00:31
<zcorpan_>
Hixie: re http://forums.whatwg.org/viewtopic.php?t=51 , i don't see "custom bbcodes" in the admin ui...
00:31
<zcorpan_>
perhaps my permissions are restricted?
00:32
<Hixie>
you're the highest level i can set
00:32
<Hixie>
where would the custom bbcodes be?
00:33
<zcorpan_>
dunno
00:33
<Hixie>
i don't see anything called custom bbcodes
00:33
<Hixie>
maybe it's in a new version
00:34
<Hixie>
we've got 2.0.22
00:34
<zcorpan_>
could be
00:34
<Hixie>
that's the latest stable release, though, so who knows
00:34
Hixie
doesn't really know about phpBB
00:35
<Hixie>
i just clicked the "install phpBB" button on the dreamhost panel :-)
00:35
<zcorpan_>
"DOWNLOAD LATEST RELEASE Version 2.0.22" says phpbb.com
01:03
<zcorpan_>
http://forums.whatwg.org/viewtopic.php?t=51#280
01:04
zcorpan_
wonders if that will result in stray <code> tags if the input isn't balanced
01:04
<Hixie>
if i do that, we'll have to redo it every time we update
01:05
<zcorpan_>
yeah, i don't think it's robust either
01:05
<zcorpan_>
probably better to upgrade to phpbb3 later on and see if it has a custom bbcodes section
01:08
<Hixie>
k
01:23
<Lachy>
Hixie, http://returnable.org/ is another attempt at creating custom tooltips, this time using a custom element and strangely abusing URIs for a non-URI purpose
01:24
<Lachy>
we really need to find a sensible way to address the problem in HTML, CSS, etc.
01:26
<Hixie>
o_O
01:26
<Hixie>
HTML5 says newlines in tooltips are real newlines
01:29
<Lachy>
ok. I didn't realise that. it's a shame only IE supports it though
01:30
<zcorpan_>
safari too iirc
01:30
<othermaciej>
I'd take a patch for that
01:30
<othermaciej>
if we don't do it already
01:31
<Hixie>
i don't get this tina person
01:31
<Hixie>
i'm actively trying to get her input and make the spec work for her
01:32
<Hixie>
but every e-mail she asserts that i'm going to ignore her and that i shouldn't worry as she won't join the wg and so on
01:32
<othermaciej>
can't you see that HTML5 is /not/ an appropriate specification because it /allows/ things that are /wrong/ and should be /forbidden/?
01:33
<Hixie>
yeah, it doesn't help that what she's advocating for doesn't even handle the e-mails she's writing
01:33
<Lachy>
but she's writing plain text emails
01:35
<othermaciej>
it's ok to use presentational markup in plaintext, because it conveys information
01:35
<othermaciej>
at least that's the explanation someone gave me
01:35
<othermaciej>
plaintext, like email, is for iformation
01:35
<othermaciej>
but HTML is for semantics, not information
01:35
<Lachy>
oh, do you mean the way she's /emphasising/ /everything/ /she/ /says/?
01:37
<Hixie>
it's not all emphasis
01:37
<Hixie>
as in, <em> would not be appropriate for all her /markup/
01:37
<Hixie>
but whatever
01:37
Hixie
throws her e-mail onto the "phrasing" pile for now
01:37
<bewest>
Hixie: that's because she saw some teasing in the logs here and didn't like it
01:37
<Hixie>
uh huh
01:37
<Lachy>
I couldn't be bothered replying her any more. She's in the XHTML2 group, so she's got her own spec to mess up
01:38
<bewest>
Hixie: you could talk to her in #web
01:38
<Hixie>
i think i'll watch my latest fix of atlantis instead
01:38
<othermaciej>
she uses both *stars* and /slashes/, which one normally takes to be bold and italic in plaintext
01:38
<zcorpan_>
she'll probably read today's log too, so, hi, tina! :)
01:39
<Lachy>
Hi Tina Holomboe (she won't see it unless we mention her full name)
01:39
<Lachy>
now it will turn up in searches ;-)
01:39
<zcorpan_>
oh right
01:39
<Philip`>
Lachy: That would work better if you used the correct spelling
01:40
<zcorpan_>
Tina Holmboe
01:40
<Lachy>
oops
01:43
<bewest>
I will probably now be banned from #web
01:43
<bewest>
oh well
01:43
<zcorpan_>
why?
01:43
<Philip`>
For an entertaining reason?
01:43
Lachy
joins #web to see what's going on
01:44
<Lachy>
... nothing yet.
01:45
zcorpan_
is tempted to say "can anyone help me with a javascript problem?"
01:48
<Dashiva>
You just did
01:49
<zcorpan_>
in #web, that is
01:57
<Dashiva>
It's probably a spiderman channel anyhow
02:00
<Lachy>
zcorpan_, about that quote you just pasted in #web, where did that come from?
