:02.0000] <annevk>
yeah, just say that authors should not assume a specific prefix invocation algorithm or something

[00:25:03.0000] <Hixie>
bit like the leading / in /foo/bar

[00:26:00.0000] <Hixie>
saying authors "should not" is just a waste of time :-)

[00:26:01.0000] <Hixie>
won't do anything

[00:26:02.0000] <Hixie>
especially for APIs

[00:26:03.0000] <annevk>
should create a side-effect resolver...

[00:26:04.0000] <Lachy>
but I don't see what problems it solves by me taking it out

[00:27:00.0000] <annevk>
it makes things simpler

[00:27:01.0000] <Lachy>
so you think I should?

[00:27:02.0000] <annevk>
well, i was convinced

[00:27:03.0000] <annevk>
so yes

[00:28:00.0000] <annevk>
it's unfortunate jonas is so late with catching up with e-mail otherwise I would have fixed this ages ago (or never specced it)

[00:29:00.0000] <Lachy>
I suppose it doesn't really help authors much knowing their function is only going to be invoked once per prefix either

[00:30:00.0000] <annevk>
i suppose it's nice to point it out in interop concerns so people who are designing something similar know what to do

[00:31:00.0000] <annevk>
(same for people who wonder why it differs between browsers)

[00:40:00.0000] <Hixie>
annevk: "the ToASCII algorithm must  apply successfully (without errors)"

[00:40:01.0000] <Hixie>
how can it fail?

[01:13:00.0000] <Lachy>
Hixie, this has a good list of IE bugs you could add to acid3 http://easy-designs.stikipad.com/ie-next-wishlist

[01:15:00.0000] <Hixie>
IE bugs are easy to find :-)

[01:15:01.0000] <Hixie>
it's other browsers that are the problem

[01:16:00.0000] <Lachy>
right

[01:16:01.0000] <Hixie>
in fact, i've added all of _one_ test to the page so far and it breaks in IE drastically.

[01:17:00.0000] <Lachy>
yeah, I saw that.

[01:17:01.0000] <Lachy>
I don't understand what causes it to break though

[01:18:00.0000] <Lachy>
is it because there's no font-size specified for body?

[01:19:00.0000] <Hixie>
i'm guessing it doesn't support inline-block

[01:19:01.0000] <Hixie>
though i thought it did

[01:20:00.0000] <Lachy>
it does for inline level elements only

[01:22:00.0000] <Lachy>
do you have some sort of plan for the final rendering, or are you just making it up as you go?

[01:27:00.0000] <Hixie>
it'll look mostly like it does now in firefox

[01:27:01.0000] <Hixie>
but i'm making it up as i go along

[01:27:02.0000] <Hixie>
and i want to add animation

[01:37:00.0000] <Hixie>
why is opera not using arial

[01:37:01.0000] <Hixie>
wtf

[01:40:00.0000] <Lachy>
heh, I keep forgetting to switch to FF3 when looking at test cases :-)

[01:41:00.0000] <Lachy>
maybe opera doesn't support :root

[01:42:00.0000] <Hixie>
seems unlikely

[01:42:01.0000] <Hixie>
but maybe

[02:06:00.0000] <Hixie>
last chance to see what the test is intended to look like before i break it

[02:06:01.0000] <Hixie>
http://www.hixie.ch/tests/evil/acid/003/index.html

[02:06:02.0000] <Lachy>
yeah, I just looked at it.  It looks ok

[02:07:00.0000] <Hixie>
it already shows bugs in every browser except firefox

[02:07:01.0000] <Hixie>
and the only reason it works in firefox is i changed 'document' to 'window'

[02:07:02.0000] <Hixie>
for hte load event

[02:07:03.0000] <Lachy>
are you going to change it back to document?

[02:07:04.0000] <Hixie>
yes

[02:07:05.0000] <Hixie>
there's no window in 2003 standards

[02:08:00.0000] <Lachy>
yeah

[03:46:00.0000] <Hixie>
http://www.hixie.ch/tests/evil/acid/003/ and http://www.hixie.ch/tests/evil/acid/003/reference.html

[03:46:01.0000] <Hixie>
should look the same everywhere

[03:46:02.0000] <Hixie>
i haven't even tried adding tests yet and they don't

[03:46:03.0000] <Lachy>
/me checks every browser he has...

[03:48:00.0000] <Lachy>
nice, FF3 gets it right :-)

[03:48:01.0000] <Hixie>
it does?

[03:48:02.0000] <Lachy>
it did a second ago, not now

[03:48:03.0000] <Hixie>
i just get 0%

[03:48:04.0000] <Hixie>
oh you probably tested it before i switched it back to document

[03:48:05.0000] <Lachy>
yeah

[03:49:00.0000] <Hixie>
the reference rendering should be pixel-perfect, including font rasterisation

[03:50:00.0000] <Hixie>
but really the reference rendering is a test on to its own

[03:50:01.0000] <Lachy>
Opera 9 is doing surprisingly poorly

[03:51:00.0000] <Hixie>
opera 9 is hitting two bugs that i know of

[03:51:01.0000] <Hixie>
one, it's getting screwed by not having a <head> in the DOM, because one of the selectors goes through the <head>

[03:51:02.0000] <Hixie>
and two, it doesn't seem to do :root

[03:53:00.0000] <Hixie>
/me throws in some :first-child/:last-child tests

[03:53:01.0000] <Lachy>
which selector goes through head?

[03:53:02.0000] <Hixie>
*:first-child + * > * > p

[03:53:03.0000] <Hixie>
that's equivalent to head + body > div > p

[03:53:04.0000] <Lachy>
ah, I see

[03:54:00.0000] <Hixie>
as in, the only ps that it matches will match because the first simple selector there matches the head

[03:54:01.0000] <Hixie>
and since opera has no head...

[03:57:00.0000] <Hixie>
anyway i should go to bed

[03:57:01.0000] <Hixie>
nn

[03:57:02.0000] <Lachy>
cya

[05:18:00.0000] <gsnedders>
anyone know what percentage of XHTML docs are served as XML?

[05:18:01.0000] <gsnedders>
well, per-mille may be more useful, actually :P

[07:25:00.0000] <zcorpan>
Hixie: acid2 was easier to see if it was correct or not (without looking at the reference rendering)

[07:31:00.0000] <hasather>
zcorpan: agreed

[07:31:01.0000] <Lachy>
zcorpan, any suggestions for a better picture?

[07:32:00.0000] <hsivonen>
what test are you talking about?

[07:32:01.0000] <Lachy>
acid 3

[07:32:02.0000] <Lachy>
http://www.hixie.ch/tests/evil/acid/003/

[07:32:03.0000] <Lachy>
it's still a work in progress though

[07:34:00.0000] <zcorpan>
Lachy: hm, dunno... a questionmark inside a circle?

[07:34:01.0000] <zcorpan>
/me hides

[07:34:02.0000] <Lachy>
I thought maybe a cat

[07:53:00.0000] <zcorpan>
"Awareness of future trends." ?

[08:02:00.0000] <hsivonen>
zcorpan: indeed

[08:07:00.0000] <hsivonen>
It'll be interesting to see the diff for Web Apps 1.0 from the day the HTML WG was announced to the day the WG accepts HTML5 as the starting point (assuming that happens)

[08:16:00.0000] <SpookyET>
Opera allready supports some of it according to Wikipedia.

[08:18:00.0000] <Lachy>
SpookyET, yes, it has support for Web Forms 2.0 and a few things from Web Apps 1.0 like <canvas> and <event-source>

[08:21:00.0000] <hsivonen>
SpookyET: I meant it'll be interesting to see how much real work Hixie and the folks on the WHATWG list got done while public-html generated heat

[08:25:00.0000] <SpookyET>
hsivonen: http://en.wikipedia.org/wiki/Comparison_of_layout_engines_(WHATWG)

[08:25:01.0000] <SpookyET>
see all that green on the right?

[08:25:02.0000] <Lachy>
SpookyET, what's your point?  we know that already

[08:28:00.0000] <SpookyET>
Lachy: I was talking to hsivonen

[08:28:01.0000] <Lachy>
I realise that, but he knows too

[08:29:00.0000] <SpookyET>
Let him speak for himself.

[08:31:00.0000] <Lachy>
I'm trying to determine what your point is by telling people things we already know?

[08:32:00.0000] <zcorpan>
hsivonen: yeah, that's something to blog about :)

[08:40:00.0000] <Dashiva>
At the moment it looks like a lot of colorblind scandinavian flags in Opera

[08:40:01.0000] <hsivonen>
SpookyET: I don't understand what point you are trying to make to me

[08:40:02.0000] <SpookyET>
No point. I was stating a fact.

[08:41:00.0000] <hsivonen>
ok

[08:41:01.0000] <Lachy>
you can get it to work in Firefox 3 if you just change document.addEventListener to window.addEventListener.  (that is, until Hixie adds more tests to it later)

[08:43:00.0000] <hsivonen>
gotta respect the people who send patches for Lynx. last time I looked the source wasn't pretty

[08:46:00.0000] <Dashiva>
Acid3 uses HTML 4.0 doctype... am I missing something, or was it a random choice?

[08:47:00.0000] <zcorpan>
Dashiva: Hixie probably copied it from damowmow.com or something

[08:47:01.0000] <hsivonen>
that's what happened with Acid2, IIRC

[08:48:00.0000] <zcorpan>
or wait, damowmow uses 4.01

[08:48:01.0000] <zcorpan>
then hixie.ch or something :)

[08:50:00.0000] <hsivonen>
the comment in the  damowmow source is funny

[08:50:01.0000] <zcorpan>
heh

[08:51:00.0000] <SpookyET>
Adobe's Apollo is pretty interesting. I just stumbled upon it. It seems to be a competitor to Microsofts XAML.

[08:52:00.0000] <hsivonen>
I don't understand the Adobe-internal politics that produced Apollo considering that they have Flash

[08:52:01.0000] <SpookyET>
It uses Flash, Acrobat, HTML, CSS, JavaScript.

[08:52:02.0000] <zcorpan>
the new Ajax?

[08:53:00.0000] <SpookyET>
The cool thing about Apollo is that it uses WebKit, while Microsoft's solution uses IE's rendering engine.

[08:53:01.0000] <SpookyET>
http://www.niallkennedy.com/blog/archives/2007/03/adobe-apollo.html

[08:53:02.0000] <SpookyET>
zcorpan: It's not AJAX. It's designed to merge desktop and web apps.

[08:54:00.0000] <Lachy>
it's like their own version of Widgets

[08:54:01.0000] <SpookyET>
It's not widgets, but full apps.

[08:54:02.0000] <zcorpan>
SpookyET: ok

[08:55:00.0000] <SpookyET>
https://ssl.blogs.zdnet.com/blogs/Stewart/images/apollo_mac_library_large.png

[08:56:00.0000] <SpookyET>
It looks like you create a hybrid desktop + web app instead of a desktop client that connects to a web service.

[08:56:01.0000] <Lachy>
Interesting.  I just found out that for DOCTYPE sniffing, IE ignores the language code at the end of <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">, whereas in FF it needs to be EN or it triggers standards mode 

[08:57:00.0000] <hsivonen>
Lachy: interesting indeed. I wonder why the IE team chose to look inside the identifier

[08:57:01.0000] <Dashiva>
They're big on localization? :)

[08:57:02.0000] <hsivonen>
Lachy: Gecko lowercases the string and compares it against a list of strings

[08:57:03.0000] <Lachy>
probably because there are silly people around who change the "EN" to their own native langauge

[08:58:00.0000] <Lachy>
it makes sense to do what IE does

[08:58:01.0000] <hsivonen>
Lachy: yeah, but shouldn't those silly people get the quirks mode

[08:58:02.0000] <Lachy>
yes, they do in IE.  They don't in FF, that's my point

[08:58:03.0000] <hsivonen>
Lachy: considering that sniffing is a heuristic for deciding if the author had clue

[09:00:00.0000] <Philip`>
I'd like a "-//W3C//DTD HTML 4.01 Transitional//EN-GB" with <centre> and <font colour="grey">

[09:00:01.0000] <Lachy>
and in IE, the number needs to be 2, 3 or 4 to trigger quirks mode (decimal points ignored)

[09:01:00.0000] <zcorpan>
ie's doctype sniffing is insane. the way the system id also matters iirc

[09:01:01.0000] <zcorpan>
er, the say the system id is written...

[09:01:02.0000] <zcorpan>
/me -> food

[09:02:00.0000] <Lachy>
Philip`, the langauge code is basically irrelevant even in SGML.  It's supposed to refer to the language that the DTD is written in, but I have no idea why that's relevant at all

[09:06:00.0000] <Philip`>
(Incidentally, I was actually surprised when seeing that CSS3 Color supports "grey" and that it actually works in browsers - I'm certain it used to be interpreted as "green" in the past (like, uh, Netscape 3 or something))

[09:07:00.0000] <Lachy>
wow

[09:07:01.0000] <Dashiva>
I remember some early Opera canvas bug which interpreted random strings as colors. E.g. "bogus" became pink, if I recall correctly

[09:08:00.0000] <Philip`>
I just want to know how anybody is ever going to implement the "flavor" colour

[09:08:01.0000] <hasather>
"crap" became brown in IE before, I think. Dunno if that still works

[09:09:00.0000] <Lachy>
hasather, yep, still works!

[09:09:01.0000] <Lachy>
bogus becomes ping in IE too!

[09:10:00.0000] <Dashiva>
I suspect it has to do with trying to interpret it as hex, but the details elude me

[09:11:00.0000] <Lachy>
it seems that it throws away any letters other than [A-Fa-f]

[09:13:00.0000] <Lachy>
then replaces them with 0, so "bogus" becomes "b00000"

[09:13:01.0000] <Lachy>
and "crap" becomes "c0a000"

[09:14:00.0000] <Lachy>
I wonder if that's specced in HTML5...

[09:14:01.0000] <Philip`>
It work for "bogo" but not "bog" - it seems to only do that for >= 4 characters

[09:14:02.0000] <Lachy>
for the various color attributes

[09:15:00.0000] <Dashiva>
If there's a site relying on that, I will personally hunt down the author

[09:15:01.0000] <Philip`>
At least in FF3, it's quirks-only

[09:15:02.0000] <SpookyET>
Maybe, it should support non-us spelling:-)

[09:15:03.0000] <Dashiva>
What does 'beos' give you, Lachy?

[09:15:04.0000] <SpookyET>
background-colour

[09:16:00.0000] <Lachy>
Dashiva, "beos" gives "be0000"

[09:16:01.0000] <Philip`>
Maybe the browser developers were trying to be nice to authors who misread books with bad fonts, and tried writing <body bgcolor="ffooff">?

[09:16:02.0000] <Lachy>
SpookyET, doesn't work for CSS, only for the HTML attributes

[09:16:03.0000] <Dashiva>
Lachy: Hmm, in the bug I mentioned it became green

[09:17:00.0000] <Philip`>
(Quirks-only in Opera 9 too)

[09:17:01.0000] <Lachy>
Dashiva, which bug?

[09:17:02.0000] <Lachy>
and which browser?

[09:19:00.0000] <Dashiva>
Opera in the first stages of canvas support

[09:19:01.0000] <Dashiva>
Just after the "blue is red on windows" bug :)

[09:19:02.0000] <Philip`>
(Oh, it's doing bo=#b00000, bog=#0b0000, bogo=#b00000, and cra=#0c0a00 - I can see why it's called "quirks"...)

[09:20:00.0000] <Lachy>
Philip`, which browser are you getting those results?

[09:20:01.0000] <Lachy>
there shouldn't be a leading 0 on the values for bog and cra

[09:20:02.0000] <Philip`>
Oh, that was Opera 9

[09:21:00.0000] <Philip`>
but FF3 doesn't do that

[09:21:01.0000] <Philip`>
(Yay, non-interoperability)

[09:22:00.0000] <Philip`>
(FF3 just gives #000000 for anything <= 3 characters, it seems)

[09:22:01.0000] <Dashiva>
Time to run a bruteforce scan of all possible combinations on 1-6 letters, then!

[09:23:00.0000] <Lachy>
Dashiva, you only need to do 3 to 6, and letters A-F and one other

[09:24:00.0000] <Lachy>
e.g. bgcolor="azz", "bzz", ...

[09:24:01.0000] <Philip`>
Lachy: Not quite that simple - Opera ignores up to one leading #, and any spaces

[09:26:00.0000] <Philip`>
In Opera, "f f f f f" == #fffff0; in FF, it's #f00ff0 (?!)

[09:26:01.0000] <Lachy>
oh, 1 and 2 letters give weird results too

[09:28:00.0000] <Lachy>
Safari does it in standards mode

[09:30:00.0000] <Philip`>
With <body bgcolor=" f f f "> I get grey in FF, almost-black in Opera, and green in Konqueror

[09:33:00.0000] <Philip`>
zcorpan: You said you wanted to spec quirks mode? ;-)

[09:33:01.0000] <Lachy>
nice, <body bgcolor="rainbow"> give a diff colour in nearly every browser :-)

[09:35:00.0000] <Philip`>
http://lxr.mozilla.org/mozilla/source/gfx/src/nsColor.cpp#126

[09:35:01.0000] <Philip`>
so it always does black for <= 3 chars (excluding the optional leading #)

[09:35:02.0000] <Lachy>
someone want to check webkit source, it seems to be closer to IE

[10:40:00.0000] <zcorpan>
Philip`: yes, test cases welcome

[10:45:00.0000] <SpookyET>
I keep getting disconnected.

[10:47:00.0000] <SpookyET>
Has anyone seen Flashtml? :-) www.pierinc.com  Thank you for killing my scrollbar and back button. But, for a second, it fools you into thinking that it's html and dhtml animations.

[10:50:00.0000] <zcorpan>
/me will use http://del.icio.us/zcorpan/quirks to pile up quirks

[12:28:00.0000] <zcorpan>
yay! ftp works again. i have *no* idea why it didn't work before

[12:36:00.0000] <zcorpan>
/me considers moving valid-html5.png out of temp/

[12:43:00.0000] <zcorpan>
done

[13:24:00.0000] <zcorpan>
started to draft on my presentation http://simon.html5.org/temp/html5-geekmeet.html

[13:24:01.0000] <zcorpan>
(in swedish)

[13:26:00.0000] <Philip`>
"Cross-document messanging" - spurious n?

[13:26:01.0000] <zcorpan>
oops

[13:27:00.0000] <zcorpan>
fixed

[13:33:00.0000] <hsivonen>
/me learns that boilerplate is boilerplate in Swedish :-)

[13:34:00.0000] <zcorpan>
probably isn't

[13:34:01.0000] <zcorpan>
don't know a good swedish word for it :)

[13:34:02.0000] <hsivonen>
zcorpan: looks good

[13:35:00.0000] <zcorpan>
thanks. have two canvas demos iframed, commented out for now

[13:43:00.0000] <Philip`>
Be careful that the second doesn't suck up all your CPU while you're trying to get on with the rest of the presentation :-)

[13:44:00.0000] <zcorpan>
yeah, fortunately it can be paused :)

[13:44:01.0000] <zcorpan>
not sure i'll use those at all though

[13:45:00.0000] <zcorpan>
but having a "3d" fps game in 2d canvas is cool, so i want to include it

[13:46:00.0000] <Philip`>
Oh, can it?

[13:46:01.0000] <Philip`>
Aha, the "pause" key - that makes sense

[13:46:02.0000] <zcorpan>
there's also a pause button

[13:47:00.0000] <Philip`>
Oh, okay, on the 84 version since I didn't remove the debug bit at the bottom

[13:49:00.0000] <Philip`>
(The 2d canvas is quite nasty for doing this 3d stuff in, though :-p )

[13:50:00.0000] <Philip`>
(It's nice that you can do it, but it isn't really playing to the strengths of the canvas)

[13:51:00.0000] <zcorpan>
indeed. it's like using GIF for video

[13:51:01.0000] <Philip`>
((It is quite fun, though))

[13:53:00.0000] <Philip`>
Perhaps it's more like using ASCII art over Telnet for video

[13:54:00.0000] <zcorpan>
heh, ok

[13:56:00.0000] <Philip`>
(That reminds me, I did a (rubbish) text-based multiplayer deathmatch FPS game over the IRC protocol (though not a real IRC server) once - it would be nice to integrate that with the canvas code and get browser-based multiplayer deathmatch...)

[14:35:00.0000] <Philip`>
(Hixie: I think the spec has a spare > in Scripting/Origin with "<span>browsing context</span>>")


2007-04-23
[17:08:00.0000] <Philip`>
Hmm, createRadialGradient is pretty much entirely non-interoperable, which looks like fun...

[17:13:00.0000] <Philip`>
(Oh, actually, Firefox and Safari are pretty close for at least the portion of the image which they both draw, though Opera is totally crazy. (Maybe Opera is following the spec and the spec is crazy, but I haven't tried reading it yet...))

[17:16:00.0000] <Dashiva>
radial gradients are crazy to begin with, if you ask me :)

[17:19:00.0000] <Philip`>
I wouldn't disagree with that :-)

[17:20:00.0000] <Philip`>
I should have stopped after I reached linear gradients, because everything after that is hard :-(

[17:52:00.0000] <Dashiva>
Hmm... acid3...

[17:57:00.0000] <Philip`>
Aha, the spec is indeed crazy, though less so than the three incompatible implementations

[17:58:00.0000] <Philip`>
/me tries to guess what should happen instead

[18:03:00.0000] <Dashiva>
I can't find anything limiting 'element' in CSS, so :first-child and related selectors would include <script>, <style>, etc, right?

[18:23:00.0000] <gavin>
Is Matthew Raymond on IRC?

[18:27:00.0000] <zcorpan>
Dashiva: yes, any element that is in the dom tree is selectable

[18:27:01.0000] <zcorpan>
Dashiva: try * { display:block; margin:1em; border:solid; }

[18:28:00.0000] <Dashiva>
So that tricky selector seems to be checking for whether *:first-child selects an implicit head element

[18:28:01.0000] <zcorpan>
Dashiva: yes

[19:05:00.0000] <Hixie>
can someone explain the paragraph with the highlighted text in http://groups.google.com/group/alt.html/browse_thread/thread/811af37c5b2c2a96/fda33e528fee9ad6?q=whatwg&rnum=1#fda33e528 ?

[19:08:00.0000] <zcorpan>
Hixie: seems he observes that you can work with namespaces in the dom in browsers, and hence the text/html serialization should have a declarative namespacing mechanism

[19:09:00.0000] <Hixie>
is that what he's saying?

[19:10:00.0000] <Hixie>
i couldn't really tell

[19:12:00.0000] <zcorpan>
can't say for sure but that's how i read it

[19:16:00.0000] <zcorpan>
it could also be that he thinks that the dom that comes from html parsing doesn't know about namespaces

[19:17:00.0000] <Hixie>
ah

[19:27:00.0000] <othermaciej>
his message does seem confused

[19:30:00.0000] <Philip`>
(Ooh, excellent, I get totally different behaviour in Firefox 2 vs Firefox 3)

[19:30:01.0000] <Philip`>
(and I wouldn't be surprised if it changed on different OSes...)

[19:31:00.0000] <zcorpan>
Philip`: different results for what?

[19:32:00.0000] <Philip`>
Radial gradients in <canvas>

[19:32:01.0000] <zcorpan>
ok

[19:33:00.0000] <Dashiva>
Hixie: What's the status of the acid3 thing? Just a draft, work in progress, finished?

[19:33:01.0000] <Philip`>
(FF doesn't really do anything interesting with gradients, so I assume the changes are in Cairo somewhere - and I get a third behaviour when doing radial gradients with the version of Cairo I've got installed here)

[19:34:00.0000] <Hixie>
Dashiva: extremely early work

[19:34:01.0000] <Hixie>
there are no tests in it yet, to start with

[19:34:02.0000] <Hixie>
:-)

[19:35:00.0000] <Dashiva>
Well, it reveals two opera bugs already

[19:36:00.0000] <Hixie>
cool

[19:37:00.0000] <Dashiva>
But I gather I shouldn't use it for bug reference just yet

[19:39:00.0000] <Hixie>
well feel free to narrow down the bugs to file them

[19:40:00.0000] <Dashiva>
They were both already reported, to my disappointment :)

[19:47:00.0000] <Hixie>
:-)

[20:49:00.0000] <Hixie>
can anyone think of a way to programatically cause a bubbling event to fire without actually using dispatchEvent?

[20:50:00.0000] <Hixie>
HTMLInputElement.click() maybe

[20:51:00.0000] <othermaciej>
that should work

[20:51:01.0000] <othermaciej>
submitting a form would be another way

[20:52:00.0000] <othermaciej>
also DOMFocusIn / DOMFocusOut on focus changes

[20:54:00.0000] <Hixie>
i best avoid changing the focus or submitting a form during the test, i think :-)

[20:55:00.0000] <othermaciej>
well, you can cancel the form submission, perhaps in some way unlikely to fail; the click() method does seem simpler in any case

[20:55:01.0000] <Hixie>
yeah

[20:55:02.0000] <Hixie>
might do some form stuff

[00:12:00.0000] <Hixie>
&&= isn't valid in JS?

[00:24:00.0000] <othermaciej>
I don't believe so

[00:24:01.0000] <othermaciej>
it's definitely not in C, and I don't think JS extends the set of update operators

[00:25:00.0000] <Hixie>
hm

[00:25:01.0000] <Hixie>
so how do you do a &&= b without evaluating a twice?

[00:25:02.0000] <Hixie>
var x = a; x = x && b?

[00:27:00.0000] <othermaciej>
yes

[01:24:00.0000] <met_>
2 spams http://blog.whatwg.org/charmod-norm-checking#comment-3616 http://blog.whatwg.org/feed-autodiscovery#comment-3609

[01:24:01.0000] <Hixie>
do you have a blog account?

[01:25:00.0000] <Hixie>
we should just make you a moderator or whatever it is that has access to get rid of spam :-)

[01:25:01.0000] <Hixie>
Lachy should be able to hook you up

[01:26:00.0000] <met_>
don't have account

[01:26:01.0000] <met_>
/me is just registering

[01:33:00.0000] <Lachy>
hey met_, I can set you up with admin rights if you like

[01:35:00.0000] <met_>
/me ok

[01:35:01.0000] <met_>
login is met

[01:36:00.0000] <Lachy>
I know, I got the email when you registered

[01:36:01.0000] <met_>
8-)

[01:36:02.0000] <Lachy>
I upgraded you to Editor.  That should give you sufficient rights to moderate comments, I believe

[01:36:03.0000] <Lachy>
if not, let me know

[01:37:00.0000] <othermaciej>
Lachy: obviously the blog needs a non-WHATWG Editor

[01:37:01.0000] <Lachy>
lol

[01:37:02.0000] <Lachy>
if there are any nomineeds, I'll consider them for the role also

[01:37:03.0000] <met_>
see 'Delete comment' in admin, fine

[01:37:04.0000] <Lachy>
*nominees

[01:38:00.0000] <Lachy>
can you access the moderation queue?

[01:38:01.0000] <Lachy>
in Manage > Awaiting Moderation

[01:40:00.0000] <met_>
yes I see, it's empty now

[01:40:01.0000] <Lachy>
yep

[03:14:00.0000] <hsivonen>
hmm. Unicode 5.0.0 still isn't online. the ETA is in May. is it just me or wasn't the ETA supposed to be in March?

[03:51:00.0000] <krijnh>
http://xhtml.se/2007/04/20/im-not-arguing-im-just-making-fun-of-them/#comment-1358 - haha

[03:52:00.0000] <krijnh>
rel="nofollow" removed jarvklo ;)

[04:06:00.0000] <hsivonen>
yeah, the channel should be considered as public as the mailing list

[04:06:01.0000] <hsivonen>
since it is

[04:06:02.0000] <krijnh>
Yeah

[06:00:00.0000] <hsivonen>
/me learns something new about CSS namespace rules and class selectors

[06:00:01.0000] <hsivonen>
/me considers the interaction counterintuitive

[06:01:00.0000] <Lachy>
hsivonen, what did you learn?

[06:02:00.0000] <hsivonen>
Lachy: I learned that declaring the default namespace for element selectors breaks class-only selectors for elements from another namespace

[06:03:00.0000] <zcorpan>
ditto *

[06:04:00.0000] <zcorpan>
you need *|* (and *|*.foo) for any namespace if you have declared a default namespace

[06:04:01.0000] <hsivonen>
zcorpan: so it appears

[06:04:02.0000] <Lachy>
oh, right.  That is counter-intuitive

[06:04:03.0000] <hsivonen>
I'm disappointed

[06:05:00.0000] <Lachy>
well, in a way it makes sense.  By declaring the default namespace, you're basically saying you only want to  select elements in that namespace unless you explicitly state otherwise

[06:08:00.0000] <hsivonen>
bah. now that I fixed my SVG class selectors, they broke in Prince

[06:09:00.0000] <hsivonen>
I guess I'll just get rid of the namespace declarations and rely on DWIM

[06:10:00.0000] <zcorpan>
yeah, in most cases namspace declarations in css is not necessary

[06:10:01.0000] <zcorpan>
(even if you used mixed namespaces in the document)

[06:11:00.0000] <hsivonen>
the XML API user in me feel that not declaring namespaces in wrong

[06:12:00.0000] <hsivonen>
*is wrong

[06:13:00.0000] <zcorpan>
if two elements from different namepsaces get styled but you only wanted one of them, you can always qualify the selector with a parent or descendant combinator, which might not be as elegant but this situation is very rare anyway

[06:14:00.0000] <Lachy>
doesn't the pragmatic and sensible user in you make you feel that declaring namespaces is pointless?

[06:14:01.0000] <hsivonen>
Lachy: yes

[06:15:00.0000] <hsivonen>
the moron and the asshole fighting in me

[06:15:01.0000] <hsivonen>
ok. I'm going back to namespaceless CSS

[06:15:02.0000] <Lachy>
lol

[06:15:03.0000] <hsivonen>
I guess I should file a bug about Prince, though

[06:30:00.0000] <hsivonen>
well, typical. I tried to do the right thing. I learned that it didn't work like I expected. I learned that it didn't help any. and I found a bug.

[13:37:00.0000] <hsivonen>
Hixie: do you have stats on the usage proportions of all HTML elements as opposed to just top 10?

[13:37:01.0000] <hsivonen>
specifically, am I right when I think that <dfn> is used very rarely?

[13:42:00.0000] <Hixie>
hsivonen: ask me again when i get to work (a few hours from now)

[13:42:01.0000] <Hixie>
(i have stats for more than the top 10)

[13:52:00.0000] <hsivonen>
Hixie: ok

[13:55:00.0000] <hsivonen>
btw, there's a thread about the legitimacy of html5 on www-html

[13:56:00.0000] <hsivonen>
start: http://lists.w3.org/Archives/Public/www-html/2007Apr/0053.html

[13:57:00.0000] <hsivonen>
the main feedback point is that Tina Holmboe does not like the semantic retrofitting of <i>

[13:58:00.0000] <hsivonen>
also, Lynx developers have not researched <script/> behavior thoroughly

[13:58:01.0000] <hsivonen>
Hixie: do you read www-html?

[13:58:02.0000] <Hixie>
yes

[13:59:00.0000] <hsivonen>
ok. no need to mention this then

[14:00:00.0000] <Hixie>
i saw the thread. i figured i wouldn't have any credibility from tina's point of view, so i didn't bother to reply

[14:01:00.0000] <Dashiva>
Poor Hixie, his life's work demolished by the mere flick of Tina's send button

[14:03:00.0000] <gavin_>
Tina Holmboe has some interesting opinions :)

[14:05:00.0000] <hsivonen>
I bothered to reply, because back when she was on ciwah, she was in the clueful minority

[14:17:00.0000] <zcorpan>
"  No, I meant HTML 4.1. Starting with the 4.01 specification would make perfect sense; take out deprecated elements, clean it up, remove even more presentational markup ...  Then release it. After that we can see." perfect plan for a high-quality and useful spec, why didn't we think of that?

[14:32:00.0000] <Hixie>
i tried that but when i did the "clean it up" stage it fell apart

[14:32:01.0000] <Hixie>
</sarcasm>

[14:35:00.0000] <zcorpan>
a first step could be to remove everything that isn't normative... but then the only thing left would be "UAs MUST NOT assume a default encoding"

[14:35:01.0000] <zcorpan>
:)

[14:36:00.0000] <Dashiva>
we should take the IRC approach

[14:36:01.0000] <Dashiva>
Treat everything as a sequence of octets, and let the clients decide on a per-case basis

[14:36:02.0000] <hsivonen>
Dashiva: not interoperable

[14:36:03.0000] <Dashiva>
octets are about as interoperable as it gets

[14:37:00.0000] <zcorpan>
/me thinks the xhtml2 wg would like that proposal

[14:37:01.0000] <hsivonen>
Dashiva: irssi is so good at guessing that it is hard for people who aren't experts to figure out what they are talking to clients with worse heuristics

[14:37:02.0000] <hsivonen>
Dashiva: so irssi users can talk to each other and non-irssi users complain

[14:38:00.0000] <hsivonen>
and the complaints look ok to the irssi users :-)

[14:38:01.0000] <Dashiva>
See, mIRC got recode support for utf-8 last year, so it works out somehow

[15:01:00.0000] <Hixie>
i'm thinking placeholder="" as an attribute to hide things

[15:01:01.0000] <Hixie>
<section placeholder>, <output placeholder>, that kind of thing

[15:07:00.0000] <Hixie>
man, the arguments in www-html are very polar

[15:08:00.0000] <Hixie>
i'm extremely amused by the way that people in the ivory tower say that i'm pro-presentationalism at the expense of semantics, and people in the trenches say i'm pro-semanticism at the expense of their ability to do what users want

[15:08:01.0000] <bewest>
yeah, there seems to be some idea that every possible semantic should have it's own tag

[15:09:00.0000] <bewest>
and that users somehow want to use it

[15:09:01.0000] <hsivonen>
"I think you and I, and I believe TBL, have a higher view in that it is to help people to communicate, but I don't think that that is what the WHATWG's constituency wants."

[15:09:02.0000] <othermaciej>
Hixie: Murray is in the ivory tower and he thinks you're not going far enough with presentationalism

[15:10:00.0000] <Hixie>
it's not clear to me what murray wants or where he stands

[15:10:01.0000] <Hixie>
i think he might be in the tower's subbasement

[15:10:02.0000] <Hixie>
or maybe he's on a tall ladder in the trenches

[15:10:03.0000] <Hixie>
or something

[15:10:04.0000] <hsivonen>
:-)

[15:11:00.0000] <othermaciej>
I think he built another tower on top of the tower and wrapped around to underneath the trenches

[15:11:01.0000] <Hixie>
hah

[15:11:02.0000] <othermaciej>
(damn spherical geometries)

[15:11:03.0000] <Hixie>
that's either a tall tower or our metaphor is in a universe with a very small word size

[15:11:04.0000] <hsivonen>
I got the impression that he is even more pro reflowable presentation atoms than I am

[15:14:00.0000] <othermaciej>
he wants an <indent> element

[15:14:01.0000] <othermaciej>
and an unspecified number of other weakly semantic constructs

[15:15:00.0000] <othermaciej>
we can always get Dan Connolly to tell him whether his elements are good or not

[15:15:01.0000] <hsivonen>
othermaciej: I can sympathise with <indent>, but the thread seems like a rathole. and for compatibility, <indent> has to be called <blockquote> :-)

[15:15:02.0000] <Hixie>
i didn't get what <indent> was supposed to do

[15:16:00.0000] <hsivonen>
Hixie: my understanding was that it would behave like <blockquote> on the visual media but the element name would not suggest quoting

[15:17:00.0000] <Hixie>
so like a <div> but with a margin-left: 40px in the UA style sheet?

[15:17:01.0000] <jgraham>
The problem being it doesn't work in any existing browser so no one would use it anytime soon

[15:17:02.0000] <Hixie>
that's certainly one problem

[15:17:03.0000] <hsivonen>
Hixie: yeah. something that you could map to indent/outdent buttons in Word/OOo

[15:18:00.0000] <Hixie>
i see

[15:18:01.0000] <hsivonen>
jgraham: see has to be called <blockquote> above

[15:18:02.0000] <Hixie>
well, it seems clear that that would be a step backwards in terms of the markup cleanup...

[15:18:03.0000] <jgraham>
hsivonen: Indeed

[15:19:00.0000] <hsivonen>
Hixie: it could be a step forward in admitting reality, though :-/

[15:19:01.0000] <Hixie>
hsivonen: not really, authors rarely want 40px these days

[15:19:02.0000] <Hixie>
making <style> allowed in <div>s is the way to do this

[15:20:00.0000] <Hixie>
<div> <style> div { margin: 0 2em; border: ...; background: ...; } </style> ... </div>

[15:20:01.0000] <othermaciej>
<indent> seems like a poor name for an element to add extra horizontal margin on both sides

[15:20:02.0000] <hsivonen>
Hixie: that doesn't work with copy-paste to a document with another style sheet in a way that the indent depth adapts to the style of the target doc

[15:21:00.0000] <Hixie>
hsivonen: ironically, that's the opposite of the requirement i keep hearing that copy-paste should preserve the styles exactly

[15:21:01.0000] <Hixie>
so

[15:21:02.0000] <othermaciej>
only difference between <indent> and <div class="indented"> is default style

[15:21:03.0000] <Hixie>
go figure

[15:21:04.0000] <hsivonen>
Hixie: and you've got a verbose way of saying style='' there ;-)

[15:21:05.0000] <Hixie>
style="" isn't media-independent, nor does it support alternate stylesheets

[15:22:00.0000] <hsivonen>
Hixie: I want an editor with generic indent and italics but no copying of computed style

[15:22:01.0000] <othermaciej>
hsivonen: by that do you mean an authoring tool or a spec editor?

[15:23:00.0000] <othermaciej>
(because people are getting awfully specific in their requirements for HTML spec editors)

[15:23:01.0000] <hsivonen>
glazou says I have an orphan disease

[15:23:02.0000] <hsivonen>
othermaciej: authoring tool 

[15:23:03.0000] <Hixie>
othermaciej: hah

[15:24:00.0000] <othermaciej>
I want an editor who will add <canvas> and <video>, but not XML data islands or namespaces in HTML

[15:24:01.0000] <Hixie>
hsivonen: i think editors can already do whatever they need with the current state of things, especially if it's not got to interoperate with other editors

[15:24:02.0000] <Hixie>
anyway

[15:25:00.0000] <othermaciej>
please everyone fill out this checklist of desired features before you apply to be editor

[15:25:01.0000] <Hixie>
gotta go to work


2007-04-24
[19:18:00.0000] <zcorpan>
spam? http://blog.whatwg.org/proposing-features#comment-2393

[19:34:00.0000] <zcorpan>
http://twitter.com/ddfreyne/statuses/37293072

[19:36:00.0000] <Hixie>
is he your friend?

[19:36:01.0000] <Hixie>
we should get him to send us feedback

[19:36:02.0000] <Hixie>
so we can fix the problems he ran into

[19:36:03.0000] <zcorpan>
no, he just showed up in my google blog search feed

[19:38:00.0000] <zcorpan>
reading http://stoneship.org/journal/marked-up/ ...

