2007-08-01
[17:27:00.0000] <Hixie>
http://groups.google.com/group/opera.general/browse_thread/thread/c815c26e44f91c3c/2cef47d6d17f260e?q=whatwg&rnum=1#2cef47d6d17f260e

[17:28:00.0000] <Hixie>
what an amusingly dismissive last two lines

[17:41:00.0000] <Hixie>
lol, i just saw the tagline from this channel got put on the blog

[17:42:00.0000] <zcorpan_>
that was a while ago, i think

[17:43:00.0000] <Hixie>
i don't read our blog that often :-P

[17:50:00.0000] <zcorpan_>
btw, i got employed at opera, so i can continue to work full-time on html5 :)

[17:50:01.0000] <Hixie>
nice

[18:45:00.0000] <Hixie>
/me works his way towards emptying his mailbox of anything other than whatwg and public-html mail

[18:52:00.0000] <Lachy>
hsivonen, yt?

[18:55:00.0000] <Lachy>
hsivonen, compare http://validator.nu/?doc=http%3A%2F%2Flachy.id.au%2Ftemp%2Fid.xhtml and http://validator.nu/?doc=http%3A%2F%2Flachy.id.au%2Ftemp%2Fid.html

[18:55:01.0000] <Lachy>
the first in XHTML, the other is HTML. Why is the validity of the id attribute different?

[18:56:00.0000] <Hixie>
IDs in XML can't start with numbers

[18:56:01.0000] <Hixie>
according to XML

[18:57:00.0000] <Lachy>
welcome back Hixie!

[18:57:01.0000] <Hixie>
not sure what this means in terms of HTML5 though

[18:57:02.0000] <Hixie>
thanks :-)

[18:57:03.0000] <Hixie>
/me started the day with ~8000 e-mails

[18:57:04.0000] <Hixie>
down to ~4000

[18:57:05.0000] <Lachy>
I thought that XML issue related to the validity constraint with XML DTDs, and since we're not using DTDs, it didn't matter

[18:58:00.0000] <Hixie>
yeah, i assume he's got some XML stuff in there somewhere though :-)

[18:58:01.0000] <Lachy>
did you notice the poll for the selectors api naming debate? That closes later today, if you wanted to put in your vote

[18:58:02.0000] <Hixie>
done already

[18:59:00.0000] <Lachy>
cool

[18:59:01.0000] <othermaciej>
having a poll for that was slightly silly

[18:59:02.0000] <othermaciej>
but an online poll is still better than a voice vote at a face-to-face meeting

[18:59:03.0000] <Hixie>
that was the gist of my comment on the poll

[19:00:00.0000] <deltab>
the xml:id spec requires NCName for ids (not that XHTML uses xml:id)

[19:00:01.0000] <deltab>
(so it's not just a DTD thing)

[19:01:00.0000] <Hixie>
in XML it's a DTD thing

[19:01:01.0000] <Hixie>
in XMLID it's an XML ID thing

[19:01:02.0000] <Hixie>
in RelaxNG it's a RelaxNG thing

[19:01:03.0000] <Hixie>
etc :-)

[19:01:04.0000] <Hixie>
just happens many of them agree

[19:02:00.0000] <Hixie>
(insert rant about whatwg being affected by "not invented here" syndrome here, and rant about how whatwg ignores industry practices, etc)

[19:03:00.0000] <Lachy>
then I wonder why hsivonen's validator marks them different, since I thought it uses RelaxNG for checking some validity issues for both serialisations

[19:05:00.0000] <zcorpan_>
it may be that id processing isn't done at all on the html side. or something.

[20:11:00.0000] <Lachy>
Hey, I'm looking for a couple of example sites where <meter> could be useful. Any ideas? I need the use cases for my presentation I'm doing on Friday

[20:12:00.0000] <Lachy>
oh, YouTube is a perfect example!

[20:19:00.0000] <Hixie>
Lachy: the example in the spec is from google groups

[20:20:00.0000] <Lachy>
yeah, I could show that too

[20:20:01.0000] <Hixie>
there's also the usual thinks like disk usage (google docs, webmail apps)

[20:21:00.0000] <Lachy>
ah, the spec even has a nice image for me to steal :-)

[20:21:01.0000] <Hixie>
:-)

[20:24:00.0000] <Lachy>
the spec shows this example <meter><object data="graph75.png">0.75</object></meter>, does that work with <img> too?

[20:24:01.0000] <Hixie>
yup

[20:24:02.0000] <Hixie>
uh wait

[20:24:03.0000] <Hixie>
no

[20:24:04.0000] <Hixie>
we don't yet scan alt="" iirc

[20:25:00.0000] <Hixie>
though mind you, i haven't even looked at the spec in 3 weeks

[20:25:01.0000] <Hixie>
so what do i know :-P

[20:26:00.0000] <Lachy>
ok, I'll use the object example.  Is the intention that it would render the image, instead of generating its own bargraph?

[20:27:00.0000] <Lachy>
(perhaps I should have spent more time working on this, instead of cramming at the last minute)

[20:28:00.0000] <Lachy>
ah, it probably wouldn't.The image would probably just be fallback for UAs that don't support meter

[20:28:01.0000] <Hixie>
yeah

[20:28:02.0000] <Hixie>
the latter

[20:30:00.0000] <Lachy>
google groups doesn't appear to use that bar graph any more

[20:39:00.0000] <Hixie>
heh

[22:33:00.0000] <Hixie>
/me wonders what he's supposed to do with http://www.w3.org/mid/op.tvg629ividj3kv@hp-a0a83fcd39d2

[22:34:00.0000] <Hixie>
i already examined what the browsers did when i wrote the spec

[22:34:01.0000] <Hixie>
and the e-mail doesn't point out any problems

[22:58:00.0000] <Lachy>
without evidence to show that sites depend on one particular behaviour over another, I think the spec is fine as is

[23:06:00.0000] <Lachy>
Hixie, see the comments. http://www.search-this.com/2007/07/30/html5-tables/

[23:06:01.0000] <Lachy>
Looks like a possible bug in the table header algorithm.

[23:06:02.0000] <Hixie>
yeah known issue

[23:06:03.0000] <Lachy>
ok

[23:06:04.0000] <Hixie>
my implementation fixes it

[23:07:00.0000] <Hixie>
i'll port that back to the spec in due course

[23:07:01.0000] <Lachy>
cool

[23:07:02.0000] <Hixie>
(the implementation that i used to test whether headers="" helped, that is)

[23:07:03.0000] <Hixie>
at least if i understand this right

[23:07:04.0000] <Hixie>
the second comment confused me

[23:08:00.0000] <Lachy>
I noticed that algorithm doesn't seem to deal with <table><col>, only with <table><colgroup><col>

[23:08:01.0000] <Hixie>
yeah

[23:08:02.0000] <Hixie>
didn't know <table><col> was allowed in XHTMl1

[23:08:03.0000] <Lachy>
it's not, but the algorithm should still deal with it

[23:08:04.0000] <Lachy>
oh, in XHTML 1, yes it is allowed

[23:09:00.0000] <Hixie>
yeah we'll have to fix that

[23:09:01.0000] <Hixie>
there was a mail about it, iirc, i think i saved it to my semantics-table pile

[23:10:00.0000] <Lachy>
I started writing a JS implementation of it last night.

[23:14:00.0000] <Hixie>
/me fears the "conflation of issues or convergence of interests?" thread

[23:14:01.0000] <Lachy>
actually, that thread has turned out to be quite constructive, for the most part

[23:14:02.0000] <Hixie>
how unlikely

[23:15:00.0000] <othermaciej>
hola

[23:15:01.0000] <othermaciej>
Hixie: there are some offshoots which are not completely insane

[23:16:00.0000] <othermaciej>
and in any case it is way better than "Re: Formal Recorded Complaint"

[23:16:01.0000] <Hixie>
hey maciej

[23:16:02.0000] <othermaciej>
hello Hixie 

[23:16:03.0000] <othermaciej>
back from vacation?

[23:16:04.0000] <Hixie>
yeah

[23:17:00.0000] <othermaciej>
cool, did you have a good holiday?

[23:17:01.0000] <Hixie>
it was... different from work

[23:17:02.0000] <othermaciej>
well yes, that's the bare minimum one should expect

[23:18:00.0000] <Hixie>
mostly it was tiring :-)

[23:18:01.0000] <Hixie>
i started today with 8000 e-mails

[23:18:02.0000] <Hixie>
i now have 1100

[23:19:00.0000] <othermaciej>
that's good progress

[23:19:01.0000] <Lachy>
how on earth do you read so many emails that quickly?