02:01
<zcorpan_>
topic
02:03
<Lachy>
woah! looks like David Dorward set that topic last. I can't believe he would have said that, I thought he had more sense than that
02:03
Philip`
wonders what it says, but is too lazy to go and look
02:04
<zcorpan_>
"Rules: http://hashweb.sekrit.co.uk/ | Validate FIRST: http://validator.w3.org/ | Ask first, PM later | Don't use XHTML unless you understand it: http://www.webdevout.net/articles/beware-of-xhtml | #web is for authoring web pages, not a replacement for google or an index of irc channels | WA1 accepted as HTML 5 - R.I.P. HTML"
02:05
<Philip`>
Aha
02:06
<Philip`>
If people disagree with the direction of HTML, they could always form their own group and develop a Web Document Semantics 1.0 specification
02:06
<Lachy>
Philip`, they have. It's called XHTML 2.0
02:06
<zcorpan_>
or join the xhtml2 wg
02:09
<Dashiva>
I'm confused
02:09
<Dashiva>
Beware of XHTML, but RIP HTML?
02:11
<zcorpan_>
probably different people who wrote that
02:12
<Lachy>
Dashiva, that's because authors should use WML or cHTML instead
02:25
<Hixie>
wow, that link is somewhat rough on the xhtml guys http://www.webdevout.net/articles/beware-of-xhtml
02:32
<zcorpan_>
http://www.thewebcreator.net/2007/04/16/why-you-should-be-using-html-401-instead-of-xhtml/#comment-23 (same article)
02:52
<Hixie>
the last comment on that blog entry highlights one of the weirdest things i've repeatedly seen on the web
02:52
<Hixie>
"HTML 5.0 vs XHTML 2.0 (commercials companies vs W3C)"
02:54
<Hixie>
the idea that the W3C, which you have to pay thousands of dollars to to become a member, and which is entirely member-driven, is somehow less "commercial" than the WHATWG, which can be joined by anyone
02:57
<ianloic>
hehe
03:23
<zcorpan_>
Hixie: the spec says that <a>.href must reflect the href content attribute. but browsers return the resolved uri on getting
03:24
<zcorpan_>
oh, nevermind
03:24
<Hixie>
heh
03:31
<zcorpan_>
ok, for boolean attributes, it says that setting must remove or set the content attribute, but afaict none of safari, firefox or opera do that. (ie7 does because the content attribute === the dom attribute)
03:34
<zcorpan_>
although i only tested with <input checked>
03:35
<Hixie>
checked is a special case
03:35
<Hixie>
it doesn't reflect the content attribute
03:35
<Hixie>
defaultChecked is the DOM attribute that reflects the checked content attribute
03:35
<zcorpan_>
oh
03:39
<zcorpan_>
for .disabled, opera returns "disabled", mozilla and safari return "", and ie7 returns "true"
03:40
<Hixie>
in what context?
03:40
<zcorpan_>
when setting to .disabled to true and then reading getAttribute("disabled")
03:40
<Hixie>
and what do they do when you set it to fals?
03:40
<Hixie>
false?
03:41
<Hixie>
(afk brb)
03:41
<zcorpan_>
all return null
03:43
<zcorpan_>
which means that they remove the content attribute
03:44
<zcorpan_>
and that DOM3 Core is incompatible with the real world
03:50
<zcorpan_>
or wait, ie7 returns "false", but the content attribute is absent in ie's web developer toolbar (and live dom viewer)
03:52
<zcorpan_>
uploaded to LDV
03:52
<Hixie>
so in conclusion, the spec is close enough to reality, right? :-)
03:53
<zcorpan_>
yeah. except perhaps that it should set the value to the empty string instead (to match two implementations instead of one) :)
03:53
<Hixie>
*shrug*
03:54
<Hixie>
we'll see what the implementors say when they start implementing this stuf
03:54
<Hixie>
so in other news
03:54
<Hixie>
how are we supposed to distinguish <h1>s that represent site-wide headers, from <h1>s that are the header of the page only?
03:54
<Hixie>
or do we need to?
03:57
<zcorpan_>
if there's exactly one <article> (ignoring nested <article>s), then the heading of that <article> is the page's heading, and the <body>'s heading is the site-wide heading. otherwise, the <body>'s heading is the page's heading
04:02
<Hixie>
oo, good call
04:03
<zcorpan_>
:)
04:05
<zcorpan_>
and yes, i think it needs to be distinguished, given how much this issue has been discussed at different forums
04:05
<zcorpan_>
if only for the sake of content producers
04:20
<zcorpan_>
should i mail that to the list?