[19:45:00.0000] <zcorpan>
/me added him on msn

[19:47:00.0000] <zcorpan>
nn

[22:35:00.0000] <Hixie>
ok i've officially run out of things to test on http://hixie.ch/tests/evil/acid/003/

[22:36:00.0000] <Hixie>
and i have up to 87 more tests to write :-)

[22:37:00.0000] <Hixie>
testing DOM2 Core, DOM2 Events, DOM2 HTML, ECMAScript 3, HTTP

[22:38:00.0000] <Hixie>
and anything else you can think of testing via JS without it actually rendering

[00:53:00.0000] <hsivonen>
Hixie: are you at work? ping about dfn usage

[01:00:00.0000] <Hixie>
just got back from work, heh

[01:00:01.0000] <Hixie>
let me get my vpn card, hang on

[01:00:02.0000] <Hixie>
what was the question?

[01:06:00.0000] <hsivonen>
Hixie: what's the relative usage frequency of <dfn> compared to <i> (or perhaps <em>, too)

[01:08:00.0000] <Hixie>
on a per-page basis?

[01:08:01.0000] <Hixie>
as in, pages using dfn vs pages using i/em?

[01:08:02.0000] <hsivonen>
for example

[01:10:00.0000] <Hixie>
<i> is used on 178 times more pages than <dfn>, and <em> is used on 80 times more pages than <dfn>

[01:10:01.0000] <Hixie>
as of last september

[01:10:02.0000] <hsivonen>
Hixie: thanks. this is "several billion pages", right?

[01:11:00.0000] <Hixie>
<dfn> is used more than <del>, <ins>, <var>, <kbd>, <samp>, etc

[01:11:01.0000] <Hixie>
yeah

[01:12:00.0000] <hsivonen>
thanks. <dfn> being more common than <var> and <kbd> is surprising to me

[01:12:01.0000] <hsivonen>
and <del>

[01:12:02.0000] <hsivonen>
lots of bloggers use <del> and <ins> these days

[01:12:03.0000] <Hixie>
<dfn> is used less than <zeroboard>, <st1:place>, <noembed>, <blink>, <bgsound>, <ilayer>, etc

[01:12:04.0000] <othermaciej>
<zeroboard>?

[01:12:05.0000] <hsivonen>
othermaciej: indeed

[01:12:06.0000] <Hixie>
don't ask me

[01:12:07.0000] <Hixie>
i just work here

[01:12:08.0000] <othermaciej>
<blink>??

[01:12:09.0000] <hsivonen>
Hixie: thanks

[01:13:00.0000] <virtuelv>
/me wonders wtf <zeroboard> is

[01:13:01.0000] <othermaciej>
omg Sam Ruby posted on public-html

[01:13:02.0000] <othermaciej>
has IBM decided to join the Conspiracy of Light?

[01:14:00.0000] <virtuelv>
othermaciej: http://www.intertwingly.net/blog/2007/04/23/HTML-WG-me

[01:14:01.0000] <virtuelv>
"I’m the first to represent IBM, but there undoubtedly will be others."

[01:14:02.0000] <othermaciej>
nobody tells me these things

[04:43:00.0000] <Philip`>
<zeroboard> seems to come from e.g. http://www.nzeo.com/bbs/zboard.php?id=cgi_zskins2&no=19641 which has code like

[04:43:01.0000] <Philip`>
  <table border=0 cellspacing=0 cellpadding=0 width=99%>

[04:43:02.0000] <Philip`>
  <tr>

[04:43:03.0000] <Philip`>
    <td >

[04:43:04.0000] <Philip`>
      <Zeroboard???</a>

[04:43:05.0000] <Philip`>
      <Zeroboard????</a>

[04:43:06.0000] <Philip`>
      ....

[04:44:00.0000] <Philip`>
(where the "?" are Korean)

[04:45:00.0000] <Philip`>
so it looks like a markup error rather than an actual tag

[04:45:01.0000] <Philip`>
(but it's on (apparently) Korea's most popular message board software, so it gets around a lot)

[05:08:00.0000] <hsivonen>
where should I look for mod_negotiation for dummies?

[05:09:00.0000] <hsivonen>
/me is puzzled by the difference between http://hsivonen.iki.fi/thesis/html5-conformance-checker and http://hsivonen.iki.fi/thesis/html5-conformance-checker.var in Opera

[05:09:01.0000] <hsivonen>
Apache docs say that .var should take precedence over automatic MultiViews

[05:22:00.0000] <virtuelv>
Hixie: I think Acid3 needs a cat picture

[05:22:01.0000] <zcorpan>
virtuelv: agreed

[05:23:00.0000] <hsivonen>
/me gives up on conneg for today

[05:28:00.0000] <hsivonen>
it appears that the html60 guy hasn't read HTML5

[05:30:00.0000] <met_>
do you mean people in html-wg?

[05:34:00.0000] <Dashiva>
No, there's this russian with html60 as his email userpart. He keeps coming up with very dramatic declarative suggestions

[05:35:00.0000] <met_>
oh this

[05:36:00.0000] <Lachy>
oh wow!  This has to be the longest subject line ever: "Now there is table-element with predetermined quantity of column.But sometimes we need to visualize three tables(for example, database tables) as one table:first table contains names of rows,second table contains names of columns (this table determine quantity of columns),third table contains data.New html-element: table-dim"

[05:58:00.0000] <Philip`>
No, the longest subject line ever is the one that says "Re: {all that}"

[06:19:00.0000] <krijnh>
Anybody in here know if it's possible to turn off the black outline when navigating to form elements in Opera?

[06:34:00.0000] <MikeSmith>
Hixie: is there any chance you might be able to set up an archived mailing list to which webapps checkin messages get echoed?

[06:34:01.0000] <MikeSmith>
in my workflow at least (at work and in open-source projects), I'm used to scanning checkin message over e-mail

[06:35:00.0000] <MikeSmith>
maybe some other people are used to the same thing

[06:38:00.0000] <met_>
/me is, it's in the /svn/hooks/post-commit sending mails with diff of commits

[07:08:00.0000] <Lachy>
MikeSmith, those checkin messages get posted to twitter

[07:09:00.0000] <MikeSmith>
Lachy - which is great, but I think it could also be useful for them to go to a mailing list

[07:10:00.0000] <Lachy>
you could hook into the twitter API with your own program that sends you an email.  Perhaps someone has already written such a tool

[07:11:00.0000] <met_>
MikeSmith propably want also diffs

[07:12:00.0000] <met_>
which are included in svn mails

[07:12:01.0000] <MikeSmith>
yeah, would want to have the diffs

[07:14:00.0000] <zcorpan>
might be able to hook something to web-apps-tracker

[07:15:00.0000] <zcorpan>
s/able/possible/

[07:16:00.0000] <met_>
 /svn/hooks/post-commit has it ready

[07:16:01.0000] <MikeSmith>
right, what met_ said

[07:17:00.0000] <met_>
there is one line needed to add mail address of the receiver

[07:17:01.0000] <zcorpan>
ok

[07:17:02.0000] <MikeSmith>
yep

[07:18:00.0000] <met_>
like this /var/lib/svn/svn/hooks/commit-email.pl -h example.com -s '[SVN]' "$REPOS" "$REV" name⊙ec

[07:18:01.0000] <zcorpan>
but web-apps-tracker has a filter

[07:22:00.0000] <Lachy>
the easy way to get twitter updates by email would be to use http://www.rssfwd.com/ and subscribe to the twitter feed but that wouldn't give you the diffs

[07:23:00.0000] <zcorpan>
/me still thinks about making a feed for web-apps-tracker that is aware of the filter

[11:22:00.0000] <annevk>
http://html456.blogspot.com/

[11:29:00.0000] <zcorpan_>
anyone know of a czech translator that works?

[11:31:00.0000] <annevk>
use a real person?

[11:31:01.0000] <annevk>
works for Swedish stuff I need translated :p

[11:32:00.0000] <zcorpan_>
anyone here know czech? :)

[11:33:00.0000] <hasather>
didn't we have a Czech person here? met_? Not here now though

[11:34:00.0000] <Philip`>
There's one that doesn't really work at http://www.tranexp.com:2000/Translate/result.shtml

[11:34:01.0000] <zcorpan_>
ddfreyne: welcome

[11:34:02.0000] <Philip`>
"WHATWG tender sleight propagujc altogether singly plus acquaintance thinking , that 5 2", etc

[11:34:03.0000] <hasather>
http://krijnhoetmer.nl/irc-logs/whatwg/20070413#l-164

[11:35:00.0000] <ddfreyne>
zcorpan_: hi

[11:36:00.0000] <zcorpan_>
Hixie: ddfreyne is the http://twitter.com/ddfreyne/statuses/37293072 guy :)

[11:36:01.0000] <Philip`>
(Those log links would work better if the page didn't rearrange itself to remove the join/part/quit messages after it's scrolled down to the right line...)

[11:36:02.0000] <ddfreyne>
Hixie: hi :)

[11:36:03.0000] <zcorpan_>
Philip`: (indeed)

[11:38:00.0000] <annevk>
krijnh, see the bit between ( and ) above from Philip`

[11:39:00.0000] <annevk>
krijnh, preferably, fix it, by resetting location.hash or something

[11:41:00.0000] <zcorpan_>
ddfreyne: so, do you have any feedback about why html5 wasn't so great after all? :)

[11:42:00.0000] <ddfreyne>
zcorpan_: I do… not written down though

[11:42:01.0000] <ddfreyne>
(yet)

[11:43:00.0000] <zcorpan_>
ok

[12:04:00.0000] <ddfreyne>
Interesting — google said "Did you mean: bioy _casares_" when I googled for HTML 5.

[12:12:00.0000] <annevk>
was there inside the Acid3 directory a list with tests?

[12:13:00.0000] <annevk>
well, a list with browser bugs to exploit?

[14:02:00.0000] <jgraham>
gsnedders: It would be nice to define how _anything_ is parsed first ;) Chris Wilson was very much against the idea of defining parsing

[14:03:00.0000] <gsnedders>
jgraham: that's true, but aren't we really bound by the charter to define how to? (Or rather, describing the language in such a way that it will imply how to parse it)

[14:03:01.0000] <jgraham>
(But I generally agree, although I would have thought that a HTML fragment would be parsed like the .innerHTML case)

[14:03:02.0000] <gsnedders>
Then I think we should state that.

[14:04:00.0000] <jgraham>
I entirely agree

[14:18:00.0000] <zcorpan_>
http://forums.whatwg.org/viewtopic.php?t=38#164

[14:18:01.0000] <met_>
zcorpan_, http://html456.blogspot.com/ yes? 8-)

[14:18:02.0000] <zcorpan_>
met_: yeah, care to translate? :)

[14:19:00.0000] <met_>
yes I can, but it is not necessary i am author of html456 8-)

[14:19:01.0000] <met_>
if you want anything to know, ask

[14:19:02.0000] <zcorpan_>
ok :)

[14:21:00.0000] <Hixie>
zcorpan_: not exactly sure what he's asking in that forum post

[14:21:01.0000] <Hixie>
did someone want some sort of mail to go out on-commit or something?

[14:22:00.0000] <Hixie>
and how do i do that? :-)

[14:23:00.0000] <zcorpan_>
Hixie: not sure either, perhaps he wants guidelines for authors as to what to use

[14:23:01.0000] <Hixie>
there's a syntax section, maybe we should add informative text to that or something

[14:23:02.0000] <Hixie>
if you can get him to tell you in more detail what he wants, let me know

[14:23:03.0000] <met_>
svn commit mails it was in discuccesed today yes

[14:23:04.0000] <Hixie>
(or mail the list with his request)

[14:24:00.0000] <zcorpan_>
ok

[14:44:00.0000] <zcorpan_>
Hixie: do you have data about how many pages use non-color values for <body bgcolor>? (it seems browsers do funny things with them in the parser)

[14:45:00.0000] <Hixie>
not off-hand no

[14:45:01.0000] <zcorpan_>
ok

[14:46:00.0000] <zcorpan_>
/me would write some testcases about that to find out exactly what happens but it would be nice to know if the web relies on it or not

[14:47:00.0000] <Hixie>
define non-colour

[14:48:00.0000] <Hixie>
if you mean whether pages depend on the handling of bogus colours, like bgcolor="2138a", then yes, they do

[14:50:00.0000] <zcorpan_>
ok

[15:03:00.0000] <zcorpan_>
hm. < in unquoted attribute values. have i bugged about that before? the parsing section says < in an unquoted attribute value terminates the tag but all browsers i've tested just append it to the attribute value

[15:07:00.0000] <zcorpan_>
data:text/html,%3C!doctype%20html%3E%3Cp%20class%3Dfoo%3Ci%3EThis%20text%20should%20not%20be%20italic.

[15:08:00.0000] <zcorpan_>
(or well, per spec currently it should be italic)

[15:09:00.0000] <Hixie>
feel free to send mail about that again

[15:09:01.0000] <Hixie>
have you checked what browsers do with end tags and <?

[15:10:00.0000] <zcorpan_>
you mean </p class=foo<i> ?

[15:11:00.0000] <zcorpan_>
or <p class=foo</i> ?

[15:13:00.0000] <zcorpan_>
in both cases (and with <p class=foo<?>) it's part of the attribute value

[15:15:00.0000] <Hixie>
or </p< or <p< for that matter

[15:18:00.0000] <zcorpan_>
those are dealt with correctly in the spec afaict, it's just < in unquoted attribute values

[15:18:01.0000] <Hixie>
k

[15:33:00.0000] <met_>
strange on http://www.w3.org/2000/09/dbwg/details?group=40318&public=1 is only D. Connolly marked as chair, not Ch. Wilson

[16:10:00.0000] <Hixie>
www-html:

[16:10:01.0000] <Hixie>
  I /expect/ it to be shaped by my participation, alongside others, in

[16:10:02.0000] <Hixie>
  the /standardization organization/ tasked with creating it the

[16:10:03.0000] <Hixie>
  specification.

[16:11:00.0000] <Hixie>
/me wonders if Tina is aware that she is not participating in the right group

[16:12:00.0000] <othermaciej>
I wonder if Tina knows that she is using /too much/ punctuation to indicate *emphasis*

[16:17:00.0000] <Dashiva>
othermaciej: It's probably proprietary markup to indicate semantically different kinds of emphasis

[16:57:00.0000] <Dashiva>
acid3 got brutal overnight...


2007-04-25
[17:00:00.0000] <Philip`>
At least it looks much better in Opera than when I last looked - it's very broken, but not so broken that you can't see how broken it is :-)

[17:01:00.0000] <Philip`>
(I think the rendering in IE6 counts as 'too broken' - it just looks like a really boring and slightly buggy page, and doesn't tempt you with the promise of interestingness if only a few small browser bugs were fixed)

[17:02:00.0000] <Hixie>
i'm not really targetting IE

[17:02:01.0000] <Hixie>
they have bigger fish to fry, imho

[17:02:02.0000] <Hixie>
i'd be very interested in suggestions for more tests

[17:02:03.0000] <Hixie>
i've run out of the things i wanted to test :-)

[17:04:00.0000] <Philip`>
(Konqueror gives an Opera-like level of very-brokenness)

[17:14:00.0000] <Philip`>
(Ooh, KDE4's Konqueror does surprisingly well - it appears to get the whole layout correct, and it just fails the 1st and 3rd buckets and says "0%" at the end)

[17:15:00.0000] <zcorpan_>
time to hunt for khtml bugs then? :)

[17:15:01.0000] <Hixie>
click the "Acid3" text to get a report of what failed

[17:17:00.0000] <Philip`>
34, 35, 37, it says

[17:18:00.0000] <Philip`>
but half the time it asks me if I want to Save As / Open / Cancel the empty.txt file

[17:19:00.0000] <Philip`>
(The other half the time, it fails doesn't ask but it fails 48 and 49)

[17:19:01.0000] <othermaciej>
WebKit gets to 95%, positions the boxes slightly wrong, gets two box colors wrong, and gets the color of the text "Acid 3" wrong

[17:19:02.0000] <Philip`>
s/fails//

[17:20:00.0000] <othermaciej>
fails 34, 35, 37, 48, 49

[17:20:01.0000] <othermaciej>
I feel like it would be cheating to start fixing before the test is done

[17:21:00.0000] <nlogax>
where can i find the test? (sorry, new in here)

[17:23:00.0000] <othermaciej>
I bet 34 is because comments are not preserved in the DOM

[17:23:01.0000] <othermaciej>
http://hixie.ch/tests/evil/acid/003/

[17:32:00.0000] <zcorpan_>
othermaciej: presuming that the test uses sgml parsing rules as the spec -- what does sgml say about comments? (may they be dropped as in xml?)

[17:33:00.0000] <othermaciej>
zcorpan_: I think the DOM spec says that implementations may include comment nodes in the DOM or not

[17:35:00.0000] <othermaciej>
SGML does not say anything about parsing into a DOM afaik

[17:35:01.0000] <othermaciej>
HTML5 may require preserving comment nodes; not sure if this is a compat issue

[17:37:00.0000] <zcorpan_>
ok... surprised that the dom spec would say anything about it

[17:38:00.0000] <Dashiva>
Is the reference rendering updated?

[17:44:00.0000] <Hixie>
hm, good point, the test shouldn't rely on comments

[17:44:01.0000] <Hixie>
feel free to fix the bugs, btw

[17:44:02.0000] <Hixie>
for the love of kittens don't _not_ fix bugs because acid3's early versions trigger them :-P

[17:45:00.0000] <Dashiva>
/me promotes test as an example of 'corner cases with no relevance to the real web, which can safely be ignored'

[17:45:01.0000] <Hixie>
oh?

[17:46:00.0000] <othermaciej>
zcorpan_: SGML (and I think XML) predate the existence of the DOM spec, so I don't think they had the opportunity to talk about parsing into a DOM

[17:46:01.0000] <Dashiva>
Just to be the first to do so ;)

[17:47:00.0000] <zcorpan>
othermaciej: yeah, but the xml spec says that comments may be dropped

[17:47:01.0000] <Hixie>
removed the comment thing from the test

[17:48:00.0000] <zcorpan>
Hixie: will there be 100 tests when it's done?

[17:48:01.0000] <othermaciej>
zcorpan: yeah, I think it is talking about processing in a more general way

[17:48:02.0000] <othermaciej>
WebKit is borking the layout much worse now, though still failing the same number of tests

[17:48:03.0000] <Hixie>
really?

[17:48:04.0000] <Hixie>
that's weird

[17:49:00.0000] <Hixie>
i only removed the comment

[17:49:01.0000] <Hixie>
zcorpan: if y'all give me enough things to test :-)

[17:49:02.0000] <zcorpan>
Hixie: ok :)

[17:49:03.0000] <Lachy>
good morning

[17:49:04.0000] <zcorpan>
Lachy: morning

[17:49:05.0000] <othermaciej>
Hixie: it's pretty weird to me too - I have a saved copy of the older test though, so pretty sure it's a real phenomenon

[17:49:06.0000] <Hixie>
odd

[17:49:07.0000] <Hixie>
wonder what i did

[17:50:00.0000] <Hixie>
reload?

[17:50:01.0000] <Hixie>
i put the comment back

[17:50:02.0000] <Hixie>
does it render right now?

[17:50:03.0000] <othermaciej>
Hixie: yes

[17:50:04.0000] <Lachy>
/me wonders how a comment could be used in a test

[17:50:05.0000] <othermaciej>
I wonder if this is a parsing bug of some sort

[17:50:06.0000] <Hixie>
othermaciej: looks like your html parser fails to imply <head> or something

[17:51:00.0000] <othermaciej>
Hixie: that's possible

[17:51:01.0000] <Hixie>
brb, getting food

[17:51:02.0000] <othermaciej>
I bet looking for browser-specific quirks in Dojo, Prototype, MochiKit and similar would be a good source of tests

[18:10:00.0000] <Lachy>
Hixie, your definition of the placeholder attribute conflicts with the <input placeholder> attribute implemented by Safari

[18:11:00.0000] <zcorpan>
<input placeholder> is something that should be part of html5 imho, the lack of that feature is worked around all over the place (mostly abusing value="")

[18:12:00.0000] <othermaciej>
we plan to submit <input type="search"> and related stuff for consideration in HTML5

[18:12:01.0000] <othermaciej>
I asked someone on my team to write it up

[18:13:00.0000] <Lachy>
class="search" was predefined for that, but type=search has a cool rendering in safari

[18:14:00.0000] <othermaciej>
class="search" is for search forms, not search fields, afaik

[18:14:01.0000] <Lachy>
zcorpan, placeholder was discussed back in 2004 or 2005, and I think Hixie said it would be considered for Web Forms 3

[18:14:02.0000] <othermaciej>
"It must only be used on the following elements: aside, body, form, p, section, span"

[18:15:00.0000] <Lachy>
oh, I thought it could be used on input too

[18:15:01.0000] <othermaciej>
<input type="search"> and placeholder on non-search fields (which we originally intended only for search but accidentally enabled for text fields too) are both in some actual use on the web

[18:15:02.0000] <othermaciej>
mostly Apple-oriented sites but not exclusively

[18:16:00.0000] <Lachy>
I would love to use placeholder, but I've been forced to fake it with js in the past

[18:16:01.0000] <Hixie>
Lachy: d'oh!

[18:18:00.0000] <Lachy>
hixie, maybe try relevant=""

[18:19:00.0000] <Hixie>
placeholder="" had the great property that it would sound silly to put it on a hidden tab panel

[18:19:01.0000] <Hixie>
relevant="" doesn't really hav ethat

[18:23:00.0000] <zcorpan>
"Well-formed document requirement—HTML5 defines a well-formed document according to the requirements specified in section 4 Appendix C of the XHTML 1.0 specification." -- http://www.devx.com/webdev/Article/34389/0/page/2

[18:29:00.0000] <othermaciej>
I think hyatt is just emotionally attached to a doctype that looks the way doctypes do today

[18:30:00.0000] <othermaciej>
because his post on the matter makes a hash of the relevant arguments

[18:39:00.0000] <Philip`>
Does SGML/XML/anything care about the actual content of the "-//W3C//DTD ..." string in the doctype, and become unhappy if it was e.g. "5.0"? (If not, has anybody suggested any reasons why there's any point in using that string, given that it's impossible to remember and is confusing since there's no DTD?)

[18:44:00.0000] <othermaciej>
Philip`: I don't know if it is needed or not - some other people commented on it, but I think there are strong reasons to prefer a version attribute over a doctype anyway

[23:34:00.0000] <othermaciej>
hmm, I still can't figure out why the missing comment in Acid3 breaks our display

[23:34:01.0000] <annevk>
you do reparsing?

[23:35:00.0000] <annevk>
if that doesn't make much sense in this context, feel free to ignore

[23:36:00.0000] <annevk>
Maybe the innerHTML parsing mode should be renamed to HTML fragment with context parsing

[23:37:00.0000] <annevk>
so people don't get the feeling it has to do with scripting...

[23:40:00.0000] <othermaciej>
that could be good yeah

[23:46:00.0000] <othermaciej>
hmm, looks like WebKit does in fact keep comments in the DOM (though strictly speaking it is optional)

[00:09:00.0000] <othermaciej>
annevk: it is a parsing issue, but not due to reparsing - <script> in an implicit head ends up between head and body

[00:19:00.0000] <Hixie>
othermaciej: what does your dom look like below <html?

[00:20:00.0000] <othermaciej>
Hixie: I made a reduction which illustrates the issue

[00:20:01.0000] <othermaciej>
just a sec

[00:21:00.0000] <othermaciej>
Hixie: http://paste.lisp.org/display/40272

[00:21:01.0000] <othermaciej>
weird bug

[00:21:02.0000] <othermaciej>
I'm not sure why that hoses the rendering though

[00:22:00.0000] <Hixie>
/me looks

[00:22:01.0000] <Hixie>
there's a style rule

[00:22:02.0000] <Hixie>
that matches head + body > div > p

[00:22:03.0000] <Hixie>
so you'll start/stop matching it

[00:22:04.0000] <annevk>
whoa, that's some quirky parsing right there :)

[00:23:00.0000] <othermaciej>
annevk: I think Opera puts both of those scripts between head and body

[00:23:01.0000] <othermaciej>
firefox puts both in head

[00:23:02.0000] <Hixie>
opera doesn't have a head

[00:23:03.0000] <Hixie>
in that document

[00:23:04.0000] <annevk>
right

[00:23:05.0000] <Hixie>
it doesn't imply <head> tags at all

[00:23:06.0000] <othermaciej>
ok, that, then

[00:23:07.0000] <annevk>
who needs <head> anyway

[00:23:08.0000] <Hixie>
(causes all kinds of subtle problems, that's why acid3 has an implied <head>)

[00:23:09.0000] <othermaciej>
my test is not very discriminating, just tells me if it went in <head> or <html>

[00:24:00.0000] <Hixie>
the live dom viewer is probably what you want to be using

[00:24:01.0000] <Hixie>
http://software.hixie.ch/utilities/js/live-dom-viewer/

[00:25:00.0000] <othermaciej>
mainly I'm interested in the WebKit bug which I can figure out w/ the built-in Web Inspector, but that is indeed useful for comparisons

[00:25:01.0000] <annevk>
Hixie, placeholder clashes with <input placeholder> from WebKit, do you need an e-mail on that?

[00:25:02.0000] <Hixie>
yeah lachy said

[00:25:03.0000] <Hixie>
what i need is a better name :-)

[00:27:00.0000] <othermaciej>
what is it supposed to do?

[00:28:00.0000] <annevk>
irrelevant, not-relevant-now, hidden

[00:29:00.0000] <othermaciej>
so placeholder is basically a presentational attribute for display: none

[00:30:00.0000] <annevk>
othermaciej, http://html5.org/tools/web-apps-tracker?from=776&to=777

[00:30:01.0000] <annevk>
othermaciej, made "semantic" and only for specific use cases, yes

[00:33:00.0000] <othermaciej>
it seems like "placeholder" is the wrong meaning for the given example

[00:33:01.0000] <othermaciej>
the input is a placeholder for the video

[00:33:02.0000] <othermaciej>
the video isn't a placeholder for anything while hidden

[00:33:03.0000] <othermaciej>
at least as I understand placeholder (temporary substitute for something else)

[00:34:00.0000] <othermaciej>
it seems like "hidden" would be the best word, but then you have to wonder why not just use the CSSOM, or a user-defined class that styles to display: none

[00:36:00.0000] <othermaciej>
placeholder does seem like it captures a common use case, but it's pretty limited, since it is harder to hook in author-controlled animation of the new content appearing/disappearing, which might be desirable

[00:36:01.0000] <othermaciej>
it's also hard to tell the cases where it should be used from ones where it shouldn't so it seems easy to abuse

[00:37:00.0000] <othermaciej>
I assume it would be considered wrong to use this for hover menus, for instance

[00:37:01.0000] <annevk>
it does mention that at the end

[00:39:00.0000] <annevk>
animations are a good point

[00:45:00.0000] <othermaciej>
so yeah, it doesn't sound like a compelling feature to me just yet

[00:50:00.0000] <Hixie>
othermaciej: we need a way to do what the examples show without any css (i.e. non-presentational hiding of content that isn't usable/relevant)

[00:51:00.0000] <Hixie>
othermaciej: i'm all in favour of having ways to hook animation to it

[00:52:00.0000] <othermaciej>
Hixie: can you clarify why being able to do it without CSS is a requirement? (I don't necessarily disagree, I am just unaware of the reasoning behind it)

[00:52:01.0000] <Hixie>
because css is a presentation-layer thing, not semantic

[00:52:02.0000] <Hixie>
but this is a semantic

[00:53:00.0000] <Hixie>
or more pragmatically, because css can be disabled

[00:53:01.0000] <Hixie>
because lync has no css

[00:53:02.0000] <Hixie>
lynx

[00:53:03.0000] <Hixie>
etc

[00:53:04.0000] <othermaciej>
does lynx have JavaScript?

[00:53:05.0000] <Hixie>
it could do

[00:54:00.0000] <Hixie>
the point is css is the wrong way to do this

[00:54:01.0000] <Hixie>
it's a semantic thing

[00:54:02.0000] <Hixie>
i need to say here's a section that isn't relevant, hide it

[00:54:03.0000] <Hixie>
it's not a stylistic thing

[00:55:00.0000] <annevk>
non-stylistic-hidden

[00:55:01.0000] <othermaciej>
I guess I'm not sure how to tell what reasons to make something hidden are semantic and which aren't

[00:57:00.0000] <Hixie>
if you can think of a way to apply an alternate style sheet that would show that section, then it's stylistic

[00:57:01.0000] <Hixie>
if you would never want to show it, it's semantic

[00:58:00.0000] <annevk>
/me wonders if these type of explanations should be in the spec

[00:59:00.0000] <annevk>
maybe a section that explains markup semantics...

[00:59:01.0000] <othermaciej>
seems to me like something that's hidden in the markup could just be in a comment, if it is truly not part of the semantics of the document, and if your UA has JS it probably has CSS too

[01:00:00.0000] <Hixie>
the problem is that a comment isn't parsed

[01:00:01.0000] <Hixie>
you want something that's in the DOM

[01:00:02.0000] <Hixie>
e.g. so it can be conformance checked

[01:02:00.0000] <othermaciej>
maybe the attribute should name should be "irrelevant" or "notrelevant"

[01:03:00.0000] <othermaciej>
it still doesn't sound terribly compelling to me, but I can see why you want it

[01:04:00.0000] <Hixie>
i've switched it to irrelevant for now

[01:05:00.0000] <Hixie>
i would love to make it more useful

[01:05:01.0000] <annevk>
so with animations you would set this after the animation has run I suppose

[01:05:02.0000] <othermaciej>
I'm trying to think how you'd make it tie in well with animation approaches, which are generally CSS-based

[01:05:03.0000] <Hixie>
we basically need a Core Animation of Web technologies

[01:06:00.0000] <Hixie>
i suppose SMIL is intended to be it, but i'm not impressed by SMIL

[01:06:01.0000] <othermaciej>
I sure hope someone works on such a completely hypothetical thing

[01:06:02.0000] <Hixie>
indeed

[01:07:00.0000] <annevk>
othermaciej, was that sarcastic?

[01:07:01.0000] <othermaciej>
it was a fancy way of saying "no comment"

[01:07:02.0000] <annevk>
if it was, I don't get it

[01:08:00.0000] <annevk>
Hixie, the problem is that SVG inflicts SMIL upon the world

[01:09:00.0000] <annevk>
And SVG animation is being implemented (supported)...

[01:09:01.0000] <Hixie>
there's room for more than one solution

[01:09:02.0000] <Hixie>
especially when one is as bad as SVG

[01:14:00.0000] <annevk>
I suppose

[01:14:01.0000] <annevk>
I think it would have to be significantly better though to persuade implementors, but maybe that's not hard

[01:17:00.0000] <Hixie>
how do you mean?

[01:18:00.0000] <annevk>
Well, having two sets of technologies to do the same thing doesn't seem very desirable

[01:19:00.0000] <annevk>
IE already has SMIL working for HTML (HTML+Time proposal to the W3C)

[01:19:01.0000] <annevk>
otoh, they had VML too...

[01:19:02.0000] <annevk>
s/had/have/

[01:24:00.0000] <othermaciej>
they also have WPF/E now

[01:25:00.0000] <annevk>
Yeah, this Flash competitor

[01:25:01.0000] <annevk>
Silverlight iirc

[01:35:00.0000] <hsivonen>
Is Flash still bundled with Vista?

[01:36:00.0000] <hsivonen>
/me is very surprised about hyatt's opinions on versioning

[01:36:01.0000] <othermaciej>
Flash is not in fact bundled with Vista

[01:37:00.0000] <hsivonen>
othermaciej: ooh. radical

[01:37:01.0000] <hsivonen>
othermaciej: that means MS is ready to break the Web!

[01:38:00.0000] <hsivonen>
Silverlight in Windows-only and IE/Firefox-only, right?

[01:38:01.0000] <annevk>
I think Safari too

[01:38:02.0000] <hsivonen>
does the Firefox version work with Opera?

[01:38:03.0000] <hsivonen>
annevk: oh. where do I get a Mac version?

[01:38:04.0000] <annevk>
Maybe I was misinformed

[01:39:00.0000] <annevk>
I only read about it in w3c-svg-wg archives

[01:39:01.0000] <othermaciej>
I think they intend to have it work on Safari, dunno if that has been released yet

[01:40:00.0000] <hsivonen>
it seems that there is a "Silverlight CTP for Macintosh"

[01:42:00.0000] <hsivonen>
the Silverlight FAQ uses user-unfriendly styling/scripting. have to use Lynx to read

[01:44:00.0000] <hsivonen>
apparently, "/E" in the MS world view means Windows and Mac

[01:47:00.0000] <annevk>
The '<' doesn't seem to be specialcased at all in IE's tokenizer

[01:48:00.0000] <annevk>
For element names, attribute names and unquoted attribute values that is

[01:58:00.0000] <othermaciej>
hsivonen: it means that at least as long as it takes them to kill Flash

[02:02:00.0000] <annevk>
This basically means that the SHORTTAG TAGC OMISSION feature of SGML is obsolete...

[02:03:00.0000] <othermaciej>
what's the SHORTTAG TAGC OMISSION feature?

[02:03:01.0000] <annevk>
<test</test> being parsed as <test></test>

[02:03:02.0000] <annevk>
<p><img src=""</p> like <p><img src=""></p> etc.

[02:04:00.0000] <annevk>
it basically means special casing <

[02:04:01.0000] <virtuelv>
heh, I now have a new showcase of how to make shiny accessible JS+DOM-based UI, http://www.satama.nl/ </offtopic>

[02:06:00.0000] <othermaciej>
ah

[02:06:01.0000] <othermaciej>
does any browser do that?

[02:07:00.0000] <annevk>
Firefox

[02:07:01.0000] <annevk>
maybe Safari?

[02:08:00.0000] <annevk>
It also means that <p</p>test will give you a 'p<' element with a 'p' attribute with 'test' as textContent

[02:31:00.0000] <annevk>
othermaciej, what's the DOM view in Safari for http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%3Ctest%3C/test%3E

[02:31:01.0000] <othermaciej>
annevk: in webkit trunk, it's HTML --> BODY --> TEST

[02:32:00.0000] <othermaciej>
with no text content

[02:34:00.0000] <annevk>
IE has a "TEST<" element which has a "test" attribute

[02:35:00.0000] <annevk>
"TEST<" also occurs before "HTML" but we shouldn't follow that I think

[02:35:01.0000] <annevk>
This is pretty trivial to fix in the html5lib tokenizer though it breaks a bunch of tests

[02:36:00.0000] <othermaciej>
it

[02:36:01.0000] <annevk>
maybe http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%3Cbody%3E%3Cp%20%3C/p%3Etest is a better test

[02:36:02.0000] <othermaciej>
it's really hard to tell how much of IE's crazier parsing rules should be followed

[02:37:00.0000] <othermaciej>
annevk: the text is sibling to the P and the P has no funny attributes

[02:38:00.0000] <annevk>
Well, cwilso is right with suggesting there's an IE web and an "other browser" web I think. Some application frameworks are relying on the IE namespaced DOM for instance and have some patched version for other browsers.

[02:38:01.0000] <annevk>
othermaciej, I think WebKit supports the SGML feature in that case, at least partially

[02:39:00.0000] <annevk>
In IE "test" is a child text node of <p> which has <="" and p="" attributes

[02:41:00.0000] <othermaciej>
it is sometimes hard to withhold my aesthetic judgment on these parsing oddities

[02:42:00.0000] <annevk>
I would personally prefer IEs behavior

[02:42:01.0000] <annevk>
Less work for Opera and simplifies the code of html5lib

[02:43:00.0000] <annevk>
As the affect of having less characters that do something special

[02:44:00.0000] <othermaciej>
the only part that seems nasty is having an attribute named "<"

[02:44:01.0000] <othermaciej>
"<" is clearly already a special character, question is only whether this is also true while parsing a tag or only when parsing text

[03:04:00.0000] <annevk>
Hixie, the spec points to http://lists.whatwg.org/admin.cgi/commit-watchers-whatwg.org

[03:04:01.0000] <annevk>
Hixie, it should point to http://lists.whatwg.org/listinfo.cgi/commit-watchers-whatwg.org

[03:08:00.0000] <MikeSmith>
annevk - just tried to subscribe to the list but got a non-delivery message:

[03:08:01.0000] <MikeSmith>
[[

[03:09:00.0000] <MikeSmith>
<commit-watchers-request⊙lwo>: host

[03:09:01.0000] <MikeSmith>
    lists.whatwg.org[66.33.216.179] said: 550

[03:09:02.0000] <MikeSmith>
    <commit-watchers-request⊙lwo>: Recipient address rejected:

[03:09:03.0000] <MikeSmith>
    User unknown in virtual alias table (in reply to RCPT TO command)

[03:09:04.0000] <MikeSmith>
]]

[03:09:05.0000] <annevk>
Through the interface?

[03:10:00.0000] <MikeSmith>
anne - yeah. sent me a confirm message, I replied, got the non-delivery notification

[03:10:01.0000] <annevk>
Interesting. I'd suggesting waiting a couple of hours (like nine, ten) for Hixie so he can fix it

[03:12:00.0000] <MikeSmith>
k

[05:35:00.0000] <zcorpan>
anyone know spanish?

[05:37:00.0000] <nlogax>
a little

[05:37:01.0000] <citoyen>
Only enough to order beer and dinner

[05:37:02.0000] <zcorpan>
can you translate this to spanish? :) "This is not correct. Well-formedness is a concept defined in XML 1.0, and namespace-well-formedness is a concept defined in Namespaces in XML 1.0. The concepts do not apply to HTML, and HTML5 does not introduce such a concept for text/html."

[05:37:03.0000] <annevk>
cerveza

[05:38:00.0000] <zcorpan>
(which is a response to "HTML5 define como documento bien formado según las especificaciones de la section 4 Appendix C of the XHTML 1.0 specification." -- http://mistervertigo.es/blogs/introduccion-html-5-i-2/ )

[05:38:01.0000] <citoyen>
/me agrees with annevk's translation

[05:38:02.0000] <citoyen>
it's all you'll ever need

[05:38:03.0000] <annevk>
I believe he got that from somewhere else (in English)

[05:38:04.0000] <zcorpan>
annevk: yes

[05:39:00.0000] <annevk>
in which case replying in English should cause no problems

[05:39:01.0000] <zcorpan>
/me wanted to reply in spanish mostly for fun :P

[05:40:00.0000] <zcorpan>
but ok

[05:42:00.0000] <zcorpan>
commented in english

[05:50:00.0000] <hsivonen>
/me just talked to a reporter about the conformance checker project; wonders what will end up in print

[05:56:00.0000] <annevk>
A promising student has been doing his master thesis on a conformance checker for XHTML2, a technology frequently used on the web.

[06:03:00.0000] <Dashiva>
Oh yes

[06:04:00.0000] <Dashiva>
hsivonen is a role="model" for web-interested people everywhere

[07:54:00.0000] <krijnh>
annevk, Philip`: I think it's fixed now

[07:56:00.0000] <Lachy>
Hixie should rename placeholder="" to role="placeholder"!

[07:56:01.0000] <Lachy>
/me ducsk

[07:56:02.0000] <Lachy>
*ducks

[07:58:00.0000] <Philip`>
krijnh: Looks to me like it's working - thanks!