[23:20:00.0000] <Hixie>
i'm on a LOT of mailing lists, and i don't do more than scan the subject lines of most of them

[23:20:01.0000] <Hixie>
e.g. i don't do more than scan the subject lines of www-tag

[23:21:00.0000] <Lachy>
lol, that's what I do with www-tag too

[23:21:01.0000] <Hixie>
most of the remainder are those that i have to actually read

[23:37:00.0000] <Lachy>
in what cases would it be advantageous to use <progress value="60"> instead of just <progress>60%</progress> (or some other appropriate content within the element)?

[23:39:00.0000] <Lachy>
oops, that should be <progress value="0.6">

[23:40:00.0000] <othermaciej>
Lachy: is <progress> expected to render the content text anywhere?

[23:40:01.0000] <othermaciej>
if so, you might want an amount of kilobytes or something

[23:40:02.0000] <Lachy>
only as fallback

[23:40:03.0000] <othermaciej>
instead of a percentage

[23:44:00.0000] <Lachy>
oh, I see, it's an advantage when you want to be able to get progress.value, since that would return 0 without the attributes.  It never reflects the value parsed from textContent

[00:04:00.0000] <hsivonen>
Lachy: your HTML doc was HTML5. your XHTML doc was versionless and the validator picked XHTML 1.0 because XHTML5 is not near CR yet

[00:05:00.0000] <Lachy>
oh

[00:05:01.0000] <Lachy>
I see, it validates if I select HTML5 manually

[00:06:00.0000] <Lachy>
s/HTML5/XHTML5/

[00:08:00.0000] <hsivonen>
Hixie: btw, IDs in XML are crazier than just being a DTD thing or an xml:id thing. the permitted charecters in IDs are different in DTDs and in xml:id.

[00:08:01.0000] <hsivonen>
validator.nu has a long-standing bug here

[00:09:00.0000] <hsivonen>
(inherited from elsewhere)

[00:09:01.0000] <hsivonen>
the bug being that the colon is not allowed in IDs for XHTML 1.0

[00:10:00.0000] <hsivonen>
because the schema uses the XSD definition of ID, but XHTML 1.0 should be subject to the DTD definition of ID

[00:26:00.0000] <Lachy>
hsivonen: do you have any idea about the origin of the character restrictions in ID and why they are the way they are?

[00:57:00.0000] <hsivonen>
Lachy: I don't *know* but I imagine the restriction to Name in XML 1.0 comes from SGML tradition and the production for Name comes from a feeling the for aesthetics Names should capture a certain letter-like format

[00:58:00.0000] <hsivonen>
Lachy: as for why XSD and xml:id used NCName instead, I can only guess that the writes of the specs didn't want IDs to look like QNames-in-content

[00:59:00.0000] <hsivonen>
Lachy: anyway, Real Software needs to check those string for equality and both the Name and NCName production are arbitrary

[00:59:01.0000] <Hixie>
hsivonen: fun

[01:21:00.0000] <MikeSmith>
I suspect the reason for character restrictions in ID values is for reasons similar to character restrictions in symbol names in programming languages

[01:21:01.0000] <MikeSmith>
and in filesystems

[01:21:02.0000] <annevk>
wb Hixie

[01:23:00.0000] <hsivonen>
MikeSmith: and in filesystems, the historical convergence has been towards allowing everything except the character reserved for separating path segments

[01:24:00.0000] <hsivonen>
MikeSmith: because eventually someone out there wants to use characters that don't fit the sense of aesthetics of the spec writer

[01:24:01.0000] <hsivonen>
MikeSmith: and no techical badness ensues so it is hard to deny them

[01:32:00.0000] <MikeSmith>
hsivonen - you won't get any disagreement from me about that. Just was chiming in to speculate on what the thought behind the restriction might have been. FWIW, I've personally run into real-world instances where constraints on ID values are completely counter productive. So I'm not suggesting I think they're sound.

[01:33:00.0000] <MikeSmith>
I think it's enough of a problem that any tool or processing app that does constraint-checking on ID/xml:id values should offer a switch for disabling it.

[03:58:00.0000] <annevk>
jgraham, should we tackle it for HTML5 though or let it rest for HTML6 to tackle?

[03:58:01.0000] <annevk>
HTML5 already tackles so many issues and there hasn't been much experimental stuff happening yet with namespaces in HTML except one failed attempt at Opera (recognizing xmlns)

[04:15:00.0000] <hsivonen>
annevk: can you say whether Opera has tried hardwiring well-known prefixes?

[04:29:00.0000] <annevk>
hsivonen, I don't think we did, but what exactly do you mean?

[04:38:00.0000] <hsivonen>
annevk: I mean hardwiring the rdf prefix to the RDF namespace, the dc prefix to dublin core namespace, svg to svg namespace, etc.

[04:39:00.0000] <annevk>
so you would have to write <svg:svg> etc. ?

[04:40:00.0000] <annevk>
/me does not really like that approach

[04:44:00.0000] <hsivonen>
annevk: I'd like to have scope-based ns mapping for <svg> and <math> subtrees, yes.

[04:45:00.0000] <hsivonen>
but yes, I did mean doing hardwired things with colonified names

[04:46:00.0000] <annevk>
and how do you deal with empty elements?

[04:47:00.0000] <hsivonen>
annevk: /> would have to close the element when in the tag name is colonified or in <svg> or <math> scope

[04:47:01.0000] <annevk>
hmm

[04:48:00.0000] <hsivonen>
I care more about <svg> and <math> scoping than about colonified names (except xlink:href)

[04:48:01.0000] <zcorpan_>
or we make certain tags void?

[04:49:00.0000] <hsivonen>
zcorpan_: that would miss an opportunity to be forward-compatible with new empty elements added later to MathML or SVG

[04:49:01.0000] <zcorpan_>
true

[04:50:00.0000] <annevk>
I'm afraid of breakage

[04:50:01.0000] <annevk>
I'd rather let this wait until HTML5 parsing itself is reasonably interoperable

[04:52:00.0000] <hsivonen>
annevk: I agree we should probably not tackle this in the next 6 months, but I think it is important to make SVG more competitive vs. proprietary closed web technologies

[04:56:00.0000] <Lucian>
hi

[04:56:01.0000] <annevk>
hi

[04:58:00.0000] <annevk>
hsivonen, yeah, that's a fair point I suppose

[05:13:00.0000] <hsivonen>
annevk: fwiw, the text scaling issue applies to font zoom and em-sized canvases anyway

[05:14:00.0000] <annevk>
well, I meant that if the canvas size depended on the size of the div

[05:14:01.0000] <annevk>
that is, if the canvas grid depended on the rendered size of the <div>

[05:14:02.0000] <annevk>
(the canvas grid for <canvas> depends on width and height, which simply take integers)

[05:15:00.0000] <hsivonen>
oh. I missed the point

[05:17:00.0000] <annevk>
I suppose it was not clear

[06:08:00.0000] <hsivonen>
whoa! the XHTML2 WG namespace minutes are 11 months old!

[06:13:00.0000] <zcorpan_>
how is "version 1.0 or later of XML" different or more specific than "some version of XML"?

[06:17:00.0000] <hsivonen>
zcorpan_: it looks more rigorous than "some"

[06:18:00.0000] <annevk>
/me made http://html5.org/ more usable

[06:37:00.0000] <annevk>
hsivonen, isn't the table sorted?

[06:39:00.0000] <Philip`>
It isn't - it puts "AElig;" before "AElig"

[06:39:01.0000] <annevk>
Hixie, help-whatwg and implementors-whatwg are affected as well

[06:41:00.0000] <annevk>
ah, I see, oh well, it doesn't realy matter as long as we keep a dictionary implementation

[06:41:01.0000] <Philip`>
(http://canvex.lazyilluminati.com/misc/parser/tokeniser.html uses a big entityNameMatch regexp, which sorts the names by descending length, but my C++ one uses a lexicographically-sorted table instead since I copied hsivonen's idea)

[06:47:00.0000] <annevk>
a \ exposes a bug in your JS tokenizer

[06:48:00.0000] <Philip`>
?

[06:49:00.0000] <annevk>
it outputs \\

[06:50:00.0000] <hsivonen>
aren't you supposed to escape \ in JSON?

[06:50:01.0000] <Philip`>
You are supposed to

[06:50:02.0000] <Philip`>
which is why it does :-)

[06:50:03.0000] <annevk>
oh

[06:50:04.0000] <annevk>
/me hides

[06:50:05.0000] <Philip`>
The "Serialised HTML" view gives a single \

[07:08:00.0000] <grimeboy>
You know it'd be pretty interesting if HTMLAudioElements had an attribute of type string called waveform, or even better an attribute of type byte_array called waveform.