04:35
<Hixie>
nah, i got it
04:35
<zcorpan_>
ok
04:36
zcorpan_
will go to bed now
04:37
<zcorpan_>
nn
04:38
<Hixie>
nn
05:04
Hixie
nukes the predefined class names
05:50
Philip`
wonders why commit-watchers hasn't seen r823
06:22
<Philip`>
r284 is there, but 283 hasn't come through email or into the archives :-(
06:25
<Hixie>
weird
06:26
<Hixie>
maybe the script crashed when i submitted it
06:56
<Hixie>
othermaciej: assuming you could iterate over classList
06:56
<Hixie>
othermaciej: would you want to see dupes, or would you want them collapsed?
06:58
<othermaciej>
Hixie: hmm, haven't thought about it
07:14
<Hixie>
sorted it is
07:27
<hendry>
i'm going to try it...
07:27
<hendry>
woops, wrong channel
07:45
<jruderman>
lol at Hixie's twitter status "Dashed lines are so Web 1.0. <canvas> is Web 2.0. Solid colours and transparency!"
07:46
<Hixie>
:-)
07:53
<othermaciej>
where are the rounded corners and reflections?
07:54
<Hixie>
reflections are here: http://cow.neondragon.net/stuff/reflection/
07:55
<Philip`>
http://www.netzgesta.de/corner/ - rounded corners
07:55
<Philip`>
What more could you want?
07:55
<othermaciej>
that reflection thing is pretty hot
07:56
<Hixie>
wow i hadn't seen the corners
07:56
<Hixie>
that's pretty awesome
07:56
<Hixie>
ok
07:56
<Hixie>
we're set!
07:56
<Hixie>
web 2.0 compliance! check!
08:02
<othermaciej>
does that use canvas?
08:03
<othermaciej>
I suppose adding NEW! and BETA badges can be an excercise to the reader
08:03
<Hixie>
yeah
08:03
<Hixie>
they both use canvas
08:04
<othermaciej>
that's pretty hot
08:04
<Hixie>
http://h-master.net/web2.0/index.php will add the BETA badge, but that's not canvas
08:05
<Philip`>
You can do nice animated rippley reflective water with the 3D canvas, but I guess that's probably a bit too distracting for use on normal web pages
08:06
<othermaciej>
cool, I just made the logo for Specr
08:06
<othermaciej>
it even colored the r red
09:52
<mikeday>
Prince 6.0 has been released at long last
09:53
<mikeday>
be the first kid on the block to download it, try it out, find bugs, and complain to us, today! :)
10:13
<mikeday>
hmm, quiet in here. You're all at XTech, aren't you.
11:15
<jdandrea>
Yipe: http://www.thewhir.com/blogs/Paul-Hirsch/index.cfm/2007/4/17/HTML-5-the-next-generation-or-largely-a-pointless-effort
11:17
jdandrea
sees others have already commented. ahh.
11:28
<Lachy>
does anyone have any stats on how often blockquote is actually used for indenting these days? I've never seen any, it seems to be taken as axiom
11:41
<ROBOd>
Lachy: i have some experience with cleaning up HTML documents saved with OpenOffice Writer and Microsoft Word
11:42
<ROBOd>
Lachy: based on this experience, yes, blockquote is used for indenting, but definition lists are the champions
11:43
<ROBOd>
almost any document i've worked with contains definition lists wrongly used for indenting text. blockquote is not really often
11:49
<mikeday>
blockquote is used a lot on blogs, for... block quoting :)
12:16
<mikeday>
hi zcorpan_
13:35
<met_>
http://my.opera.com/hallvors/blog/2007/05/16/quick-spec-for-ies-document-activeelement
16:09
<zcorpan_>
Hixie: hmm, does the #distinguishing text look for <section>s that are descendants of <aside> and <nav>? surely they should be ignored?
17:53
<zcorpan_>
Hixie: why does it look for <section>s in the first place?
17:58
<Dashiva>
Hixie: The links in DOMTokenList interface are kinda messed up. add links to remove, and remove doesn't link
18:01
<Dashiva>
Personally I would define toggle in terms of has, add, remove, since that's kind of what it does
18:03
<clotman>
:bn
18:08
<Lachy>
does anyone remember what the use cases for class="warning" were?
18:09
<Lachy>
I remember one for class="error", which was for indiciating error messages, such as for form controls, which has benefits particularly for ATs
18:45
zcorpan_
continues to test bgcolor handling
18:46
<zcorpan_>
everyone do slightly different things. i wonder what is sanest, and what is required
18:50
zcorpan_
is amused by the insanity
19:11
<Philip`>
Hmm, the spec currently seems to have the thin green SCS line down pretty much its entire length, but only in Opera and not Firefox
19:15
<zcorpan_>
yeah
19:16
<zcorpan_>
the status xml file needs an update
19:19
<ddfreyne>
Is there a feed for the specs' recent changes?