[07:58:01.0000] <krijnh>
Np

[07:58:02.0000] <krijnh>
Can't test too much now, so if it breaks somewhere, let me know :)

[07:58:03.0000] <Lachy>
oh, there's an XHTML2 WG telcon going on!  This should be fun to watch :-)

[07:59:00.0000] <Lachy>
in #xhtml on w3 IRC server

[07:59:01.0000] <krijnh>
http://krijnhoetmer.nl/irc-logs/xhtml/20070425 ;)

[08:00:00.0000] <krijnh>
Oki, bye :)

[08:00:01.0000] <Philip`>
You should add extra AJAX for live streaming IRC logs ;-)

[08:00:02.0000] <Philip`>
(I suppose anybody who wants live streaming IRC could just join the channel, but that's no fun)

[08:01:00.0000] <Lachy>
just reload every ~30 seconds

[08:02:00.0000] <Lachy>
I'll let you know when something interesting happens

[08:02:01.0000] <Philip`>
"when" rather than "if"?

[08:03:00.0000] <Lachy>
yeah, sorry. I meant "if"

[08:04:00.0000] <Lachy>
they're not typing the minutes :-(

[08:04:01.0000] <Lachy>
oh, now they are

[08:19:00.0000] <virtuelv>
we should stream the logs using server-sent events

[08:19:01.0000] <Lachy>
that'd be awesome

[08:22:00.0000] <Philip`>
"[16:24] * Zakim Steven, you typed too many words without commas" - Zakim seems like quite a peculiar bot

[08:23:00.0000] <Lachy>
I didn't understand that message either.  I had no idea zakim was a grammar checker

[08:24:00.0000] <Lachy>
wow, IE will apparently be implementing role=""

[08:25:00.0000] <virtuelv>
I could point out that words themselves don't have commas

[08:40:00.0000] <annevk>
Zakim is the telcon bot

[08:41:00.0000] <MikeSmith>
that message from Zakim prolly just means it was expecting the word "to"

[08:42:00.0000] <MikeSmith>
q+ to talk about blah

[08:42:01.0000] <MikeSmith>
instead of

[08:42:02.0000] <MikeSmith>
q+ talk about blah

[08:43:00.0000] <MikeSmith>
(q+ is command to tell Zakim you want to added to the speaker queue during a telcon)

[08:43:01.0000] <MikeSmith>
s/to added/to be added/

[08:50:00.0000] <annevk>
This RDF nonsense makes role= confusing

[09:30:00.0000] <Philip`>
Hmm, how odd - Safari (at least in old versions) seems to fail if I define a "function native(...)", while no other browser minds :-(

[09:31:00.0000] <annevk>
-> #webkit

[09:34:00.0000] <hasather>
Philip`: they probably don't allow future reserved words either then

[11:20:00.0000] <Hixie>
/me looks at the responses to his e-mail on www-html and decides to just leave it alone

[11:21:00.0000] <Hixie>
i don't understand why people think that something that the spec says is non-conformant is somehow a "feature"

[11:23:00.0000] <annevk>
Bjoern had a nice response

[11:23:01.0000] <Hixie>
not really

[11:23:02.0000] <Hixie>
it was a "look at me, i'm so clever" response that totally missed the point

[11:29:00.0000] <annevk>
/me filed a bug on something Acid3 tests today

[11:29:01.0000] <Hixie>
heh

[11:29:02.0000] <Hixie>
which one?

[11:29:03.0000] <annevk>
exposing DOCTYPE in text/html

[11:29:04.0000] <Hixie>
ah

[11:30:00.0000] <annevk>
I think we got bugs on the other failures

[11:30:01.0000] <Hixie>
probably filed by me :-P

[11:31:00.0000] <Hixie>
   * Presentational elements are kept. This includes the HR element,

[11:31:01.0000] <Hixie>
     SMALL, B, and I. This is 2007; NONE of them should be kept.

[11:31:02.0000] <Hixie>
/me wonders how those are presentational anymore

[11:32:00.0000] <Hixie>
mind you the same person thought <m> was presentational

[11:32:01.0000] <Hixie>
so...

[11:33:00.0000] <Hixie>
maybe they're confused and think "presentational" means "easy" and "semantic" means "hard"

[11:33:01.0000] <Hixie>
that would explain it

[11:33:02.0000] <Hixie>
lol the same person wants us to remove indeterminate progress bars

[11:34:00.0000] <annevk>
I was just reading that e-mail

[11:34:01.0000] <annevk>
I think I agree with his last point

[11:34:02.0000] <annevk>
but I don't have a better solution

[11:35:00.0000] <annevk>
(last point of the list)

[11:35:01.0000] <annevk>
http://www.w3.org/mid/tkrat.c998060f76b1e71c⊙gn is the e-mail, fwiw

[11:36:00.0000] <Hixie>
not sure why APIs shouldn't be defined

[11:36:01.0000] <Hixie>
it's not a markup language

[11:36:02.0000] <Hixie>
it's "web apps 1.0"

[11:37:00.0000] <annevk>
The APIs that are markup language neutral

[11:38:00.0000] <Hixie>
i guess i don't really care about "language neutral". the web is html.

[11:39:00.0000] <Philip`>
It's not WA1 - it's WA1-in-the-context-of-becoming-HTML5 (at least in that email discussion), so the HTML name gives some justification for wanting it to be a markup language

[11:43:00.0000] <Hixie>
i guess my point of view on that is that i have bigger fish to fry than trying to get the ideal specification silos sorted out

[11:54:00.0000] <annevk>
Just got an e-mail that says the following:

[11:54:01.0000] <annevk>
Would save the trouble of going through and adding alt="" to all the

[11:54:02.0000] <annevk>
non-content elements when validating. The alt attribute should be

[11:54:03.0000] <annevk>
automatically applied to <object> and <img> at least.

[11:54:04.0000] <annevk>
(plus quoting my last message to whatwg⊙wo

[12:56:00.0000] <annevk>
Hixie, Google was a typo? :)

[12:56:01.0000] <annevk>
being picky, that comment should probably say icon="y.png" now you changed / fixed the name of the company

[12:57:00.0000] <annevk>
oh, and it's "Yahoo!"

[13:21:00.0000] <Philip`>
Hixie: Was "Fix a couple of typos" really meant to change the hex pattern to say "<!1DOCTYPE HTML"?

[14:18:00.0000] <Hixie>
annevk: it's a comment. deal with it. :-P (i only changed it to help track down a bug with the postprocessor)

[14:22:00.0000] <Hixie>
i love this sentence from htm5: "Implementations that use ECMAScript to implement the APIs defined in this specification must implement them in a manner consistent with the ECMAScript Bindings for DOM Specifications specification, as this specification uses that specification's terminology. [EBFD]"

[14:22:01.0000] <Hixie>
/me notes EBFD doesn't exist yet

[14:24:00.0000] <annevk>
House on ice

[14:34:00.0000] <nickshanks>
what was with that Tina woman?

[14:34:01.0000] <nickshanks>
on the mailing list

[14:36:00.0000] <bewest>
opinionated

[14:36:01.0000] <bewest>
like everyone else :-)

[14:39:00.0000] <annevk>
hah, encoding error on http://www.greytower.net/about/

[14:40:00.0000] <bewest>
she does seem to have strange notions of "pragmatic" and what consitutes a sane solution

[14:40:01.0000] <bewest>
evidently billions of previously authored pages don't constitute evidence suitable for decision making

[14:42:00.0000] <nickshanks>
bewest: i think a combination of carrot and stick is needed to get web developers to fix things like raw ampersands in documents.

[14:43:00.0000] <nickshanks>
of course that depends on your definition of 'broken' and 'fixed' in this context

[14:44:00.0000] <nickshanks>
people here seem to prefer the view that the web is correct and the specs are broken, not the other way around.

[14:45:00.0000] <annevk>
I would think that most people here think both are broken

[14:45:01.0000] <Hixie>
indeed

[14:45:02.0000] <annevk>
But realize the web is way harder to fix

[14:45:03.0000] <Hixie>
surely to fix the web we just have to educate the web developers

[14:46:00.0000] <annevk>
They're not the only people creating content, but yes

[14:46:01.0000] <annevk>
Reducing the need to rely on browser specific features is another important thing

[14:46:02.0000] <jgraham>
Everything is broken / everyone is broken...

[14:47:00.0000] <annevk>
Fostering interop and such

[14:47:01.0000] <Hixie>
(i was being sarcastic)

[14:47:02.0000] <annevk>
oh, duh :)

[14:47:03.0000] <annevk>
/me was wondering about it

[14:48:00.0000] <Hixie>
you should get your sarcasm detector checked, i think it's off-calibration ;-)

[14:48:01.0000] <annevk>
in my defense, I still feel jetlagged

[14:48:02.0000] <annevk>
/me should probably get some more sleep

[14:49:00.0000] <Hixie>
:-)

[14:49:01.0000] <jgraham>
So, if I wanted to find out how a HTML5 DOM tree for some content looked in an existing browser what would be the most sane way to do it?

[14:49:02.0000] <jgraham>
/me was thinking generating a script that built the DOM

[14:49:03.0000] <annevk>
http://software.hixie.ch/utilities/js/live-dom-viewer/

[14:49:04.0000] <Hixie>
http://software.hixie.ch/utilities/js/live-dom-viewer/

[14:49:05.0000] <Hixie>
hah

[14:49:06.0000] <annevk>
I'm still fast though

[14:49:07.0000] <annevk>
:p

[14:49:08.0000] <Hixie>
i beat you that time

[14:50:00.0000] <annevk>
oh, not in my client...

[14:50:01.0000] <annevk>
weird

[14:50:02.0000] <Hixie>
oh, actually, the logs say you did

[14:50:03.0000] <Hixie>
nevermind :-)

[14:50:04.0000] <bewest>
yeah, annevk won over here too

[14:50:05.0000] <Hixie>
clearly google should upgrade from our modem connection to dsl

[14:51:00.0000] <bewest>
btw, I read some of annevk's and maciej's work on sunday... it was pretty good

[14:51:01.0000] <bewest>
Window

[14:51:02.0000] <bewest>
and

[14:51:03.0000] <bewest>
uh

[14:51:04.0000] <jgraham>
Does that actually do what I want? I want to take some content, parse it with the HTML5 algorithm and then see how the browser redering of the resulting tree would compare with the rendering of the same source using the browser's native parser

[14:51:05.0000] <bewest>
evidenty by "read" I mean scanned

[14:51:06.0000] <Hixie>
jgraham: oh, i misunderstood

[14:51:07.0000] <othermaciej>
I need to update Window RSN

[14:52:00.0000] <jgraham>
I don't know if it's a *useful* think to want to do but it seems like one way to see how much HTML5 parsing breaks

[14:52:01.0000] <jgraham>
s/think/thing/

[14:52:02.0000] <Hixie>
jgraham: you'd want something similar, but which had two iframes, one which just got the given markup rendered straight in, and the other which round-tripped through an html5lib script to convert it to valid markup and sent that back

[14:52:03.0000] <Hixie>
jgraham: it would be extremely useful to have, yes

[14:53:00.0000] <Hixie>
othermaciej: fwiw i've collapsed most of the window spec back into the html5 spec

[14:53:01.0000] <Hixie>
othermaciej: there were many things that i needed to define that were just too closely tied to things Window defined

[14:53:02.0000] <annevk>
we should have html5lib in ECMAScript...

[14:53:03.0000] <Hixie>
othermaciej: e.g. session history, navigation, content sniffing, etc

[14:53:04.0000] <Hixie>
browsing contexts

[14:53:05.0000] <Hixie>
javascript security model

[14:54:00.0000] <othermaciej>
Hixie: yeah, that makes me feel less concerned, OTOH I still would like to help SVG and CDF with crack removal

[14:54:01.0000] <othermaciej>
(then the trick is making sure what they say is compatible)

[14:54:02.0000] <othermaciej>
if Web Apps 1.0 becomes a W3C spec, I can freely plagiarize

[14:54:03.0000] <Hixie>
yeah, dunno how to solve it really. maybe they should just require html support ;-)

[14:54:04.0000] <Hixie>
or they should keep out of that level altogether

[14:55:00.0000] <annevk>
so the thing is that "HTML5" not just defines a markup language but also the web navigation, security etc. model

[14:56:00.0000] <Hixie>
it _is_ called Web Applications 1.0

[14:56:01.0000] <annevk>
right

[14:56:02.0000] <othermaciej>
Hixie: btw, have you thought about the tension between history state objects and ability to have a URI for your current state that you can bookmark or send to others?

[14:56:03.0000] <othermaciej>
it seems like they solve the back problem, but not the bookmark or send problem

[14:56:04.0000] <annevk>
I suppose this ok since to support the web you have to support all of that, but some people would like those features to be implementable without having to implement HTML5

[14:56:05.0000] <Hixie>
othermaciej: yeah, there's a red box in there about it. i think pushState() will need two more arguments, URL and title

[14:57:00.0000] <Hixie>
othermaciej: the problem is with whether it raises any security or usability issues, since the URI might not represent the state at all

[14:57:01.0000] <Hixie>
annevk: yeah, i'd love it if they weren't so closely related

[14:58:00.0000] <Hixie>
annevk: that doesn't mean it's easy to do

[14:58:01.0000] <othermaciej>
Hixie: not being able to change the protocol, host or port from the real one probably addresses the security issues

[14:58:02.0000] <Hixie>
probably

[14:58:03.0000] <othermaciej>
agreed there could be usability issues

[14:58:04.0000] <Hixie>
but i need to make sure i look at it very closely before i spec it

[14:59:00.0000] <Hixie>
and i havnen't gotten around to it yet

[14:59:01.0000] <jgraham>
Hixie: just rendering the output of html5lib doesn't work in general e.g. http://wordsandpictures.dyndns.org/cgi-bin/parsetree/parsetree.py?uri=&source=%3C%21Doctype+html%3E%0D%0A%3Chtml%3E%0D%0A%3Chead%3E%0D%0A%3Cstyle%3E%0D%0Afoo+p+%7Bcolor%3Agreen%3B%7D%3E%0D%0A%3C%2Fstyle%3E%0D%0A%3C%2Fhead%3E%0D%0A%3Cbody%3E%0D%0A%3Cfoo%3E%0D%0A%3Cp%3EThis+should+be+green%3C%2Fp%3E%0D%0A%3C%2Ffoo%3E%0D%0A%3C%2Fbody%3E%0D%0A%3C%2Fhtml%3E (click View in B

[14:59:02.0000] <jgraham>
rowser)

[14:59:03.0000] <othermaciej>
the current de facto approach to this is using a fragment ID

[14:59:04.0000] <Hixie>
it's on my radar for this quarter, fwiw

[14:59:05.0000] <annevk>
you can also have usability issues by just giving someone a URL which does browser sniffing

[14:59:06.0000] <othermaciej>
Hixie: ok, noted

[14:59:07.0000] <jgraham>
(In firefox)

[14:59:08.0000] <Hixie>
(but let me know if it becomes more urgent for you and it'll become on my radar for the week in question)

[15:00:00.0000] <othermaciej>
it's not immediately urgent, I just thought of it because you mentioned session history

[15:00:01.0000] <Hixie>
jgraham: good point

[15:00:02.0000] <Hixie>
jgraham: maybe write a script that recreates the DOM using JS?

[15:01:00.0000] <jgraham>
Right. I guess that's the way forward :)

[15:01:01.0000] <Hixie>
let me know if there's anything i can do to help, short of actually writing it, because this is something i'd really love to have

[15:01:02.0000] <Hixie>
i encourage you to check it into the html5 repo :-)

[15:02:00.0000] <annevk>
does <video> already cover video-support-disabled, can't-do-video, etc.?

[15:02:01.0000] <annevk>
it doesn't seem like it, but maybe I missed something

[15:02:02.0000] <Hixie>
what do you mean by "cover"?

[15:02:03.0000] <annevk>
tell implementors what to do

[15:02:04.0000] <annevk>
it seems to just say that video should always be rendered

[15:03:00.0000] <Hixie>
if the implementor supports <video>, then yes, it has to be rendered (though it might not play until the user starts it).

[15:03:01.0000] <annevk>
where for Lynx it might make sense to offer a download link or something

[15:04:00.0000] <Hixie>
if you think the spec isn't clear enough, or requires something bad, send a mail with your suggestion, i'll add it to the list 

[15:04:01.0000] <Hixie>
i think you're right and i haven't yet really taken care of handling alternative <video>/<audio> modes

[15:17:00.0000] <annevk>
btw, you could also modify Hixie's dom viewer script to output the same serialization as html5lib

[15:18:00.0000] <annevk>
yet another way would be adding a new flag to html5lib to output the hixie html format

[15:19:00.0000] <annevk>
that shouldn't be that hard either

[15:46:00.0000] <fantasai>
"User agents should net render elements" s/net/not/

[16:10:00.0000] <Hixie>
hm, cool, tbl said he'd be on the call tomorrow

[16:12:00.0000] <bewest>
hrm email says 5pm. I wonder which timezone

[16:13:00.0000] <bewest>
oh I see

[16:13:01.0000] <bewest>
who's invited?

[16:13:02.0000] <bewest>
/me is guessing invited experts aren't invited in general

[16:13:03.0000] <Hixie>
didn't you get the mail on the list?

[16:13:04.0000] <bewest>
I'm looking at list email

[16:13:05.0000] <Hixie>
http://lists.w3.org/Archives/Public/public-html/2007Apr/1458.html

[16:14:00.0000] <Hixie>
http://www.w3.org/2002/09/wbs/40318/tel26Apr/

[16:14:01.0000] <gavin_>
to more directly answer your question, and WG member is invited

[16:14:02.0000] <gavin_>
that includes Invited Experts

[16:15:00.0000] <bewest>
yeah, I'm looking at that email... following the references now

[16:15:01.0000] <bewest>
oh WG Participants are invited

[16:15:02.0000] <bewest>
I thought member means something special

[16:16:00.0000] <Dashiva>
Member with capital m does

[16:16:01.0000] <bewest>
ok thanks

[16:16:02.0000] <Dashiva>
W3C Member, as opposed to WG members

[16:20:00.0000] <bewest>
Hixie: when someone submits some feedback, there are basically two possibilities, correct? 1.) you respond relatively quickly  or 2.) it goes on some queue 

[16:21:00.0000] <bewest>
if there is no immediate response, is there some way to know if my feedback has been noticed and that I should simply wait?

[16:21:01.0000] <Hixie>
it's always noticed

[16:21:02.0000] <bewest>
ah

[16:21:03.0000] <Hixie>
but if you really want me to check, i can check :-)

[16:22:00.0000] <Hixie>
(anything sent to whatwg⊙wo that isn't replied to and isn't inane ends up in my folders)

[16:23:00.0000] <bewest>
ah

[16:23:01.0000] <bewest>
I'm specifically talking about http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-April/010924.html

[16:24:00.0000] <Hixie>
yeah, that's e-mail 109 of 119 in INBOX.input-for-whatwg-WF2

[16:24:01.0000] <bewest>
ok

[16:25:00.0000] <ajnewbold_>
/me sends the latest hot pharmaceutical offer to whatwg⊙wo

[16:25:01.0000] <Hixie>
qv. inane. :-)

[16:25:02.0000] <ajnewbold_>
:P


2007-04-26
[01:24:00.0000] <annevk>
Flex is open source now...

[01:25:00.0000] <annevk>
Product competition: Company A does C. Really big company B starts doing C as well. A makes their version of C open source.

[01:28:00.0000] <virtuelv>
annevk: well, I don't really think Really big company B will do the same

[02:36:00.0000] <hsivonen>
annevk: whoa! URL to Flex announcement? what about Flash player?

[02:36:01.0000] <annevk>
http://labs.adobe.com/wiki/index.php/Flex:Open_Source

[02:36:02.0000] <annevk>
(not sure about announcements)

[02:36:03.0000] <annevk>
(not sure about Flash either)

[02:36:04.0000] <annevk>
MPL for those who care

[02:37:00.0000] <hsivonen>
/me hopes they mean the tri-license like Dirac when they say MPL

[02:38:00.0000] <hsivonen>
annevk: thanks

[02:38:01.0000] <annevk>
not sure about that

[02:39:00.0000] <annevk>
but i didn't investigate much

[02:39:01.0000] <hsivonen>
considering that Flash player in itself is not sold, Flash benefits from maximum deployment, MS is doing Silverlight and GNU Gnash is getting serious, it would make sense to open-source Flash Player

[02:39:02.0000] <annevk>
you are mistaken

[02:39:03.0000] <annevk>
Flash is sold

[02:40:00.0000] <hsivonen>
annevk: the player? to handset makers?

[02:40:01.0000] <annevk>
yes, yes

[02:41:00.0000] <othermaciej>
so what exactly does Flex do?

[02:42:00.0000] <hsivonen>
othermaciej: dunno exactly. looks like a Flash app development kit.

[02:42:01.0000] <othermaciej>
but it has a markup lanaguage

[02:42:02.0000] <othermaciej>
and isn't just the normal flash authoring

[02:42:03.0000] <othermaciej>
afaik

[02:43:00.0000] <hsivonen>
hmm. looks like they are releasing parts that their Flex customers would see and build on anyway

[02:44:00.0000] <krijnh>
MXML!

[02:44:01.0000] <othermaciej>
corporate-driven open source is hard

[02:44:02.0000] <annevk>
WebKit is not?

[02:45:00.0000] <annevk>
or Mozilla for that matter... or do you mean something else?

[02:46:00.0000] <othermaciej>
WebKit and Mozilla are hard, yes

[02:46:01.0000] <othermaciej>
and I think Adobe has less clue about how to run an open source project

[02:46:02.0000] <krijnh>
So now we can all use Flex for free?

[02:48:00.0000] <hsivonen>
krijnh: only the libs and the compiler, it seems. The IDE will stay proprietary

[02:48:01.0000] <krijnh>
We can write MXML ourselves, don't we? :)

[03:47:00.0000] <met_>
my friend David Majda prepared 5 > 2 wallpapers http://whatwg.majda.cz/wallpapers/

[03:48:00.0000] <annevk>
lol, deployed

[03:49:00.0000] <met_>
my destop  already has it 8-)

[03:49:01.0000] <met_>
there is also svg original

[04:05:00.0000] <a-ja>
hsivonen: ping

[04:50:00.0000] <Lachy>
www-html is just getting funnier every day :-) http://www.w3.org/mid/FBABED17-C3FA-4AB8-8942-4BD169FE298A⊙nc

[04:51:00.0000] <nickshanks>
hilarious isnt it?

[04:52:00.0000] <Lachy>
his argument is self defeating

[04:53:00.0000] <Lachy>
"... and moreover, that [versioning] cannot be retro-actively be applied to an extant format."

[04:53:01.0000] <nickshanks>
his = "Philip & Le Khanh" or me?

[04:53:02.0000] <annevk>
you, I think

[04:53:03.0000] <Lachy>
oh, that's you!  Yes, your argument :-)

[04:53:04.0000] <krijnh>
lol

[04:53:05.0000] <annevk>
but I'm not sure Lachy realized that

[04:53:06.0000] <annevk>
right

[04:53:07.0000] <Lachy>
I didn't notice the name on the email

[04:54:00.0000] <annevk>
/me guessed it by the link Lachy pasted

[04:54:01.0000] <annevk>
users don't blame sites very often

[04:54:02.0000] <annevk>
they are not actually aware that sites are the problem

[04:55:00.0000] <zcorpan>
 really?

[04:55:01.0000] <nickshanks>
then make users aware "This website is broekn, it's not an MSIE probelem"

[04:55:02.0000] <zcorpan>
i hear general complaints about sites sucking or not working all the time

[04:55:03.0000] <Lachy>
nickshanks, attempting to introduce draconian error handling like you suggest into HTML5 simply wouldn't work, even if it were a good idea

[04:55:04.0000] <nickshanks>
gah, can't spell :)

[04:57:00.0000] <nickshanks>
whilst it doesn't have to prevent the site for being displayed, it helps if it is inconvenient for the owners and preferably insults them soehow

[04:57:01.0000] <Lachy>
but your solution affects the end users, not just developers

[04:57:02.0000] <annevk>
sounds lovely

[04:58:00.0000] <Lachy>
give developers tools and browser extensions that notify them of such errors, but don't inflict such errors upon unsuspecting users

[04:59:00.0000] <annevk>
halting problem makes lots of errors unlikely to be catched anyway

[04:59:01.0000] <Lachy>
a lot of errors can be caught by tools, though

[05:00:00.0000] <annevk>
sure

[05:00:01.0000] <nickshanks>
develoeprs shouldn't need seperate rools

[05:00:02.0000] <zcorpan>
wow, first spam since we modified the register page

[05:00:03.0000] <zcorpan>
in the forums

[05:00:04.0000] <annevk>
nickshanks, what do you mean with separate?

[05:00:05.0000] <Lachy>
nickshanks, end users don't need developer tools in their browsers

[05:01:00.0000] <annevk>
people should make it more clear I think that quirks/standards is just an if else switch at some specific places in the code

[05:01:01.0000] <nickshanks>
the browser *should* inconvenience users, or at least make it clear to them that the owners of the website are stupid

[05:01:02.0000] <annevk>
not an if else switch for the entire rendering engine

[05:01:03.0000] <nickshanks>
then the website owners will be inundated with complaints, and they will fix it!

[05:01:04.0000] <annevk>
nickshanks, maybe in nickshanks pipe dream land

[05:02:00.0000] <Lachy>
nickshanks, why?  You won't help anyone, just inconvenience users and tech support staff

[05:02:01.0000] <annevk>
handling errors is pretty trivial anyway

[05:02:02.0000] <nickshanks>
for 1 day, yes, then once the error is corrected there won't be a problem anymore

[05:02:03.0000] <annevk>
errors are not really the problem

[05:03:00.0000] <annevk>
the problem is browser sniffing

[05:03:01.0000] <nickshanks>
and most importantly, web devs will learn to write better code

[05:03:02.0000] <annevk>
and browser sniffing we won't get around by silly editing tools because it's needed as long as we don't get better interop

[05:03:03.0000] <virtuelv>
hm. there is one thing irking me with new elements

[05:03:04.0000] <virtuelv>
given <unknwonelement><p>Other Element</p></unknownelement>

[05:04:00.0000] <annevk>
virtuelv, that should give the DOM you want

[05:04:01.0000] <Lachy>
just look at the kind of nonsense that end users ask when presented with errors they don't understand http://lists.w3.org/Archives/Public/www-html-editor/2007JanMar/0053.html

[05:04:02.0000] <virtuelv>
annevk: not in IE

[05:04:03.0000] <annevk>
virtuelv, right

[05:04:04.0000] <annevk>
that's indeed problematic

[05:04:05.0000] <annevk>
but workarounds have been found (you made one yourself, no?)

[05:04:06.0000] <virtuelv>
you get unknownelement, p #text, /unknownelement

[05:04:07.0000] <nickshanks>
Lachy: well the task of writing error messages that peasants can understand i leave to someone else

[05:05:00.0000] <virtuelv>
annevk: if you're talking about my DOM autocorrector, that's a horrible, horrible, horrible hack not deemed fit for any consumption

[05:05:01.0000] <Lachy>
nickshanks, you can't just present an idea and expect someone else to do all the work

[05:05:02.0000] <annevk>
the idea that things can always be error free is just flawed imo

[05:05:03.0000] <virtuelv>
it involves recreating the entire DOM where elements whose names start with / occur

[05:05:04.0000] <annevk>
virtuelv, people have put more horrible hacks in real world usage

[05:05:05.0000] <nickshanks>
Lachy: i'm happy to do the work if a web vendor will hire me :)

[05:06:00.0000] <annevk>
if it's such a great idea maybe you should implement it yourself and earn millions...

[05:06:01.0000] <annevk>
</sarcasm>

[05:06:02.0000] <Lachy>
you could always write up a detailed proposal, explain the pros and cons of the ideas and do research to support your claims.

[05:07:00.0000] <nickshanks>
strictness has to be implemented in existing browsers, basically MSIE and Firefox, others can follow later

[05:07:01.0000] <Lachy>
If you can prove beyond all reasonable doubt that browsers should implement your idea, it might get done

[05:07:02.0000] <nickshanks>
Lachy: i was hoping the common sense in my argument would be enough

[05:07:03.0000] <Lachy>
what common sense?

[05:07:04.0000] <zcorpan>
98 new messages in one day :| seems it just gets more and more... 

[05:07:05.0000] <annevk>
98?

[05:07:06.0000] <annevk>
/me wonders if he missed them all

[05:08:00.0000] <zcorpan>
that's what i'm fetching now

[05:08:01.0000] <Lachy>
maybe zcorpan's spam filter blocked out some of the more annoying contributors :-)

[05:08:02.0000] <nickshanks>
Lachy: that if we let web developer continue producing tag soup, and not actively prevent them from reaching their customers until they conform, web developers will continue to output crap HTML 5, 6, 7 etc

[05:08:03.0000] <zcorpan>
Lachy: if only :)

[05:09:00.0000] <Lachy>
oh, 1495 mails on public-html this month!

[05:09:01.0000] <zcorpan>
a 0 decisions made so far, right?

[05:09:02.0000] <zcorpan>
s/a /and /

[05:10:00.0000] <Lachy>
/me checks his tally

[05:10:01.0000] <Lachy>
yep, 0!

[05:10:02.0000] <zcorpan>
now that's productive

[05:10:03.0000] <annevk>
nickshanks, just like browsers will continue with crap interop for the newer features

[05:11:00.0000] <annevk>
although <canvas> interop is way better than <object>, to be fair

[05:12:00.0000] <Lachy>
nickshanks, your argument seems to make the false assumption that all HTML is produced by competent web developers

[05:12:01.0000] <welly>
lol

[05:12:02.0000] <nickshanks>
i think going forward vendor interop will be much better than before, the problem lies with HTML authors not realising they are making mistakes because their web browser doesn't yell at them - to many devs, opening the page in IE *is* validation of their efforts

[05:13:00.0000] <annevk>
nickshanks, having some weird error handling on syntax doesn't help with that

[05:13:01.0000] <Lachy>
nickshanks, yep

[05:14:00.0000] <Lachy>
nickshanks, the best we can do is encourage authoring tools to produce better content and promote web standards more among professional web developers

[05:15:00.0000] <nickshanks>
i disagree: the best we can do is persuade MS that silent recovery from errors is detrimental to both users and IE itself in the longer term

[05:15:01.0000] <annevk>
that's not the problem with IE

[05:15:02.0000] <annevk>
the problem is that web authors do browser sniffing

[05:15:03.0000] <annevk>
(and have to, atm)

[05:16:00.0000] <nickshanks>
the problem is it's 80% market share? ;)

[05:16:01.0000] <annevk>
no

[05:16:02.0000] <Lachy>
how is silent recovery detrimental to users?

[05:16:03.0000] <annevk>
the problem is lack of interop

[05:16:04.0000] <annevk>
error recovery is just some silly side debate  that's hardly relevant imo

[05:16:05.0000] <Lachy>
getting interop in browsers for all content, whether it's valid or not, is a much more realisting approach than saying they should reject invalid content

[05:17:00.0000] <Lachy>
s/realisting/realistic/

[05:17:01.0000] <nickshanks>
Lachy: because when a web dev only tests in IE, and doesn't see that he has errors, he won't know that other browsers handle things correctly and what he sees is incorrect

[05:17:02.0000] <Lachy>
that's why we need interop!

[05:17:03.0000] <Lachy>
not why we need draconian error handlnig

[05:18:00.0000] <nickshanks>
i would rather have both

[05:18:01.0000] <zcorpan>
nickshanks: use xml

[05:18:02.0000] <Lachy>
and then serialise as HTML5 before sending to the end users

[05:18:03.0000] <nickshanks>
zcorpan: i can't force the creators of every website I visit to do that

[05:19:00.0000] <annevk>
you can't force them to use your utopia language either

[05:19:01.0000] <Lachy>
nickshanks, why do you want to force your development tools and opinions on the creators of every web site

[05:19:02.0000] <zcorpan>
right. if you bring drocanian error handling to html, how do you force the creators of every website I visit to fix their errors?

[05:19:03.0000] <Lachy>
how does someone else's broken code affect you?

[05:19:04.0000] <nickshanks>
not my tools, they can use any tools they like (as long as they are black)

[05:20:00.0000] <annevk>
you're just being silly

[05:20:01.0000] <virtuelv>
"nickshanks> Lachy: that if we let web developer continue producing tag soup, and not actively prevent them from reaching their customers until they conform, web developers will continue to output crap HTML 5, 6, 7 etc"

[05:20:02.0000] <nickshanks>
Lachy: because websites don't work in Safari, but work in MSIE which I can't use

[05:20:03.0000] <virtuelv>
nickshanks: they will continue outputting crap no matter what we tell them

[05:20:04.0000] <Lachy>
nickshanks, what does the colour of the tool have to do with anything?

[05:20:05.0000] <Lachy>
/me doesn't understand "as long as they are black"

[05:20:06.0000] <nickshanks>
virtuelv: you miss my point, it's not what *we* tell them, it's what the browser they test in tells them

[05:20:07.0000] <virtuelv>
and if we go to draconian error handling, all we do is lose content contributed to the web

[05:20:08.0000] <annevk>
nickshanks, that has to do with browser sniffing

[05:21:00.0000] <annevk>
nickshanks, not with IE error correction, most likely

[05:21:01.0000] <virtuelv>
nickshanks: and I'm saying it won't work

[05:21:02.0000] <annevk>
you should try to find out what the problem is first

[05:21:03.0000] <Lachy>
nickshanks, that's to do with lack of interop

[05:21:04.0000] <annevk>
before screaming you found the perfect solution

[05:21:05.0000] <nickshanks>
Lachy: it's a refernce to the Model T Ford, , Henry Ford said "You can have any colour you like, as long as it's black." (which was the only colour he offered it in)

[05:21:06.0000] <zcorpan>
/me goes read his mail instead

[05:22:00.0000] <Lachy>
/me suggests that nickshanks join the XHTML2WG, who share similar ideas

[05:23:00.0000] <annevk>
good point

[05:23:01.0000] <nickshanks>
it doesn't have to prevent display of the site XML-style, just display enough of an error message that CEOs won't allow their web developers to get away with it

[05:24:00.0000] <annevk>
IE has done that for quite some time

[05:24:01.0000] <annevk>
and that has been widely ignored

[05:24:02.0000] <Lachy>
yes, with javascript errors

[05:24:03.0000] <Lachy>
and so many sites still have errors

[05:25:00.0000] <nickshanks>
then it doesn't inconvenience a website's customers enough

[05:25:01.0000] <Lachy>
why do you want to hurt the customers?

[05:25:02.0000] <nickshanks>
because that's unacceptable to the people in control

[05:25:03.0000] <nickshanks>
so they won't let it happen on their site

[05:26:00.0000] <nickshanks>
(i am mostly thinking of commercial sites here)

[05:26:01.0000] <Lachy>
browsers will not implement such a user hostile approach

[05:27:00.0000] <nickshanks>
(since they drive a lot of the money in web development)

[05:27:01.0000] <nickshanks>
Lachy: that's why it needs to be co-ordinated and mandatory (both for vendors and users)

[05:27:02.0000] <annevk>
you can't mandate UI

[05:27:03.0000] <Lachy>
inflicting error messages upon end users is simply unacceptable

[05:28:00.0000] <annevk>
that browsers have implemented the UI for XML that they did, well... dunno why they did that

[05:28:01.0000] <Lachy>
unless you have a solution that doesn't involve innocent by-standers, this debate is over

[05:28:02.0000] <nickshanks>
would you rather limp around on a gangrenous leg for the rest of your life, or have it amputated before it gets infected?

[05:28:03.0000] <annevk>
I suppose it keeps people away from using XML

[05:28:04.0000] <nickshanks>
HTML will last for 100 years or more

[05:29:00.0000] <annevk>
yes, including all old content

[05:29:01.0000] <nickshanks>
anne: no, wrong

[05:29:02.0000] <Lachy>
nickshanks, I don't understand your analogy

[05:29:03.0000] <nickshanks>
most old content gets updated

[05:29:04.0000] <Lachy>
wrong!

[05:29:05.0000] <virtuelv>
nickshanks: most content on the web is static

[05:29:06.0000] <nickshanks>
a few years of hurt between 2008 and 2010 won't matter to people in 2050, but continuing to permit tag soup in perpetuity will

[05:30:00.0000] <Lachy>
/me doesn't think nickshanks understands all the implications of his own ideas

[05:30:01.0000] <nickshanks>
virtuelv: most content on the web going forward is generated by content management systems

[05:31:00.0000] <Lachy>
nickshanks, do you have evidence to support that claim?

[05:31:01.0000] <nickshanks>
updating the CMS to output better code will cause all pages it generates (including old content) to immediatly take on the new code

[05:32:00.0000] <Lachy>
I can assure you that there is a heck of a lot of content that is not output by a CMS

[05:32:01.0000] <nickshanks>
Lachy: obviously I haven't collected data from 2050, but lets draw an alalogy from the rail industry

[05:32:02.0000] <nickshanks>
there used to be many competing gauges of track that were interoperable

[05:32:03.0000] <Lachy>
nickshanks, I'm asking for evidence from the web, as it is *today*

[05:32:04.0000] <nickshanks>
they converged on one size

[05:33:00.0000] <Lachy>
what's your point?

[05:34:00.0000] <annevk>
so the other problem with your idea about showing an error for <input type=silly> is that we can then never introduce such a control in the future

[05:34:01.0000] <annevk>
because it would break older clients

[05:34:02.0000] <nickshanks>
we currently have many websites that use non-interoperable code, i want to get them to converge, rather than have browsers try and handle everything

[05:34:03.0000] <annevk>
silent error recovery implies a nice extensibility story

[05:34:04.0000] <Lachy>
nickshanks, pave the cowpaths

[05:35:00.0000] <annevk>
having silent error recovery for the syntax means the same thing

[05:35:01.0000] <nickshanks>
annevk: in a versioned document format you can, which is why I advocated versioning along with error display

[05:36:00.0000] <annevk>
versioning is a very silly idea

[05:36:01.0000] <annevk>
implementing 38 variants of HTML is just not nice

[05:36:02.0000] <Lachy>
nickshanks, versioning implies that UAs will implement different processing for each version, and that becomes unworkable because you get an increasing number of versions to maintain

[05:36:03.0000] <Lachy>
to see how harmful versioning is, see the 11 (or more) incompatible versions of RSS!

[05:36:04.0000] <nickshanks>
then you can start to drop older versions when they become uncommercial

[05:37:00.0000] <annevk>
i'd suggest you read a bit more into the subject before suggesting these type of things

[05:37:01.0000] <Lachy>
and how that led to liberal feed parsers that just don't care

[05:37:02.0000] <annevk>
nickshanks, drop support for content?!

[05:37:03.0000] <annevk>
while we can actually support it with a saner extensbility story?

[05:37:04.0000] <nickshanks>
PDF has versioning, Word has versioning, GIF has versioning, practuically every format under the sun does

[05:37:05.0000] <annevk>
which is easier for authors, user agents, etc.