[07:10:00.0000] <grimeboy>
Actually javascript needs a ByteArray.

[07:10:01.0000] <Philip`>
For reading the waveform, or for altering it?

[07:11:00.0000] <annevk>
ES4 has a byte array

[07:11:01.0000] <grimeboy>
Both.

[07:11:02.0000] <annevk>
XMLHttpRequest level 2 is going to use it

[07:11:03.0000] <grimeboy>
Oh that's good.

[07:11:04.0000] <Philip`>
Byte arrays wouldn't be very useful for 16-bit audio

[07:12:00.0000] <grimeboy>
No, probably not.

[07:13:00.0000] <Philip`>
For just generating audio files dynamically, I guess you can already do audio.src = 'data:audio/wav;base64,...'

[07:14:00.0000] <grimeboy>
Oh right. That's a interesting idea.

[07:19:00.0000] <Philip`>
Are there specific cases for which it'd be useful to read the waveform?

[07:20:00.0000] <Philip`>
(I can't think of anything obvious that wouldn't be horribly slow to do in JavaScript...)

[07:31:00.0000] <grimeboy>
No, I suppose most things would be too slow (although javascript implementations are getting faster). I was thinking the traditional stuff, changing pitch, averaging with a sine wave, etc. etc.

[08:05:00.0000] <zcorpan>
annevk: the thread is about wrap=off specifically

[08:05:01.0000] <annevk>
does that map to soft?

[08:06:00.0000] <zcorpan>
submission-wise yes, rendering-wise no

[08:06:01.0000] <annevk>
oh, I see

[08:07:00.0000] <annevk>
spec doesn't seem to deal with invalid values

[08:08:00.0000] <annevk>
edited my reply

[08:09:00.0000] <zcorpan>
"For other attributes that contain invalid values"

[08:10:00.0000] <annevk>
so like soft

[14:50:00.0000] <Hixie>
annevk: IE supports draggable=""??

[15:35:00.0000] <Xsss4hell>
hi

[15:35:01.0000] <zcorpan_>
hi

[15:35:02.0000] <Xsss4hell>
http://webforms2.org/ is offline

[15:35:03.0000] <Xsss4hell>
or are you just redesigning?

[15:36:00.0000] <zcorpan_>
probably just down temporarily

[15:36:01.0000] <Xsss4hell>
yep, hope it's back in some minutes =)

[15:37:00.0000] <Xsss4hell>
I'm developing a framework and wanted to use this nice xforms

[15:56:00.0000] <Xsss4hell>
which xpath and xforms frameworks do you recommend for clients having browser which do not support xforms and xpath

[15:56:01.0000] <Xsss4hell>
I'm talking about IE :P

[15:58:00.0000] <zcorpan_>
are there browsers that natively support xforms?

[15:59:00.0000] <zcorpan_>
gsnedders: your test case sucks ;) "FAIL" should be "This text should be striked out"

[16:00:00.0000] <zcorpan_>
gsnedders: or better yet, "This line should have a green background" along with background:lime

[16:01:00.0000] <zcorpan_>
and p { background:red }

[16:01:01.0000] <Philip`>
(Is it striked or struck? Or maybe it could be strike outed)

[16:03:00.0000] <zcorpan_>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%3Cstyle%3Ep%7Bbackground%3Ared%7D%23t%E91%5C%24t%7Bbackground%3Alime%7D%3C/style%3E%3Cp%20id%3D%22t%E91%24t%22%3EThis%20line%20should%20have%20a%20green%20background.

[16:06:00.0000] <gsnedders>
zcorpan_: meh. throwing stuff together while half asleep :P

[16:06:01.0000] <jgraham>
annevk: I think we will have to at least look at the issue (SVG+etc. in text/html) for HTML 5. I have no idea what the outcome will be.

[16:06:02.0000] <zcorpan_>
gsnedders: fair enough :)

[16:06:03.0000] <gsnedders>
zcorpan_: and also in the middle of doing other stuff

[16:18:00.0000] <Xsss4hell>
no script that enabled xpath and xforms crossbrowser?

[16:24:00.0000] <zcorpan_>
Xsss4hell: i don't know of any. there is an xforms player plugin for ie and an extension or something for firefox, though, iirc

[16:24:01.0000] <zcorpan_>
Xsss4hell: there are however scripted implementations of web forms 2.0

[16:25:00.0000] <zcorpan_>
Xsss4hell: and opera supports wf2 natively

[16:35:00.0000] <Xsss4hell>
I've found some xpath and xforms player but thought you know better, however thankx

[16:35:01.0000] <Xsss4hell>
sf.net^^

[16:55:00.0000] <Hixie>
only 772 e-mails to go


2007-08-02
[17:17:00.0000] <Hixie>
Robert Burns and I interpret XHTML2 and related issues in rather dramatically different ways.

[17:24:00.0000] <Hixie>
man, some people are taking this stuff way too seriously

[17:24:01.0000] <Hixie>
lighten up people, it's only a markup language

[18:01:00.0000] <Hixie>
blimey, what's with the obsession with xhtml

[18:01:01.0000] <Hixie>
didn't we already establish that was a waste of time?

[18:04:00.0000] <zcorpan_>
no, it's the future and the answer to all problems :)

[18:04:01.0000] <Hixie>
sheesh

[18:05:00.0000] <Hixie>
/me couldn't care less about which syntax people use

[18:05:01.0000] <Hixie>
but really

[18:05:02.0000] <Hixie>
arguing about what the syntax should be is a waste of time

[18:05:03.0000] <Hixie>
since it's not going to change

[18:10:00.0000] <grimboy>
semantics > syntax

[18:28:00.0000] <kingryan>
grimboy: don't you mean "semantics &gt; syntax"?

[18:29:00.0000] <grimboy>
Heh

[23:48:00.0000] <Hixie>
robert burns has too much free time

[23:48:01.0000] <Hixie>
i swear he accounts for like half the volume to public-html

[23:49:00.0000] <Hixie>
and his e-mails are rarely short

[23:49:01.0000] <Hixie>
i don't really know if his posts are especially good, because my eyes start to glaze over when i get to an e-mail from him, because subconsciously i realise that i can get through the mail much faster if i just skip his

[23:50:00.0000] <Hixie>
i need to read them more carefully

[23:52:00.0000] <othermaciej>
he is not so clueless that you could immediately dismiss him, but his understanding of many issues seems murky

[23:53:00.0000] <karlUshi>
http://www.robburns.com/

[00:30:00.0000] <Hixie>
wow, some people really have trouble with namespaces

[00:31:00.0000] <Hixie>
at least one of these threads has a number of people repeatedly misreading what I thought was a pretty simple statement about namespaces

[01:10:00.0000] <Hixie>
http://slashdot.org/comments.pl?sid=253027&threshold=-1&commentsort=3&mode=nested&pid=20064769

[01:10:01.0000] <Hixie>
slashdot is weird

[01:10:02.0000] <Hixie>
(me and zcorpan trying to explain something to someone commenting there)

[01:39:00.0000] <MikeSmith>
slashdot is hopeless

[01:41:00.0000] <MikeSmith>
except for the entertainment value provided by some people who post there

[01:41:01.0000] <MikeSmith>
much of which entertainment is unintended on the part of the people posting there

[01:43:00.0000] <karlUshi>
what is the color of Henri IV's white horse?   (translated from French say for children)  cf "slashdot is weird"

[02:51:00.0000] <annevk>
Hixie, I thought it did

[02:52:00.0000] <annevk>
Seems I was wrong: http://www.google.com/search?q=msdn+draggable

[02:52:01.0000] <annevk>
(that's not a testcase, but it comes pretty close...)

[07:08:00.0000] <Xsss4hell>
Do you have recommendations related to css typography and layouts?

[07:10:00.0000] <annevk>
you probably want something like #css

[07:12:00.0000] <Xsss4hell>
haha I'm already in that chanell ^^

[07:12:01.0000] <Xsss4hell>
And I'm pretty good with css

[07:12:02.0000] <Xsss4hell>
But I'm open to learn new things

[07:13:00.0000] <annevk>
from http://www.whatwg.org/specs/web-apps/current-work/#scope "The scope of this specification does not include addressing presentation concerns (although default rendering rules for Web browsers are included at the end of this specification)."

[07:14:00.0000] <Xsss4hell>
Isn't WHATWG a consortion that heps standards that were left alone by W3C. So that's why I'm here, to learn..