19:20
<zcorpan_>
http://twitter.com/WHATWG
19:20
<met_>
or mailing list
19:21
<Philip`>
http://html5.org/tools/web-apps-tracker and http://lists.whatwg.org/pipermail/commit-watchers-whatwg.org/2007/date.html have information on changes too
19:21
<Dashiva>
Sign up for the commit-watchers
19:21
<ddfreyne>
thanks!
19:23
<Philip`>
Also http://dev.w3.org/cvsweb/html5/spec/Overview.html
19:24
<Philip`>
It's hard to avoid accidentally running into a list of recent changes :-)
19:24
<ddfreyne>
yeah, but there still is no rss feed. :P
19:25
<ddfreyne>
(or atom)
19:25
<zcorpan_>
there is for twitter
19:25
<zcorpan_>
though you'll only get the messages
19:25
<ddfreyne>
true, but that doesn't include the diff
19:26
<zcorpan_>
we can add a feed to web-apps-tracker. contributors welcome: http://html5.googlecode.com/svn/trunk/web-apps-tracker/web-apps-tracker
20:50
<Philip`>
http://tapper-ware.net/canvas3d/ - I hadn't seen that before, but it seems to work fairly nicely
21:27
<Hixie>
aww, how cute
21:27
<Hixie>
public-html is still arguing about whether we should have predefined classes
21:29
<gsnedders>
awwwwww… so sweet, isn't it? :P
21:30
<met_>
Hixie you sent the notice about classes only to whatwg list
21:32
<Philip`>
Maybe commit-watchers could be forwarded to public-html so people don't get lost arguing about points that were already addressed weeks ago
21:32
<Philip`>
(Well, that's not actually a sensible idea, but maybe something to make sure people stay up to speed)
21:37
<Hixie>
that's Dan's problem, IMHO
23:04
<zcorpan_>
Hixie: yt?
23:21
<Hixie>
zcorpan_: yo
23:22
<zcorpan_>
why does #distinguishing look for <section> elements?
23:24
<Hixie>
it looks for anything that can be considered "in flow"
23:24
<zcorpan_>
but a <section> in an <aside> shouldn't make a difference, should it?
23:25
<zcorpan_>
or <nav>
23:25
<Hixie>
true
23:31
<zcorpan_>
btw, Roger said that the links made the text in the spec harder to read. this might make it easier to read: [href^="#"] { color: inherit; text-decoration: underline; }
23:33
<Hixie>
maybe
23:35
<Hixie>
ugh, there are so many things wrong with the #distinguishing thing
23:35
<Hixie>
how to fix
23:35
<Hixie>
how to fix
23:37
<Hixie>
what i want to say is that if the structure is 1 top level header and 1 header below that, ignoring headers in <nav> and <aside>, and that the header below it is in an <article>, then that's the site/page case
23:37
<Hixie>
and all other cases are standalone pages
23:38
<othermaciej>
what's #distinguishing?
23:38
<zcorpan_>
othermaciej: http://html5.org/tools/web-apps-tracker?from=825&to=826
23:40
<Lachy>
othermaciej, see the spec. it's a fragment identifier for the section
23:40
<Lachy>
http://www.whatwg.org/specs/web-apps/current-work/#distinguishing
23:40
<zcorpan_>
Hixie: perhaps phrase it as "for the purposes of finding the page's heading..." or so, instead of having it as authoring conformance requirements
23:40
othermaciej
looks
23:40
<Hixie>
zcorpan_: i want it as both authoring and ua reqs
23:41
<zcorpan_>
ok
23:41
<Hixie>
othermaciej: it's basically an experimental solution to an issue that has caused many arguments
23:41
<Hixie>
othermaciej: i'm trying to see if i can make the spec clear it up once and for all
23:41
<Lachy>
the way it is right now, it's extremely complex. I don't even understand it
23:42
othermaciej
is confused reading that
23:43
<zcorpan_>
wouldn't it clear up the issue even if it didn't actually contain authoring conformance requirements?
23:43
<Hixie>
yeah don't worry about the text there now
23:43
<Hixie>
zcorpan_: there's no point having the author be allowed to do things that the UA will interpret as something else
23:44
<zcorpan_>
well, you can't check the author's intent anyway, right?
23:44
<Hixie>
actually in this case you could
23:45
<Hixie>
if you validated multiple pages at once
23:45
<zcorpan_>
oh
23:45
<zcorpan_>
right
23:45
<Hixie>
(somewhat)
23:45
<Hixie>
anyway there are plenty of requirements that can't be checked
23:45
<zcorpan_>
indeed
23:56
<Hixie>
ok, redid it
23:56
<Hixie>
is it any better?
23:57
zcorpan_
checks
23:58
<Hixie>
(eventually i'll have examples of course)
23:59
<Lachy>
in the first paragraph, shouldn't it say "must be nested as the heading of an article [or section] element,"?