[05:37:06.0000] <annevk>
nickshanks, that doesn't imply it's a good thing

[05:37:07.0000] <virtuelv>
nickshanks: have you even looked at the three most popular CMSes? 

[05:38:00.0000] <annevk>
Word especially is horrible

[05:38:01.0000] <virtuelv>
There's not a hint of XML in their toolchain, and its not even possible to plug in to the toolchains

[05:38:02.0000] <annevk>
to open older versions of Word files in Word 2007 you apparently need to change registry settings to circumvent security warnings

[05:38:03.0000] <virtuelv>
(That goes for blogger, Movable Type and in particular Wordpress

[05:38:04.0000] <nickshanks>
virtuelv: only wordpress, which i occasionally fix XHTML bugs on

[05:38:05.0000] <Lachy>
Word is prohibitively expensive, even for MS, to maintian with support for all those different versions

[05:38:06.0000] <virtuelv>
wordpress is a cancer

[05:39:00.0000] <annevk>
no

[05:39:01.0000] <annevk>
wordpress is development as its typically done

[05:39:02.0000] <annevk>
and probably much better than that

[05:39:03.0000] <annevk>
we have to take that into account

[05:39:04.0000] <annevk>
perfection is not attainable

[05:39:05.0000] <nickshanks>
virtuelv: but think about this - every site running wordpress can update to the latest version and have better quality HTML applied to all their old content

[05:39:06.0000] <virtuelv>
nickshanks: patentably false

[05:40:00.0000] <nickshanks>
uhh, no, that's a fact

[05:40:01.0000] <Lachy>
ah, no, it's not

[05:40:02.0000] <virtuelv>
there's bunches of wordpress-based hosting services where the users don't control the installs

[05:40:03.0000] <nickshanks>
i am not saying every site will, but they could

[05:40:04.0000] <virtuelv>
besides, upgrading wordpress is applying gold paint to a turd

[05:40:05.0000] <annevk>
heh

[05:40:06.0000] <Lachy>
/me is tired of this poinless debate

[05:40:07.0000] <nickshanks>
virtuelv: i'm not aguing about the merits of one CMS over others here

[05:40:08.0000] <Lachy>
*pointless

[05:41:00.0000] <virtuelv>
my point is still that "No, CMSes can't easily leverage XML, and in particular the popular end-user ones"

[05:42:00.0000] <nickshanks>
i'm not  talking about XML, just conformant HTML

[05:42:01.0000] <virtuelv>
that would inflict serious harm on the users of said tools

[05:42:02.0000] <virtuelv>
which means they would stay with the older versions.

[05:42:03.0000] <virtuelv>
eod

[05:42:04.0000] <annevk>
doesn't matter

[05:43:00.0000] <nickshanks>
how would making their HTML output more conformant cause any harm?

[05:43:01.0000] <annevk>
checking if something is conforming HTML is mathematically impossible btw

[05:43:02.0000] <mpt>
nickshanks, if you want Web browsers to switch to draconian error handling, I recommend you take it up with your elected representatives

[05:43:03.0000] <mpt>
encouraging them to introduce a law requiring that

[05:44:00.0000] <annevk>
maybe they can solve the halting problem :)

[05:44:01.0000] <mpt>
because in an unregulated browser market, it will never happen.

[05:44:02.0000] <nickshanks>
as i have said, draconian error handling is not necessary, just enough to make CEOs prevent their web developers from outputting tag soup

[05:44:03.0000] <annevk>
i'm still not sure why you think it's a good idea

[05:45:00.0000] <annevk>
what exactly is the problem we currently have you think would be solved by having this?

[05:45:01.0000] <krijnh>
Caring developers, and not caring developers

[05:45:02.0000] <krijnh>
Who cares

[05:46:00.0000] <annevk>
heh, stevenp promoting XForms on www-html

[05:46:01.0000] <mpt>
And no specification, whether from the W3C or the WhatWG or anyone else without force of law, can make it happen.

[05:47:00.0000] <mpt>
(I am not at all suggesting that draconian error handling is a good idea.)

[05:47:01.0000] <nickshanks>
annevk: two problems: browser interop, and future development of web specs being hindered by backwards compatibility issues (like that blasted alt tag!)  i don't subscribe to XHTML2's view of dropping backwards compatibility, i just want it to be easy to maintain going forward. currently it is not

[05:47:02.0000] <annevk>
alt tag?

[05:48:00.0000] <annevk>
1 can be solved by creating testsuites

[05:48:01.0000] <annevk>
2 not sure what you mean

[05:48:02.0000] <annevk>
seems that development of specs is going forward

[05:48:03.0000] <annevk>
see Web Applications 1.0

[05:48:04.0000] <nickshanks>
re alt tag: the fact that img is empty and so cannot have fallback content, thus we're stuck with the alt tag

[05:49:00.0000] <annevk>
there's no such thing is an alt tag

[05:49:01.0000] <nickshanks>
oh, bugger

[05:49:02.0000] <mpt>
That's what I was going to say

[05:49:03.0000] <annevk>
and the alt attribute works good enough

[05:49:04.0000] <nickshanks>
attribute :-)

[05:49:05.0000] <nickshanks>
i meant to say attribute

[05:49:06.0000] <mpt>
but then I thought, no, Hixie probably knows of a few occurrences of an alt tag :-)

[05:49:07.0000] <annevk>
heh

[05:49:08.0000] <nickshanks>
the alt attribute doesn't work good enough at all

[05:49:09.0000] <annevk>
Maciej proposed that <source> would be named <alt>

[05:50:00.0000] <annevk>
use <object>

[05:50:01.0000] <krijnh>
http://www.technorati.com/tag/alt <-- look, an alt tag :p

[05:50:02.0000] <mpt>
and will have some cheerful factoid such as "<alt> is used 146.2 times more often than <samp>"

[05:50:03.0000] <nickshanks>
lol @ krijnh 

[05:50:04.0000] <annevk>
no need to debate much about this

[05:50:05.0000] <annevk>
mpt, hehe

[05:51:00.0000] <mpt>
nickshanks, the poor fallback for <img> is almost entirely TimBL's fault

[05:51:01.0000] <nickshanks>
no, it's Marc Anddressen's fault

[05:51:02.0000] <nickshanks>
as is the "src" attribute. i like vowels damnit

[05:51:03.0000] <mpt>
nono

[05:52:00.0000] <nickshanks>
http://1997.webhistory.org/www.lists/www-talk.1993q1/0182.html

[05:52:01.0000] <mpt>
You see, if TimBL hadn't been so worried about people putting pictures of nekkid ladies on the Web, he would have introduced it himself, with a proper fallback model

[05:52:02.0000] <nickshanks>
oops, one 'd' two 'e's

[05:52:03.0000] <mpt>
and then marca wouldn't have needed to do it, and so wouldn't have done it hamfistedly

[05:53:00.0000] <nickshanks>
mpt: yes, but I lay most of the blame at marc's feet

[05:53:01.0000] <mpt>
History doesn't repeat itself, but it does rhyme: trying to prevent nekkid ladies from appearing on the Web, and trying to introduce draconian (or even partially draconian) error handling on the Web, are both examples of standardistas fighting economic forces

[05:54:00.0000] <nickshanks>
with hindsight I'm sure TimBL would have created object, image with fallback, video, audio and all the WA 1.0 elements we have now :P

[05:54:01.0000] <Lachy>
what's this about TimBL being against nekkid ladies on the web?

[05:54:02.0000] <nickshanks>
i don't blame him for not seeing what the web would become, no-one did

[05:55:00.0000] <annevk>
<A HREF="..." INCLUDE>See photo</A> is nice

[05:55:01.0000] <krijnh>
annevk: Nice for?

[05:56:00.0000] <nickshanks>
i don't think <A INCLUDE> had any more or less merit than <IFRAME>

[05:57:00.0000] <krijnh>
Ow, duh

[05:57:01.0000] <krijnh>
annevk: Never mind :)

[05:57:02.0000] <mpt>
"At a conference, Berners-Lee yelled at Andreessen, telling him that adding images to the Web was going to bring in a flood of new users who would do things like post photos of nude women. 'He was right,' Andreessen now says with a shrug."

[05:58:00.0000] <mpt>
http://www.usatoday.com/tech/news/2003-03-09-internet_x.htm

[05:58:01.0000] <nickshanks>
TimBL wouldn't have got a knighthood without internet porn

[05:59:00.0000] <annevk>
nickshanks, it's the backwards compatible alternative to <img>, <object>, etc. as seen from HTML1 perspective

[05:59:01.0000] <krijnh>
IE6 wouldn't be here without internet porn..

[05:59:02.0000] <Lachy>
heh, that's kind of evidence against Bruce Lawson's theory about TimBL

[05:59:03.0000] <mpt>
which is?

[05:59:04.0000] <Lachy>
look it up on brucelawson.co.uk

[05:59:05.0000] <Lachy>
http://www.brucelawson.co.uk/2005/internet-porn/

[06:00:00.0000] <mpt>
oh, found it

[06:00:01.0000] <mpt>
@#$^%!

[06:00:02.0000] <zcorpan>
annevk: but doesn't work nice with linked images

[06:01:00.0000] <mpt>
When HTML5 becomes widely used, remind me to add "m {everything: inherit}" to my user style sheet

[06:01:01.0000] <mpt>
These retarded sites who think I want them to highlight the Google search terms -- no, that's what Google Cache is for

[06:01:02.0000] <zcorpan>
mpt: you forgot !important

[06:02:00.0000] <nickshanks>
i think the "m" element has too short a name, it's not clear enough to people what it does

[06:02:01.0000] <annevk>
zcorpan, nested <a>

[06:02:02.0000] <Lachy>
mpt, that's also what Google Toolbar is for, and many other extensions

[06:02:03.0000] <mpt>
zcorpan, that'd be necessary only if the author set !important too, wouldn't it?

[06:02:04.0000] <Lachy>
mpt, no

[06:02:05.0000] <zcorpan>
mpt: no

[06:03:00.0000] <annevk>
mpt, yes

[06:03:01.0000] <Lachy>
the order goes user normal -> author normal -> author !imporant -> user !important

[06:03:02.0000] <Lachy>
doesn't it?

[06:03:03.0000] <annevk>
no

[06:03:04.0000] <Lachy>
/me should look it up

[06:03:05.0000] <annevk>
yes

[06:03:06.0000] <annevk>
user -> author -> author !important -> user !important

[06:04:00.0000] <Dashiva>
UA -> Author -> User -> Author imp -> user imp

[06:04:01.0000] <zcorpan>
authors style sheet wins over user stylesheet wins over ua style sheet, regardless of specificity... and !important can't be overridden

[06:04:02.0000] <Lachy>
annevk, that's what I said

[06:04:03.0000] <mpt>
ok, I was rusty on that part of the cascade

[06:04:04.0000] <annevk>
whoa

[06:04:05.0000] <Dashiva>
You switched user and author normal. I wonder if there are any UA rules with !important

[06:04:06.0000] <annevk>
/me messed it up again

[06:05:00.0000] <mpt>
though my current USS does use !important

[06:05:01.0000] <annevk>
Dashiva, there's no such thing as UA style sheet

[06:05:02.0000] <annevk>
Dashiva, although some UAs do have such a concept

[06:05:03.0000] <krijnh>
annevk: Don't follow my example to much plz k10x

[06:05:04.0000] <Dashiva>
Well, even if it's not a stylesheet, they do provide default styles

[06:05:05.0000] <Lachy>
Dashiva, UA don't have !important rules officially (though FF uses them for things that can't be overridden)

[06:05:06.0000] <zcorpan>
user style sheet doesn't need to be a style sheet either

[06:06:00.0000] <mpt>
input[type="file"] {... !important}, I guess

[06:06:01.0000] <annevk>
krijnh, heh, I got the theory right, but didn't match it accurately against the given case...

[06:06:02.0000] <Lachy>
Dashiva, you're the one who switch the user and author.  annevk and I got it right

[06:06:03.0000] <annevk>
which is better than the last time I talked about this part of the cascade...

[06:06:04.0000] <Lachy>
http://www.w3.org/TR/CSS21/cascade.html#cascading-order

[06:06:05.0000] <Dashiva>
Lachy: Yeah, I noticed just after writing it

[06:07:00.0000] <Dashiva>
Mixed up user and user agent :/

[06:07:01.0000] <annevk>
and more...

[06:09:00.0000] <annevk>
heh, bac then in 93 they also wanted to generalize

[06:09:01.0000] <annevk>
afraid of people proposing <aud> next week

[06:09:02.0000] <nickshanks>
annevk: what are you reading?

[06:09:03.0000] <annevk>
this thread: http://1997.webhistory.org/www.lists/www-talk.1993q1/0202.html

[06:10:00.0000] <annevk>
that e-mail contains probably the worst suggestion from an authoring perspective

[06:10:01.0000] <annevk>
from timbl...

[06:10:02.0000] <krijnh>
<aud> would be bad cause people could put sounds of nekkid women online ;p

[06:10:03.0000] <Lachy>
krijnh, accessible porn for the blind! :-)

[06:10:04.0000] <annevk>
SVG: http://1997.webhistory.org/www.lists/www-talk.1993q1/0209.html

[06:10:05.0000] <annevk>
(or VML, if you wish)

[06:11:00.0000] <krijnh>
<dialog><dt>Girl</dt><dd>Ah!</dd><dt>Boy</dt><dd>Oeh!</dd><dt>Girl2</dt><dd>Even more ah!</dd></dialog>

[06:11:01.0000] <krijnh>
Would be great!

[06:11:02.0000] <Lachy>
/me notes that there is actually an accessible porn site ;-)

[06:11:03.0000] <Lachy>
http://www.brucelawson.co.uk/index.php/2006/blind-accessible-porn/

[06:11:04.0000] <krijnh>
I had a possible client mail me for a porn site 2 weeks ago :]

[06:12:00.0000] <krijnh>
"Cause I knew how to make GoogleBot happy"

[06:12:01.0000] <mpt>
Then there's the porn sites proudly sporting "W3C valid HTML 4.01" and "W3C valid CSS" badges...

[06:12:02.0000] <krijnh>
I think I should have accepted it an build the first HTML5 porn site ;p

[06:13:00.0000] <nickshanks>
krijnh: i'll do it, i need some work

[06:13:01.0000] <Lachy>
zcorpan, you should make an Valid HTML5! logo specifically for porn sites.  (replace the cat with something slightly more appropriate)

[06:13:02.0000] <krijnh>
nickshanks: he hasn't mailed me back since :)

[06:13:03.0000] <krijnh>
Yeah, replace it with a pussy

[06:13:04.0000] <Lachy>
lol

[06:13:05.0000] <nickshanks>
send him another email for me

[06:14:00.0000] <nickshanks>
there's a pussy in the corner of ln.hixie.ch

[06:14:01.0000] <krijnh>
Ow, crap, this chat is logged in public ;]

[06:14:02.0000] <krijnh>
nickshanks: That's a cat

[06:14:03.0000] <Lachy>
nickshanks, that's the cat used on the Valid HTML5 logo already

[06:14:04.0000] <nickshanks>
replace it with a WebKitten ?

[06:15:00.0000] <Lachy>
/me notified twitter users of our discussion ;-)

[06:16:00.0000] <mpt>
-cringe-

[06:17:00.0000] <mpt>
Oh no, Lachy prefers Vegemite to Marmite -- we must now be sworn enemies

[06:18:00.0000] <Lachy>
mpt, yes

[06:18:01.0000] <annevk>
/me vaguely recalls he dislikes both

[06:18:02.0000] <nickshanks>
well he's an aussie, right?

[06:18:03.0000] <Lachy>
nickshanks, yes

[06:19:00.0000] <mpt>
So that's what this Twitter thing is

[06:19:01.0000] <nickshanks>
vegermite is the national food

[06:19:02.0000] <Lachy>
annevk, did you remember to buy a pack of tim tams and a jar or vegemite before you left Aus?

[06:19:03.0000] <mpt>
I just spent half a minute trying to figure out why people use it

[06:19:04.0000] <nickshanks>
(Lachy: it was a rhetoriccal question ;-)

[06:19:05.0000] <annevk>
Lachy, I got the cookies

[06:19:06.0000] <annevk>
Lachy, marcos gave them

[06:20:00.0000] <Lachy>
cool

[06:20:01.0000] <nickshanks>
i think twitter is some "listen in to everyone's conversations all over the world thing, because one IRC channel at a time is not enough for most people"

[06:20:02.0000] <annevk>
Lachy, I also got two toy animals (wonbat(sp?) and platypus) and a boomerang...

[06:20:03.0000] <annevk>
not sure what to do with those, but they looked funny

[06:21:00.0000] <Lachy>
*wombat

[06:21:01.0000] <annevk>
right

[06:21:02.0000] <mpt>
... And then I realized "ah, the same reason people write Weblogs, but worse"

[06:22:00.0000] <mpt>
Wow, Guido van Rossum chimed in on the "proposed new tag: IMG" thread

[06:22:01.0000] <mpt>
small world

[06:22:02.0000] <nickshanks>
who's he?

[06:23:00.0000] <annevk>
python

[06:23:01.0000] <nickshanks>
he's a snake. gotcha.

[06:23:02.0000] <mpt>
... bdfl

[06:24:00.0000] <mpt>
http://en.wikipedia.org/wiki/Guido_van_Rossum

[06:25:00.0000] <mpt>
annevk, iirc when British zoologists first received a sample of a platypus they thought it was a hoax

[06:26:00.0000] <mpt>
http://www.museumofhoaxes.com/platypus.html

[06:26:01.0000] <nickshanks>
heh, not surprising! what would you have thought?

[06:28:00.0000] <annevk>
first time I heard about the animal was at http://www.ragingplatypus.com/

[06:28:01.0000] <annevk>
"Raging Platypus - Geeks drink it for breakfast"

[06:28:02.0000] <nickshanks>
didn't you get taught at school about it?

[06:29:00.0000] <annevk>
don't think so

[06:30:00.0000] <nickshanks>
odd. being the only mammal to lay eggs it usually gets a mention in Biology classes here in the UK

[06:30:01.0000] <krijnh>
Blinky Bill!

[06:30:02.0000] <nickshanks>
btw: that site has an inordinate number of RSS links!

[06:31:00.0000] <annevk>
that's part of the joke, iirc

[06:55:00.0000] <annevk>
http://weblogs.mozillazine.org/roadmap/archives/2007/04/openness.html

[07:22:00.0000] <annevk>
what's the short URL for web forms 2?

[07:23:00.0000] <Philip`>
http://whatwg.org/wf2/ ?

[07:24:00.0000] <annevk>
oh, duh

[07:24:01.0000] <annevk>
/me was trying web-forms, web-forms-2, webforms2, web-forms2, etc.

[07:24:02.0000] <met_>
wf2 ?

[07:33:00.0000] <zcorpan>
Delivery to the following recipient failed permanently:

[07:33:01.0000] <zcorpan>
     commit-watchers-request⊙lwo

[07:33:02.0000] <zcorpan>
hm

[07:35:00.0000] <zcorpan>
confirming the subscription by following the link worked though

[07:37:00.0000] <Philip`>
http://lists.w3.org/Archives/Public/public-xhtml2/2007Apr/0032.html seems to be considering the same kind of use case as http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-April/010800.html

[07:42:00.0000] <annevk>
zcorpan, was reported on the list

[07:42:01.0000] <zcorpan>
Hixie: "We have three lists:" s/three/four/

[07:42:02.0000] <zcorpan>
annevk: yeah noticed

[07:43:00.0000] <zcorpan>
fwiw, "+1" to http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-April/011070.html

[07:44:00.0000] <Lachy>
yeah, I somewhat agree with that too

[07:44:01.0000] <annevk>
except for requiring a particular UI, sure

[07:44:02.0000] <Lachy>
perhaps the pragmatic decision is to make it conforming and just let UAs disable it (like they already do)

[07:45:00.0000] <Lachy>
disabling target=_blank is definately easier than disabling window.open

[07:45:01.0000] <zcorpan>
indeed

[07:45:02.0000] <Philip`>
zcorpan: Or, for improved ease of future extensibility, s/three/several/

[07:45:03.0000] <Lachy>
although I disable both, having .open() disabled causes problems on a few sites

[07:51:00.0000] <zcorpan>
annevk: seems firefox does "<html><body marginheight="0" marginwidth="0"><embed type="application/x-shockwave-flash" src="foo" name="plugin" height="100%" width="100%"></body></html>" (no HEAD)

[07:52:00.0000] <annevk>
wow, why name and type?

[07:52:01.0000] <zcorpan>
don't ask me :)

[07:53:00.0000] <annevk>
Opera uses style= for the <body> margin stuff

[07:53:01.0000] <annevk>
not that it matters

[07:57:00.0000] <annevk>
krijnh, http://krijnhoetmer.nl/stuff/html/input-type-image-value/ WF2 actually mandates that behavior...

[07:57:01.0000] <annevk>
krijnh, from MSIE

[07:57:02.0000] <annevk>
(and Opera)

[08:01:00.0000] <zcorpan>
krijnh: aren't the images presentational?

[08:02:00.0000] <zcorpan>
/me thought type=image was for server side image maps... but acknowledges that buttons are hard to style in some browsers

[08:02:01.0000] <Philip`>
krijnh: The "unobtrusive JavaScript" is a little obtrusive since the button looks like a button momentarily before it's replaced - could it be hidden with CSS before it's rendered to avoid the flash?

[08:04:00.0000] <zcorpan>
annevk: speaking of which, if a form control is replaced with an image with css3 generated content, but images are disabled, shouldn't the button be rendered as normal? opera renders it like an <img> without an alt attribute

[08:05:00.0000] <zcorpan>
annevk: your blog is a good test case for that

[08:07:00.0000] <annevk>
per the definition of 'content' that's not part of css3-content but will be at some point, yes

[08:07:01.0000] <annevk>
iirc

[08:09:00.0000] <zcorpan>
should i file a bug?

[08:10:00.0000] <annevk>
we have one

[08:10:01.0000] <zcorpan>
ok

[08:13:00.0000] <krijnh>
Back

[08:15:00.0000] <zcorpan>
http://www.robertnyman.com/2007/04/26/geek-meet-may-2007-html-5-and-xhtml/

[08:17:00.0000] <annevk>
have fun

[08:17:01.0000] <zcorpan>
:)

[08:18:00.0000] <krijnh>
annevk: so.. what do you mean?

[08:19:00.0000] <annevk>
krijnh, I mean that what IE does is correct

[08:19:01.0000] <krijnh>
zcorpan: they probably are

[08:19:02.0000] <annevk>
krijnh, and that Firefox and Konquerer etc. are wrong

[08:19:03.0000] <annevk>
per the current specs

[08:19:04.0000] <krijnh>
annevk: i know

[08:20:00.0000] <annevk>
okay

[08:20:01.0000] <krijnh>
But that's what makes it hard to use :0

[08:20:02.0000] <krijnh>
Or better yet, unusable

[08:20:03.0000] <annevk>
/me goes back to figuring out what the tag name productions are for XML 1.1

[08:20:04.0000] <zcorpan>
krijnh: usable for it's intended purpose or for presentational purposes? :)

[08:21:00.0000] <krijnh>
Usable as in, imho, the nicest solution :)

[08:21:01.0000] <krijnh>
Philip`: That's possible I think

[08:21:02.0000] <krijnh>
Philip`: But JS has to hide it in some way

[08:22:00.0000] <annevk>
so it seems that XML 1.1 pretty much allows any character

[08:22:01.0000] <annevk>
awesome

[08:22:02.0000] <krijnh>
zcorpan: How would you do it?

[08:23:00.0000] <zcorpan>
krijnh: <input type=submit name=option value=Edit> ... input[value=Edit] { content:url(edit.png); }

[08:24:00.0000] <krijnh>
That'd be cool

[08:24:01.0000] <zcorpan>
works in opera

[08:24:02.0000] <krijnh>
Even though my very few customers don't use Opera that much

[08:24:03.0000] <krijnh>
I know ;)

[08:24:04.0000] <krijnh>
Had to find a way to fix up http://qontent.nl/_img/screenshots/paginas.png

[08:25:00.0000] <annevk>
/me is amazed with Eliotte Harold

[08:25:01.0000] <annevk>
the guy with double l

[08:26:00.0000] <annevk>
guess I had too many silly assumptions about some XML people

[08:27:00.0000] <MikeSmith>
annevk - for instance?

[08:28:00.0000] <MikeSmith>
/me is also amazed by EHR sometimes, though maybe not for same reasons as annevk 

[08:29:00.0000] <annevk>
he's promoting it where I thought he would find HTML5 quite a silly adventure

[08:29:01.0000] <MikeSmith>
who knows, maybe he did

[08:30:00.0000] <MikeSmith>
and maybe he changed his mind after actually reading the spec

[08:30:01.0000] <krijnh>
annevk: assumptions are the mother of all fuckups ;)

[08:30:02.0000] <jdandrea>
/me is late to the chat. Link for EHR?

[08:30:03.0000] <Philip`>
Maybe he dislikes the HTML5 serialisation but likes the rest of it?

[08:30:04.0000] <MikeSmith>
Elliotte Rusty Harold

[08:30:05.0000] <annevk>
jdandrea, see whatwg⊙wo

[08:30:06.0000] <jdandrea>
reading it now in fact - thx

[08:30:07.0000] <jdandrea>
/me is prolly not caught up - will keep at it ...

[08:31:00.0000] <Philip`>
(given that he mentioned in http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-April/010954.html being very pro-XHTML)

[08:31:01.0000] <MikeSmith>
not saying that it's necessarily so for his case, but people do change their minds about things sometimes

[08:32:00.0000] <annevk>
Philip`, ah right, I guess he's still pro XHTML

[08:33:00.0000] <MikeSmith>
is that XHTML in the sense of well-formed XML serialization of HTML, or XHTML as in the XHTML 1.1/2

[08:33:01.0000] <MikeSmith>
/me reads and sees now

[08:33:02.0000] <annevk>
former

[08:35:00.0000] <MikeSmith>
/me wonders how many people might change their minds about the webapps/HTML5 spec if they actually took time to read it

[08:36:00.0000] <Philip`>
Are you only considering people who currently object to it, or also the people who currently think they like HTML5 but haven't actually read much of it? :-)

[08:36:01.0000] <krijnh>
Hehe

[08:37:00.0000] <MikeSmith>
heh. the former, actually

[08:37:01.0000] <MikeSmith>
but I guess there probably are a few of the latter as well

[08:37:02.0000] <MikeSmith>
bible-thumpers on both sides who haven't actually taken the time to read the bible ... 

[08:39:00.0000] <MikeSmith>
but I'd guess that there are far more of the former than latter

[08:40:00.0000] <krijnh>
/me should log IP addresses of people flagging useless lines; http://krijnhoetmer.nl/irc-logs/whatwg/20070426#l-472

[08:40:01.0000] <krijnh>
Or dump that feature anyway :/

[08:40:02.0000] <Philip`>
At least parts of the bible don't keep getting rewritten every few days - it's harder to keep up with the latest version of WA1

[08:41:00.0000] <zcorpan>
krijnh: you could perhaps highlight lines that get linked to (if that's possible to do)

[08:41:01.0000] <MikeSmith>
krijnh - I like that feature. but I guess like every other useful interactive feature, it's open to abuse

[08:41:02.0000] <annevk>
krijnh, i'd like a checkbox

[08:41:03.0000] <krijnh>
Easy on the requests people ;)

[08:41:04.0000] <annevk>
krijnh, the current UI doesn't work at all in my Opera installation

[08:42:00.0000] <Philip`>
annevk: I think it works if you double-click the non-text part of the line

[08:42:01.0000] <Philip`>
(at least when I last tried it)

[08:42:02.0000] <annevk>
yeah, sometimes...

[08:42:03.0000] <krijnh>
Yeah, I had trouble with that as well

[08:42:04.0000] <zcorpan>
krijnh: as a compromise you could do all of the above ;)

[08:42:05.0000] <krijnh>
zcorpan: fair enough

[08:42:06.0000] <zcorpan>
krijnh: (i was joking :P)

[08:43:00.0000] <krijnh>
;)

[08:43:01.0000] <annevk>
don't tell him

[08:43:02.0000] <zcorpan>
heh

[08:43:03.0000] <krijnh>
Next time put it in a <sarcasm> tag(!) then ;)

[08:43:04.0000] <jdandrea>
LOL

[08:43:05.0000] <zcorpan>
right, forgot :(

[08:44:00.0000] <krijnh>
zcorpan: so how would you say that's possible?

[08:44:01.0000] <annevk>
lets introduce <joke> on April 1

[08:44:02.0000] <krijnh>
If there's a referrer that's not the current document and the location.hash is set

[08:44:03.0000] <annevk>
or would that  be too obvious?

[08:44:04.0000] <krijnh>
Flag that line

[08:44:05.0000] <krijnh>
Right?

[08:44:06.0000] <annevk>
yeah

[08:44:07.0000] <zcorpan>
something like that

[08:44:08.0000] <zcorpan>
i have del.icio.used a few lines already

[08:45:00.0000] <annevk>
maybe that should be the only way to mark lines...

[08:46:00.0000] <annevk>
kind a like page rank

[08:46:01.0000] <krijnh>
zcorpan: Yeah, I noticed

[08:46:02.0000] <krijnh>
Hmm

[08:46:03.0000] <krijnh>
Would be cool

[08:46:04.0000] <krijnh>
But that flagging should be done with XHR

[08:46:05.0000] <krijnh>
Err, AJAX, sorry

[08:46:06.0000] <krijnh>
So that's probably possible by hand anyway :)

[08:47:00.0000] <krijnh>
zcorpan: your delicious/zcorpan/zcorpan line is so interesting for the rest of the world btw ;)

[08:48:00.0000] <zcorpan>
who said del.icio.us musn't contain personal stuff? :)

[08:48:01.0000] <krijnh>
I don't, but using the new fluffy referrer spam technique that line would be flagged :)

[08:49:00.0000] <zcorpan>
right

[08:50:00.0000] <annevk>
maybe the if the referring page contains the name of the person who said the message it should not be marked

[08:50:01.0000] <krijnh>
Riiight

[08:50:02.0000] <annevk>
(although it should not be unmarked either, of course, if it already is)

[08:50:03.0000] <krijnh>
What do you think I am? A competent developer or something?

[08:50:04.0000] <annevk>
it's not like you got something better to do :p

[08:50:05.0000] <Philip`>
You could have a Bayesian learning network that decides which lines are interesting, given their content and the replies in IRC and any external links to that line

[08:51:00.0000] <annevk>
oh, +1

[08:51:01.0000] <krijnh>
annevk: because I have irc open on one monitor?

[08:52:00.0000] <zcorpan>
obviously linked to doesn't imply "important"

[08:52:01.0000] <krijnh>
Nope

[08:52:02.0000] <zcorpan>
but flagging them as "linked to" (perhaps with a link back) could be nice

[08:53:00.0000] <zcorpan>
it's not a request, just an idea :)

[09:01:00.0000] <krijnh>
annevk: checkbox added

[09:01:01.0000] <krijnh>
Well, somewhat checkboxy

[09:02:00.0000] <annevk>
cool

[09:03:00.0000] <Lachy>
krijnh, can you remove the "are you sure?" confirmation for unmarking a line?

[09:03:01.0000] <krijnh>
No, that's impossible, it will break backwards compat!

[09:03:02.0000] <krijnh>
Done ;)

[09:06:00.0000] <krijnh>
I think dynamically adding 1000 checkboxes isn't nice, so this will do

[09:06:01.0000] <Philip`>
It's odd how the little yellow/red box gets a few pixels taller when you put the mouse over a link on that line

[09:07:00.0000] <krijnh>
Yeah

[09:07:01.0000] <krijnh>
In Opera

[09:07:02.0000] <krijnh>
Oddness removed

[09:08:00.0000] <zcorpan>
krijnh: you need a version switch then

[09:08:01.0000] <krijnh>
zcorpan: Perhaps Lachy can tell me about it in a private mail..

[09:09:00.0000] <Philip`>
krijnh: Ah, it works more evenly now - thanks :-)

[09:11:00.0000] <krijnh>
Yeah, with this I dropped support for IE6

[09:11:01.0000] <krijnh>
\o/

[09:12:00.0000] <MikeSmith>
krijnh - this is not an official request, but given that the /topic for the #html-wg channel references http://krijnhoetmer.nl/irc-logs/ ... 

[09:12:01.0000] <krijnh>
I didn't put it there..

[09:13:00.0000] <MikeSmith>
I'm glad it's there... it's just that part about "(for entertainment ;-)" next to the #xhtml channel

[09:13:01.0000] <Philip`>
krijnh: Add some <!--[if IE]--> hacks to your code

[09:14:00.0000] <krijnh>
Yeah, I wondered how long it would take for somebody to fall over that ;)

[09:14:01.0000] <krijnh>
MikeSmith: removed

[09:15:00.0000] <MikeSmith>
krijnh - cheers

[09:15:01.0000] <MikeSmith>
perhaps some would say that numbers there speak for themselves ... 

[09:16:00.0000] <MikeSmith>
making the message in a perhaps more subtle way

[09:17:00.0000] <zcorpan>
krijnh: heh.. a page listing referrers comes up in your list of referrers :)

[09:24:00.0000] <krijnh>
zcorpan: that .cz page?

[09:24:01.0000] <zcorpan>
yeah

[09:30:00.0000] <annevk>
hmm, writing down a parser for XML in ASCII-art is harder than I expected

[09:31:00.0000] <annevk>
actually, tokenizer

[09:32:00.0000] <Philip`>
Would Unicode-art be easier? It has lots of nice lines and double-lines and corners and things

[09:32:01.0000] <krijnh>
You can make hamburgers with unicode!

[09:39:00.0000] <gsnedders>
a parser in ASCII-art!?

[09:39:01.0000] <zcorpan>
krijnh: d(*⌒▽⌒*)b

[09:39:02.0000] <Philip`>
Bonus points if it's a correctly-operating Befunge program as well as a diagram

[09:40:00.0000] <krijnh>
?--------?

[09:40:01.0000] <krijnh>
Bah

[09:40:02.0000] <krijnh>
Crap client :)

[09:41:00.0000] <zcorpan>
heh

[09:41:01.0000] <annevk>
gsnedders, just the various states indented etc.

[09:41:02.0000] <annevk>
but I stepped away from that

[09:41:03.0000] <gsnedders>
aaahhhhh.

[09:41:04.0000] <gsnedders>
sounds saner :)

[09:43:00.0000] <zcorpan>
Hixie: the first example on irrelevant="", shouldn't <section id="game"> be <section id="game" irrelevant>?

[09:48:00.0000] <Lachy>
Hixie, typo in that section too "User agents should net render elements that have the irrelevant  attribute specified" -- s/net/not/

[09:49:00.0000] <nickshanks>
<html irrelevant>

[09:50:00.0000] <zcorpan>
<!doctype irrelevant>

[09:50:01.0000] <nickshanks>
doctype is implicitly irrelevent

[09:50:02.0000] <Lachy>
if only that triggered standards mode, I'd use it :-(

[09:51:00.0000] <Philip`>
<!doctype html system "irrelevant">?

[09:51:01.0000] <Lachy>
could use <!doctype html publick "irr

[09:51:02.0000] <Lachy>
/me was too slow

[09:53:00.0000] <nickshanks>
how about application/html MIME type as the switch to stop supporting quirks mode?

[09:53:01.0000] <nickshanks>
then you could leave off the doctype and still be in standards mode

[09:53:02.0000] <annevk>
not backwards compatible

[09:54:00.0000] <nickshanks>
anne: that's the point

[09:54:01.0000] <Philip`>
Is it implicit that 'irrelevant' should cause UAs not to render any child elements of the irrelevant element?

[09:54:02.0000] <annevk>
nickshanks, the point is adding a few bytes to the MIME type which is way harder to change for people to save a few bytes within the document (which is easy to change)?

[09:54:03.0000] <krijnh>
Philip`: as I understand it, it's like [irrelevant]{display: none;}

[09:55:00.0000] <annevk>
and besides that making the whole thing backwards incompatible?

[09:55:01.0000] <annevk>
seems like way more cost than it's worth to me

[09:55:02.0000] <nickshanks>
but if you're not going to support tag soup browsers anyway, it doesn't really matter

[09:56:00.0000] <zcorpan>
Hixie: i don't get the second example of irrelevant=""... the second paragraph seems to be fallback, why don't put that as contents of the <video> (and make <video> fallback on error)?

[09:56:01.0000] <annevk>
if you don't want to play friendly with the web there's always XHTML2

[09:56:02.0000] <nickshanks>
what is with you guys keep suggesting that?

[09:56:03.0000] <Philip`>
I expect XHTML5 will always be rendered in really-standards mode too

[09:57:00.0000] <annevk>
nickshanks, maybe you should ask yourself?

[09:57:01.0000] <annevk>
Philip`, yeah

[09:58:00.0000] <Philip`>
I assume IE will support XHTML(5) at some point, and I assume it'll do it in really-standards mode (since it's not like there's any existing content to break), and if you don't mind backward compatibility then you can just use that instead of a new MIME type

[09:59:00.0000] <nickshanks>
that's my point, but no-one seems to listen when i say it

[09:59:01.0000] <nickshanks>
there is no existing HTML 5 content to break

[09:59:02.0000] <annevk>
the idea of HTML5 is that it's compatible with older UAs

[10:00:00.0000] <nickshanks>
it will never be

[10:00:01.0000] <annevk>
the idea is also that UAs implementing HTML5 will automatically support older versions of HTML

[10:00:02.0000] <annevk>
as in, no versioning

[10:00:03.0000] <nickshanks>
that is achievable

[10:00:04.0000] <nickshanks>
but mosaic will never support canvas

[10:00:05.0000] <nickshanks>
or any other new element

[10:00:06.0000] <annevk>
compatible does equal support

[10:00:07.0000] <annevk>
does not*

[10:00:08.0000] <annevk>
oops

[10:01:00.0000] <annevk>
Mosaic will render the fallback content

[10:01:01.0000] <nickshanks>
i'm not suggesting we change anything that makes mosaic unable to parse the document

[10:01:02.0000] <krijnh>
And it will render irrelevant elements :)

[10:01:03.0000] <nickshanks>
like replace <a> with [a]

[10:01:04.0000] <annevk>
(there may be exceptions to this I suppose, <event-source> isn't really backwards compatible at all, but there's no need for it to be)

[10:02:00.0000] <annevk>
you're suggesting something that will make Mosaic crash or maybe offer a download dialog

[10:02:01.0000] <annevk>
for no real benefit

[10:03:00.0000] <annevk>
krijnh, what do you mean?

[10:03:01.0000] <nickshanks>
well mosaic won't accept application/html so it will get the text/html version

[10:03:02.0000] <annevk>
you want to author two versions?

[10:03:03.0000] <krijnh>
annevk: I think it won't know about the irrelevant attribute?

[10:03:04.0000] <nickshanks>
if i care about old clients

[10:03:05.0000] <annevk>
to save <!doctype html> in one?!

[10:03:06.0000] <annevk>
please...