[07:14:01.0000] <annevk>
well, we're working together with the W3C again

[07:14:02.0000] <annevk>
s/again/now/

[07:15:00.0000] <annevk>
currently "we"'re not doing anything related to typography or layouts

[07:15:01.0000] <annevk>
just HTML, APIs, etc.

[07:15:02.0000] <Xsss4hell>
I know you're defining new standards

[07:18:00.0000] <annevk>
well, then I'm not sure how I can help you :)

[07:21:00.0000] <Xsss4hell>
OK, I've once tried an aplha of XHTML2, it was pretty powerful, but it power was just limited to things that were intended to preset in the preview, it was a closed preview, or it was very hard to find any information about it, I don't know. But can you tell me any news related to XHTML2, can I start websites with it now? I mean is the draft finished but just needs approvement by the w3c, or is it not ready yet.

[07:23:00.0000] <annevk>
We're not working on XHTML2

[07:23:01.0000] <Xsss4hell>
only HTML5?

[07:23:02.0000] <Xsss4hell>
hmm..

[07:24:00.0000] <Xsss4hell>
so why XHTML2 and HTML5? I don't understand

[07:24:01.0000] <annevk>
yes, see http://www.whatwg.org/specs/web-apps/current-work/#relationship0

[07:25:00.0000] <zcorpan>
http://blog.whatwg.org/faq/#why-html5-xhtml2

[07:36:00.0000] <Xsss4hell>
ok I've read it all

[07:38:00.0000] <Xsss4hell>
I just hope, microsoft doesn't buy html5.

[07:40:00.0000] <annevk>
?

[07:40:01.0000] <Xsss4hell>
oh http://webforms2.org/ is still down

[07:40:02.0000] <Xsss4hell>
somebody informed the site owner?

[07:44:00.0000] <Xsss4hell>
I want to use webform2, but they're offline. aarrrgh http://webforms2.org/

[07:44:01.0000] <Xsss4hell>
since 2days or more..

[07:44:02.0000] <Xsss4hell>
a w3c site

[07:45:00.0000] <Philip`>
webforms2.org doesn't seem to have ever existed

[07:46:00.0000] <Philip`>
Oh, maybe it did, but Google's not very good at finding it

[07:46:01.0000] <zcorpan>
Xsss4hell: what did you expect to find at webforms2.org?

[07:47:00.0000] <annevk>
Philip`, http://whois.domaintools.com/webforms2.org

[07:47:01.0000] <annevk>
It most certainly existed

[07:47:02.0000] <mpt>
Xsss4hell, it's probably impossible for Microsoft to "buy" html5

[07:48:00.0000] <Philip`>
http://web.archive.org/web/20070125014928/http://webforms2.org/ - it doesn't seem to have existed in terms of having actual content

[07:48:01.0000] <Xsss4hell>
I just wanted to use webforms2 on my websites with all browsers that support it, and serve other browsers an alternative

[07:49:00.0000] <zcorpan>
was it this? http://sourceforge.net/projects/wf2/

[07:49:01.0000] <Xsss4hell>
so before..I use it: webforms2 or xforms?? 

[07:49:02.0000] <mpt>
Xsss4hell, you may be looking for http://www.whatwg.org/specs/web-forms/current-work/

[07:50:00.0000] <Xsss4hell>
Thanks for your help :)

[07:51:00.0000] <annevk>
zcorpan, a frontpage for that I believe, yes

[07:51:01.0000] <Xsss4hell>
omg (2005-09-28 13:08)

[07:51:02.0000] <Philip`>
Oh, it did have content once - http://web.archive.org/web/20070327030455/http://webforms2.org/download/wf2_0_1.zip

[07:51:03.0000] <Xsss4hell>
yes, it is the frontpage for weforms2.org

[07:52:00.0000] <annevk>
Xsss4hell, Web Forms 2 has been stable for quite some time, but it has had bug fixes since that date (latest is October 2006), eventually it will be integrated into HTML5

[07:53:00.0000] <Xsss4hell>
oh, so I don't need to use it until 2022?

[07:53:01.0000] <Xsss4hell>
oh my..

[07:54:00.0000] <annevk>
??

[07:55:00.0000] <Xsss4hell>
the docs say html5 will be ready earlier but we expect it getting fully supported or whatever in ~2022

[07:56:00.0000] <zcorpan>
Xsss4hell: html4 is not interoperably implemented. css level 2 is not interoperably implemented. can you use those today anyway?

[07:56:01.0000] <Xsss4hell>
yes I can, with lotta hacks and quirks. until then 20.000 test nees to be written an passsed, they say

[07:57:00.0000] <annevk>
prolly more

[07:58:00.0000] <zcorpan>
Xsss4hell: yeah. so you can use new features in html5 as they get implemented (although they may well have some bugs initially)

[07:59:00.0000] <Xsss4hell>
What I find is kinda weird is that the us military already have a working prototype of a new web-protocol that is thousand times faster then http and compresses so good that you can get some gb/s with a normal broadband cable connection..

[08:00:00.0000] <zcorpan>
that's cool

[08:00:01.0000] <Xsss4hell>
and thats old news..^^

[08:00:02.0000] <Xsss4hell>
they've things oh lord...

[08:02:00.0000] <Xsss4hell>
u know that there is already a cpu with 16cores and two 10gb/s lanes for network? It was a military chip producer, until they got dissmissed due to better concurennce

[08:02:01.0000] <Xsss4hell>
now they make high-end pc parts

[08:04:00.0000] <Xsss4hell>
Where to find practical examples of HTML5, with and without webforms or xforms or whatever =)

[08:05:00.0000] <zcorpan>
http://dev.opera.com/articles/view/improve-your-forms-using-html5/

[08:05:01.0000] <zcorpan>
http://simon.html5.org/presentations/html5-geekmeet.en

[08:06:00.0000] <zcorpan>
http://simon.html5.org/sandbox/html/suggest/

[08:12:00.0000] <Xsss4hell>
thaaaaaaaaaanks

[10:21:00.0000] <zcorpan>
annevk: http://quuz.org/xml5/play?source=%3Cxml%3Afoo%3E

[10:21:01.0000] <zcorpan>
should be in the http://www.w3.org/XML/1998/namespace namespace

[10:22:00.0000] <zcorpan>
similarly <xmlns:foo>

[10:23:00.0000] <zcorpan>
or wait

[10:24:00.0000] <zcorpan>
/me got unsure about xmlns

[10:25:00.0000] <zcorpan>
"Element names MUST NOT have the prefix xmlns."

[10:27:00.0000] <zcorpan>
well, as far as xml5 goes, the logical thing to do would be to just act as if it was declared, i.e. same as the xml prefix

[10:42:00.0000] <Philip`>
Is there any documentation on how IE handles XML-like tags outside of a <xml>?

[10:42:01.0000] <gsnedders>
I would hedge a bet at "no"

[10:43:00.0000] <zcorpan>
Philip`: define XML-like tags

[10:43:01.0000] <Philip`>
Tags with colons

[10:43:02.0000] <Philip`>
or I think ([a-zA-Z][^:>/\s]*):([^>/\s]+) in particular

[10:46:00.0000] <Philip`>
<x:y>...</x:y> makes an element with actual content, <x:y/> makes a void element, etc

[10:46:01.0000] <zcorpan>
i haven't seen documentation, but i've played with it a bit

[10:46:02.0000] <Philip`>
<x:y>a<x:z>b</x:y>c</x:y> does the usual non-tree thing

[10:46:03.0000] <Philip`>
Oops

[10:46:04.0000] <Philip`>
<x:y>a<x:z>b</x:y>c</x:z>

[10:46:05.0000] <zcorpan>
stray end tags are ignored

[10:46:06.0000] <zcorpan>
a start tag with a non-declared prefix implies a PI before it... or something

[10:46:07.0000] <zcorpan>
<?xml:namespace prefix = x />

[10:46:08.0000] <zcorpan>
at least when getting innerHTML

[10:48:00.0000] <Philip`>
<html xmlns:x="foo">Test<x:y> gives innerHTML with <BODY>Test<?xml:namespace prefix = x ns = "foo" /><x:y></x:y></BODY>

[10:48:01.0000] <Philip`>
but only when the xmlns is on <html>, not any other element

[10:49:00.0000] <zcorpan>
ah indeed

[10:49:01.0000] <Philip`>
(though you can put in multiple <html> tags anywhere and it uses them all)

[10:50:00.0000] <zcorpan>
innerHTML will output the PI if the declarations are on the element itself or an ancestor

[10:51:00.0000] <zcorpan>
attributes are parsed the same way as on other tags

[10:52:00.0000] <zcorpan>
end tags can have attributes in the tokenizer (</x:x y=">">)

[11:00:00.0000] <Philip`>
document.namespaces gives an array of them, including explicit (via <html xmlns:x>) and implicit (via <x:y>) ones

[11:01:00.0000] <Philip`>
(The namespace object has properties name, urn, and tagNames except the last one just gives me "not implemented" errors)

[11:02:00.0000] <Philip`>
(and onreadystatechange and readyState (?!))