[10:03:07.0000] <annevk>
krijnh, oh, duh

[10:03:08.0000] <nickshanks>
no, you are not understanding

[10:03:09.0000] <annevk>
krijnh, didn't get that

[10:04:00.0000] <nickshanks>
stop thinking in terms of now

[10:04:01.0000] <nickshanks>
and think in terms of 2025

[10:04:02.0000] <krijnh>
annevk: Np, I don't the irrelevant attribute either :)

[10:04:03.0000] <krijnh>
*don't get even

[10:04:04.0000] <annevk>
nickshanks, i don't see the point of breaking backcompat then either

[10:04:05.0000] <nickshanks>
do you really still want to be stuck supporting all the crap that was in HTML 3.2 that far ahead?

[10:04:06.0000] <annevk>
nickshanks, yes

[10:04:07.0000] <zcorpan>
nickshanks: yes

[10:04:08.0000] <annevk>
nickshanks, that's the whole point of HTML5

[10:05:00.0000] <annevk>
the other option is XHTML2

[10:05:01.0000] <annevk>
it has been a conscious decision to not break backcompat

[10:05:02.0000] <annevk>
if we expect to do that anyway in 20 years we might as well do it now

[10:06:00.0000] <nickshanks>
but it's not possible to both retain backwards compat and add conflicting features without a switch of some kind

[10:06:01.0000] <nickshanks>
either doctype, mime type or something else

[10:07:00.0000] <annevk>
we won't add conflicting features

[10:07:01.0000] <annevk>
maybe you should read up on the design principles

[10:08:00.0000] <nickshanks>
there's no way you can know what will be desired in the future

[10:09:00.0000] <zcorpan>
nickshanks: if we need conflicting features in the future then we can introduce versioning/new mime type/whatever *then*

[10:09:01.0000] <zcorpan>
nickshanks: we haven't done so know (afaik) so no need to do so now

[10:09:02.0000] <zcorpan>
er, s/know/now/

[10:10:00.0000] <nickshanks>
well, there is an need for versioning now, as shown by doctype switching

[10:10:01.0000] <zcorpan>
nickshanks: no, there was a need back when doctype switching was introduced. you're suggesting yet another switch, no?

[10:11:00.0000] <nickshanks>
that need still exists

[10:11:01.0000] <annevk>
doctype switching is not versioning

[10:11:02.0000] <zcorpan>
nickshanks: yes, we're stuck with it and probably will be forever

[10:11:03.0000] <annevk>
it's a few specific rendering differences that depend on a particular string at the start of the document

[10:11:04.0000] <nickshanks>
no, because HTML doesn't support versioning so we had to come up with a hack

[10:11:05.0000] <annevk>
it's mostly versioning for the rendering engine, not HTML

[10:12:00.0000] <nickshanks>
right

[10:12:01.0000] <annevk>
it should never have been introduced

[10:12:02.0000] <nickshanks>
but so is any document version

[10:12:03.0000] <annevk>
but unfortunately it was

[10:12:04.0000] <nickshanks>
you don't have a better solution

[10:12:05.0000] <annevk>
i've no idea what you're talking about

[10:13:00.0000] <nickshanks>
well the application reads a document, and behaves differently depending on what the version number at the start of the document is

[10:13:01.0000] <annevk>
well yes, as mentioned above, if that's needed we can introduce it

[10:13:02.0000] <nickshanks>
whether there's one giant switch statement of little if..else's here and there is an implementation detail and not relavent here

[10:13:03.0000] <annevk>
i don't think we'll ever need it

[10:14:00.0000] <nickshanks>
i think "we can't be sure we'll never need it, so better to include it and be on the safe side"

[10:14:01.0000] <annevk>
i don't think quirks mode was needed either, but supposedly it seemed like a good idea back then

[10:14:02.0000] <annevk>
including it and not letting it have effect will just confuse people

[10:14:03.0000] <annevk>
as mentioned

[10:14:04.0000] <nickshanks>
i've worked with plenty of C structs that have version numbers at the start, where the only version was 0

[10:14:05.0000] <annevk>
we can always introduce it

[10:14:06.0000] <annevk>
we don't need to be safe

[10:15:00.0000] <annevk>
C is not comparable with HTML

[10:15:01.0000] <nickshanks>
HTML is instructions to be interpreted and cause an effect

[10:15:02.0000] <nickshanks>
same as any other computer language

[10:16:00.0000] <hasather>
nickshanks: no, HTML is no programming language

[10:17:00.0000] <nickshanks>
i don't see the difference

[10:20:00.0000] <othermaciej>
nickshanks: how many C programs have you seen that declare the version of C they are using at the top of the file?

[10:21:00.0000] <nickshanks>
the version of C is declared outside of the code files in the IDE

[10:21:01.0000] <Philip`>
HTML6 can say the default value of the 'version' attribute is '5', and an HTML6 browser will apply that rule to every page it sees. You can't do anything like that in C - a change to the struct's layout won't be picked up by all the existing binary code that uses it, so the version has to be explicit

[10:22:00.0000] <annevk>
html-wg looks like chaos

[10:23:00.0000] <othermaciej>
nickshanks: when I type cc myprogram.c at the command line, I don't include a mention of the C version anywhere

[10:23:01.0000] <nickshanks>
the version of C is declared outside of the code files in the IDE (you may have missed that)

[10:24:00.0000] <nickshanks>
and if you compile it with a subset or incompatible version, you get errors

[10:24:01.0000] <Philip`>
The version there is optional - you could run "gcc -ansi" or "gcc -std=c89"

[10:24:02.0000] <nickshanks>
but it has a default value

[10:24:03.0000] <Philip`>
but it seems most people don't use either of those, because the language is largely backward-compatible and so you can compile all your code as C99-plus-GCC-extensions

[10:25:00.0000] <nickshanks>
anyway, i wasn't talking about the version of C (good example, btw) i was talking about the version of some data structure that your app is reading off disk

[10:25:01.0000] <Philip`>
but when it doesn't work, you can modify the source code, which is not possible with HTML because it's not your source code

[10:26:00.0000] <othermaciej>
nickshanks: but HTML is a language, not an on-disk binary formatted data structure

[10:27:00.0000] <annevk>
Philip`, nice comparison; (as opposed to modify we fix the source code)

[10:27:01.0000] <nickshanks>
but how that language is to be interpreted is not consistent across all versions of HTML

[10:29:00.0000] <annevk>
it's whatever the latest version says

[10:29:01.0000] <nickshanks>
mostly due to errors in specs

[10:30:00.0000] <nickshanks>
anne: but not all UAs follow the "latest version" so you lose your back compat right there

[10:30:01.0000] <annevk>
no

[10:30:02.0000] <annevk>
because the latest version is designed with that in mind

[10:31:00.0000] <nickshanks>
unless it has an error in it

[10:31:01.0000] <annevk>
in which case that will be pointed out and fixed

[10:31:02.0000] <nickshanks>
after it has been implemented

[10:31:03.0000] <nickshanks>
at which point it's too late

[10:31:04.0000] <annevk>
this also goes for theoretical versioned HTML so I'm not sure what your point is

[10:32:00.0000] <annevk>
yes, it could also be implemented incorrectly... if that's done and not much breaks it's apparently acceptable

[10:32:01.0000] <annevk>
languages evolve

[10:32:02.0000] <nickshanks>
versioning in HTML would basically be about how to recover from mistakes in specs or mistakes in how developers wrote for that version

[10:32:03.0000] <nickshanks>
so that when it was fixed, you could do it the "old way" when encountering an old file

[10:32:04.0000] <annevk>
what fixed?

[10:33:00.0000] <nickshanks>
whatever the original error was

[10:33:01.0000] <annevk>
the current algorithm is largely compatible with the web

[10:33:02.0000] <annevk>
it needs more impl experience of course

[10:33:03.0000] <nickshanks>
usemap in XHTML is a good example

[10:33:04.0000] <Philip`>
annevk: But evolution requires death, and nobody wants to kill any old HTML content, so HTML can't evolve - it just grows and mutates :-)

[10:33:05.0000] <annevk>
nickshanks, usemap in XHTML was a bug in the spec

[10:34:00.0000] <nickshanks>
right

[10:34:01.0000] <nickshanks>
you can't guarantee that all future versions of HTML won't be bug free

[10:34:02.0000] <annevk>
and hardly relevant given that 0.00% of the sites out there use XML

[10:34:03.0000] <nickshanks>
it's an EXAMPLE for god's sake

[10:34:04.0000] <annevk>
nickshanks, no, you can't

[10:34:05.0000] <nickshanks>
it could happpen to HTML 5

[10:34:06.0000] <annevk>
yes, the current spec is imperfect

[10:34:07.0000] <annevk>
i don't really see the problem

[10:35:00.0000] <annevk>
if such a bug is found the bug is fixed

[10:35:01.0000] <nickshanks>
even if it wasn't, the next version might be

[10:35:02.0000] <annevk>
specs and implementations require many iterations

[10:35:03.0000] <othermaciej>
if you make a mistake in the spec, you have a problem

[10:35:04.0000] <annevk>
Philip`, I'm not sure I agree with that

[10:35:05.0000] <othermaciej>
if you use versioning to address it, you have two problems

[10:35:06.0000] <annevk>
Philip`, but I can't really refute it either

[10:35:07.0000] <annevk>
othermaciej, :)

[10:36:00.0000] <nickshanks>
if I write a site and put <html version="5.0"> at the top, web browsers will know that i mean b when i write a, and if i put version="5.01" they'll know not to change the meaning

[10:36:01.0000] <annevk>
browsers will ignore that information...

[10:37:00.0000] <nickshanks>
then they won't know which I meant and will have to guess

[10:37:01.0000] <nickshanks>
i, the author, provide the answer so no need to guess

[10:37:02.0000] <annevk>
i've no idea what you're talking about

[10:37:03.0000] <annevk>
html will just be interpreted by whatever the latest spec says on the matter

[10:38:00.0000] <Philip`>
Is the (theoretical) problem only that HTML5 might specify some behaviour, and lots of sites are built to rely on it, then HTML6 accidentally specifies a different incompatible behaviour, but nobody notices the bug and lots of sites are built that rely on it, and then somebody notices and HTML7 is stuck because it'll break half the sites?

[10:38:01.0000] <othermaciej>
version info is not generally used to switch anything that is based on the spec

[10:38:02.0000] <nickshanks>
Philip`: yeah, basically

[10:38:03.0000] <othermaciej>
doctype switching for standards vs. quirks mode is solely about some CSS rendering differences for compatibility

[10:38:04.0000] <othermaciej>
HTML 2, HTML 3.2 and HTML 4 are all treated the same

[10:38:05.0000] <annevk>
nickshanks, that's not a realstic scenario though

[10:39:00.0000] <nickshanks>
i don't see why you're against having a safty net

[10:39:01.0000] <nickshanks>
+e

[10:39:02.0000] <annevk>
it's not needed at this point

[10:39:03.0000] <Philip`>
I would expect that situation to be unlikely, because people will test their HTML6 sites in HTML5 UAs too, and their HTML5 sites in HTML6 UAs, and so somebody will notice the bug soon (before the next version of market-dominant-browser-of-the-day is released) and the spec can be fixed

[10:39:04.0000] <annevk>
adding things that are not needed is just adding bloat

[10:39:05.0000] <annevk>
and sets the wrong precedents

[10:40:00.0000] <nickshanks>
until next year, when you realise, "oh shit, we screwed up there, and now there are two different interpretations floating about, and no way to tell which one is which because we didn't include differentiation"

[10:40:01.0000] <annevk>
Philip`, implementors will surely notice such an incompatibility

[10:40:02.0000] <annevk>
nickshanks, if two UAs do something different introducing versioning doesn't help

[10:40:03.0000] <nickshanks>
Philip`: a lot of people don't test in more than one browser

[10:40:04.0000] <annevk>
nickshanks, versioning doesn't solve the UA problem

[10:41:00.0000] <annevk>
how the hell would it solve that?!

[10:41:01.0000] <annevk>
/me has the feeling he's wasting his time here

[10:41:02.0000] <nickshanks>
sorry, which problem?

[10:41:03.0000] <annevk>
the problem that two UAs might do different things

[10:42:00.0000] <nickshanks>
i wasn't talking about two USa, i was meaning two subsequent versions of HTML

[10:42:01.0000] <nickshanks>
UAs *

[10:42:02.0000] <Philip`>
The problem of two versioned sites using <html version="5.0.6">, each tested in different UAs that have incompatible behaviour

[10:43:00.0000] <nickshanks>
cross compatibility on the same version of HTML is not what's being discussed

[10:43:01.0000] <annevk>
nickshanks, such a thing hasn't occured yet, so i don't really see that as a potential problem

[10:43:02.0000] <nickshanks>
that's hardly a defence

[10:43:03.0000] <nickshanks>
and it occured with usemap

[10:43:04.0000] <annevk>
i think it is

[10:43:05.0000] <annevk>
XHTML is irrelevant to this discussion

[10:43:06.0000] <annevk>
there's no usage

[10:44:00.0000] <annevk>
and people did notice and reported the error

[10:44:01.0000] <annevk>
the spec was just never fixed

[10:44:02.0000] <annevk>
fortunately we now have more competent people running HTML

[10:45:00.0000] <nickshanks>
right, my point is that there was an error, and it could happen again (no matter how competent we are)

[10:45:01.0000] <Philip`>
Was it reported before any implementations were released that used the incompatible behaviour?

[10:46:00.0000] <nickshanks>
future HTML developers might also realise they have to break back compat in order to move forward (in which case versioning will be required, instead of just being a reserve parachute)

[10:46:01.0000] <nickshanks>
Philip: it was reported before the spec was finished, they just said "yeah, so what?"

[10:47:00.0000] <annevk>
nickshanks, in which case they can introduce versioning then

[10:47:01.0000] <annevk>
nickshanks, introducing versioning and unnoticely introducing a backcompat change doesn't solve this problem either

[10:47:02.0000] <annevk>
nickshanks, because UAs won't know the behavior has suddenly changed

[10:48:00.0000] <nickshanks>
anne: do you not see how having a version might help future UAs resolve a dilemma ?

[10:48:01.0000] <annevk>
nickshanks, in fact, they won't change their current behavior...

[10:48:02.0000] <annevk>
nickshanks, I don't think the dilemma will ever occur

[10:48:03.0000] <nickshanks>
as i said, that's no defence

[10:48:04.0000] <annevk>
how is it not?

[10:49:00.0000] <nickshanks>
because you could quite easily be wrong!

[10:49:01.0000] <annevk>
so could you!

[10:49:02.0000] <zcorpan>
nickshanks: you still haven't said why we need to introduce it *now*. why can't it be introduced when we find that it is needed (if at all)?

[10:49:03.0000] <annevk>
and then we've added bloat for nothing and all kinds of other crap

[10:49:04.0000] <nickshanks>
it should have been in every version of html

[10:49:05.0000] <zcorpan>
why?

[10:49:06.0000] <zcorpan>
so far there haven't been a need afaict

[10:49:07.0000] <Philip`>
You could both be wrong - maybe the future is LaTeX

[10:49:08.0000] <nickshanks>
so we can drop old elements or change their functionality

[10:49:09.0000] <nickshanks>
like menu

[10:50:00.0000] <annevk>
Philip`, LaTeX would presumably not use text/html

[10:50:01.0000] <nickshanks>
menu in HTML 5 bears no relation to menu in html 1

[10:50:02.0000] <annevk>
nickshanks, the upgrade is done in a compatible way

[10:50:03.0000] <nickshanks>
and in order to retain back compat there is a whole load of unnecessary bloat in the spec

[10:50:04.0000] <annevk>
versioning doesn't solve that either

[10:51:00.0000] <zcorpan>
nickshanks: what's the benefit of dropping support for old elements under a new switch but keeping support for them for pages without the switch? it just adds implementation complexity

[10:51:01.0000] <annevk>
versioning just introduces more bloat

[10:51:02.0000] <annevk>
so far versioning just introduces more problems as I see it

[10:55:00.0000] <Philip`>
nickshanks: The goal of the WHATWG work is to be an implementation comparable with current browsers - if the browsers have to implement <menu> in a certain way because content requires it, then the spec will have to do the same, so it can't just drop it and become less bloaty

[10:56:00.0000] <Philip`>
(i.e. it's necessary bloat, given the requirements which are placed on the spec)

[10:57:00.0000] <Philip`>
and that seems like quite a fundamental goal, so I don't see it ever being changed within this group

[10:57:01.0000] <nickshanks>
on another topic: i think it would idea to take the error handling from WA 1 and apply it to a HTML 4.1, so that we have a version of HTML 4 that is interoperable and standardised as such

[10:58:00.0000] <nickshanks>
whilst leaving all the new stuff for HTML 5

[10:58:01.0000] <Philip`>
Like an HTML4-featured profile of the HTML5 spec?

[10:58:02.0000] <Philip`>
(People don't seem to like profiles that much...)

[10:58:03.0000] <nickshanks>
whatever you want to call it

[10:59:00.0000] <annevk>
everything was broken in HTML4

[10:59:01.0000] <annevk>
it would still be a lot of work

[10:59:02.0000] <nickshanks>
now you're exaggerating a little :)

[11:00:00.0000] <nickshanks>
yeah, it would be work

[11:00:01.0000] <nickshanks>
but we can get the W3C to do that

[11:00:02.0000] <annevk>
not really...

[11:00:03.0000] <annevk>
why?

[11:00:04.0000] <Philip`>
Is a conformant interoperable HTML 4.2 UA more useful than a non-conformant (because it's missing all the new features) interoperable (because the implemented ones follow the spec) HTML 5 UA?

[11:00:05.0000] <Philip`>
*the implemented features

[11:01:00.0000] <zcorpan>
more to the point, what's the benefit of such a spec?

[11:01:01.0000] <nickshanks>
philip: that would depend on what content you were using, primarily 4.2 or 5.0 sites

[11:02:00.0000] <nickshanks>
zcorpan: it would (i assume) be out more quickly than HTML 5 and web authors can start using it

[11:02:01.0000] <nickshanks>
and vendors can follow it

[11:02:02.0000] <annevk>
you need a testsuite for that

[11:02:03.0000] <nickshanks>
without haing to implement the rest of HTML 5.0 befor ethey release too

[11:02:04.0000] <nickshanks>
is one not being written?

[11:02:05.0000] <annevk>
people are already implementing HTML5

[11:02:06.0000] <annevk>
not at lightspeed

[11:02:07.0000] <nickshanks>
it would be the HTML 5 test suite with new stuff removed

[11:03:00.0000] <nickshanks>
just a stop-gap measure to sure up HTML 4 interop before HTML 5 arives

[11:03:01.0000] <annevk>
the best way to get interop is create tests

[11:04:00.0000] <annevk>
not by creating specs

[11:04:01.0000] <nickshanks>
well the spec exists 99% already

[11:04:02.0000] <zcorpan>
nickshanks: why can't UAs just implement a subset of html5, and authors use a subset of html5? i don't see the benefit of having a de jure subset rather than de facto subset for a given point in time

[11:04:03.0000] <Philip`>
Are the new HTML5 features much easier/harder to implement than the bugfixes for HTML4 features?

[11:04:04.0000] <annevk>
...

[11:04:05.0000] <nickshanks>
okay, alternative plan: create HTML4-compatible tests before the HTML5-only ones

[11:05:00.0000] <nickshanks>
and forego a version 4.1 spec

[11:05:01.0000] <zcorpan>
since the subset changes over time, when you're done with your html4 subset the implementations have added support for a number of other features which are part of the de facto subset

[11:05:02.0000] <zcorpan>
rendering the html4 subset irrelevant

[11:06:00.0000] <annevk>
Philip`, implementing <canvas> might be easier than fixing some of the bug... patching legacy code can be tricky

[11:06:01.0000] <Philip`>
nickshanks: But then the early HTML5 implementations would have no tests and wouldn't be interoperable, which wouldn't be good, unless the implementations are held back until the tests are finished (which isn't going to happen, since they're not even waiting for the spec to be finished)

[11:06:02.0000] <nickshanks>
zcorpan: the tests would never become irrelevent

[11:06:03.0000] <zcorpan>
nickshanks: did i say that?

[11:07:00.0000] <nickshanks>
yes you did, a few lines up\

[11:07:01.0000] <zcorpan>
i meant the de jure html4 subset of html5 you're advocating, not tests

[11:07:02.0000] <nickshanks>
i am talking about tests

[11:07:03.0000] <zcorpan>
oh, sorry

[11:07:04.0000] <othermaciej>
zakim, unmute me

[11:07:05.0000] <othermaciej>
doh

[11:07:06.0000] <zcorpan>
creating tests for a subset of html5 is good

[11:08:00.0000] <zcorpan>
but doesn't mean we have to create a separate spec

[11:08:01.0000] <nickshanks>
right

[11:08:02.0000] <nickshanks>
specs help HTML authors, tests help UA authors

[11:08:03.0000] <zcorpan>
tutorials help authors

[11:09:00.0000] <zcorpan>
specs and tests help implementors

[11:09:01.0000] <Philip`>
Apart from <!doctype html>, I'd expect most HTML5 tests to not be using any new HTML5 features, so they'd work fine for testing HTML4-featured UAs

[11:09:02.0000] <nickshanks>
tutorials are usually written by people who've read and comprehended a specification first though :P

[11:10:00.0000] <zcorpan>
nickshanks: right, specs help tutorial authors

[11:10:01.0000] <nickshanks>
i can't imagine anyone proposing HTML 6 have no spec and just a bunch of tests :)

[11:10:02.0000] <zcorpan>
and authors who know how to read specs

[11:11:00.0000] <Philip`>
(Maybe the tests that use new features could just be labelled, rather than intentionally writing tests to a specific subset of the spec)

[11:11:01.0000] <nickshanks>
i think anyone with a basic grasp of english can read the spec, and if it's well written, understand it

[11:11:02.0000] <nickshanks>
trouble is not many authors do!

[11:12:00.0000] <nickshanks>
and even, apparently, some people who write "Lern Yerself HTML Kwik!" type books don't read the spec :-o

[11:12:01.0000] <Philip`>
Not many authors do have a basic grasp of English? Maybe there should be an official translation of the spec into Chinese :-)

[11:12:02.0000] <zcorpan>
nickshanks: not really, you have to know about rfc2119 terms and be able to distinguish between what applies to authors or not, and what is normative or not

[11:12:03.0000] <nickshanks>
Philip: yes, that would be a good idea too

[11:13:00.0000] <nickshanks>
I think readers SHOULD know about rfc2199

[11:13:01.0000] <nickshanks>
but it makes pretty good sense to an english speaker even if they don't

[11:14:00.0000] <nickshanks>
i would expect most errors on web pages are trivial though, not conceptual

[11:15:00.0000] <nickshanks>
whether something is normative or not isn't too critical

[11:15:01.0000] <Philip`>
Someone (can't quite remember who) had some confusion over http://www.whatwg.org/specs/web-apps/current-work/#lists where it says you need comma-only separators to be valid, but the algorithm accepted spaces too

[11:15:02.0000] <jdandrea>
That was me.

[11:15:03.0000] <Philip`>
because it's not particularly obvious that the former applies to authors, and the latter should never be looked at by authors and only applies to UAs

[11:16:00.0000] <jdandrea>
Bingo.

[11:16:01.0000] <jdandrea>
Of course now when I read it, I grok. :)

[11:16:02.0000] <jdandrea>
(well, maybe not grok but you get the idea - it makes sense now)

[11:17:00.0000] <Philip`>
I think that kind of subtlety makes it hard to read the spec properly, even after you can read all the English

[11:17:01.0000] <jdandrea>
/me is reminded of: "It's easy when you know how!" :)

[11:17:02.0000] <Philip`>
jdandrea: Ah, sorry, I wasn't sure who it was :-)

[11:18:00.0000] <nickshanks>
should the WA spec wrap implementor only details with a display: none div?

[11:18:01.0000] <nickshanks>
and have a little button to show them

[11:18:02.0000] <nickshanks>
it would make it an awful lot shorter for most users! :)

[11:19:00.0000] <Philip`>
(jdandrea: Also sorry, can't respond to PMs, since I've been too lazy to register on Freenode :-) )

[11:20:00.0000] <zcorpan>
nickshanks: might be nice but how would it be done? should Hixie markup everything with classes?

[11:20:01.0000] <jdandrea>
Philip` - no worries. It's a good example (comma-only separators)!

[11:20:02.0000] <zcorpan>
nickshanks: or is there a way to programmatically determinate whether something applies to authors or not?

[11:20:03.0000] <jdandrea>
There was discussion on this at the time of that chat, IIRC. Checking logs ...

[11:21:00.0000] <nickshanks>
zcorpan: well i'm not sure, i'd have to look at the page's code

[11:21:01.0000] <zcorpan>
nickshanks: if you figure out a good way, let me know

[11:21:02.0000] <nickshanks>
even if it has to be done by hand, it won't take as long as writing the spec down in the first place did

[11:22:00.0000] <nickshanks>
zcorpan: were you the one responsible for the multipage script?

[11:22:01.0000] <jdandrea>
http://krijnhoetmer.nl/irc-logs/whatwg/20070420

[11:22:02.0000] <zcorpan>
nickshanks: no

[11:22:03.0000] <jdandrea>
Relevant lines:

[11:23:00.0000] <zcorpan>
nickshanks: i have contributed to the status script

[11:23:01.0000] <Philip`>
nickshanks: Blame me for that :-)

[11:23:02.0000] <jdandrea>
zcorpan_: "people asked for a view of the spec that only had the information that applied to authors, i think." Philip`: "Could we add a new HTML element for it? <relevance who="authors"> "  zcorpan: "it would be nice but it's hard to do programmatically as it looks now"

[11:23:03.0000] <nickshanks>
zcorpan: <p>The <dfn id=rules3>rules for parsing a list of integers</dfn> are as follows:

[11:24:00.0000] <nickshanks>
so it looks like matching on "<dfn id=rules" and taking the enclosing block-level element might work as a stating point

[11:25:00.0000] <nickshanks>
and why is the webapps spec written in HTML 4 :P

[11:25:01.0000] <Philip`>
Trying to simply chop out parts of the document would probably be too much of a constraint on the prose if it was going to be precise, and it'd become hard to read and write - a separate author-level not-officially-normative spec may be easier to write

[11:26:00.0000] <Philip`>
nickshanks: Apparently just because of the tools used to generate it

[11:26:01.0000] <Philip`>
The multipage one is HTML5, since that has nicer tools ;-)

[11:26:02.0000] <nickshanks>
fair engough

[11:26:03.0000] <nickshanks>
-g

[11:27:00.0000] <Philip`>
(I hope the multipage one is using <nav> correctly - I never really checked what the spec says...)

[11:27:01.0000] <zcorpan>
nickshanks: ok, but that wouldn't catch very much... if we want to filter out things that don't apply to authors, we have to filter out everything that doesn't apply to authors and find a sane way to do so

[11:28:00.0000] <nickshanks>
doing it by hand may be boring, but is probably the only way

[11:28:01.0000] <zcorpan>
Philip`: i think it does

[11:28:02.0000] <zcorpan>
nickshanks: ok. so who would do it and how exactly?

[11:29:00.0000] <nickshanks>
probably hixie, and by putting a div around them

[11:29:01.0000] <nickshanks>
or a class on the <p>

[11:29:02.0000] <nickshanks>
it need not be hidden at the moment

[11:30:00.0000] <nickshanks>
but given a new bg colour

[11:30:01.0000] <nickshanks>
or some such

[11:30:02.0000] <zcorpan>
/me thinks Hixie's time is better spent on other things

[11:30:03.0000] <nickshanks>
zcorpan: does anyone else have authorship access to that document?

[11:31:00.0000] <nickshanks>
also, it can wait until the spec gets finalised

[11:31:01.0000] <zcorpan>
nickshanks: define finilised

[11:31:02.0000] <nickshanks>
but UA implementation instructions should not be shown by default to HTML authors who just want to learn about this new spec that just got announced.

[11:32:00.0000] <zcorpan>
agreed, problem is how to do it

[11:32:01.0000] <nickshanks>
zcorpan: either CR or REC depending on how busy the editor is

[11:32:02.0000] <nickshanks>
well the alternative is cutting up the document into a UA version and a HTML version

[11:33:00.0000] <nickshanks>
which would probably take just as long

[11:33:01.0000] <nickshanks>
(but wouldn't require javascript, so...)

[11:34:00.0000] <zcorpan>
nickshanks: i was thinking of somehow applying a mask on top of the spec with javascript

[11:34:01.0000] <Philip`>
I imagine there'd be extra text that probably should be shown only in the author version, not the real-spec version - e.g. the parsing section would be replaced with "you write <tag attr="value">...</tag> but you can skip the closing one on this list of elements, and you have to nest them properly, etc" (but written far better), which (as far as I can tell) is not in the real spec today

[11:34:02.0000] <nickshanks>
or painting the screen with TippEx

[11:34:03.0000] <zcorpan>
haven't figured out if or how it would work exactly though

[11:35:00.0000] <zcorpan>
but perhaps in a similar way as the status markers

[11:35:01.0000] <nickshanks>
the red fortresses?

[11:36:00.0000] <Philip`>
/me has currently been identifying sentences in the spec by using regexps that match them, which is probably quite fragile and inefficient, but at least it works and doesn't need the spec itself to be modified

[11:36:01.0000] <zcorpan>
nickshanks: yes

[11:36:02.0000] <nickshanks>
how would that hide them from potentially confusable HTML newbies?

[11:37:00.0000] <zcorpan>
nickshanks: you locate the elements that are irrelevant to authors, then hide them with css

[11:38:00.0000] <nickshanks>
am investingating....

[11:38:01.0000] <nickshanks>
okay, so it seems we have code like:

[11:38:02.0000] <nickshanks>
 <h3 id=forms><span class=secno>3.16. </span>Forms</h3>

[11:38:03.0000] <nickshanks>
  <!-- XXX everything in WF2 -->

[11:38:04.0000] <nickshanks>
  <p class=big-issue>This section will contain definitions of the

[11:38:05.0000] <nickshanks>
   <code>form</code> element and so forth.

[11:39:00.0000] <nickshanks>
so is CSS  or JS responsible for inserting the [TBW] bit there?

[11:39:01.0000] <zcorpan>
nickshanks: look at http://status.whatwg.org/annotate-web-apps.js 

[11:39:02.0000] <nickshanks>
thx

[11:41:00.0000] <nickshanks>
okay, so that isn't how the HTML4 version works

[11:43:00.0000] <nickshanks>
zcorpan: seems like it would work in HTML5 with some simple changes for what we want

[11:43:01.0000] <nickshanks>
but that means the current HTML4 version would need to be a secondary document, one that HTML authors aren't going to come across easily

[11:43:02.0000] <Philip`>
/me isn't sure what the difference is

[11:44:00.0000] <nickshanks>
Philip`: ? difference with what

[11:45:00.0000] <Philip`>
Between the HTML4 version and HTML5 version

[11:46:00.0000] <nickshanks>
well the js linked above operated on <section staus=""> tags

[11:47:00.0000] <nickshanks>
it would be simpler to write a different javascript IMO

[11:47:01.0000] <nickshanks>
not that it matters too much at this point anyway

[11:47:02.0000] <Philip`>
Ah, no, that's the <section>s in http://www.whatwg.org/specs/web-apps/current-work/annotate-data.xml

[11:48:00.0000] <nickshanks>
ooh, thanks :)

[11:57:00.0000] <zcorpan>
the magic of hiding stuff that doesn't apply to authors could be done in the multipage script instead, i don't know which would be simpler to do

[11:57:01.0000] <Philip`>
Depends on whether you like JavaScript or Python, I guess

[11:58:00.0000] <Philip`>
I suppose JS has the problem that you wouldn't want authors to download 1.5MB of spec and then have their JS engine chug over it for a minute and then throw most of the document away

[11:59:00.0000] <zcorpan>
that's why it's multipage, no? :)

[11:59:01.0000] <zcorpan>
but indeed

[12:00:00.0000] <zcorpan>
if done with js, it could work with both the single page version and the multipage version

[12:00:01.0000] <Philip`>
Oh, I forgot about that one :-)

[12:00:02.0000] <zcorpan>
if done with python, it would only work with the multipage version, which is ok because people who would read the subset view would probably prefer multipage anyway

[12:01:00.0000] <Philip`>
The Python one could generate a stripped-down version of the single-page version easily

[12:01:01.0000] <zcorpan>
ok

[12:01:02.0000] <nickshanks>
just hope it doesn't strip away everything :)

[12:01:03.0000] <Philip`>
(just by having it load the original, do the stripping, then write the single-page output, then split it into sections and write all the sections)

[12:02:00.0000] <zcorpan>
right

[12:02:01.0000] <Philip`>
I suppose at least some sections would involve stripping everything away

[12:07:00.0000] <zcorpan>
we could perhaps just write css that hides stuff

[12:08:00.0000] <zcorpan>
#status+p+p, ...all other thigns that are to be hidden... { display:none; }

[12:08:01.0000] <zcorpan>
/me looks into whether that would work

[12:09:00.0000] <Philip`>
We really should use the 'irrelevant' attribute, but then it couldn't easily change relevance depending on the audience

[12:11:00.0000] <zcorpan>
i was thinking of just an alternate style sheet

[12:12:00.0000] <zcorpan>
you get the full spec by default, but can change to the subset view using the browser's style sheet switcher

[12:12:01.0000] <zcorpan>
or with a scripted link or #hash in the uri or whatever

[12:13:00.0000] <annevk>
Acid3 should test :target

[12:14:00.0000] <nickshanks>
what about :target needs an acid test?

[12:14:01.0000] <nickshanks>
seems to work fine for me

[12:15:00.0000] <Philip`>
Maybe HTML5 should allow <div audience="authors">...<p irrelevant="authors">...<p>...<p irrelevant="authors">...</div>

[12:15:01.0000] <zcorpan>
:target+p doesn't work correctly in safari

[12:15:02.0000] <Philip`>
then you could fake it in legacy browsers with CSs like  [irrelevant] { display: none }   [audience="authors"] [irrelevant="authors"] { display: inherit }

[12:15:03.0000] <Philip`>
*CSS

[12:15:04.0000] <zcorpan>
Philip`: let's not make it harder than it needs to be :)

[12:16:00.0000] <Philip`>
zcorpan: CSS is for presentation, but this is altering the semantics of the document so it's a guide for authors instead of a normative spec, hence the need for an HTML attribute rather than CSS :-)

[12:17:00.0000] <zcorpan>
Philip`: when there are implementations of irrelevant="" i might change my mind, but for now i'm trying to find the simplest solution possible

[12:17:01.0000] <Philip`>
/me likes making things harder than they need to be, as long as he doesn't actually have to do any work for them

[12:17:02.0000] <nickshanks>
alternativlyly, you caould say it is the "don't present me with too much info" viewing

[12:17:03.0000] <zcorpan>
it's not like authors will read the subset view with lynx or anything... and even if they did lynx doesn't support irrelevant=""

[12:18:00.0000] <Philip`>
Authors could submit patches to lynx

[12:18:01.0000] <zcorpan>
if selecting already present elements is good enough, it seems to me that just having an alternative stylesheet is the simplest

[12:19:00.0000] <zcorpan>
that stylesheet could be created by the status script

[12:19:01.0000] <zcorpan>
or inserted by the status script... if we want to maintain the stylesheet separately

[12:28:00.0000] <zcorpan>
this seems to work better than i expected

[12:39:00.0000] <hsivonen>
/me just got more convinced about the relative efficiency of the WHATWG working methods

[12:39:01.0000] <annevk>
heh, could have told you that beforehand

[12:41:00.0000] <hsivonen>
annevk: John Boyer's reply to you was most interesting

[12:41:01.0000] <om_phone>
hsivonen: I can't believe he said "we don't have to answer that"

[12:41:02.0000] <annevk>
hsivonen, to me or someone else?

[12:41:03.0000] <hsivonen>
annevk: to you, I think

[12:42:00.0000] <annevk>
hsivonen, to be honest, I missed his interesting reply I'm afraid

[12:42:01.0000] <annevk>
at least, I don't recall any

[12:42:02.0000] <hsivonen>
annevk: what othermaciej quoted above

[12:42:03.0000] <othermaciej>
I think that was to MatthewRaymond

[12:43:00.0000] <annevk>
didn't DanC say that?

[12:43:01.0000] <hsivonen>
oh

[12:43:02.0000] <annevk>
in reply to me?

[12:43:03.0000] <othermaciej>
I'm pretty sure John Boyer said that

[12:43:04.0000] <jdandrea>
<DanC> re "no answer necessary"... I'll have to elaborate in email. that's not a very good record of what I said. or meant. ???

[12:43:05.0000] <annevk>
That they're not obliged to answer that question and that we're in this together or something...

[12:43:06.0000] <annevk>
That seems to confirm it was indeed the chair

[12:43:07.0000] <othermaciej>
well, I was a bit confused about who said what

[12:44:00.0000] <annevk>
Although othermaciej raised the question again in a different form and he didn't say it again...

[12:44:01.0000] <hsivonen>
I thought it was Boyer's voice

[12:44:02.0000] <hsivonen>
anyway, it bothers me that XForms Transitional is taken for granted

[12:44:03.0000] <annevk>
from the minutes: "DanC: good question to ask, no answer necessary"

[12:44:04.0000] <annevk>
to me it seemed like a reasonable question...

[12:45:00.0000] <annevk>
as it seems that the HTML WG generally agrees WF2 is a good idea

[12:45:01.0000] <othermaciej>
oh, per the minutes it says DanC, I assume he'll correct it if it wasn't him

[12:55:00.0000] <annevk>
it was him (per #html-wg)

[13:09:00.0000] <zcorpan>
perhaps the status markers would be easier to implement with a stylesheet too

[13:09:01.0000] <zcorpan>
it's already implemented though but we still need a way to maintain it and extend it with "implemented" markers etc

[13:10:00.0000] <zcorpan>
a stylesheet seems to be a lot simpler

[13:11:00.0000] <Philip`>
Does it have to work in IE7?

[13:12:00.0000] <zcorpan>
ie7 doesn't support generated content... so ms would have to implement that for ie8 :)

[13:12:01.0000] <hsivonen>
I wonder what browser the people who complain about the single-page spec size use

[13:13:00.0000] <Philip`>
Will the IE developers be reading the HTML5 spec in IE? :-)

[13:13:01.0000] <hsivonen>
WFM in Firefox and Opera

[13:13:02.0000] <zcorpan>
hsivonen: yeah. in ie7 it is a real pain. in firefox it freezes for a second when you do autoscrolling. in opera it works fine.

[13:14:00.0000] <zcorpan>
don't know about safari

[13:14:01.0000] <hsivonen>
works fine in Opera even a device with 64 MB of RAM (haven't tried in Minimo, yet)

[13:14:02.0000] <Philip`>
It seems to work alright for me in IE6 - it takes twenty seconds to load then has a JS error (missing XMLHttpRequest), but otherwise it seems perfectly fine

[13:34:00.0000] <zcorpan>
ponder, some things in the spec are only defined in terms of ua conformance requirements, if i were to hide all ua conformance requirements then authors wouldn't know how e.g. document.URL works

[13:34:01.0000] <annevk>
conformance for APIs is a red herring

[13:34:02.0000] <annevk>
I'm not entirely sure what "red herring" implies, but I think my usage of it here is correct :)

[13:35:00.0000] <zcorpan>
annevk: what i said was that authors want to know what the APIs do

[13:36:00.0000] <annevk>
fair enough

[13:36:01.0000] <othermaciej>
I don't think "red herring" applies here

[13:36:02.0000] <othermaciej>
"red herring" implies that something was brought up as a relevant point but is actually not applicable to the issue at hand

[13:36:03.0000] <othermaciej>
a meaningless distraction

[13:36:04.0000] <annevk>
hah, ignore me then!

[13:37:00.0000] <zcorpan>
so some UA conformance requirements would have to stay in the "author subset view", or would have to be rewritten as statements of facts for authors in that view

[13:37:01.0000] <othermaciej>
conformance requirements for API use are in general dubious, since no kind of automated verification is possible

[13:37:02.0000] <othermaciej>
however, other languages such as C do have such a concept, basically any program that is syntactically valid and does not rely on undefined behavior

[13:37:03.0000] <othermaciej>
I think many C programs implicitly rely on technically undefined behavior though

[13:38:00.0000] <zcorpan>
like encoding? (just guessing here)

[13:38:01.0000] <othermaciej>
the fact that char is exactly 8 bits

[13:38:02.0000] <othermaciej>
is one of the more common ones

[13:39:00.0000] <zcorpan>
ok

[13:39:01.0000] <othermaciej>
assuming that C strings are in ASCII encoding is another

[13:40:00.0000] <zcorpan>
before i asked my teacher in programming what encoding C used. "extended ascii" he said.

[13:41:00.0000] <zcorpan>
i said i didn't know what extended ascii meant -- is it windows-1252, iso-8859-1, utf-8, or what? he couldn't answer that

[13:41:01.0000] <zcorpan>
google didn't help me either

[13:42:00.0000] <zcorpan>
it seems to me that it depends on the compiler, but i'm not sure

[13:43:00.0000] <hsivonen>
hrm. Opera doesn't do SVG arrowheads

[13:43:01.0000] <othermaciej>
it depends on the compiler and on the C library

[13:43:02.0000] <hsivonen>
Prince doesn't either. grr

[13:43:03.0000] <hsivonen>
Firefox and WebKit support them

[13:43:04.0000] <hsivonen>
yay for interop

[13:44:00.0000] <zcorpan>
othermaciej: ok

[13:55:00.0000] <Philip`>
zcorpan: Do you mean the encoding of the source code files, or of the strings passed to/from the C standard library?

[13:55:01.0000] <zcorpan>
the source code files

[13:56:00.0000] <Philip`>
C99 says "Physical source file multibyte characters are mapped, in an implementation-defined manner, to the source character set (introducing new-line characters for end-of-line indicators) if necessary."

[13:56:01.0000] <Philip`>
so that's not undefined behaviour

[13:57:00.0000] <zcorpan>
just implementation-defined?

[13:58:00.0000] <othermaciej>
implementation-defined behavior is undefined, if you assume it is being done in some specific way

[13:58:01.0000] <othermaciej>
for instance, assuming your physical source is UTF-16 and is translated to a source character set of UTF-8 would be wrong

[13:58:02.0000] <Philip`>
The source character set is letters, digits, and 29 graphic characters, and some whitespace

[13:58:03.0000] <Philip`>
"If any other characters are encountered in a source file (except in an identifier, a character constant, a string literal, a header name, a comment, or a preprocessing token that is never converted to a token), the behavior is undefined."

[13:59:00.0000] <zcorpan>
ah

[13:59:01.0000] <zcorpan>
ok

[13:59:02.0000] <zcorpan>
our teacher did printf("hallå där"); all the time

[13:59:03.0000] <Philip`>
but inside a string literal, it sounds like it allows any character (i.e. single byte)

[14:00:00.0000] <Philip`>
othermaciej: The difference is that implementation-defined behaviour is documented by the implementation, whereas undefined behaviour isn't

[14:01:00.0000] <Philip`>
(in their terminology)

[14:01:01.0000] <Philip`>
but it's as good as undefined if you can't find the documentation for it :-)

[14:03:00.0000] <annevk>
zcorpan, so to get CSS fixed some browsers need to change behavior...

[14:03:01.0000] <Philip`>
Oh, I'm wrong about it being a single byte - only the source character set must be, but you can have other multibyte characters

[14:03:02.0000] <Philip`>
(as far as I can see)

[14:04:00.0000] <zcorpan>
annevk: what are you referring to?

[14:04:01.0000] <annevk>
xhtml body

[14:04:02.0000] <zcorpan>
ah

[14:04:03.0000] <zcorpan>
yes

[14:04:04.0000] <Philip`>
but it all seems like implementation-defined behaviour - the only requirement is that '0'..'9' are sequential character codes, which I guess is true even in EBCDIC

[14:05:00.0000] <Philip`>
(In the code I've worked on, we've avoided the problems by rewriting German names to have nice ASCII "ae" instead, though that was problems with editors and SVN rather than with the compiler...)

[14:06:00.0000] <zcorpan>
annevk: so it's time to file bugs about this, eh?

[14:06:01.0000] <annevk>
think so

[14:06:02.0000] <zcorpan>
/me goes ahead and does so

[14:12:00.0000] <annevk>
whoa, w3.org is slow

[14:12:01.0000] <zcorpan>
indeed

[14:13:00.0000] <annevk>
guess the telcon took all their resources :p

[14:26:00.0000] <annevk>
although I don't entirely agree with the CSS WG (which I'm in...)

[14:27:00.0000] <annevk>
this is not about what implementations do, but about what's better for XHTML

[14:33:00.0000] <zcorpan>
is the css21 spec in cvs? where?

[14:34:00.0000] <annevk>
Member only

[14:35:00.0000] <zcorpan>
ok

[14:35:01.0000] <annevk>
not really :p

[14:36:00.0000] <zcorpan>
didn't imply that it was ok :)

[14:36:01.0000] <zcorpan>
meant ok as in "so now i know"

[14:37:00.0000] <zcorpan>
...and don't need to search further

[14:39:00.0000] <annevk>
i know (hence the smiley)

[14:44:00.0000] <zcorpan>
/me considers creating equivalent html and xhtml test cases that are the same byte by byte

[14:45:00.0000] <zcorpan>
either that or i should run a script that converts html tests to equivalent xhtml

[14:45:01.0000] <annevk>
that's doable now

[14:46:00.0000] <Philip`>
How do you test that the converter is correct?

[14:46:01.0000] <annevk>
not talking about converting

[14:46:02.0000] <annevk>
talking about byte identical tests with a different MIME type

[14:46:03.0000] <annevk>
where both are conforming

[14:47:00.0000] <zcorpan>
Philip`: with the parsing test suite?

[14:47:01.0000] <zcorpan>
annevk: [22:50] <zcorpan> either that or i should run a script that converts html tests to equivalent xhtml

[14:47:02.0000] <annevk>
/me replies to Philip`

[14:47:03.0000] <Philip`>
Ah, are xmlns and xml:lang and stuff now conforming HTML5?

[14:47:04.0000] <annevk>
replied* damn it

[14:47:05.0000] <annevk>
Philip`, xmlns is

[14:48:00.0000] <annevk>
the cool thing about that is not byte identical files though

[14:49:00.0000] <annevk>
it's that you can write <html xmlns=http://www.w3.org/1999/xhtml> and do a happy dance

[14:49:01.0000] <zcorpan>
lol

[14:53:00.0000] <Philip`>
annevk: Aha, okay. (I was remembering what I'd done for HTML->XHTML conversion, and forgot the (xml:)lang is optional)

[14:54:00.0000] <annevk>
I think we should get rid of xml:lang and just make it lang throughout

[14:54:01.0000] <annevk>
same reasons as for not adopting xml:id

[14:55:00.0000] <annevk>
actually, we still have to define that xml:lang takes precedence, when specified

[14:55:01.0000] <zcorpan>
annevk: isn't that defined already?

[14:56:00.0000] <zcorpan>
or nm

[14:56:01.0000] <annevk>
(it was in reply to me suggesting getting rid of it altogether)

[14:57:00.0000] <zcorpan>
(figured)

[14:59:00.0000] <Philip`>
The spec says the xml:lang attribute takes precedence over the lang attribute, but is that talking about the "lang" attribute in the "xml" namespace, as distinct from the "xml:lang" attribute in the default namespace (like what an HTML5 parser would produce)?

[14:59:01.0000] <hsivonen>
offtopic: do I understand correctly that <svg height="50%"> in a compound document refers to the height of the containing block if the containing block has a non-auto height and to the width of the containing block if the height of the containing block is auto?

[15:00:00.0000] <annevk>
Philip`, oh, lang in xml:

[15:00:01.0000] <annevk>
hsivonen, that sounds wrong

[15:00:02.0000] <zcorpan>
Philip`: see #terminology

[15:00:03.0000] <annevk>
hsivonen, unless it has an aspect ratio of 1:1

[15:00:04.0000] <annevk>
hsivonen, or something

[15:01:00.0000] <hsivonen>
annevk: ok. what's the right answer? what do height percentages on outer <svg> refer to?

[15:01:01.0000] <Philip`>
annevk: So that precedence rule can only apply to document deserialised from XHTML5 and not HTML5? (I think that makes sense now)

[15:02:00.0000] <annevk>
hsivonen, I'm not an expert on this, but I believe it becomes a CSS height of 50%.

[15:02:01.0000] <zcorpan>
Philip`: if you don't do scripting then yes

[15:02:02.0000] <annevk>
hsivonen, http://www.w3.org/TR/CSS21/visudet.html#inline-replaced-width and http://www.w3.org/TR/CSS21/visudet.html#inline-replaced-height would then clarify what happens

[15:03:00.0000] <annevk>
hsivonen, the problem is though that I believe the CSS WG agreed to change those sections (again!) so it's likely something different in a couple of weeks

[15:04:00.0000] <hsivonen>
annevk: WebKit and Gecko seem to follow my hypothesis

[15:04:01.0000] <hsivonen>
annevk: can't figure out what Opera does

[15:04:02.0000] <annevk>
Gecko has the wrong implementation of SVG height and width

[15:04:03.0000] <annevk>
dunno about WebKit

[15:05:00.0000] <annevk>
Opera had something close to CSS 2.1 (but now that changed!) in some build (parts might be in Opera 9)

[15:05:01.0000] <hsivonen>
annevk: so what's the intrinsic ratio for SVG? the ratio of viewBox?

[15:05:02.0000] <hsivonen>
annevk: I've got 9.20

[15:05:03.0000] <annevk>
I think so

[15:06:00.0000] <hsivonen>
considering that I'm making a document that is supposed to last years unmodified, I don't like this

[15:06:01.0000] <annevk>
SVG is either really vague about all this or it's just me

[15:06:02.0000] <annevk>
which is certainly possible

[15:06:03.0000] <annevk>
Besides 42, I suppose the answer is PNG

[15:07:00.0000] <hsivonen>
basically, I wan't to say "give me 100% and the height that is right for the viewBox aspect ratio"

[15:07:01.0000] <hsivonen>
s/100%/100% width/

[15:08:00.0000] <hsivonen>
annevk: one possibility is that height can compute to auto, but height='auto' is an error and falls back to height='100%'

[15:08:01.0000] <annevk>
/me wonders how CSS 2.1 can go to LC while changing intrinsic widths

[15:08:02.0000] <annevk>
and heights

[15:08:03.0000] <annevk>
a block with percentage height within one with height auto becomes 0

[15:08:04.0000] <hsivonen>
and it certainly doesn't help that em-sized SVG in horked in Gecko

[15:12:00.0000] <hsivonen>
rough interop with the moron method achieved between WebKit, Firefox trunk and Prince

[15:12:01.0000] <hsivonen>
unwanted results in Opera

[15:13:00.0000] <hsivonen>
I have to say that SVG/CSS cooperation has a serious counterintuitiveness or downright brokenness in this very reasonable use case

[15:13:01.0000] <annevk>
post it to www-style

[15:13:02.0000] <annevk>
maybe www-svg

[15:14:00.0000] <annevk>
I would like to be more of assistence, but people just tell me how they think it should work, without actually saying where that's defined

[15:14:01.0000] <annevk>
(for instance, the SVG WG said that percentage widths / heights on svg:svg elements don't contribute to the intrinsic width / height mechanism in CSS)

[15:15:00.0000] <annevk>
(where implementors and testcase writers assumed it they would)

[15:15:01.0000] <hsivonen>
annevk: current attempt: http://hsivonen.iki.fi/thesis/html5-conformance-checker.xhtml#back-end

[15:15:02.0000] <hsivonen>
bed time

[15:17:00.0000] <hsivonen>
annevk: thanks. I will have to investigate more

[15:18:00.0000] <hsivonen>
annevk: perhaps use the Distler em hack and use ems that equal 100% under my particular Prince conditions

[15:41:00.0000] <zcorpan>
is http://simon.html5.org/test/css/magic-body/003.htm too evil?

[15:42:00.0000] <Hixie>
too evil for what?

[15:42:01.0000] <Hixie>
it's blank on ff2

[15:43:00.0000] <Hixie>
seems quite reasonable to me

[15:43:01.0000] <zcorpan>
i mean, too much an edge case for being relevant for web content

[15:43:02.0000] <zcorpan>
however the situation is more likely in xml so

[15:44:00.0000] <Hixie>
it's not something that will be run into frequently, but it should work

[15:44:01.0000] <Hixie>
it's not a P1, but it should work

[15:44:02.0000] <zcorpan>
ok

[15:46:00.0000] <nickshanks>
i get a white page with no text in safari (title works though)

[15:46:01.0000] <Hixie>
do you get a js error in safari?

[15:47:00.0000] <Hixie>
(if anything fails at replaceChild() you'll get a blank page)

[15:47:01.0000] <zcorpan>
blank page in firefox 2 is weird. the dom looks ok and it works in firefox 3 (but it still fails the test)

[15:47:02.0000] <Hixie>
i wonder why it doesn't render anything in ff2

[15:47:03.0000] <Hixie>
weird

[15:47:04.0000] <Hixie>
oh it works on trunk? ok

[15:48:00.0000] <Hixie>
then who cares about ff2 :-)

[15:48:01.0000] <zcorpan>
indeed :)

[15:48:02.0000] <Hixie>
k

[15:48:03.0000] <Hixie>
well

[15:48:04.0000] <Hixie>
i told the htmlwg that r1000 was the version we were putting forward

[15:48:05.0000] <Hixie>
i guess i'd better get cracking on commiting changes

[15:49:00.0000] <Hixie>
since we're just under r700 at the moment

[15:50:00.0000] <zcorpan>
if you want me to bug you about typos then shout ;)

[15:50:01.0000] <Hixie>
nah

[15:50:02.0000] <Hixie>
still too early for typos 

[15:50:03.0000] <zcorpan>
</kidding>

[15:50:04.0000] <Hixie>
i think i'll just start responding to feedback

[15:50:05.0000] <Hixie>
oh there'll come a time where typos will be critical :-)

[15:50:06.0000] <Hixie>
that's when the spec is done :-P

[15:50:07.0000] <zcorpan>
yeah

[15:57:00.0000] <jgraham>
http://wordsandpictures.dyndns.org/html5/html5view.html  (this is very much unpolished)

[15:57:01.0000] <jgraham>
(and probably very buggy)

[16:00:00.0000] <zcorpan>
no type=url?

[16:01:00.0000] <jgraham>
zcorpan: Did I not say unpolished ;)

[16:01:01.0000] <zcorpan>
you did :)

[16:01:02.0000] <Philip`>
Hmm, I guess google.com is a bad page to try it on, since it has an extra link (Scholar) in the top one...

[16:02:00.0000] <Philip`>
yahoo.com is bad too since they ask html5lib to upgrade to a newer browser

[16:03:00.0000] <zcorpan>
got an error for http://simon.html5.org/sandbox/html/w3c-home-in-html5

[16:04:00.0000] <jgraham>
So I guess UA spoofing would help...

[16:04:01.0000] <Philip`>
And microsoft.com seems to generate no response from the lower window :-(

[16:05:00.0000] <jgraham>
zcorpan: Known bug in html5lib. Hixie hasn't specced what to do with <header> and other new tags yet...

[16:05:01.0000] <zcorpan>
jgraham: aren't they covered by "anything else" then?

[16:05:02.0000] <Philip`>
msdn.microsoft.com causes "<type 'exceptions.UnicodeEncodeError'>: 'ascii' codec can't encode character u'\u222b' in position 217: ordinal not in range(128) "%(value, node.attributes[key]) UnicodeEncodeError: 'ascii' codec can't encode character u'\u222b' in position 217: ordinal not in range(128) -->"

[16:05:03.0000] <zcorpan>
(for now, that is)

[16:05:04.0000] <bewest>
jgraham: is this just two iframes?

[16:06:00.0000] <Philip`>
Oops, the latter half of the error is caused by printing the traceback unescaped when it contains a line "print "<!-- link %s changed to %s-->"%(value, node.attributes[key])"

[16:07:00.0000] <jgraham>
bewest: Yeah two iframes. However the html5lib frame has to build up the DOM using js to avoid browser parsing errors

[16:07:01.0000] <bewest>
ah

[16:08:00.0000] <zcorpan>
jgraham: or use xhtml5? although i guess that won't work in ie :)

[16:08:01.0000] <bewest>
so it renders a tree constructed using html5 rules in the browser?

[16:08:02.0000] <jgraham>
s/errors/oddities/

[16:08:03.0000] <jgraham>
bewest: Yeah, that's the idea

[16:10:00.0000] <jgraham>
Hmm. Clearly I need to work on I18n

[16:10:01.0000] <Philip`>
w3.org looks different in the two - is this a quirks vs standards issue?

[16:10:02.0000] <Philip`>
(The spacing around the logo and between some lines is different)

[16:10:03.0000] <bewest>
ah... it makes sense when I hit the url from the command line

[16:11:00.0000] <bewest>
that loadom=1 is insane

[16:12:00.0000] <bewest>
oh loaddom=1

[16:13:00.0000] <jgraham>
Philip`: possibly. Firefox has helpfully taken the rendering mode out of the page info dialog...

[16:13:01.0000] <bewest>
jgraham: why show the rendered version?  from the initial description I thought it would show a comparative visualization of the two trees as opposed to a rendering of them 

[16:13:02.0000] <zcorpan>
it is

[16:13:03.0000] <bewest>
hmmmm

[16:15:00.0000] <zcorpan>
jgraham: the second iframe should probably either always use <!doctype html> or use the same prolog as the input document

[16:15:01.0000] <jgraham>
bewest: http://wordsandpictures.dyndns.org/html5/parsetree.html almost does what you want (it shows the html5 tree which you can compare with the live DOM viewer)

[16:15:02.0000] <Philip`>
jgraham: Oh, I hadn't noticed they'd changed that, but it does like quite <sarcasm>helpful</sarcasm> now

[16:17:00.0000] <jgraham>
zcorpan: It should keep the same doctype (but obviously doesn't)

[16:18:00.0000] <bewest>
jgraham: nifty

[16:19:00.0000] <Philip`>
Looks like it adds <!DOCTYPE HTML> for HTML 4.01 pages, but not for XHTML pages (or at least ones with <?xml ...?>)

[16:20:00.0000] <Philip`>
(Doctype sniffing sounds like fun - let's do more of it!)

[16:22:00.0000] <zcorpan>
/me wonders if he should have a third set of tests -- html in quirks mode

[16:23:00.0000] <zcorpan>
and almost standards mode? man, all these modes makes creating test cases trivial indeed

[16:23:01.0000] <zcorpan>
let's introduce more rendering modes while we're at it

[16:25:00.0000] <zcorpan>
/me will try to get the "almost" and "full" modes merged, i.e. get the "almost" mode specced in css21

[16:25:01.0000] <zcorpan>
wish me luck

[16:25:02.0000] <Hixie>
we will shortly be getting more rendering modes!

[16:25:03.0000] <zcorpan>
yay! :(

[16:27:00.0000] <othermaciej>
ah, rendering modes

[16:28:00.0000] <zcorpan>
i wish there was only quirks mode

[16:28:01.0000] <Hixie>
i wish there was only one mode and it matched the specs

[16:29:00.0000] <zcorpan>
me too, which effectively means the specs should have adopted to quirks mode a few years back

[16:32:00.0000] <zcorpan>
i'll aim to merge almost and full, and reduce the number of html vs xhtml differences to a minimum, and then spec quirks... i have no idea what to do with whatever ms comes up with

[16:32:01.0000] <zcorpan>
hopefully other vendors won't have to implement new modes

[16:39:00.0000] <Hixie>
hopefully

[16:40:00.0000] <zcorpan>
and hopefully the web will use new features in quirks mode and "html4-mode" so ms have to implement them there as well :)

[16:42:00.0000] <zcorpan>
not that i think it will affect them, i mean most authors have to make it work in current IEs somehow (so long as they are majority vendor anyway)

[16:43:00.0000] <othermaciej>
Chris Wilson is definitely giving other vendors a lot more motive to reduce IE market share

[16:43:01.0000] <Hixie>
in a "win or die" kind of way, yeah

[16:45:00.0000] <Philip`>
zcorpan: People using SVG or canvas today don't seem to worry much about making it work in current IEs

[16:45:01.0000] <Hixie>
/me comes across an old thread where someone is asking for something, an xforms person replies with "xforms can do that!", and the original commentor dismisses that as being irrelevant to web browsers

[16:45:02.0000] <Philip`>
though there aren't exactly many of those people

[16:45:03.0000] <Hixie>
not one of the usual suspects, either

[16:46:00.0000] <zcorpan>
Philip`: it's possible to make both work in ie using script

[16:46:01.0000] <zcorpan>
i.e., "work"

[16:48:00.0000] <zcorpan>
to address the xslt use-case, we should make the html4 strict doctype conforming :)

[16:49:00.0000] <zcorpan>
(and perhaps xhtml1 strict too)

[16:49:01.0000] <Hixie>
wow, we basically hit almost everything on this list: http://www.joelonsoftware.com/oldnews/pages/June2004.html

[16:52:00.0000] <jdandrea>
/me reads the 17 June 2004 entry - ha-HA! Nice.

[16:53:00.0000] <othermaciej>
nice :-)

[16:53:01.0000] <othermaciej>
can datagrid be used as a tree/outline view?

[16:53:02.0000] <Hixie>
yeah

[16:54:00.0000] <othermaciej>
compressed JS can already be done via http 1.1

[16:54:01.0000] <othermaciej>
(trying to find what is missing)

[16:54:02.0000] <Hixie>
modal dialogs are missing

[16:54:03.0000] <Hixie>
but that's intentional

[16:54:04.0000] <bewest>
not sure why he wants... yeah exactly

[16:54:05.0000] <jdandrea>
(See also the 25 June 2004 entry with quotes from Brendan Eich and hixie!)

[16:54:06.0000] <Hixie>
contentEditable isn't perfect, but that's hard to do, i don't know what else to do about it

[16:54:07.0000] <bewest>
why would we want a modal windowing model on the web?

[16:56:00.0000] <Philip`>
(If datagrid is for trees, could it do something like the 'expand' links in http://canvex.lazyilluminati.com/tests/tests/, maybe with some scripting/styling? Er, I should probably read it myself some time to see...)

[16:57:00.0000] <Hixie>
i don't think it trivially does that, no

[16:57:01.0000] <Hixie>
it only supports showing all or nothing at a particular node and level


2007-04-27
[17:05:00.0000] <zcorpan>
http://simon.html5.org/test/css/magic-body/ -- are there any cases i'm missing for body-magicness wrt background-color in html?

[17:11:00.0000] <zcorpan>
oh yes, a body that is not child of root

[17:15:00.0000] <othermaciej>
I have an increasingly hard time understanding anything David P. Dailey says

[17:15:01.0000] <bewest>
/me too

[17:18:00.0000] <zcorpan>
"and must not paint a background for that BODY element" -- guess i have to make the HEAD visible and position it behind the BODY to test this (without using background-image that is)

[17:21:00.0000] <Dashiva>
The www-html noise is still going? Wow

[17:22:00.0000] <zcorpan>
/me has unchecked the "check for new messages every [5__] minutes" option

[17:45:00.0000] <Hixie>
wow, now we have people claiming XForms is an HTML language

[18:03:00.0000] <kingryan>
Hixie:  it *does* have angle brackets, so they *must* be compatible

[18:03:01.0000] <kingryan>
right?

[18:04:00.0000] <zcorpan>
it has *more* angle brackets... and it is xml... so it must also be more semantic

[18:04:01.0000] <zcorpan>
and free of bugs in implementations

[18:05:00.0000] <Dashiva>
They're both *ML

[18:05:01.0000] <kingryan>
...and easy to parse (there'll be libraries!) and it will save kittens

[18:05:02.0000] <zcorpan>
and all XForms on the web will be conforming

[18:06:00.0000] <zcorpan>
mobile-friendly!

[18:06:01.0000] <zcorpan>
poor mobiles can't afford to parse old HTML forms

[18:08:00.0000] <othermaciej>
Hixie: better yet, we have people claiming that it is ok for a spec to tell you how to process a model, but not how to choose the components that go into that model

[18:08:01.0000] <zcorpan>
or perhaps xforms advocates don't talk about mobiles?

[18:08:02.0000] <Hixie>
/me sighs as more xforms FUD is sent to the public-html list

[18:08:03.0000] <Hixie>
othermaciej: yes, that entire e-mail was baffling to me

[18:08:04.0000] <othermaciej>
I hope that claim is mistaken

[18:08:05.0000] <othermaciej>
otherwise XForms is worse than I thought

[18:09:00.0000] <Hixie>
certainly the claim that xforms is ok in text/html is completely bogus

[18:09:01.0000] <othermaciej>
but it works in IE in his plugin!

[18:09:02.0000] <zcorpan>
/me goes back to focus on creating test cases

[18:10:00.0000] <Hixie>
mark has been defending it for years, i don't know if he even realises how far from the standards we consider that

[18:12:00.0000] <othermaciej>
well, apparently he thinks standards are grab bags of ideas to use as building blocks

[18:12:01.0000] <Hixie>
then why be so anti-wf2?

[18:15:00.0000] <Hixie>
othermaciej: so in bug 3958, support for showModalDialog was added to webkit. Should it be in the spec?

[18:16:00.0000] <othermaciej>
Hixie: we added it at the request of a specific vendor, they make some enterprise app which has a web interface (generally used on intranet sites)

[18:16:01.0000] <othermaciej>
I don't know of significant use on the public web

[18:16:02.0000] <Hixie>
k

[18:16:03.0000] <othermaciej>
I don't know if Firefox has it

[18:16:04.0000] <Hixie>
it doesn't to my knowledge

[18:16:05.0000] <Hixie>
i'll leave it out for now then

[18:16:06.0000] <zcorpan>
hmm, i will have about 8-10 tests for each of background-color, background-image, and overflow, then duplicated as XML (perhaps even duplicated again for quirks)... should i submit meta-bugs to vendors or separate bugs where they fail?

[18:16:07.0000] <othermaciej>
it might be worth speccing anyway, but it's messy and hard to implement

[18:17:00.0000] <Hixie>
oh?

[18:17:01.0000] <Hixie>
normal nested event loop problem?

[18:17:02.0000] <Hixie>
or worse?

[18:18:00.0000] <othermaciej>
I'll look up the bug in our internal bug tracker

[18:19:00.0000] <othermaciej>
I don't know how it would be for other engines, but in our case this is the only situation where a nested event loop is actually supposed to display web content, and it caused all sorts of quirks and bugs

[18:19:01.0000] <Hixie>
ah interesting

[18:19:02.0000] <Hixie>
makes sense

[18:20:00.0000] <othermaciej>
Mozilla apparently has a window.open argument of "modal"

[18:20:01.0000] <Hixie>
re-ally

[18:20:02.0000] <Hixie>
interesting

[18:20:03.0000] <Hixie>
/me pokes

[18:21:00.0000] <othermaciej>
apparently some chinese sites use it too

[18:21:01.0000] <othermaciej>
or at least did a while ago

[18:21:02.0000] <othermaciej>
this was cited as using it in 2003: http://www.dangdang.com

[18:21:03.0000] <Hixie>
wouldn't surprise me

[18:21:04.0000] <othermaciej>
it is apparently in some common shopping cart JS used in china

[18:21:05.0000] <Hixie>
interesting

[18:21:06.0000] <Hixie>
i wouldn't have any idea how to trigger it on that site

[18:21:07.0000] <othermaciej>
2) Go to the web site: http://www.dangdang.com

[18:21:08.0000] <othermaciej>
3) Change the text encoding to Simplified Chinese (Mac OS)

[18:21:09.0000] <othermaciej>
4) Click the button - buy

[18:21:10.0000] <othermaciej>
* RESULTS

[18:21:11.0000] <othermaciej>
Expected: the shoppingcart window will be opened in a new Safari window.

[18:22:00.0000] <Hixie>
everything i click opens a new tab for me

[18:25:00.0000] <Hixie>
i can't get modal to do anything magic in firefox

[18:28:00.0000] <Hixie>
wow

[18:28:01.0000] <Hixie>
showModalDialog() is really annoying in safari :-)

[18:34:00.0000] <kingryan>
yeah, no resize, no address bar, no status bar

[18:38:00.0000] <Hixie>
can't even interact with the window below it

[18:38:01.0000] <othermaciej>
well, it's modal

[18:38:02.0000] <othermaciej>
a modal dialog

[18:38:03.0000] <othermaciej>
that's what modal dialogs do

[18:39:00.0000] <othermaciej>
ok, I guess most don't block window dragging

[18:39:01.0000] <othermaciej>
anyway, this was done as a checklist feature, not out of love

[18:40:00.0000] <Hixie>
i bet

[18:45:00.0000] <Hixie>
wow, showModalDialog is crazy

[18:45:01.0000] <Hixie>
context menus don't work

[18:45:02.0000] <Hixie>
the user can't touch the parent browsing context

[18:45:03.0000] <Hixie>
selection doesn't work

[18:45:04.0000] <Hixie>
links open in new windows

[18:45:05.0000] <Hixie>
i'm amazed more sites aren't abusing this

[18:45:06.0000] <othermaciej>
don't give them any ideas!

[18:46:00.0000] <Hixie>
wow, mark's e-mail totally dodged the original question

[18:47:00.0000] <Hixie>
he said something, he was asked to show a spec for it, he said something nonsensical, he was asked to clarify it, and he said something unrelated to the original question

[18:48:00.0000] <othermaciej>
I was really confused by his messages

[18:48:01.0000] <Hixie>
it's a kind of fud

[18:48:02.0000] <othermaciej>
I couldn't tell if he was being evasive, or was just ignorant of fairly basic things

[18:48:03.0000] <Hixie>
well, more like smokes and mirrors

[18:49:00.0000] <Hixie>
hm, i hadn't considered that  he might just be ignorant

[18:49:01.0000] <Hixie>
i guess that's possible too

[18:49:02.0000] <othermaciej>
I like to give the benefit of the doubt

[18:50:00.0000] <Hixie>
me too but i guess i figured him being ignorant was what i was giving him the benefit of not being :-)

[18:51:00.0000] <othermaciej>
fair enough

[19:09:00.0000] <Dashiva>
Hixie: acid3 test 37 expects attribute node's .specified to be false after removeAttributeNode. I can't find anything about that in dom-3-core.

[19:10:00.0000] <zcorpan>
wow, <html:body> actually is magic in opera again it seems: http://simon.html5.org/test/css/magic-body/background-color/001.xht

[19:11:00.0000] <Philip`>
Would anyone happen to be able to look in a recent version of WebKit at http://canvex.lazyilluminati.com/misc/radial/radial3.html and see if the left column is the same as http://canvex.lazyilluminati.com/misc/radial/webkit419.png ? (I'm hoping the behaviour hasn't changed in newer versions, else I'll be unhappy...)

[19:11:01.0000] <Hixie>
Dashiva: um yeah, that's bogus

[19:11:02.0000] <Hixie>
Dashiva: can you mail me a reminder?

[19:11:03.0000] <Hixie>
ian⊙hc

[19:11:04.0000] <othermaciej>
Lachy: I wish you hadn't encouraged Tina Holmboe to join the HTMLWG -- she might actually do it

[19:11:05.0000] <Hixie>
that was me, i think

[19:11:06.0000] <Hixie>
why don't we want her input?

[19:11:07.0000] <Hixie>
some of what she said was quite useful

[19:12:00.0000] <othermaciej>
you recommended it to someone else (Laura Carlson)

[19:12:01.0000] <bewest>
she thinks the w3 is out to get her

[19:12:02.0000] <Lachy>
othermaciej, sorry :-(

[19:12:03.0000] <othermaciej>
she seems more disruptive than helpful

[19:12:04.0000] <bewest>
wrt dropping mailing list subscriptions

[19:12:05.0000] <Dashiva>
Hixie: done

[19:12:06.0000] <Hixie>
thanks!

[19:12:07.0000] <Hixie>
othermaciej: ah

[19:13:00.0000] <Hixie>
well i meant to invite all of them :-)

[19:13:01.0000] <othermaciej>
doesn't matter much, we have all sorts of random sources of disruption as it is

[19:13:02.0000] <Lachy>
well, we don't want her input till after HTML5 is accepted by the HTMLWG, cause she'll object to including it

[19:14:00.0000] <Hixie>
so i'm replying to this e-mail from sebastian

[19:14:01.0000] <bewest>
careful... it's statements like those that objecters object to as evidence of who knows what

[19:14:02.0000] <Hixie>
it ends with asking whether we want to migrate xforms over time, and says:

[19:14:03.0000] <othermaciej>
oh good, then maybe I don't have to

[19:14:04.0000] <Hixie>
> If the answer is yes, we have to think about an architectural strategy

[19:14:05.0000] <Hixie>
> of introducing new features that have made XForms successful over time

[19:15:00.0000] <Hixie>
> in HTML. 

[19:15:01.0000] <othermaciej>
ah yes, if only HTML could have as much success as XForms

[19:15:02.0000] <Lachy>
I wonder what makes people think XForms has been successful

[19:16:00.0000] <Dashiva>
I think that in the long run, it's better to get them in right away, so there can't be later arguments about "They snuck WA1.0 in before I could object" later

[19:17:00.0000] <Hixie>
yeah

[19:17:01.0000] <Hixie>
i agree

[19:17:02.0000] <othermaciej>
I think the straw poll will carry in any case

[19:17:03.0000] <Hixie>
also, people are more likely to stand up for you when tehy feel part of your clique

[19:17:04.0000] <Hixie>
why do you think i invited you all to this channel?

[19:17:05.0000] <Hixie>
/me ducks

[19:17:06.0000] <Lachy>
they've had since March 7 to join, I think 2 months is sufficient time and we can dismiss such arguments

[19:18:00.0000] <othermaciej>
Hixie: yeah, we better not tell them about #whatwg-seeekrit

[19:18:01.0000] <othermaciej>
ooops!

[19:18:02.0000] <Dashiva>
#whatwg-secret-tree-house-no-patents-allowed

[19:18:03.0000] <Lachy>
/me joins that channel

[19:20:00.0000] <zcorpan>
opera still differentiates between html and xhtml wrt to 'overflow' though

[19:21:00.0000] <Dashiva>
/me ponders typos

[19:22:00.0000] <Dashiva>
Hixie: Typo in HIERARCHY (HIEARCHY) in test 35, sending you a new mail

[19:23:00.0000] <Hixie>
thanks

[19:24:00.0000] <Dashiva>
No wonder so many browsers were failing those two tests :)

[19:26:00.0000] <Hixie>
can someone think of a technology that succeeded despite being incompatible with preceeding products when it came out?

[19:26:01.0000] <Dashiva>
CDs?

[19:26:02.0000] <Philip`>
Every programming language except C++?

[19:27:00.0000] <Dashiva>
OS X, at least partially

[19:27:01.0000] <Hixie>
people made tape-deck adapters with audio leads to connect cd players to tape players, and most hifi systems just took a cd player as an additional unit

[19:27:02.0000] <zcorpan>
otoh, gecko doesn't differentiate between html and xhtml wrt 'overflow', which is nice

[19:27:03.0000] <Hixie>
os x shipped with classic for years

[19:27:04.0000] <Hixie>
programming languages all have cross-language shims

[19:28:00.0000] <bewest>
what do you mean by incompatible

[19:28:01.0000] <bewest>
and what is the scope of technology? :-)

[19:28:02.0000] <Philip`>
HDMI with HDCP? (Not sure that counts as successful yet, though)

[19:28:03.0000] <Hixie>
bewest: i guess those are good questions to which i don't really have answers

[19:29:00.0000] <Hixie>
Philip`: screens with hdmi inputs have other inputs too (e.g. dvi or vga)

[19:29:01.0000] <bewest>
does the wheel count?

[19:29:02.0000] <Dashiva>
Nobody's mentioned XHTML2 yet

[19:29:03.0000] <Dashiva>
/me ducks

[19:29:04.0000] <Hixie>
Dashiva: i said successful :-)

[19:29:05.0000] <Lachy>
XForms!

[19:30:00.0000] <bewest>
qwerty

[19:30:01.0000] <Lachy>
oh, wait, do you mean *really* successful?

[19:30:02.0000] <zcorpan>
quirks mode

[19:30:03.0000] <Hixie>
bewest: how are wheels incompatible with what came before them?

[19:30:04.0000] <Hixie>
Lachy: i guess

[19:30:05.0000] <Hixie>
Lachy: i dunno, i'm just trying to work out if the argument that backwards compatible is a requirement to be successful is true

[19:30:06.0000] <bewest>
dunno.. it was a bad joke

[19:31:00.0000] <bewest>
qwerty was a more serious entry

[19:31:01.0000] <Philip`>
Canned food?

[19:31:02.0000] <Lachy>
what?

[19:31:03.0000] <Hixie>
or if you can sidestep that requirement if the technology is suitably groundbreaking

[19:31:04.0000] <bewest>
Hixie: there are some good books on this

[19:31:05.0000] <Dashiva>
Hixie: I would say that if backwards compatability is required, and not provided, and the users cannot provide it...

[19:31:06.0000] <bewest>
I just recommended one to tantek.. let me see if it's still on his wishlist

[19:31:07.0000] <Philip`>
(Not sure if people put up with knives before they had compatible can openers, though)

[19:31:08.0000] <Hixie>
bewest: i guess compatibility isn't required for ui, then, interesting

[19:31:09.0000] <Dashiva>
Like the hifi example, users were able to work around it, even though the tech itself was incompatible

[19:32:00.0000] <Hixie>
yeah

[19:32:01.0000] <Hixie>
hmm

[19:32:02.0000] <Hixie>
i guess also cds took a decade to take off

[19:32:03.0000] <Dashiva>
Sort of like IE and other browsers :)

[19:33:00.0000] <Lachy>
BitTorrent, which was incompatible with previous P2P networks, maybe

[19:33:01.0000] <Hixie>
maybe

[19:33:02.0000] <Hixie>
hm

[19:33:03.0000] <bewest>
Hixie: http://www.amazon.com/Evolution-Useful-Things-Artifacts-Zippers-Came/dp/0679740392/ref=sr_1_1/102-7106207-0155322?ie=UTF8&s=books&qid=1177637884&sr=8-1

[19:34:00.0000] <Dashiva>
You could argue the horde aspect of BT makes it a new market, not a replacement

[19:35:00.0000] <bewest>
Hixie: anyway, the basic idea the author proposes is that technology always evolves by fixing problems and that because of this "form does not follow function"

[19:35:01.0000] <Hixie>
ah

[19:35:02.0000] <Hixie>
interesting

[19:36:00.0000] <Hixie>
well thanks for the input. i've sent the e-mail now. :-)

[19:36:01.0000] <Hixie>
i'm gonna go get some food

[19:36:02.0000] <bewest>
(eg it took the zipper over 100 years to succeed)

[19:37:00.0000] <Lachy>
really? When was it invented?