[11:08:00.0000] <Philip`>
It's fun how MSDN is missing much of the information about these things, and much of the information that it does have is incorrect

[13:36:00.0000] <annevk>
Hixie, since you're online, the help-whatwg and implementors-whatwg archives are still not public

[13:50:00.0000] <Hixie>
fixed

[13:53:00.0000] <annevk>
cool

[14:09:00.0000] <annevk>
Interoperability is such a great thing: http://lists.w3.org/Archives/Public/www-archive/2007Aug/att-0002/offset-mess.htm

[14:10:00.0000] <othermaciej>
I wonder if Safari matches any of those

[14:10:01.0000] <annevk>
no access to Safari here, would be nice to have results

[14:11:00.0000] <othermaciej>
annevk: http://www.apple.com/safari/download/

[14:11:01.0000] <annevk>
for Ubuntu?

[14:11:02.0000] <othermaciej>
did someone else do the IE7 tests?

[14:11:03.0000] <othermaciej>
I assumed you had Windows based on that

[14:11:04.0000] <annevk>
ah, I have some hack to run IE7

[14:12:00.0000] <othermaciej>
or I can try it for you, but I don't know how to determine the table results from the test case that's referenced

[14:13:00.0000] <annevk>
check body or the html element; put the doc in quirks or standards and put lots of content in it (that causes a scrollbar) and few and observe the changes on the side

[14:13:01.0000] <annevk>
roughly

[14:18:00.0000] <annevk>
the main problem is deciding what is correct

[14:19:00.0000] <annevk>
as it will break stuff; in fact, it is already breaking stuff

[14:21:00.0000] <othermaciej>
I'm not smart enough to know offhand how to test whether <body> is getting the viewport height or ICB height or the max

[14:21:01.0000] <othermaciej>
if you point me to specific test cases to run I can report results

[14:22:00.0000] <annevk>
the viewport height of the box is roughly 256, you'll see it directly if body doesn't contain any content

[14:22:01.0000] <annevk>
say it's 256 without content and something larger than 256 with content that causes a scrollbar you know it's max

[14:22:02.0000] <annevk>
if it stays 256 it's vh

[14:23:00.0000] <annevk>
if it's some small without content and increments with new content it's icb

[14:23:01.0000] <othermaciej>
with the default contents of the test page

[14:23:02.0000] <othermaciej>
http://tc.labs.opera.com/tools/cssom/layout-dom-attributes

[14:23:03.0000] <othermaciej>
I have a height of 18

[14:23:04.0000] <othermaciej>
for the div

[14:23:05.0000] <othermaciej>
body is 18/256/18

[14:24:00.0000] <othermaciej>
html is 18/18/256

[14:24:01.0000] <othermaciej>
if I take off the doctype entirely is that good enough to test quirks mode

[14:24:02.0000] <othermaciej>
?

[14:24:03.0000] <annevk>
yeah

[14:24:04.0000] <othermaciej>
quirks div - 18/18/18

[14:24:05.0000] <othermaciej>
quirks body - 18/256/18

[14:25:00.0000] <othermaciej>
quirks html - 18/256/18

[14:25:01.0000] <othermaciej>
er

[14:25:02.0000] <othermaciej>
wrong

[14:25:03.0000] <othermaciej>
quirks html is 18/18/256

[14:25:04.0000] <othermaciej>
so there don't appear to be quirks/strict differences in webkit

[14:25:05.0000] <annevk>
put in lots of content that causes a scrollbar and test again?

[14:26:00.0000] <annevk>
(and are these results in testpage order or result page order?)

[14:27:00.0000] <othermaciej>
result page order

[14:27:01.0000] <othermaciej>
top to bottom

[14:27:02.0000] <othermaciej>
offsetHeight / scrollHeight / clientHeight

[14:30:00.0000] <othermaciej>
annevk: ok, I added enough content to make it scroll

[14:30:01.0000] <othermaciej>
standards mode div w/ scroll - 486/486/486

[14:30:02.0000] <annevk>
then just do the same tests again and report the new results (<div> is not needed fwiw)

[14:30:03.0000] <mpt>
zcorpan, is <http://simon.html5.org/sandbox/html/suggest/> supposed to work in any released browsers?

[14:30:04.0000] <othermaciej>
standards mode body w/ scroll - 486/486/486

[14:31:00.0000] <othermaciej>
standards mode html w/ scroll - 486/486/256

[14:31:01.0000] <othermaciej>
quirks mode is the same

[14:32:00.0000] <mpt>
zcorpan, never mind, I see it works in Opera :-)

[14:34:00.0000] <annevk>
so you guys didn't need any quirks standards mode differences...

[14:34:01.0000] <annevk>
interesting

[14:34:02.0000] <annevk>
othermaciej, this is WebKit from what day?

[14:38:00.0000] <othermaciej>
annevk: today

[14:40:00.0000] <othermaciej>
annevk: we certainly could be having compat issues due to our behavior - just none that have bubbled to the top of the fix pile

[14:41:00.0000] <annevk>
seems you guys have reverse engineered standards mode behavior in other browsers

[14:42:00.0000] <annevk>
I e-mailed a new version to http://lists.w3.org/Archives/Public/www-archive/2007Aug/ but it doesn't seem to have appeared just yet

[14:44:00.0000] <annevk>
hmm, maybe I did something wrong; oh well, if it's not there tomorrow I'll e-mail it again

[15:41:00.0000] <kingryan>
/me is working on bringing the ruby port of html5lib up to the current test suite

[15:53:00.0000] <kingryan>
a question about the html5lib test suite:

[15:54:00.0000] <kingryan>
anyone know what state the tokenizer should be in after the '2' in "<x x=1 x=2 X=3>" ?

[15:59:00.0000] <Philip`>
Just after the 2 it's still in the attribute value (unquoted) state, I think

[16:37:00.0000] <kingryan>
Philip`: yeah, you're right. I was looking for the wrong thing, though. :(

[16:57:00.0000] <kingryan>
anyone know if there's a way to browse the SVN changesets on code.google.com ?

[16:58:00.0000] <takkaria>
nope, there isn't

[16:58:01.0000] <takkaria>
the best you'll do is getting a local GUI client and browsing that logs/changes that way


2007-08-03
[17:01:00.0000] <kingryan>
I have an email mailbox that contains most of them. that works alright for me

[17:01:01.0000] <kingryan>
but I wish there were something better

[17:08:00.0000] <Philip`>
You can use svnsync to replicate the repository onto your own machine, then run a local web server with ViewSVN or Trac or something

[17:15:00.0000] <kingryan>
hmm, that sounds useful

[17:17:00.0000] <takkaria>
IIRC google have said they'll do something about it in an unspecified amount of time

[17:18:00.0000] <Philip`>
/me already has some other copy of Trac so he might see if it's trivial to do an html5lib one, since it's a reasonably nice source browser...

[18:24:00.0000] <Philip`>
Oh, good, it is easy to set up

[18:24:01.0000] <Philip`>
http://canvex.lazyilluminati.com/html5lib/

[18:29:00.0000] <kingryan>
nice, Philip` 

[18:53:00.0000] <welly>
is the faq response to "when will html 5 be finished" a joke?

[19:01:00.0000] <Philip`>
welly: No - it just has a more complete meaning of "finished", like requiring there to be multiple correct implementations of the whole specification before it's considered finished

[19:02:00.0000] <Philip`>
The writing of the specification will be finished much sooner, and implementations of some features are already available and usable

[19:02:01.0000] <welly>
Philip`: so in 15 years time, html 5 will be a complete specification? I can't help but think the internet moves on a little bit quicker than that

[19:03:00.0000] <welly>
i can't even imagine what kind of technology or standards we're going to be using in 15 years time

[19:04:00.0000] <Philip`>
HTML4 is almost ten years old already, and not that much has changed since then

[19:05:00.0000] <welly>
it seems like an interesting project but wow.. lol... 15 years. no, you're right but all the same. 

[19:05:01.0000] <Philip`>
(HTML4 and CSS2 (I think) wouldn't be considered finished yet, if they used the same criteria as HTML5)

[19:05:02.0000] <welly>
ok

[19:06:00.0000] <welly>
what's the likelyhood of html being taken up and implemented by the browser developers?

[19:07:00.0000] <welly>
er html 5

[19:07:01.0000] <Philip`>
Some parts have already been implemented - http://wiki.whatwg.org/wiki/Implementations_in_Web_browsers

[19:08:00.0000] <welly>
ahh ok.. well that's promising

[01:56:00.0000] <Lachy_>
My presentation went well this morning :-)

[03:12:00.0000] <ROBOd>
hey guys

[03:14:00.0000] <ROBOd>
is it a problem if i sent two emails about the HTML 5 spec to www-html and not to public-html? i sent two emails yesterday by mistake to www-html. i forgot of public-html, eh

[03:27:00.0000] <annevk>
you could provide a pointer to your e-mails on www-html in your e-mail to public-html

[03:29:00.0000] <ROBOd>
annevk: eh, two mailing lists with almost the same name. i think should unsubscribe from www-html, just for avoiding these situations

[03:30:00.0000] <annevk>
that's what I did, though for different reasons

[03:31:00.0000] <ROBOd>
should i resend the two emails to public-html? or just a pointer?

[03:33:00.0000] <annevk>
whatever suits you

[03:34:00.0000] <hsivonen>
ROBOd: I unsubscribed from www-html after I ended up sending email to public-html when public-html was in recess.

[03:35:00.0000] <hsivonen>
ROBOd: also, www-html is more or less in /dev/null mode as far as actually getting yourself heard goes

[03:35:01.0000] <ROBOd>
hsivonen: yeah, too much confusion

[03:35:02.0000] <ROBOd>
hsivonen: precisely. i only got one reply :P

[03:35:03.0000] <annevk>
I believe Hixie still reads it

[03:39:00.0000] <ROBOd>
that's very nice, if Hixie still reads www-html

[11:57:00.0000] <G0k>
hey all

[11:57:01.0000] <zcorpan_>
hey G0k

[11:58:00.0000] <G0k>
i'm a little confused about the RemoteEventTarget interface in html 5

[11:58:01.0000] <G0k>
"Any object that implements the EventTarget interface must also implement the RemoteEventTarget interface."

[11:58:02.0000] <zcorpan_>
which part is confusing?

[11:59:00.0000] <G0k>
i mean is that really the same as saying that the EventTarget interface is being amended to include some new methods?

[12:00:00.0000] <zcorpan_>
not quite but i guess the end result would be pretty much the same

[12:04:00.0000] <zcorpan_>
the difference being that it would be one interface instead of two... you can check whether an object implements a certain interface and which members it has

[12:04:01.0000] <G0k>
right but if any object that implements EventTarget "MUST" also implement RemoteEventTarget....

[12:05:00.0000] <G0k>
i guess this allows you to detect legacy behavior better

[12:06:00.0000] <G0k>
ok the other thing

[12:06:01.0000] <G0k>
"For connections to domains other than the document's domain, the semantics of the Access-Control HTTP header must be followed. [ACCESSCONTROL]"

[12:06:02.0000] <G0k>
[ACCESSCONTROL] seems to be a dead link?

[12:06:03.0000] <zcorpan_>
yes, refs haven't been written yet

[12:06:04.0000] <G0k>
heh. oki

[12:08:00.0000] <zcorpan_>
http://dev.w3.org/cvsweb/~checkout~/2006/waf/access-control/Overview.html?content-type=text/html;%20charset=utf-8

[12:12:00.0000] <G0k>
ok there's one other part that has me slightly confused

[12:12:01.0000] <G0k>
"If the Namespace is null and the Event field is message (including if it was not specified explicitly), then the MessageEvent interface must be used.

[12:12:02.0000] <G0k>
Otherwise, the Event interface must be used."

[12:12:03.0000] <G0k>
so that would seem to suggest for instance, that if you got an event like

[12:12:04.0000] <G0k>
Event: foo

[12:12:05.0000] <G0k>
data: bar

[12:13:00.0000] <G0k>
it would create an Event event with type "foo", then ignore the data field, since data isn't a defined attribute for the Event interface

[12:15:00.0000] <G0k>
which isn't bad, it's just precisely the opposite of how Opera has implemented it

[12:15:01.0000] <zcorpan_>
it is?

[12:15:02.0000] <G0k>
well like...see their sample code http://labs.opera.com/news/2006/09/01/

[12:16:00.0000] <G0k>
they have it sendings stuff like

[12:16:01.0000] <G0k>
Event: the-answer

[12:16:02.0000] <G0k>
data: 42

[12:16:03.0000] <G0k>
in their implementation, it sets the data attribute on the event

[12:16:04.0000] <G0k>
but that's not technically what the spec says, as the Event interface has no data attribute

[12:18:00.0000] <G0k>
for that matter, the pending patch for mozilla also implements it the other way

[12:18:01.0000] <zcorpan_>
hmm, might be a bug in the spec then

[12:20:00.0000] <G0k>
k. is there an issue tracker somewhere?

[12:20:01.0000] <zcorpan_>
no

[12:20:02.0000] <G0k>
should I use the forum?

[12:21:00.0000] <zcorpan_>
it would be great if you could send it to the list directly

[12:21:01.0000] <G0k>
which? html-dev or whatwg?

[12:21:02.0000] <zcorpan_>
whatwg

[12:22:00.0000] <G0k>
k

[12:22:01.0000] <zcorpan_>
what is html-dev?

[12:22:02.0000] <G0k>
er i dunno the one w3c runs

[12:22:03.0000] <zcorpan_>
aha. public-html

[12:22:04.0000] <G0k>
that one

[12:23:00.0000] <zcorpan_>
if you're subscribed then you can send it to that list if you want. if you do, include "detailed review of" in the subject line

[12:24:00.0000] <G0k>
k

[12:47:00.0000] <G0k>
sent

[12:47:01.0000] <zcorpan_>
thanks

[12:47:02.0000] <G0k>
np

[13:34:00.0000] <annevk>
it defaults to the MessageEvent actually, not Event

[13:34:01.0000] <G0k>
except if the Event: field is specified

[13:35:00.0000] <G0k>
(to something other than "message")

[13:35:01.0000] <G0k>
or at least that's what the spec seems to say right now

[13:36:00.0000] <annevk>
no

[13:36:01.0000] <annevk>
oh, maybe

[13:36:02.0000] <G0k>
"If the Namespace is null and the Event field is message (including if it was not specified explicitly), then the MessageEvent interface must be used.

[13:36:03.0000] <G0k>
Otherwise, the Event interface must be used."

[13:36:04.0000] <annevk>
/me looks at the spec

[13:36:05.0000] <annevk>
that seems wrong

[13:37:00.0000] <G0k>
yeah :)

[13:37:01.0000] <G0k>
it should just say

[13:37:02.0000] <G0k>
"Otherwise, the MessageEvent interface must be used"

[13:38:00.0000] <G0k>
(imho)

[13:38:01.0000] <annevk>
yeah, that's what we do anyway and it's proven useful

[13:38:02.0000] <G0k>
yep

[13:38:03.0000] <G0k>
it's what i did in my webkit implementation too, then i just re-read that

[13:41:00.0000] <annevk>
ah, you're implementing this in WebKit? awesome

[13:42:00.0000] <G0k>
yeah well...it's a little sketchy right now but with some work yeah it'll be cool. :)

[13:42:01.0000] <annevk>
:)

[13:42:02.0000] <G0k>
i'll be able to stream events to my iphone. :)

[14:02:00.0000] <annevk>
html5lib is linked from dailypythonurl, whatever that is

[14:02:01.0000] <annevk>
http://www.pythonware.com/daily/

[14:40:00.0000] <Hixie>
(yes i still read www-html)

[14:40:01.0000] <Hixie>
if anyone sees G0k again, let him know I agree with his proposal

[14:41:00.0000] <Hixie>
i'd mail him straight back but his mail is lost somewhere in my event-source feedback pile

[14:41:01.0000] <zcorpan_>
"Henry Mason" <hmason⊙mc>

[15:05:00.0000] <Hixie>
thanks

[15:22:00.0000] <Jero>
hey whats up

[15:23:00.0000] <Jero>
I was wondering when the WHATWG will start on the HTML5 spec again

[15:26:00.0000] <zcorpan_>
Jero: you mean when Hixie will start edit the spec again?