[19:37:01.0000] <bewest>
it was started in the 1800's

[19:37:02.0000] <bewest>
didn't hit critical mass till well into 1900's

[19:38:00.0000] <bewest>
there were numerous problems

[19:38:01.0000] <bewest>
it took 3 generations of dedicated inventors passionate about the future of zippers before they succeeded as a common fastening technology

[19:40:00.0000] <zcorpan>
can someone check 001.htm and 001.xht in each of the folders of http://simon.html5.org/test/css/magic-body/ in webkit and say whether there is any difference between html and xhtml handling?

[19:44:00.0000] <Lachy>
zcorpan, for background-*: the .xht files did not have green canvases

[19:44:01.0000] <Lachy>
for overflow, the .xht one had scrollbars on the box

[19:45:00.0000] <zcorpan>
Lachy: ok, thanks

[19:47:00.0000] <zcorpan>
my swift filled the canvas for the .xht, but now swift doesn't seem to work anymore (until i reboot or something)

[19:47:01.0000] <othermaciej>
so I have a question

[19:47:02.0000] <othermaciej>
what does the phrase "tag soup" actually mean?

[19:48:00.0000] <othermaciej>
I used to think it specifically meant noncomforming markup that relies on error handling and so works anyway

[19:48:01.0000] <othermaciej>
but some people seem to use it in different ways, for example, referring to any non-XML markup or markup designs they don't like as "tag soup" 

[19:52:00.0000] <Dashiva>
Hixie: test 97 returns 1 on failure (success is 7). This makes the red box turn white, which does indicate failure, but it still causes score to increase, so the test will still display 100%!!!

[19:52:01.0000] <Dashiva>
Intentional?

[19:53:00.0000] <Dashiva>
Um, test 96, not 97. 

[19:56:00.0000] <zcorpan>
/me has written 50 test cases (or 25x2) on body magicness, and calles it a day

[19:56:01.0000] <zcorpan>
s/calles/calls/

[00:05:00.0000] <MikeSmith>
Lachy - about your "well, we don't want her input till after HTML5 is accepted by the HTMLWG, cause she'll object to including it" ... 

[00:05:01.0000] <Lachy>
that was a joke

[00:06:00.0000] <Lachy>
not serious

[00:06:01.0000] <MikeSmith>
yeah, I know

[00:07:00.0000] <MikeSmith>
but was going to say it wasn't clear from the context so might be good just to have it in the channel logs that is was

[00:07:01.0000] <MikeSmith>
which it is now, so 'nuff said

[00:07:02.0000] <Lachy>
ok

[00:07:03.0000] <mpt>
/me waves to those reading the channel logs

[00:09:00.0000] <MikeSmith>
/me hopes nobody takes seriously everything he says on IRC or even on mailing lists

[00:10:00.0000] <MikeSmith>
I think maybe sarcasm and irony are pointless if you explicity mark them up as such

[00:13:00.0000] <annevk>
dude, you were not serious about that panel at XTech? :p

[00:15:00.0000] <mpt>
Maybe the channel topic should contain an unclosed <sarcasm> tag

[00:15:01.0000] <Lachy>
yeah, that would work find till someone typed </sarcasm> in a message...

[00:17:00.0000] <annevk>
http://simon.html5.org/test/css/magic-body/background-color/001.xht looks nice in Opera

[00:17:01.0000] <annevk>
Lachy, <sarcasm> is like <plaintext> though

[00:18:00.0000] <annevk>
with the difference that it does support nested elements, it just doesn't support an end tag

[00:19:00.0000] <Lachy>
ah, ok.  we should add a SARCASM value to the content model flag

[00:20:00.0000] <annevk>
it's not a tokenizing thingie

[00:20:01.0000] <MikeSmith>
annevk - the panel at XTech is an exercise in 

[00:21:00.0000] <annevk>
we just ignore the end tag

[00:21:01.0000] <MikeSmith>
surrealism

[00:21:02.0000] <annevk>
in so far that HTML is surrealistic, sure

[00:21:03.0000] <annevk>
besides ignoring the end tag, we probably have to ensure that nothing else can close it either

[00:22:00.0000] <Lachy>
but how does it work with bad nesting?  does it just keep getting reopened like unclosed inline elementts?

[00:22:01.0000] <annevk>
yeah, it needs some more thought

[00:30:00.0000] <mpt>
Make it part of the doctype

[00:32:00.0000] <Lachy>
or <meta name=sarcasm value=true>

[00:35:00.0000] <othermaciej>
explaining compatibility is frustrating

[00:35:01.0000] <othermaciej>
it's like trying to prove to someone that 2 + 2 = 4

[00:35:02.0000] <othermaciej>
it's so obvious that it is hard to explain

[00:36:00.0000] <mpt>
2 + 2 < 5

[00:37:00.0000] <Lachy>
othermaciej, try using: if you have 2 apples and you buy another 2 apples, how many apples do you have?

[00:39:00.0000] <othermaciej>
Lachy: it depends on what kind of apples!

[00:39:01.0000] <othermaciej>
Lachy: what if two are granny smiths...

[00:39:02.0000] <othermaciej>
Lachy: and two are MacBook Pros

[00:39:03.0000] <othermaciej>
or what if I plant the seeds of one of the apples?

[00:39:04.0000] <Lachy>
Apple Macs ofcourse

[00:40:00.0000] <Lachy>
the question isn't how many apples can you grow

[00:40:01.0000] <Hixie>
wait if i bury my mac i can get more?

[00:41:00.0000] <Lachy>
if you have iLife, yes

[00:42:00.0000] <othermaciej>
I think people don't understand that the set of documents HTML5 UAs must accept is considerably larger than the set of conforming HTML5 documents

[00:43:00.0000] <othermaciej>
maybe this needs to be clarified more

[00:44:00.0000] <Hixie>
wait didn't i answer that before leaving work?

[00:44:01.0000] <Hixie>
with the whole 93% of documents are syntactically non-conforming thing?

[00:44:02.0000] <Lachy>
othermaciej, maybe Don't Reinvent the Wheel could talk about use cases instead. Something like, if an existing feature already solves some related problems/use cases, consider defining and/or enhancing that, instead of inventing something new

[00:44:03.0000] <Lachy>
/me is surprised 7% are conforming!

[00:44:04.0000] <Hixie>
btw i recommend removing "break" from the page

[00:45:00.0000] <Hixie>
say it the way the second sentence does

[00:45:01.0000] <Hixie>
Lachy: i didn't say 7% were conformin

[00:45:02.0000] <Hixie>
g

[00:46:00.0000] <Lachy>
well, at least syntactically correct then

[00:47:00.0000] <othermaciej>
Hixie: I'm going to take your review comments into account

[00:51:00.0000] <othermaciej>
Hixie: I agree the second sentence is much more clear - I liked dbaron's punchy statement of the idea, but "Support Existing Content" or something dry like that seems much more ambiguous

[00:51:01.0000] <annevk>
Lachy, I added an example to don't reinvent the wheel

[00:51:02.0000] <annevk>
Lachy, contenteditable=

[00:52:00.0000] <othermaciej>
I think talking about use cases is also a good idea

[00:54:00.0000] <annevk>
heh, nice one hsivonen

[00:55:00.0000] <annevk>
/me wasn't thinking in that direction

[00:55:01.0000] <hsivonen>
annevk: you mean reinventing the wheel? what was your direction?

[00:56:00.0000] <Hixie>
Lachy: well, that's just the 7% that my script didn't catch bugs on

[00:56:01.0000] <Hixie>
i doubt many of them are fully conforming

[00:56:02.0000] <annevk>
when I made the principle up it was more about new features for which UAs had already a solution which was widely used and known

[00:56:03.0000] <annevk>
but this explanation fits as well, I think

[00:57:00.0000] <annevk>
s/this/your/

[00:57:01.0000] <annevk>
/me should update his slides to say 93% as opposed to 95%...

[00:59:00.0000] <othermaciej>
Hixie: ok, made changes per your comments

[01:00:00.0000] <Lachy>
it'd be interesting to do a survey of 3 billion documents using tool based on hsivonen's checker (perhapsh when it's more mature and less buggy)

[01:00:01.0000] <othermaciej>
hsivonen: I think he means XForms in HTML rendering XForms Transitional redundant

[01:00:02.0000] <annevk>
btw, saying that XForms in HTML wouldn't work is wrong

[01:01:00.0000] <annevk>
defining some kind of parsing algorithm for specific elements could make it work

[01:01:01.0000] <annevk>
the syntax would be slightly different, but it could probably be made to work

[01:01:02.0000] <annevk>
such a design just fails to meet many of the design principles

[01:02:00.0000] <hsivonen>
othermaciej: yeah, it seems to me, too, that XForms in text/html removes the premise of XForms Transitional

[01:04:00.0000] <hsivonen>
annevk: I think the most productive way forward would be making Web Forms 2.0 part of the HTML WG spec and extending the parsing algorithm to handle XForms in text/html without requiring UAs to actually implement XForms on top of the parser-produced DOM

[01:05:00.0000] <hsivonen>
(and dropping XForms Transitional and not doing any Task Force stuff)

[01:07:00.0000] <othermaciej>
hsivonen: perhaps a more general namespaces-in-HTML thingie could cover that - though I'm highly ambivalent about such a thing

[01:08:00.0000] <hsivonen>
othermaciej: namespaces-in-HTML is such a recurring theme that perhaps it needs addressing

[01:08:01.0000] <othermaciej>
I'm trying to come up with an example for "Media Independence"

[01:08:02.0000] <othermaciej>
I have a negative example:

[01:09:00.0000] <hsivonen>
either by speccing or documenting why the Opera attempt failed

[01:09:01.0000] <othermaciej>
"A hyperlink can't be actuated in a printed document, but that is no reason to omit the <a> element."

[01:09:02.0000] <othermaciej>
but I'd also like a positive one

[01:09:03.0000] <othermaciej>
hsivonen: well, IE seems to have a sort of namespacing in HTML

[01:09:04.0000] <othermaciej>
there's also XML Data Islands

[01:09:05.0000] <othermaciej>
I wonder if we want to spec that

[01:10:00.0000] <othermaciej>
/me thinks "data island" is a hilarious mental image

[01:10:01.0000] <othermaciej>
anyway, what's a decision made for reason of media/platform/device independence?

[01:10:02.0000] <othermaciej>
ideally something new in HTML5, but doesn't have to be

[01:11:00.0000] <hsivonen>
the general reflowability of HTML text

[01:11:01.0000] <hsivonen>
do we have an aural default rendering suggestion for <aside>?

[01:12:00.0000] <othermaciej>
great, how about an example for "Support World Languages"?

[01:12:01.0000] <othermaciej>
does HTML5 still have the bdo element?

[01:12:02.0000] <othermaciej>
or, heck, just being based on unicode is good for that

[01:14:00.0000] <hsivonen>
othermaciej: based on Unicode including astral planes, <i> should be kept, because it is used a lot for bicameral scripts even though italics per se may be inapplicable to some scripts. <ruby> is legitimate even though it has a strong CJK focus.

[01:14:01.0000] <othermaciej>
We don't have <ruby> yet so I am not sure if I should cite it

[01:14:02.0000] <hsivonen>
iirc, bdo was still there

[01:15:00.0000] <othermaciej>
I just mentioned supporting different character sets

[01:16:00.0000] <hsivonen>
othermaciej: I don't think we should emphasize supporting different encodings. we should emphasize Unicode being the one size fits all solution

[01:16:01.0000] <othermaciej>
ok

[01:16:02.0000] <othermaciej>
does HTML5 explain the cross-site scripting security model yet?

[01:17:00.0000] <hsivonen>
othermaciej: <ruby> is a good example to balance <i>, because only mentioning <i> makes the example look Latin-centric while <ruby> is the "look it applies to something non-Latin" example

[01:17:01.0000] <othermaciej>
oh, wait, cross-document messaging is a good eample

[01:17:02.0000] <othermaciej>
hsivonen: ok

[01:25:00.0000] <virtuelv>
othermaciej: for what purpose are you looking for "Media independence" 

[01:25:01.0000] <virtuelv>
and what do you mean by it?

[01:25:02.0000] <Hixie>
we need to support <ruby>

[01:25:03.0000] <Hixie>
we don't have suggested aural renderings yet. we also don't have suggested visual renderings.

[01:25:04.0000] <Hixie>
(we need both)

[01:26:00.0000] <Hixie>
an example of something that was done for media independence would be the way drag-and-drop is defined

[01:26:01.0000] <Hixie>
it's defined in such a way that it actually handles most ui paradigms, including copy/paste, using the same mechanism

[01:27:00.0000] <Lachy_>
I thought there were suggested renderings for <ruby> in its existing spec

[01:28:00.0000] <othermaciej>
I added a bunch of examples to http://esw.w3.org/topic/HTML/ProposedDesignPrinciples

[01:28:01.0000] <othermaciej>
feel free to suggest better or just edit them

[01:28:02.0000] <Lachy_>
(or maybe I just misunderstood what hixie wrote above)

[01:28:03.0000] <virtuelv>
Hixie: is there anything done?

[01:29:00.0000] <virtuelv>
(Ok, I'll stop playing devil's advocate here)

[01:29:01.0000] <Hixie>
Lachy_: i meant in html5

[01:29:02.0000] <Hixie>
the ruby spec is inadequate for our needs, sadly

[01:30:00.0000] <Hixie>
and since i don't know anything about it, it's hard for me to spec it

[01:30:01.0000] <Hixie>
though anne sent some useful data

[01:30:02.0000] <Hixie>
so maybe it'll be possible to do

[01:30:03.0000] <Lachy_>
I assumed that :-)

[01:30:04.0000] <Hixie>
virtuelv: ?

[01:31:00.0000] <virtuelv>
Hixie: nm, I was just pointing out that not much has been achieved in terms of media independence in HTML of any version, except the underlying principles of the markup language itself

[01:33:00.0000] <Hixie>
virtuelv: how so? i use html regularly on three media (screen, tty, print) and i know other people who use others (e.g. aural) exclusively

[01:35:00.0000] <virtuelv>
Hixie: Yes, I am aware of all this, and I use screen, handheld, print regularily, I am just pointing out that media independence is inherit to HTML

[01:35:01.0000] <virtuelv>
s/inherit/inherent/

[01:35:02.0000] <Hixie>
it doesn't have to be

[01:36:00.0000] <Hixie>
layout tables and <font> make it hard to use html across media

[01:36:01.0000] <Hixie>
bed time i think

[01:36:02.0000] <Hixie>
bbl

[01:37:00.0000] <Lachy_>
othermaciej, othermaceij, "prefer specifying that technology as opposed to inventing something new..." might be better rephrased as "consider specifying that technology in preference to inventing something new..."

[01:38:00.0000] <othermaciej>
Lachy_: done

[01:39:00.0000] <othermaciej>
(well, as soon as it posts)

[01:39:01.0000] <Lachy_>
cool

[03:01:00.0000] <othermaciej>
/me sighs at all the email from schepers

[05:32:00.0000] <Dashiva>
"Surely any browser manufacturer is always going to have a mode that will render older pages. What is preventing them having an HTML5 mode, which may or may not build upon their previous engine."

[05:32:01.0000] <Dashiva>
The modes just won't go away

[05:36:00.0000] <Lachy>
adding a new mode just for HTML5 doesn't help improve interop with legacy content

[05:40:00.0000] <virtuelv>
Lachy: quite the opposite

[05:40:01.0000] <Lachy>
what?

[05:40:02.0000] <virtuelv>
at least the MS "solution" as proposed is going to perpetuate non-interop

[05:41:00.0000] <Lachy>
yeah

[05:41:01.0000] <Lachy>
perhaps you misread what I wrote

[05:41:02.0000] <nickshanks>
are there any multilinguists in here? i need help with plural forms for as many languages as possible

[05:42:00.0000] <Dashiva>
I think his 'opposite' was 'not just no help, it actually makes it worse'

[05:42:01.0000] <zcorpan>
swedish, english, dutch, some german

[05:42:02.0000] <virtuelv>
nickshanks: for what purpose?

[05:42:03.0000] <MikeSmith>
nickshanks - as far as Japanese, plural forms are not normally used

[05:42:04.0000] <virtuelv>
Dashiva: correct assumption

[05:42:05.0000] <MikeSmith>
or used only very occasionally

[05:42:06.0000] <nickshanks>
virtuelv: for selecting correct translations

[05:43:00.0000] <nickshanks>
see the table at the bottom of http://doc.trolltech.com/qq/qq19-plurals.html

[05:43:01.0000] <MikeSmith>
and in Japanese, plurals are used only for people, as far as I know

[05:44:00.0000] <Dashiva>
You can use it for other things too, with some 'besjelening', not sure of the English word

[05:44:01.0000] <nickshanks>
i want that table in as many languages as possible

[05:45:00.0000] <zcorpan>
nickshanks: i don't understand the table

[05:45:01.0000] <nickshanks>
it's logic for selecting the correct translation when given an input number n

[05:46:00.0000] <nickshanks>
e.g. english has two forms, a singular for n==1, and a plural used otherwise

[05:46:01.0000] <nickshanks>
0 houses, 1 house, 2 houses...

[05:46:02.0000] <MikeSmith>
so what is Form 3?

[05:46:03.0000] <Dashiva>
I can vouch for the table in English and Japanese

[05:46:04.0000] <Dashiva>
Some languages have more than just singular and plural

[05:47:00.0000] <nickshanks>
MikeSmith: english has no form three

[05:47:01.0000] <zcorpan>
swedish: 0 hus, 1 hus, 2 hus (but this is not generalizable)

[05:47:02.0000] <Dashiva>
Think 'one, both, all'

[05:48:00.0000] <Dashiva>
/me doesn't try to explain polish

[05:48:01.0000] <nickshanks>
well if anyone can provide expansions to that table, just post them here and I'll add them to my implementation

[05:49:00.0000] <Dashiva>
Norwegian is like English

[05:49:01.0000] <virtuelv>
generalizing that is going to be hard

[05:50:00.0000] <nickshanks>
Dashiva: nn, no, or nb ?

[05:50:01.0000] <zcorpan>
swedish is sometimes +ar or +or or +er or no change for plural, and sometimes a vowel changes (e.g. foot/feet -- fot/fötter)

[05:50:02.0000] <Dashiva>
zcorpan: He's not asking about how the plural changes, just for which numbers plural is used

[05:51:00.0000] <nickshanks>
zcorpan: well the translators may have to get creative, i am just implementing the switching logic.

[05:51:01.0000] <zcorpan>
ok

[05:51:02.0000] <Dashiva>
nickshanks: all of them, I'd say

[05:52:00.0000] <nickshanks>
Dashiva: thx

[05:52:01.0000] <Dashiva>
Considering trolltech is norwegian, wouldn't the default code include it though?

[05:52:02.0000] <nickshanks>
i am trying to port the logic to Mac

[05:53:00.0000] <nickshanks>
the Qt page i found via google

[08:11:00.0000] <Voluminous>
it

[10:18:00.0000] <mpt>
nickshanks, we've been collecting plural forms for Launchpad Translations, unfortunately we don't have a one-page list but you can search at https://translations.launchpad.net/+languages

[10:20:00.0000] <nickshanks>
how does it work?

[10:21:00.0000] <nickshanks>
i'm typing in things and getting nothing :)

[12:30:00.0000] <deltab>
nickshanks: search for e.g. Polish

[12:30:01.0000] <nickshanks>
i looked for gaelic, dutch and français then gave up :)

[12:30:02.0000] <deltab>
it's case sensitive too

[12:31:00.0000] <deltab>
https://translations.launchpad.net/+languages/gd

[12:31:01.0000] <deltab>
https://translations.launchpad.net/+languages?language_search=Dutch

[12:38:00.0000] <nickshanks>
how odd, no-one speaks middle dutch

[16:11:00.0000] <Dashiva>
Hixie: More acid3. Test 96 removes the lastchild, making white-space:pre-wrap apply to #instructions. This doesn't seem to be reflected in the reference rendering

[16:11:01.0000] <Hixie>
oh?

[16:11:02.0000] <Hixie>
oh

[16:11:03.0000] <Hixie>
yeah, i should make the paragraph be one long line in the source

[16:11:04.0000] <Hixie>
duh

[16:11:05.0000] <Hixie>
can i just say that you're amazingly helpful here.

[16:12:00.0000] <Hixie>
fixed the bug

[16:12:01.0000] <Dashiva>
I'm just trying to keep up with how opera fails it :)

[16:12:02.0000] <Hixie>
:-)

[16:13:00.0000] <Hixie>
what's your full name again?

[16:13:01.0000] <Hixie>
so i can add you to the acknowledgments

[16:13:02.0000] <Dashiva>
Magnus Kristiansen

[16:13:03.0000] <Hixie>
thanks

[16:59:00.0000] <Hixie>
so... anyone against making line feeds in title="" be relevant?

[16:59:01.0000] <Hixie>
as in, have them be real line breaks that you're supposed to honour?


2007-04-28
[17:01:00.0000] <Dashiva>
I can see the use, but at the same time it might lead to a slippery slope of more styling of titles

[17:01:01.0000] <othermaciej>
how does attribute parsing normally deal with whitespace in attributes?

[17:02:00.0000] <othermaciej>
I don't mind if it doesn't have to be a parser special case

[17:02:01.0000] <Hixie>
html5 preserves everything

[17:02:02.0000] <Hixie>
(html4 did per-attribute special casing through the fancy sgml rules)

[17:13:00.0000] <hsivonen>
and XML normalizes line feeds to spaces unless escaped

[17:21:00.0000] <Hixie>
ok so people can subscribe to the commit-watchers mailing list, and i can post to it...

[17:21:01.0000] <Hixie>
so why doesn't it work

[17:21:02.0000] <Hixie>
grr

[17:55:00.0000] <a-ja>
Hixie, hsivonen: details' content model says legend element required first child (which is what the conformance checker wants), but details' dom interface text says legend is *optional* 1st element.  difference between conforming document and what UA should put in dom in case of non-conforming document, or ?  

[18:20:00.0000] <Hixie>
where does the dom interface say it's optional?

[18:29:00.0000] <a-ja>
sorry, Hixie....was on another channel

[18:29:01.0000] <a-ja>
The first element child of a details  element, if it is a legend element, represents the summary of the details.

[18:29:02.0000] <a-ja>
If the first element is not a legend element, the UA should provide its own legend (e.g. "Details"). 

[18:29:03.0000] <Hixie>
that doesn't say it's optional

[18:30:00.0000] <Hixie>
it just says what happens when the element is missing

[18:31:00.0000] <a-ja>
gotcha...UA dom requirement vs what makes a conforming document.

[18:31:01.0000] <Hixie>
right

[18:32:00.0000] <othermaciej>
there are many cases where UAs are required to handle particular non-conforming cases

[18:33:00.0000] <Dashiva>
Hixie is too diligent at work. He's impeding my daily lggwg fix :)

[18:34:00.0000] <Hixie>
heh

[18:36:00.0000] <a-ja>
dunno if it's just me finding this sorta thing a wee bit confusing...from some of the html-wg comments i've been seeing.  almost wondering if separate "conforming document" and "UA behavior" doc specs might help that.

[18:38:00.0000] <a-ja>
any clue how the vote's going so far?

[18:38:01.0000] <Dashiva>
These? http://www.w3.org/2002/09/wbs/40318/htmlbg/results

[18:39:00.0000] <othermaciej>
a-ja: I think it's true that a lot of people don't quite get it

[18:40:00.0000] <a-ja>
tks...wasn't aware running vote total was anywhere public

[18:40:01.0000] <Hixie>
a-ja: i think we'll have to have separate tutorial pages that help with that, but if you have suggestions for how to make the spec clearer please let me know

[18:40:02.0000] <othermaciej>
I think what is needed is a good clear explanation

[18:40:03.0000] <othermaciej>
not whole separate docs

[18:40:04.0000] <othermaciej>
for WG members

[18:40:05.0000] <Hixie>
yeah

[18:40:06.0000] <Hixie>
that too

[18:40:07.0000] <Dashiva>
I don't know if UA and author sections are completely separate. If they are, they could be given classes so they could be styled with e.g. different backgrounds

[18:41:00.0000] <a-ja>
yeah....maybe just a "how to read this spec" intro section or somesuch

[18:41:01.0000] <Hixie>
there is a "how to read the spec" section

[18:41:02.0000] <a-ja>
doh!

[18:41:03.0000] <Hixie>
http://www.whatwg.org/specs/web-apps/current-work/#how-to

[18:41:04.0000] <a-ja>
hard keeping up with you :)

[18:42:00.0000] <Hixie>
:-)

[18:42:01.0000] <Dashiva>
/me considers that spec an atheist error

[18:42:02.0000] <Dashiva>
*that section

[18:44:00.0000] <a-ja>
Hixie: abstained?  sheesh!

[18:45:00.0000] <Hixie>
hm?

[18:46:00.0000] <a-ja>
you 'abstained' in the vote? 

[18:47:00.0000] <wilhelm>
Probably the appropriate option to choose..(c;

[18:47:01.0000] <Hixie>
google abstained from the vote

[18:47:02.0000] <Hixie>
(i'm one of several people representing google)

[18:47:03.0000] <a-ja>
ah

[18:47:04.0000] <Hixie>
i figured that it would be inappropriate for me to vote for myself or for my work

[18:48:00.0000] <Hixie>
and google, as a company, doesn't have a strong opinion either way

[18:48:01.0000] <Hixie>
(we have people in the xforms wg too, e.g.)

[18:50:00.0000] <a-ja>
noticed any WAI-types in html-wg?

[18:54:00.0000] <a-ja>
my impression is that accessibility specialists are kinda conspicuous by their absense.....wonder why.

[19:13:00.0000] <zcorpan>
does anyone know which spec defines that <!-- is ignored by JS engines? couldn't find anything in ecma262

[19:13:01.0000] <othermaciej>
zcorpan: I don't think it is in any spec

[19:15:00.0000] <Hixie>
it's not defined in any spec but it should be in ecma262

[19:16:00.0000] <zcorpan>
http://forums.whatwg.org/viewtopic.php?t=38#179

[19:18:00.0000] <zcorpan>
is ecma262 being developed?

[19:19:00.0000] <Hixie>
yejs

[19:19:01.0000] <Hixie>
yes

[19:19:02.0000] <Hixie>
even

[19:20:00.0000] <Dashiva>
As far as I know, <!-- counting as // in JS is simply a result of browsers not wanting two different JS parsers for external and internal scripts

[19:25:00.0000] <zcorpan>
ah. i thought it was only special on the first line. it is indeed equivalent to //

[19:25:01.0000] <Hixie>
correct

[19:25:02.0000] <Dashiva>
It's an outrage, I use the valid JS a<!--b all the time (not) :)

[19:26:00.0000] <Hixie>
heh

[19:27:00.0000] <othermaciej>
zcorpan: I'm not sure it is fixed in the latest ecma262

[19:27:01.0000] <othermaciej>
zcorpan: I need to make a list of JS compat issues that are missing from or contrary to the spec

[19:28:00.0000] <othermaciej>
like for..in enumeration order, reserved keyword leniency, <!--, etc

[19:28:01.0000] <Dashiva>
othermaciej: regexp has some nasty ones

[19:28:02.0000] <zcorpan>
ok

[19:29:00.0000] <othermaciej>
Dashiva: good point

[19:30:00.0000] <Hixie>
hm

[19:30:01.0000] <Hixie>
in cross-document messaging

[19:31:00.0000] <Hixie>
do we want to be passing Documents around, or Windows?

[19:31:01.0000] <Hixie>
if we pass Documents around we have to let XSS access postMessage()

[19:31:02.0000] <Hixie>
rigth now i think Documents are just locked down totally, right?

[19:31:03.0000] <Dashiva>
Windows, please

[19:32:00.0000] <Dashiva>
Unless you can recovery window from document

[19:32:01.0000] <Dashiva>
-y

[19:32:02.0000] <Hixie>
you can go both ways these days

[19:32:03.0000] <Hixie>
window.document.defaultView == window

[19:33:00.0000] <Dashiva>
Then it's not critical which way. I'd rather prefer window, though. It's the natural root element when scripting is concerned

[19:33:01.0000] <Hixie>
need to get a hold of the opera guys

[19:33:02.0000] <Hixie>
since this changes their implementation

[19:35:00.0000] <othermaciej>
Hixie: in Safari we don't allow xss access to the document at all

[19:51:00.0000] <zcorpan>
http://simon.html5.org/temp/author-view-of-html5.css -- this was how i planned to hide (or mark) parts of the spec that don't apply to authors (as an alternative stylsheet)... it seems to be the simplest way to me

[19:53:00.0000] <othermaciej>
Hixie: what do you think about adding a toggle() operation to classList? adds the token if not present, removes it if present

[20:25:00.0000] <Hixie>
othermaciej: seems reasonable; send a mail?

[20:30:00.0000] <othermaciej>
Hixie: sure

[20:31:00.0000] <Hixie>
thanks

[20:31:01.0000] <othermaciej>
done

[20:49:00.0000] <Hixie>
horrah, the commit-watchers mailing list works!

[20:49:01.0000] <zcorpan>
yay!

[20:50:00.0000] <Hixie>
turns out i'd forgotten to add the e-mail address it was sending from to the commit-watchers list

[20:50:01.0000] <Hixie>
so the list was like "hey stop spamming me"

[20:50:02.0000] <Hixie>
but since i'd turned off moderation...

[21:13:00.0000] <Hixie>
why the fuck does mozilla support target="_new"

[21:13:01.0000] <Hixie>
where did they get that from

[21:15:00.0000] <othermaciej>
Hixie: netscape legacy I'd guess?

[21:15:01.0000] <Hixie>
IE doesn't do it

[21:15:02.0000] <othermaciej>
that's why I think it is just netscape legacy, and not a real compat issue

[04:17:00.0000] <annevk>
For the internal subset only entity and attribute declarations matter right?

[04:17:01.0000] <annevk>
Attribute declarations for attributes of type ID and entity declarations (and DTD level entity declarations) for entities used in the document

[04:21:00.0000] <Lachy>
I assume parameter entities would matter too

[04:22:00.0000] <annevk>
Default attributes too :(

[04:22:01.0000] <annevk>
And entities can actually contain markup?

[04:23:00.0000] <annevk>
Lachy, yeah, that's what I meant with DTD level entity decl...

[04:23:01.0000] <Lachy>
oh

[04:23:02.0000] <annevk>
/me is not too good with the terminology

[04:24:00.0000] <Lachy>
I assume external entities don't need to be fetched, though

[04:24:01.0000] <annevk>
no, never

[04:24:02.0000] <Lachy>
good

[04:24:03.0000] <annevk>
that concept is dropped from XML5

[04:24:04.0000] <annevk>
no need to be  backwards compatible with that

[04:36:00.0000] <annevk>
the internal subset legacy is painful

[04:37:00.0000] <annevk>
see the last example in http://www.w3.org/TR/xml/#sec-entexpand for instance

[04:41:00.0000] <Lachy>
yikes!

[04:41:01.0000] <Lachy>
/me wishes annevk luck and runs off to have some dinner

[04:45:00.0000] <annevk>
it seems doable, but people claiming XML is simple should learn it a bit better

[04:46:00.0000] <annevk>
also stuff like data:text/xml,<!DOCTYPE html[<!ENTITY test "<">]><html>&test;</html> not being well-formed...

[04:49:00.0000] <annevk>
but inventing yet another processor is clearly not desirable

[05:01:00.0000] <annevk>
Because of this silly stuff you also have to think about recursion

[05:01:01.0000] <annevk>
consider<!DOCTYPE html[<!ENTITY test "&test;">]>

[05:02:00.0000] <othermaciej>
/me shudders

[05:03:00.0000] <annevk>
or the more complicated <!DOCTYPE html[<!ENTITY test "&test2;"><!ENTITY test2 "&test;">]>

[05:08:00.0000] <annevk>
/me ponders about error handling for that

[05:10:00.0000] <annevk>
The XML spec says this about it: "A parsed entity MUST NOT contain a recursive reference to itself, either directly or indirectly."

[05:13:00.0000] <annevk>
/me was hoping he could just put the expansion of the entity in the inputstream

[05:13:01.0000] <annevk>
this seems to prohibit that

[05:16:00.0000] <Dashiva>
/me shakes head at http://forums.whatwg.org/viewtopic.php?t=39

[05:20:00.0000] <annevk>
heh

[05:20:01.0000] <annevk>
that guy is funny

[05:21:00.0000] <annevk>
data:text/xml,<!DOCTYPE html[<!ENTITY test "<x/>">]><x x="&test;"/> is another funny case

[05:22:00.0000] <annevk>
(non-well-formed)

[05:23:00.0000] <annevk>
data:text/xml,<!DOCTYPE html[<!ENTITY test "<x/>">]><x>&test;</x> is well-formed

[05:23:01.0000] <annevk>
with "x>x" as tree

[05:24:00.0000] <annevk>
so the value of an entity needs to be parsed in context, but it needs to be parsed in a contrained environment to make sure you don't get infinite loops

[05:25:00.0000] <Philip`>
data:text/xml,<!DOCTYPE html[<!ENTITY test "<x>">]><x>&test;</x> in Firefox complains of "asynchronous entity" - did they just make that term up? I can't see anybody else calling it that

[05:25:01.0000] <annevk>
/me is not planning to introduce the eternal XML doc

[05:25:02.0000] <annevk>
their XML parser comes from a library

[05:28:00.0000] <Philip`>
Ah - in that case, did expat just make that term up? :-)

[05:32:00.0000] <annevk>
I suppose you just put the used entities within the contrained environment on a stack and once you encounter an entity you've seen before you either ignore it or print it out (without expanding it)

[05:33:00.0000] <annevk>
maybe it's an idea that if you encounter an entity you do just that... and once you're done you just feed all that into the inputstream so it can take care of <p> sequences

[05:33:01.0000] <annevk>
and such

[07:35:00.0000] <krijnh>
Typo in the WA1 draft; 1.4. Terminology, paragraph 8, 'or the node itself is there is none' => if there is no none

[07:35:01.0000] <krijnh>
Hixie: Ping :)

[07:35:02.0000] <annevk>
typos are after the spec is done material

[07:35:03.0000] <krijnh>
Okidoi

[07:36:00.0000] <krijnh>
*Okidoki

[07:36:01.0000] <annevk>
maybe we should have a wiki page that lists them :)

[07:36:02.0000] <krijnh>
Nah

[07:37:00.0000] <krijnh>
The spec isn't done with typos btw

[07:42:00.0000] <krijnh>
Is the web forms 2 project for IE still running?

[07:43:00.0000] <krijnh>
Seems to totally not work in IE7

[07:44:00.0000] <annevk>
dunno

[08:00:00.0000] <SpookyET>
Someone was critical http://www.webdevout.net/tidings/2007/04/23/the-whimzical-world-of-html-5/

[08:03:00.0000] <annevk>
he's certainly correct to assert that bugmode is a terrible idea

[08:05:00.0000] <annevk>
the parts about SGML and DTD seem to be uninformed

[08:06:00.0000] <annevk>
same for the rest

[16:24:00.0000] <zcorpan_>
Hixie: "A date or time string is a valid date or time string if the following algorithm, when run on the string, doesn't say the string is invalid." does that mean that authors have to understand the algorithm in order to know how to write their date or time strings?

[16:36:00.0000] <Dashiva>
"If second is not a number in the range 0 ≤ hour < 60, then fail."

[16:36:01.0000] <Dashiva>
Poor second, not even inside its own range

[16:37:00.0000] <zcorpan_>
/me is working on http://simon.html5.org/temp/author-view-of-html5.css

[16:39:00.0000] <zcorpan_>
it will be a very large ruleset that

[16:42:00.0000] <zcorpan_>
i hope this will help authors read and review the spec

[16:53:00.0000] <Philip`>
zcorpan_: Looks like that might be a bit fragile when the spec changes...

[16:56:00.0000] <zcorpan_>
Philip`: yes, like the status markers


2007-04-29
[17:00:00.0000] <zcorpan_>
having Hixie mark up status markers and what applies to authors or not is not useful use of his time... so if we can set up something that anyone can update it would be better

[03:55:00.0000] <ddfreyne>
The example of the "small" element (http://www.whatwg.org/specs/web-apps/current-work/#the-small) is confusing… 

[03:55:01.0000] <ddfreyne>
If "small" is used to mark up small print, including copyright notices… why can the "copyright" class name not be used on "small"?

[03:56:00.0000] <ddfreyne>
(the "copyright" predefined class name is only applicable to p and span)

[03:57:00.0000] <annevk>
The spec is not done yet so you might have encountered an error. The best way forward is to e-mail whatwg⊙wo

[03:57:01.0000] <ddfreyne>
will do

[05:29:00.0000] <zcorpan_>
ddfreyne: just read http://stoneship.org/journal/2007/html5-sucks/

[05:30:00.0000] <ddfreyne>
ah, neat

[05:30:01.0000] <ddfreyne>
the "sucks" is just there to catch attention ;)

[05:31:00.0000] <zcorpan_>
footnote: i think this is an open issue still

[05:32:00.0000] <zcorpan_>
note: there is class=note

[05:32:01.0000] <zcorpan_>
page, comment: this is what article is for

[05:32:02.0000] <ddfreyne>
class=note… yes

[05:33:00.0000] <ddfreyne>
zcorpan_: I'm against adding new elements like note/page/comment… i'd prefer the set of elements to be reduced rather than enlarged

[05:34:00.0000] <ddfreyne>
those are very big changes I'm proposing and they'll probably be ignored…

[05:34:01.0000] <zcorpan_>
i understand that you prefer that, but i don't understand why

[05:34:02.0000] <zcorpan_>
i don't think they'll be ignored :)

[05:35:00.0000] <zcorpan_>
consider this: would you prefer if html4 did the same 10 years ago? <div role="table"> instead of <table>?

[05:35:01.0000] <zcorpan_>
<div role="object">

[05:35:02.0000] <ddfreyne>
table, object… are much more specific than <article>

[05:36:00.0000] <ddfreyne>
hm… sec

[05:37:00.0000] <ddfreyne>
<article> by itself doesn't have any article-specific attributes

[05:37:01.0000] <zcorpan_>
should there be?