[15:27:00.0000] <Jero>
or that, yes

[15:27:01.0000] <zcorpan_>
well, he's back from his vacation now so i guess he just needs to get up to speed with all email

[15:28:00.0000] <Jero>
i see, thanks

[15:30:00.0000] <zcorpan_>
http://annevankesteren.nl/2007/07/web#comment-6176

[16:39:00.0000] <Hixie>
zcorpan_: i can't get to that site, is it down?

[16:40:00.0000] <hsivonen>
oh dear. this "low quality" thing is going to be a rathole

[16:43:00.0000] <Hixie>
it's the same style="" rathole we've always had

[16:44:00.0000] <hsivonen>
Hixie: no. now it is becoming a b and i rathole as well

[16:47:00.0000] <Hixie>
othermaciej_: <font> and style="" are "low quality" because they are media-specific, and potentially hide information from non-presentational UAs.

[16:48:00.0000] <Hixie>
othermaciej_: <div>s-containing-inlines are "low quality" because they imply some sort of paragraphing structure without giving any information about why

[16:49:00.0000] <Hixie>
othermaciej_: (typically, the latter is used as a workaround for the lack of a widget abstraction language like xbl)

[16:49:01.0000] <othermaciej>
Hixie: I don't see how style="" is media-specific

[16:50:00.0000] <othermaciej>
Hixie: it lacks the ability to give different styles per media, but can still be used in media-independent ways

[16:50:01.0000] <othermaciej>
since in practice most styles are not media-scoped anyway, nor do they need to be

[16:51:00.0000] <Hixie>
ok, strike the media-specific part

[16:51:01.0000] <othermaciej>
in that case, style="" hides no more information than <style>, <link rel="stylesheet"> or <style scoped>

[16:51:02.0000] <Hixie>
(though i actually do believe they usually result in media-specific content, even if it's technically possible to use them otherwise)

[16:51:03.0000] <othermaciej>
none of which you have labelled "low quality"

[16:52:00.0000] <othermaciej>
I have to go grab a beer, brb

[16:52:01.0000] <Hixie>
i considered including <style>, but in practice if you use a style sheet with selectors you tend to be pushed towards content that can stand even without the stylesheet.

[16:53:00.0000] <Hixie>
obviously though that's not automatically the case, which is why the other mode wouldn't be called "high quality"

[16:57:00.0000] <othermaciej>
back

[16:58:00.0000] <othermaciej>
if you include <style> and <link rel="stylesheet"> it would make it pretty hard to have a useful non-low-quality document

[16:58:01.0000] <Hixie>
indeed

[16:59:00.0000] <othermaciej>
but I don't think there's any evidence of a difference in hiding information from non-presentational UAs between those and style=""

[16:59:01.0000] <Hixie>
oh come now

[16:59:02.0000] <othermaciej>
I mean, that might be the case, but I am not sure how you would quantify it


2007-08-04
[17:00:00.0000] <Hixie>
oh i don't know that it could be quantified

[17:01:00.0000] <othermaciej>
is there any way to show evidence for or against?

[17:01:01.0000] <Hixie>
but it certainly is clear to me that you can improve the quality of almost any document with <font>, style="", or <div>s-containing-inlines, whereas i don't think the same can be said of all documents that do not have those characteristics.

[17:01:02.0000] <Hixie>
objectively? i don't know

[17:01:03.0000] <Hixie>
i haven't looked at it in detail

[17:02:00.0000] <othermaciej>
obviously it's hard to argue against your proposal if it is based on an an unfalsifiable claim

[17:03:00.0000] <Hixie>
it's not really a proposal

[17:03:01.0000] <Hixie>
just an idea

[17:03:02.0000] <Hixie>
i don't think we can drop style="", <font>, and <div>s-containing-inlines altogether

[17:03:03.0000] <othermaciej>
ok, s/proposal/idea/

[17:03:04.0000] <Hixie>
but i do think that we should discourage their use in favour of the alternatives

[17:03:05.0000] <Philip`>
Could call it "Compatibility" instead of "Low quality", and have it provide the features that are compatible with implementations and content and forbid things that are likely to cause authors pain and not attempt to make subjective judgements of value

[17:03:06.0000] <othermaciej>
dropping all those would certainly make a lot of documents nonconforming for seemingly-gratuitous reasons

[17:04:00.0000] <Hixie>
Philip`: it's not really compatibility either, since if we were desigining the language from scratch, you'd still want these features

[17:04:01.0000] <othermaciej>
I just don't see any reason to discourage style="" if <style scoped> won't be similarly discouraged

[17:04:02.0000] <othermaciej>
the latter can be abused in all the same ways, the main difference is that it is more verbose

[17:05:00.0000] <Hixie>
othermaciej: i'm not even talking about legacy content, even for a brand new language with no content i think these features are necessary, e.g. for prototyping

[17:05:01.0000] <othermaciej>
and less compatible with compatible with existing UAs

[17:05:02.0000] <Hixie>
nor am i talking about abuse

[17:05:03.0000] <Hixie>
you can abuse <ins> and <del>

[17:05:04.0000] <Hixie>
and anything else

[17:05:05.0000] <Hixie>
i'm talking about completely conforming and correct uses

[17:05:06.0000] <othermaciej>
ok, let me put it this way

[17:05:07.0000] <othermaciej>
I don't think systematically replacing all style="" with <style scoped> would improve any documents

[17:05:08.0000] <Hixie>
i agree

[17:06:00.0000] <Hixie>
it's a case-by-case problem

[17:06:01.0000] <Hixie>
sometimes style="" has to be replaced with samething else, e.g. using a different element altogether

[17:06:02.0000] <othermaciej>
thus, creating an incentive to do so (badge-seeking behavior, etc) seems like a bad idea

[17:07:00.0000] <Hixie>
granted, but it seems better than giving the impression that a page that uses only <div>s and style="" is fine.

[17:08:00.0000] <Hixie>
bbl

[17:16:00.0000] <othermaciej>
I agree such documents are not great; just not sure how to make a good machine-checkable rule that detects bad style

[17:40:00.0000] <Hixie>
yeah

[17:52:00.0000] <zcorpan_>
Hixie: i think having different conformance levels is a bad idea. we should just allow style="" and <div>s containing inlines but perhaps discourage the former for authors

[17:54:00.0000] <Hixie>
how?

[17:55:00.0000] <zcorpan_>
by saying that they are conforming, and then having a note about why style="" can be harmful

[17:56:00.0000] <zcorpan_>
this way, badge hunters won't replace their style=""s with more verbose but no better alternatives, and people can cite the spec about why people shouldn't use style=""

[17:57:00.0000] <zcorpan_>
(i don't think <div>s containing inlines is harmful)

[17:59:00.0000] <Hixie>
there are pages that consist of nothing but <div>s and style=""s

[17:59:01.0000] <Hixie>
or even just <div>s and stylesheets

[18:00:00.0000] <zcorpan_>
indeed

[18:01:00.0000] <zcorpan_>
they are not going to disappear if we label them non-conforming, but some might be rewritten to use constructs that pass validation without increasing the quality or semantics

[18:02:00.0000] <zcorpan_>
i don't think labeling them non-conforming will result in them being rewritten to use semantic markup

[18:04:00.0000] <Hixie>
it might not affect _those_ pages, but i think it would help with people the same way that having "strict" vs "transitional" has helped some in terms of mindshare growth

[18:05:00.0000] <zcorpan_>
hmm, perhaps

[18:07:00.0000] <Philip`>
That seems to resulted in people choosing to use transitional twenty times more often than strict

[20:40:00.0000] <Lachy_>
<alt for=idref> is exactly the kind of explicit association that won't work in reality. I do not understand why some people have concluded explicit markup associations are the only way to solve the problem - they seem to be ignoring the expressiveness of natural language

[23:56:00.0000] <Hixie>
"Fine, so my comparison wasn't perfect. None ever are." is an interesting way of dismissing someone's counterargument that i hadn't seen before

[02:46:00.0000] <Dashiva>
Lachy_, Hixie: Grumbling in here is just going to spawn another formal objection, be careful ;)

[02:47:00.0000] <Lachy_>
Dashiva: we're still allowed to discuss issues in here

[02:48:00.0000] <gsnedders>
Lachy_: even though it's logged in the same place?

[02:49:00.0000] <Lachy>
gsnedders: the fact that it's logged doesn't matter

[02:50:00.0000] <Lachy>
as long as our comments focus on the issue and not attack anyone personally, it's fine

[03:02:00.0000] <hsivonen>
hmm. apparently the W3C survey system uses colons in ids

[03:18:00.0000] <Hixie>
i'll be damned if i'm going to let people stop me from chatting about html on irc

[03:28:00.0000] <ROBOd>
<hello world=all>

[03:29:00.0000] <ROBOd>
i have a question: is there any implementation of the outlining algorithm?