[05:38:00.0000] <ddfreyne>
something like a hAtom subset converted to elements where appropriate (<article> instead of <div class="hentry">) would be useful

[05:38:01.0000] <ddfreyne>
I think it'd be useful

[05:39:00.0000] <zcorpan_>
yes... <article> is exactly like <div class="hentry"> aiui

[05:39:01.0000] <zcorpan_>
you can use <artile> for wrapping a blog post and inner <article>s for the comments

[05:40:00.0000] <ddfreyne>
I don't see why <article> is better than <div class="hentry"> though

[05:41:00.0000] <zcorpan_>
i'll discuss more in a bit, i have to run to the gym now

[05:41:01.0000] <ddfreyne>
I sound microformats-biased :)

[05:41:02.0000] <ddfreyne>
alright, seeya

[05:41:03.0000] <zcorpan_>
later

[06:38:00.0000] <mpt>
The only reason I can think of for having <article> is to prevent accidental search results when the search keywords are spread across multiple <article>s that happen to be shown on the same page

[06:40:00.0000] <mpt>
Because search engines can't know what <div class="some random value representing an article"> means, but they can know what <article> means

[06:41:00.0000] <mpt>
but, well, blah

[06:43:00.0000] <mpt>
Reducing the number of <div>s in the world is not a useful goal

[06:54:00.0000] <ddfreyne>
that problem could be solved with microformats as well

[06:59:00.0000] <ddfreyne>
(and I'm not saying everything can be solved with microformats)

[06:59:01.0000] <gsnedders>
how can it be saved with µf?

[06:59:02.0000] <gsnedders>
*solved

[07:00:00.0000] <ddfreyne>
class="hentry" on a block element that is an article

[07:02:00.0000] <mpt>
I'm <div class="hentry" id="the-8th"> I am, I am, I'm <div class="hentry" id="the-8th"> I am...

[07:05:00.0000] <ddfreyne>
Sorry… not following you there

[07:05:01.0000] <nlogax>
:D

[07:18:00.0000] <mpt>
http://en.wikipedia.org/wiki/I'm_Henery_the_Eighth,_I_Am

[07:24:00.0000] <ddfreyne>
ah, haha

[07:26:00.0000] <gsnedders>
ddfreyne: but people make up class names, someone might (oddly) use hentry. people don't make up element names (normally)

[07:34:00.0000] <mpt>
That's an argument against microformats in general, not against <article> in particular

[07:36:00.0000] <ddfreyne>
gsnedders: yeah, that's why I actually suggested something like role=

[07:36:01.0000] <ddfreyne>
(which is similar to class, but predefined, with clear semantic meanings for each value)

[07:37:00.0000] <ddfreyne>
I think it'd be better to not meddle with class… 

[07:54:00.0000] <gsnedders>
how are unknown attributes treated?

[07:54:01.0000] <gsnedders>
%Text?

[08:00:00.0000] <Philip`>
ddfreyne: I think the selection of structural elements in HTML5 (and the lack of footnote, note, comment, etc) is partly justified by http://code.google.com/webstats/2005-12/classes.html - it's trying to cover all of the most common cases, though I guess the cutoff point is largely a subjective choice

[08:02:00.0000] <zcorpan_>
/me is back

[08:22:00.0000] <zcorpan_>
ddfreyne: i think there are several reasons for using elements rather than attributes for article &c

[08:23:00.0000] <zcorpan_>
ddfreyne: (1) it's easier to author and maintain markup with element names instead of lots of divs

[08:24:00.0000] <zcorpan_>
ddfreyne: (2) if we continue reusing div for anything new this way we would end up with DivML 1.0 or something after a while

[08:25:00.0000] <zcorpan_>
ddfreyne: (3) it prevents the case where an element is both an article and nav at the same time, or where the attribute is placed on bogus places

[08:25:01.0000] <ddfreyne>
zcorpan_: (1) You could keep on adding new elements for anything that is used a lot

[08:26:00.0000] <ddfreyne>
zcorpan_: (2) using new elements breaks in some browsers, while role=... or class=... does not

[08:26:01.0000] <ddfreyne>
and I have no 3 yet

[08:26:02.0000] <zcorpan_>
ddfreyne: (4) it's more performant to process elements than attributes in the general case (look at the headings and sections section for how article is part of the document outline algorithm)

[08:27:00.0000] <ddfreyne>
I simply think the amount of new elements is too much

[08:28:00.0000] <ddfreyne>
I like <m> and <section>, but I don't think it's worth adding 10 more elements

[08:28:01.0000] <zcorpan_>
(2) i don't think it breaks anything (the content is still accessible as if the tags weren't there, no different from say <label>), it just isn't possible to style in some browsers

[08:29:00.0000] <ddfreyne>
Hmm…  I remember someone mentioning a workaround for that in fact… I've never seen it though

[08:30:00.0000] <zcorpan_>
why is it better to use attributes instead of elements? as i said, it would complicate processing, so i don't think using attributes would make the language simpler in any way

[08:31:00.0000] <zcorpan_>
one workaround is to use xhtml5 :) another might be to fix up the dom with script afterwards

[08:32:00.0000] <ddfreyne>
Hm… what about the predefined class names, though… what is the reasoning for not putting them into a separate element?

[08:32:01.0000] <ddfreyne>
Hm, they are more like 'properties' of existing elements I suppose

[08:32:02.0000] <zcorpan_>
they can apply to multiple elements

[08:32:03.0000] <zcorpan_>
yeah

[08:34:00.0000] <zcorpan_>
it could be argued that we should combine all sectioning elements (except body) to just <section> and then use an attribute for what kind of section

[08:35:00.0000] <zcorpan_>
but i'm not sure what the advantage would be

[08:35:01.0000] <ddfreyne>
There's still the difference between <div> and <section> which isn't entirely clear

[08:36:00.0000] <ddfreyne>
If you could add a… say, type="header" attribute to <section>… the entire page would be <section>s, with very few divs

[08:36:01.0000] <zcorpan_>
indeed... i think this needs to be pointed out very explicitly in the spec... in a nutshell div doesn't affect the document outline

[08:36:02.0000] <ddfreyne>
so you end up replacing <div> by <section> which is pointless

[08:36:03.0000] <zcorpan_>
section is not intended to be a catch-all replacement of div

[08:36:04.0000] <ddfreyne>
clearly

[08:37:00.0000] <zcorpan_>
but i fear many authors will go "oh, section is the new div, i should replace all my divs with sections!"

[08:37:01.0000] <zcorpan_>
q.v. b->strong

[08:38:00.0000] <ddfreyne>
yeah

[08:38:01.0000] <ddfreyne>
I do believe many div's could be replaced by section's though

[08:38:02.0000] <zcorpan_>
sure

[08:40:00.0000] <zcorpan_>
here's an experiment: http://simon.html5.org/sandbox/html/w3c-home-in-html5

[08:41:00.0000] <ddfreyne>
Yeah, I did something similar with my site a few days ago

[08:42:00.0000] <ddfreyne>
Perhaps I simply haven't used html5 enough to judge the use of new elements

[08:43:00.0000] <ddfreyne>
What is the reasoning for giving presentational elements a semantic meaning?

[08:43:01.0000] <zcorpan_>
there isn't much that can be used today in practice :)

[08:43:02.0000] <ddfreyne>
that too

[08:44:00.0000] <Hassman>
whow http://simon.html5.org/valid-html5.png with cat inside 8-)

[08:44:01.0000] <zcorpan_>
:)

[08:44:02.0000] <ddfreyne>
that's probably the biggest reason why I switched back… HTML4 is stable and predictable

[08:44:03.0000] <Hassman>
/me met__

[08:44:04.0000] <zcorpan_>
ddfreyne: probably to not upset the semanticists or so

[08:45:00.0000] <ddfreyne>
wouldn't it have made more sense to deprecate the elements instead?

[08:45:01.0000] <zcorpan_>
that will make em and strong be the de facto italics and bold elements

[08:45:02.0000] <zcorpan_>
but they already are

[08:46:00.0000] <ddfreyne>
Redefining b, i, small, etc gives me the impression "b, i and small are no longer uncool, and you can use presentational markup again!"

[08:48:00.0000] <ddfreyne>
Hm… the use of <article> could actually bring back this old "subscription" feature IE5 had… 

[08:48:01.0000] <zcorpan_>
see http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-January/009060.html

[08:49:00.0000] <ddfreyne>
You'd bookmark a page, then tell IE to check whether the page had been updated or not… but that doesn't work on dynamic pages unless the actual content can be extracted

[08:50:00.0000] <ddfreyne>
which is possible with <article> (anything outside that could be ignored)

[08:50:01.0000] <zcorpan_>
indeed

[08:50:02.0000] <ddfreyne>
zcorpan_: ah, interesting…

[08:52:00.0000] <zcorpan_>
trying to force authors to use semantic markup doesn't work, they will instead misuse semantic elements for presentational purposes

[08:53:00.0000] <zcorpan_>
additionally, semantics is not an end of itself, it's more a means to achieve device independence

[08:53:01.0000] <ddfreyne>
zcorpan_: that raises the question… why have four elements instead now (i, em, b, strong) if the more "semantic" ones are supposed to replace the more "presentational" ones?

[08:53:02.0000] <ddfreyne>
(or the other way around)

[08:53:03.0000] <zcorpan_>
we already have 4. why forbid one pair of them?

[08:53:04.0000] <wilhelm>
Because they have different meanings.

[08:54:00.0000] <ddfreyne>
well, <i> and <b>'s meaning changed

[08:55:00.0000] <wilhelm>
Yes. To something that makes more sense.

[08:55:01.0000] <zcorpan_>
wilhelm: in practice, you can't differentiate between <i> and <em> in an app that extracts semantics, and the benefits of differentiating them on the consumer side is questionable

[08:56:00.0000] <zcorpan_>
so i'm with hsivonen (it took me a while though)

[08:56:01.0000] <ddfreyne>
How would WYSIWYG editors differentiate between <i> and <em>, and <b> and <strong>?

[08:56:02.0000] <zcorpan_>
they don't

[08:57:00.0000] <ddfreyne>
So they could use <b> for important text, and <strong> for bold?

[08:57:01.0000] <ddfreyne>
That doesn't seem to solve anything…

[08:57:02.0000] <zcorpan_>
you can use either for any purpose of why you want bold

[08:58:00.0000] <zcorpan_>
you want bold for a purpose, but most authors don't think about the purpose

[08:58:01.0000] <zcorpan_>
or they do but they don't want to mark up the purpose

[08:58:02.0000] <zcorpan_>
they just hit command-b

[08:59:00.0000] <zcorpan_>
and the reader gets the purpose because of the context

[08:59:01.0000] <zcorpan_>
search engines can't extract anything useful here

[08:59:02.0000] <ddfreyne>
I see… Why do <strong> and <b> still have different meanings, then?

[08:59:03.0000] <wilhelm>
In written Norwegian, titles of books, magazines, newspapers, movies, records, TV programs and plays are to be displayed in italics.

[08:59:04.0000] <zcorpan_>
ddfreyne: because Hixie isn't convinced yet, i think :)

[08:59:05.0000] <wilhelm>
Those italics are not emphasis. And they are not really presentational.

[09:00:00.0000] <zcorpan_>
wilhelm: indeed

[09:00:01.0000] <ddfreyne>
wilhelm: but italics there has a semantic meaning… it's not italics because it's pretty

[09:01:00.0000] <wilhelm>
When we have an element that does the job (<i>), there is no point in requring the author to use <span class='title-of-book'> or some other nonsense.

[09:01:01.0000] <wilhelm>
ddfreyne: Exactly.

[09:01:02.0000] <zcorpan_>
indeed

[09:01:03.0000] <zcorpan_>
seems we agree :)

[09:01:04.0000] <zcorpan_>
now i gotta go again

[09:01:05.0000] <ddfreyne>
seeya

[09:01:06.0000] <zcorpan_>
cya

[09:02:00.0000] <wilhelm>
Which is what the HTML5 spec says. <i> is used to markup “a span of text in an alternate voice or mood, or otherwise offset from the normal prose, such as a taxonomic designation, a technical term, an idiomatic phrase from another language, a thought, a ship name, or some other prose whose typical typographic presentation is italicized”.

[09:03:00.0000] <ddfreyne>
/me nods

[09:05:00.0000] <ddfreyne>
You wouldn't write something like this either…  the <span class="noun">title</span> of a <span class="noun">book</span> <span class="verb">is</span> <span class="adjective">italicized</span>.

[09:05:01.0000] <ddfreyne>
silly example really

[09:06:00.0000] <ddfreyne>
You just mark up what's necessary… marking up the title of a book as a separate class doesn't make sense unless you somehow want to do something with a list of book titles

[09:06:01.0000] <wilhelm>
No. I would write “<i>Menneskehetens rikdommer</i> av Leo Huberman er en bra bok.”

[09:07:00.0000] <ddfreyne>
which is pretty much why I don't like the introduction of header, footer, section, aside, article, nav, dialog, … because usually it's not necessary to explicitly mark them up that way

[09:08:00.0000] <wilhelm>
It might not be neccessary. But it is very useful.

[09:09:00.0000] <wilhelm>
Especially <article> and <section>.

[09:09:01.0000] <ddfreyne>
Apart from less keystrokes and readability… what is the use of <header> and <footer>?

[09:09:02.0000] <ddfreyne>
yeah, I like <article> and <section>

[09:12:00.0000] <wilhelm>
Standardization. I can tell my web browser to don't bother printing <header>s, <footer>s or <nav>s, because they are not useful on paper.

[09:12:01.0000] <ddfreyne>
But you could simply tell your browser to display the contents of <article> instead.

[09:14:00.0000] <wilhelm>
That is also possible, I guess. But standardization of commonly used classes of elements is useful anyway.

[09:15:00.0000] <wilhelm>
Both for page authors and browser vendors.

[09:16:00.0000] <ddfreyne>
Standardization for standardization's sake doesn't make a lot of sense though

[09:17:00.0000] <wilhelm>
Standardization makes it easier to write and parse documents.

[09:18:00.0000] <zcorpan_>
the use-case of <header> is mainly subheadings (that shouldn't be part of the document outline)

[09:18:01.0000] <zcorpan_>
"footer" was a very common class name

[09:18:02.0000] <zcorpan_>
/me shouldn't be here

[09:18:03.0000] <ddfreyne>
Ahh, hm

[09:19:00.0000] <zcorpan_>
these elements also remove the need for skip links

[09:20:00.0000] <ddfreyne>
an <article> element would eliminate the need for skip links as well though

[09:21:00.0000] <zcorpan_>
<article> and <nav> mainly

[09:23:00.0000] <wilhelm>
Sure. But why _not_ standardize the most commonly used classes? footer is the most widespread class on the web.

[09:23:01.0000] <wilhelm>
http://code.google.com/webstats/2005-12/classes.html

[09:24:00.0000] <ddfreyne>
elementitis :)

[15:40:00.0000] <zcorpan_>
http://forums.whatwg.org/viewtopic.php?t=38#190 

[15:46:00.0000] <zcorpan_>
perhaps the spec should have a non-normative appendix that lists the differences between html5 and xhtml5

[15:48:00.0000] <met__>
something like this appendix http://www.w3.org/TR/xhtml2/xhtml2-changes.html#a_changes ?

[15:48:01.0000] <zcorpan_>
heh


2007-04-30
[20:09:00.0000] <Hixie>
the level to which DOM2 HTML is underspecified is insane

[20:20:00.0000] <othermaciej>
is it worse than DOM Core?

[20:25:00.0000] <Hixie>
to some extent yes

[20:25:01.0000] <Hixie>
e.g. try to work out what createTHead() does

[20:26:00.0000] <Hixie>
or what happens when you set table.caption

[20:27:00.0000] <othermaciej>
no thanks

[20:41:00.0000] <hober>
Would you consider <legend> to be block-level or inline-level? (Its content model is inline, but what of the element itself?)

[21:57:00.0000] <a-ja>
anyone here have a clue why an <a> element wouldn't be allowed inside a <details> ? 

[00:54:00.0000] <annevk>
a-ja, probably because <details> is interactive at the moment

[00:54:01.0000] <annevk>
a-ja, I think it should be allowed though

[00:55:00.0000] <a-ja>
annevk: seems to take every other strictly inline i've tried

[00:56:00.0000] <annevk>
I'm not sure how that negates my point

[00:56:01.0000] <a-ja>
annevk: e.g. img, span, meter 

[01:01:00.0000] <a-ja>
annevk: sure would be handy having hide/show ToC's in a <details>.  in any event, if <a>'s not gonna be allowed, spec needs to say so 

[01:03:00.0000] <annevk>
It already does

[01:03:01.0000] <annevk>
If you look at the definition of <a> you see: "Where strictly inline-level content is allowed, if there are no ancestor interactive elements."

[01:03:02.0000] <annevk>
And for <details> you can see that it's an interactive element

[01:04:00.0000] <annevk>
I don't think this restriction makes much sense, but the conformance checker is correct

[01:04:01.0000] <a-ja>
ah....links are considered interactive, eh?

[01:04:02.0000] <annevk>
yes

[01:14:00.0000] <a-ja>
hmm...that'll make <details> useless for things like hiding/showing blog comments/feedback, too since they're likely to contain links :(

[01:15:00.0000] <annevk>
would you classify those as "details" though?

[01:16:00.0000] <a-ja>
yeah,,,that might be iffy

[01:16:01.0000] <annevk>
but the interactive stuff need sto be fixed, I think Hixie is aware of that (it also affects <datalist> for instance), but you might want to re-raise the point on the list

[01:17:00.0000] <a-ja>
toggling  post's full text vs. just a summary would be nice, too

[01:17:01.0000] <a-ja>
better on whatwg or html-wg, you think? 

[01:18:00.0000] <annevk>
whatwg

[01:18:01.0000] <a-ja>
/me is not a member of either, but has been considering it

[01:18:02.0000] <annevk>
public-html hasn't adopted the spec yet

[01:18:03.0000] <annevk>
and you should probably join at least one of them

[01:20:00.0000] <a-ja>
still not any "no" votes

[01:23:00.0000] <annevk>
do you have a pointer btw?

[01:24:00.0000] <a-ja>
to signup info? remember seeing html-wg signup info at hixie's blog, wasn't it?

[01:24:01.0000] <annevk>
no, to the voting proceedings

[01:25:00.0000] <a-ja>
http://www.w3.org/2002/09/wbs/40318/htmlbg/results

[01:26:00.0000] <annevk>
cool, thanks

[01:26:01.0000] <annevk>
/me couldn't find it on the homepage

[01:27:00.0000] <a-ja>
heh....compact view of the results is non-public

[01:29:00.0000] <a-ja>
someone in moz #developers gave me the link the other day....never saw it anywhere else

[01:30:00.0000] <om_afk>
<details> should definitely allow things that are normally not allowed in interactive elements since they are themselves interactive

[01:30:01.0000] <om_afk>
IMO

[01:31:00.0000] <annevk>
agreed

[01:33:00.0000] <othermaciej>
In Mac OS X UI, the control that corresponds to "details" is often used to hide extra controls for a form or dialog

[01:33:01.0000] <othermaciej>
that are normally not of interest but sometimes are

[01:36:00.0000] <annevk>
yeah, I think Hixie solved the issues now

[01:36:01.0000] <annevk>
we can now also safely allow nested links if it weren't for the parsing

[01:38:00.0000] <a-ja>
/me has been playing with <details> equivalent class-swapper thingy for a couple hours...and is frustrated getting it working via keyboard with opera

[01:44:00.0000] <annevk>
A future version of Opera will have addressed that...

[01:44:01.0000] <annevk>
(by supporting tabindex for HTMLElement)

[01:47:00.0000] <a-ja>
adding tabindex to a link child of legend let me get there, but won't fire "onactivate" for some reason, unless i add a href....but that make the page scroll  

[01:50:00.0000] <a-ja>
annevk: in Kestrel, per chance?

[03:32:00.0000] <zcorpan_>
re <a> in <details>: could be reasonable to disallow <a> in <details><legend>

[03:33:00.0000] <zcorpan_>
since it's really the legend that is interactive

[03:51:00.0000] <mpt>
precisely

[04:07:00.0000] <annevk>
note that that argument doesn't apply to <datalist>

[05:15:00.0000] <Lachy>
zcorpan_, this article might be useful for you to post in that forum thread http://lachy.id.au/log/2005/05/script-comments

[06:37:00.0000] <Philip`>
 Is anybody testing HTML5 backward-compatibility with UAs that don't understand <script>/<style> and still need the content-hiding comments? (I would have thought it's very unlikely for anyone making an HTML5 page to expect it to work decently in ancient browsers, and if it's going to break badly anyway then there's no need for the spec to explain the script-hiding syntax)

[06:38:00.0000] <othermaciej>
how ancient are we talking about

[06:38:01.0000] <othermaciej>
Netscape 1?

[06:39:00.0000] <Lachy>
the browser on the Motorolla v3 phone!

[06:40:00.0000] <Lachy>
but that one can be ignored, since it would be mostly used in walled gardnes anyway

[06:40:01.0000] <Lachy>
*gardens

[06:41:00.0000] <Philip`>
The only age-related references I've seen are saying that HTML 3.2 added script/style and would correctly not display the contents, but I've got no idea how that matches implementations from that time period

[06:46:00.0000] <Lachy>
Netscape 2.0 added support for script, so any browser released after relatively soon afterwards would have added support too

[06:47:00.0000] <Philip`>
Wow, not even Google works in Netscape 3

[06:47:01.0000] <Philip`>
(It has JavaScript error boxes coming up)

[06:47:02.0000] <Philip`>
I wonder if that suggests it's old enough to be worth abandoning...

[06:48:00.0000] <Lachy>
anything before IE6 is worth abandoning

[06:50:00.0000] <Lachy>
heh, "What's the point of having different browsers if they are all required to work the same way ... ?" -- http://lists.w3.org/Archives/Public/www-html/2007Apr/0211.html 

[06:56:00.0000] <zcorpan_>
Lachy: it's a forum, anyone can post there :)

[06:56:01.0000] <Lachy>
zcorpan_, I'm not registered

[06:57:00.0000] <Lachy>
I'm also avoiding it because I don't like forums

[06:58:00.0000] <zcorpan_>
ok

[06:59:00.0000] <Lachy>
/me responds once more to www-html (though I'm getting tired of the baseless arguments)

[06:59:01.0000] <Philip`>
Aha: Netscape Gold 2.02 and Navigator 3.04Gold shows the contents of <style>s, but not <script>s. Navigator 4.08 shows neither.

[07:00:00.0000] <Lachy>
I surprised you have all of those browsers installed

[07:00:01.0000] <Philip`>
I didn't, until ten minutes ago :-)

[07:00:02.0000] <Philip`>
http://sillydog.org/narchive/full123.php etc

[07:03:00.0000] <Philip`>
Hmm, Netscape 2 takes quite a while to load the HTML5 spec, and then it gets stuck trying to download an image and doesn't draw the rest of the page :-(

[07:04:00.0000] <Lachy>
can you put up a screenshot of google, or something in NN2?

[07:05:00.0000] <Lachy>
or the Acid2 test :-)

[07:08:00.0000] <zcorpan_>
wouldn't <noscript><img src="tracker"/></noscript> work even in xhtml5, given that images that are display:none are not fetched?

[07:09:00.0000] <Lachy>
zcorpan_, are you sure they're not fetched?

[07:09:01.0000] <Lachy>
isn't that up to the browser?

[07:10:00.0000] <Philip`>
(Hmph, Gmail makes Netscape crash...)

[07:10:01.0000] <Lachy>
lol

[07:10:02.0000] <Lachy>
can anyone else access hixie.ch?  I'm trying to get to the live dom viewer

[07:11:00.0000] <Philip`>
That and whatwg.org seem to be down

[07:14:00.0000] <Philip`>
http://canvex.lazyilluminati.com/misc/nn2/

[07:14:01.0000] <Philip`>
(Google is at least still generally usable, if you ignore the script error on every page)

[07:17:00.0000] <Philip`>
In any case, I think I feel justified in not doing <style><!--... on my sites because anybody using Netscape 3 is already going to be used to having a hard time on the web

[07:24:00.0000] <Philip`>
(I have to go back to Netscape Navigator 1.22 before there's any point in doing <script><!--...//--></script>)

[07:26:00.0000] <Lachy>
what about past versions of IE?

[07:26:01.0000] <Lachy>
or TimBL's WWW browser

[07:27:00.0000] <Philip`>
(In relation to earlier discussions about parsing bogus colour values, NN1 parses "black" as #b0ac00, which disturbs my site's colour scheme)

[07:29:00.0000] <Lachy>
really?  I thought it would have support for colour keywords

[07:29:01.0000] <Lachy>
I wonder when X11 colours were added to NN then

[07:30:00.0000] <Philip`>
NN2 handles black correctly

[07:31:00.0000] <Lachy>
try some more http://en.wikipedia.org/wiki/X11_color_names

[07:31:01.0000] <Philip`>
http://browsers.evolt.org/?worldwideweb/NeXT - maybe that would work if I had a NeXT computer, but sadly I don't :-(

[07:32:00.0000] <Philip`>
http://browsers.evolt.org/?ie/32bit - aha, handy archive

[07:34:00.0000] <Lachy>
see which colours NN2 uses for Gray, Green Maroon and Purple, which clash between X11 and CSS

[07:38:00.0000] <othermaciej>
Jukka's message seems poorly thought out

[07:38:01.0000] <othermaciej>
what's the point of having more than one browser if they all act completely different?

[07:38:02.0000] <Lachy>
did you read my response?

[07:39:00.0000] <zcorpan_>
i need to figure out exactly which color names are supported by html parsers, in order to put forward a proposal

[07:39:01.0000] <Lachy>
a proposal for what?

[07:39:02.0000] <othermaciej>
www-html seems much more insane than public-html

[07:39:03.0000] <zcorpan_>
for changing the html5 parsing algorithm

[07:39:04.0000] <othermaciej>
glad there's other people to cover it

[07:40:00.0000] <Lachy>
I thought all the supported colour names were known, you just need to determine the algorithm to use for unknown names

[07:40:01.0000] <Philip`>
Lachy: NN1 handles none of those colours; NN2 handles them all (and uses the HTML-like values rather than X11-like)

[07:40:02.0000] <zcorpan_>
don't we need to spec what the set of color names are to be known?

[07:41:00.0000] <Lachy>
CSS defines those, HTML can just reference that

[07:41:01.0000] <zcorpan_>
css3-color?

[07:42:00.0000] <Lachy>
yes

[07:42:01.0000] <zcorpan_>
could work

[07:43:00.0000] <Lachy>
oh, Tina responded to me on www-html, wanting to know which parts of HTML4 can't be implemented

[07:43:01.0000] <Lachy>
Now, where do I start?

[07:43:02.0000] <othermaciej>
<object>

[07:43:03.0000] <Philip`>
...and the Wikipedia list is precisely the same set as is listed in NN2's .exe

[07:43:04.0000] <othermaciej>
is it "can't be implemented" or "can't be implemented interoperably"

[07:44:00.0000] <zcorpan_>
it surely contradicts itself in various places, doesn't it?

[07:44:01.0000] <wilhelm>
Lachy: Error handling.

[07:44:02.0000] <othermaciej>
also are we looking for self-contradiction in the spec, or contradiction to expectations of deployed content?

[07:44:03.0000] <Lachy>
othermaciej, both

[07:44:04.0000] <othermaciej>
tag minimization is one obvious thing in the latter category

[07:44:05.0000] <zcorpan_>
</foo closing style and script

[07:45:00.0000] <Philip`>
(...and precisely the same list of colour names in NN4 too)

[07:45:01.0000] <Lachy>
we should record this stuff in the whatwg wiki too!

[07:45:02.0000] <othermaciej>
you can't implement <foo/> as SGML calls for

[07:45:03.0000] <zcorpan_>
that ua's must not assume a default encoding

[07:45:04.0000] <othermaciej>
Lachy: this isn't quite the kind of thing you asked for, but it amuses me to no end that the definition of "conforming implementation" is self-contradictiory

[07:45:05.0000] <othermaciej>
in HTML4

[07:46:00.0000] <Lachy>
I'll mention that too

[07:46:01.0000] <zcorpan_>
perhaps it implies that being conforming is optional for conformance? :)

[07:47:00.0000] <othermaciej>
http://www.w3.org/TR/html401/conform.html

[07:47:01.0000] <othermaciej>
"A conforming user agent for HTML 4 is one that observes the mandatory conditions ("must") set forth in this specification, including the following points:"

[07:47:02.0000] <othermaciej>
and then it lists one should-level, one must-level and one recommend-level requirement

[07:48:00.0000] <othermaciej>
but you can also find whole chapters where the word "must" does not appear

[07:49:00.0000] <Philip`>
Whoops, my default browser is now set to Netscape 1

[07:49:01.0000] <Philip`>
At least the later versions asked me if I wanted to change it

[07:49:02.0000] <othermaciej>
this section for instance contains no UA conformance requirements: http://www.w3.org/TR/html401/struct/lists.html

[07:52:00.0000] <Philip`>
Lachy: Unfortunately I can't test old versions of IE - "Internet Explorer 3.0 cannot be installed on a system that has Internet Explorer 4.0 installed"

[07:53:00.0000] <Philip`>
although I may actually have a spare Windows 3.1 virtual machine somewhere, which may run it...

[07:53:01.0000] <Lachy>
cool

[07:55:00.0000] <zcorpan_>
Philip`: does it need to be installed? isn't there a standalone version that just runs?

[08:01:00.0000] <Philip`>
(Hmm, installing network drivers in Win3.11 is too much of a pain so I think I'll give up)

[08:02:00.0000] <zcorpan_>
/me has or had mosaic installed... most of the web didn't work at all

[08:02:01.0000] <Philip`>
zcorpan_: Ah, hadn't tried that, but the ones at http://browsers.evolt.org/?ie/32bit/standalone seem to work

[08:03:00.0000] <zcorpan_>
...which wasn't because of the markup used but something else (probably http something)

[08:04:00.0000] <Philip`>
NN1 seems to have HTTP problems - it's not telling the server what hostname it's connecting to, or something, so it acts the same as if you typed in the IP directly (which rarely works)

[08:04:01.0000] <Philip`>
How peculiar - the HTML5 spec in IE3 has a textarea in the top left corner

[08:04:02.0000] <zcorpan_>
perhaps it sees the <textarea> in the last comment

[08:05:00.0000] <zcorpan_>
closes comments at > or something

[08:08:00.0000] <Philip`>
(Oh, there's no standalone IE2 there, and the non-standalone one fails to install with "Internet Explorer requires Windows 95.")

[08:09:00.0000] <Philip`>
(IE3 has precisely the same list of colour names as NN2)

[08:12:00.0000] <Lachy>
does <applet> also suffer from interop problems?

[08:13:00.0000] <othermaciej>
<applet> has a pretty serious flaw in Safari

[08:13:01.0000] <othermaciej>
at times, it is known to load a Java applet

[08:14:00.0000] <zcorpan_>
"If your argument was correct, then everyone would learn about mark-up simply by reading other people's mark-up, which is obviously not what happens."

[08:14:01.0000] <zcorpan_>
wow

[08:14:02.0000] <othermaciej>
that's the only way I ever learned...

[08:14:03.0000] <zcorpan_>
it might not apply to *everyone*, but surely most

[08:15:00.0000] <othermaciej>
even people who learn from books copy the examples in those books

[08:15:01.0000] <Lachy>
zcorpan_, who wrote that?

[08:15:02.0000] <zcorpan_>
Lachy: mark http://www.w3.org/mid/640dd5060704300529u70a16269xec412a16436f3e2b⊙mgc

[08:16:00.0000] <Lachy>
he is so out of touch

[08:17:00.0000] <Lachy>
even I learned by view source initially, though I started reading specs when that was too difficult to understnad

[08:23:00.0000] <Philip`>
/me wonders how different the web would be today if the early browsers didn't have a "view source" menu item

[08:24:00.0000] <Lachy>
people would have been forced to use editors to write HTML, just like very few people hand code RTF, PDF, Work docs, etc.

[08:24:01.0000] <Lachy>
s/Work/Word/

[08:27:00.0000] <Lachy>
here's the draft of my reply to Tina http://lachy.id.au/temp/html4

[08:29:00.0000] <othermaciej>
SHORTTAG syntax is kind of cool

[08:29:01.0000] <othermaciej>
too bad it can't be used

[08:37:00.0000] <Lachy>
I made a few more additions and modifications, and I think I'm ready to send it.  I'm sure there would be more, but I couldn't be bothered looking.

[08:40:00.0000] <othermaciej>
I think "do not assume a default character set" is one worth listing

[08:40:01.0000] <othermaciej>
(don't remember the precise wording of the spec)

[08:40:02.0000] <Lachy>
where does it say that in the sepc?

[08:40:03.0000] <Lachy>
found it

[08:40:04.0000] <Lachy>
it says "Therefore, user agents must not assume any default value for the "charset" parameter."

[08:40:05.0000] <Lachy>
what should I say about it?

[08:42:00.0000] <othermaciej>
user agents have to assume the default encoding is Windows Latin 1

[08:42:01.0000] <othermaciej>
(windows-1252 officially)

[08:43:00.0000] <othermaciej>
in fact, browsers have to treat latin1 as windows-1252 always, but I don't think HTML is to blame for that

[08:43:01.0000] <Philip`>
"<table border> is actually shorthand for <table frame="border">" - shouldn't that be border="border"?

[08:44:00.0000] <Lachy>
no

[08:44:01.0000] <Lachy>
in HTML5, yes

[08:44:02.0000] <Lachy>
in SGML, no

[08:44:03.0000] <Philip`>
Oh, okay, HTML4 has funny rules for that

[08:45:00.0000] <Lachy>
it's actually the attribute name that can be omitted, not the value as most people think

[08:45:01.0000] <Lachy>
but UAs only implemented it for attrs that had the same name and value

[08:46:00.0000] <wilhelm>
othermaciej: In some countries, UAs have to use other encodings as default.

[08:46:01.0000] <Lachy>
othermaciej, I added the default charset thing

[08:47:00.0000] <Lachy>
wilhelm, yes, I wrote "usually forced to default to win 1252"

[08:47:01.0000] <othermaciej>
wilhelm: well, the per-country default encoding thing is even worse - we don't actually do that in Safari because we don't want your web browsing to be affected by language setting

[08:47:02.0000] <wilhelm>
Most Japanese websites don't bother defining what charset they use. And it usually isn't windows-1252.

[08:47:03.0000] <met_>
what exactly mean "concur" in the W3C wbs? feel something like no but without canceling the approval?

[08:47:04.0000] <Lachy>
it means you agree with whatever the majority say

[08:48:00.0000] <wilhelm>
othermaciej: It is terrible, I agree. But you can't ship on Japanese devices without it.

[08:48:01.0000] <met_>
what is difference pro abstaining?

[08:49:00.0000] <othermaciej>
wilhelm: I might consider it if it was for localized hardware sold only in the given region

[08:49:01.0000] <Lachy>
met_, see the logs of #html-wg from a few hours ago for an explanation

[08:50:00.0000] <Lachy>
we just need to get japanese markets to use UTF-8 or UTF-16 for everything.

[08:50:01.0000] <wilhelm>
othermaciej: Yes. That is what we (Opera) do on the Wii and on phones for Japanese customers.

[08:51:00.0000] <Lachy>
the major problem is that OSs ship with different default charsets for different languages, instead of Unicode

[08:51:01.0000] <met_>
Lachy thx

[08:52:00.0000] <Lachy>
what does a japanese user do when they visit a site that needs to default to win 1252?

[08:52:01.0000] <wilhelm>
Lachy: Getting everyone to use Unicode would be ideal. But just getting sites to define which charset they use would solve most of these problems.

[08:52:02.0000] <othermaciej>
yeah, relying on the default encoding is bad

[08:53:00.0000] <wilhelm>
Yes. It breaks Norwegian sites in Japan, and Japanese sites in Norway.

[08:53:01.0000] <Lachy>
wilhelm, we've got more chance of fixing the situation by getting editors to use Unicode as default encoding for everyting, than by getting authors to declare charsets

[08:55:00.0000] <Lachy>
any more suggestions for my list of HTML4 issues before I send?

[08:55:01.0000] <wilhelm>
That is possible.

[08:56:00.0000] <othermaciej>
Lachy: in Asia (Japan especially), enthusiasm for unicode is not that widespread

[08:56:01.0000] <Lachy>
why not?

[08:57:00.0000] <othermaciej>
they like how compact encodings like shift-JIS are for their language

[08:57:01.0000] <othermaciej>
same reason Americans don't tend to use UTF-16

[08:57:02.0000] <MikeSmith>
the Japanese developer community is not fond of Unicode. Many Japanese mobile devices don't have proper Unicode support (or even fonts for non-ASCII Latin characters). Several Japanese desktop mail clients don't have Unicode support, and many (or most) e-mail clients on mobile devices in Japan don't either

[08:58:00.0000] <Lachy>
yeah, full support for unicode is a problem everywhere

[11:02:00.0000] <othermaciej>
good morning

[11:06:00.0000] <gsnedders>
is it possible to have a third state with a checkbox (like the line in OS X?)

[11:10:00.0000] <Dashiva>
Not as of currently, no

[11:15:00.0000] <gsnedders>
so I haven't forgotten some input type value

[11:49:00.0000] <gsnedders>
I take it the SOLIDUS is allowed in all empty elements?

[12:32:00.0000] <deltab>
gsnedders: I take it you mean inside the start tag?

[12:33:00.0000] <deltab>
void elements only

[12:33:01.0000] <deltab>
i.e. base, link, meta, hr, br, img, embed, param, area, col, input

[12:33:02.0000] <gsnedders>
yes

[13:31:00.0000] <bewest>
"the w3c should define HTML, and browser manufacturers should be willing to accept that definition (or to reject it, at their own risk: this is a free world), but it would be a great boost for standards were the "W3C HTML 5" logo to be as applicable to /browsers/ as it will be to web pages."

[14:21:00.0000] <gsnedders>
mm… looking through the HTML 4.01 Trans DTD, there are more EMPTY elements than those allowed to have a SOLIDUS character in their start tag in the parsing section of WA1.0

[14:29:00.0000] <othermaciej>
/me is tempted to +1 Hixie's email about +1 messages, but restrains his impish side

[14:32:00.0000] <bewest>
othermaciej: your buddy dave hyatt beat you to it

[14:33:00.0000] <othermaciej>
I guess he's more of an ass than I am :-)

[14:34:00.0000] <bewest>
I'm not sure how much +1 matters on the public-html list either

[14:34:01.0000] <bewest>
it appears the chairs intend to use web surveys

[14:34:02.0000] <bewest>
othermaciej: you have some nice replies on that list, btw

[14:35:00.0000] <bewest>
imo one of the voices of sanity

[14:37:00.0000] <Philip`>
Hixie's email was only explaining why +1 is not useful in discussions about technical issues, where there can be objectively wrong answers - it wasn't arguing that +1 is not useful in discussions about whether +1 is not useful in discussions, and hence it seems quite logical to +1 it

[14:37:01.0000] <jgraham>
+1

[14:38:00.0000] <jgraham>
/me has wanted to type that in response to /almost anything/ for weeks. He realises this is deeply illogical

[14:39:00.0000] <gsnedders>
+1

[14:39:01.0000] <gsnedders>
I restrained replying to Anne's (I think) call for an end of +1s on public-html, on grounds that he was citing the number of messages as a reason, making it further illogical

[14:40:00.0000] <othermaciej>
-1

[14:50:00.0000] <Hixie>
if anyone's bored and wants to argue for a while, this one's hot: http://www.w3.org/Bugs/Public/show_bug.cgi?id=1500

[14:56:00.0000] <othermaciej>
the very URL is discouraging

[16:29:00.0000] <Hixie>
annevk?