[03:33:00.0000] <krijnh>
Man, that Lachy is irritating me

[03:33:01.0000] <krijnh>
Ow, wait

[03:33:02.0000] <Lachy>
?

[04:57:00.0000] <Philip`>
ROBOd: Does the implementation in http://www.whatwg.org/specs/web-apps/current-work/#outlines count?

[05:00:00.0000] <zcorpan_>
re sam ruby's idea with cdata, isn't it better to hide the text with css than to use a specific syntactic construct?

[05:00:01.0000] <ROBOd>
Philip`: not really. i was thinking of some independent implementation

[05:01:00.0000] <ROBOd>
Philip`: one that also determines which heading and section applies to a particular node

[05:01:01.0000] <zcorpan_>
svg\:svg { display:none; } should do the trick, also for browsers that don't support declarative namespaces

[05:01:02.0000] <ROBOd>
using the given algorithm

[05:02:00.0000] <zcorpan_>
ROBOd: i don't know of any

[05:02:01.0000] <Philip`>
zcorpan_: You'd want to hide it from browsers that support namespaces but don't support SVG

[05:03:00.0000] <Philip`>
(Also svg\:svg would hide it from IE, which does support namespaces)

[05:03:01.0000] <zcorpan_>
Philip`: sam's proposal doesn't address that either

[05:03:02.0000] <zcorpan_>
(but it doesn't support svg, which is why he wanted to hide the text from ie)

[05:04:00.0000] <zcorpan_>
Philip`: however, other browsers already support svg, so that is a non-issue

[05:05:00.0000] <hsivonen>
hmm. I now see that the quality of English in my reply to Sam is really poor. fortunately, English doesn't use Draconian error handling.

[05:06:00.0000] <hsivonen>
my problem is that I don't see my own errors until I have flushed the writing flow from my head

[05:06:01.0000] <zcorpan_>
is error handling for English defined? :)

[05:06:02.0000] <zcorpan_>
hsivonen: try reading backwards, that sometimes helps finding errors

[05:07:00.0000] <zcorpan_>
but mostly just typos

[05:09:00.0000] <Philip`>
zcorpan_: I suppose SVG is already supported everywhere but if you used e.g. MathML then you wouldn't want non-MathML-supporting browsers showing all the raw textual content

[05:09:01.0000] <Philip`>
(but then you probably wouldn't want it to just show nothing at all)

[05:10:00.0000] <zcorpan_>
indeed. so don't do m\:math { display:none; } ... ;)

[05:12:00.0000] <zcorpan_>
i don't know of any good fallback story for inline mathml

[05:13:00.0000] <zcorpan_>
you either use mathml or GIF or html tables

[05:31:00.0000] <hsivonen>
it is interesting how people are complaining on www-validator about the validator now being more correct with XML

[07:07:00.0000] <annevk>
/me notices replies to ancient e-mails in his mailbox, directed to fora⊙an :-o

[08:06:00.0000] <gsnedders>
annevk: you were referencing a script by Bert Bos when discussing the formatting of the spec (pointlessly at this stage), is it available anywhere?

[08:41:00.0000] <zcorpan_>
wow http://www.w3.org/TR/pronunciation-lexicon/

[09:04:00.0000] <annevk>
gsnedders, if you're a W3C Member

[09:12:00.0000] <Lachy>
Hixie was right, Ratatouille is really good movie :-)

[10:22:00.0000] <gsnedders>
Ratatouille isn't out here till Oct :(


2007-08-05
[06:57:00.0000] <smaug>
annevk: ping

[07:28:00.0000] <annevk>
pong

[07:28:01.0000] <annevk>
smaug, ...

[07:29:00.0000] <smaug>
annevk: about DOMActivate

[07:29:01.0000] <smaug>
annevk: according to html5 the default action is to dispatch DOMActivate

[07:30:00.0000] <smaug>
where does it say that only the first default action should occur?

[07:31:00.0000] <annevk>
the default action of some form of activation (Enter, clicking, etc.) is to dispatch a 'click' event

[07:31:01.0000] <annevk>
default action of that is to dispatch DOMActivate

[07:32:00.0000] <annevk>
the first element in tree order with activation behavior gets its activation behavior triggered at that point

[07:32:01.0000] <annevk>
reverse tree order*

[07:32:02.0000] <smaug>
annevk: but where does it say that only one DOMActivate gets activated

[07:33:00.0000] <annevk>
DOMActivate is not activated

[07:33:01.0000] <smaug>
I thought all default actions should be triggered

[07:33:02.0000] <annevk>
it's just dispatched

[07:33:03.0000] <annevk>
I would suggest to read http://www.whatwg.org/specs/web-apps/current-work/#interactive1 carefully although I suppose you have already done so... :)

[07:33:04.0000] <smaug>
annevk: I've read it carefully

[07:34:00.0000] <smaug>
how does an element know whether to dispatch DOMActivate or not?

[07:34:01.0000] <annevk>
that doesn't matter

[07:34:02.0000] <smaug>
why not?

[07:34:03.0000] <annevk>
click is dispatched of which the default action is to dispatch DOMActivate

[07:35:00.0000] <annevk>
regardless of which element it is

[07:35:01.0000] <annevk>
click will always have DOMActivate triggered

[07:35:02.0000] <smaug>
what is the target of the DOMActivate?

[07:36:00.0000] <annevk>
the same element as that of the click event

[07:36:01.0000] <annevk>
"must be to fire a further DOMActivate event at the same element"

[07:36:02.0000] <smaug>
<a>foo<button>bar</button></a>

[07:36:03.0000] <smaug>
or, <a>foo<span>clickme</span></a>

[07:36:04.0000] <smaug>
span is the target of the click

[07:36:05.0000] <smaug>
how does <a> know to dispatch DOMActivate?

[07:37:00.0000] <annevk>
DOMActivate would be dispatched on <span>

[07:37:01.0000] <annevk>
and the first element with activation behavior it bubbles through (in the first case, <button>, second case, <a>) will have its activation behavior triggered (submission, following the link)

[07:37:02.0000] <annevk>
it's pretty clear in the specification

[07:38:00.0000] <smaug>
it is not clear

[07:38:01.0000] <annevk>
just read it literally without any bias of what prior specifications might have suggested

[07:38:02.0000] <annevk>
ok, what part is not clear?

[07:40:00.0000] <smaug>
the chapter talks about "interactive" elements

[07:41:00.0000] <smaug>
should DOMActivate be dispatched whether or not click happens inside an interactive element?

[07:41:01.0000] <annevk>
yes

[07:42:00.0000] <smaug>
and it also talks about default action, but not where the default action should be handled

[07:42:01.0000] <annevk>
I suppose that could be made more clear

[07:42:02.0000] <annevk>
I'm not sure I understand that

[07:43:00.0000] <smaug>
all the elements may have default actions

[07:43:01.0000] <smaug>
and also document could have default action

[07:43:02.0000] <smaug>
and it doesn't matter whether or not the target of the event is the element (or document)

[07:44:00.0000] <annevk>
it just defines what the default action of click and DOMActivate is

[07:44:01.0000] <annevk>
for click -> dispatch DOMActivate; for DOMActivate -> activate the first element with activation behavior in reverse tree order

[07:45:00.0000] <smaug>
I understand activation behavior, since that is different than default action

[07:45:01.0000] <smaug>
but default action happens in all tree levels, in reverse tree order

[07:45:02.0000] <annevk>
what do you mean?

[07:46:00.0000] <smaug>
<a> has a default action for click

[07:46:01.0000] <smaug>
in gecko it is currently to dispatch DOMActivate

[07:46:02.0000] <annevk>
it overwrites all that

[07:47:00.0000] <annevk>
as I said, you have to interpret this literally, without bias

[07:47:01.0000] <smaug>
webkit has similar meaning for default actions

[07:47:02.0000] <smaug>
at least it used to have

[07:48:00.0000] <annevk>
for instance, the <a> element is defined to have activation behavior which would be triggered if anyone click on "foo" in the following cases <a>foo</a>, <a><span>foo</span></a>, <a><span>bar<span>foo</span></a>, etc.

[07:48:01.0000] <smaug>
hmm, is the meaning of default action defined anywhere..

[07:48:02.0000] <annevk>
DOM2Events?

[07:49:00.0000] <annevk>
anyway, this new concept comes much closer to what IE (and Opera) do and is a bit more useful for cases such as <a> foo <input type=submit></a>

[07:50:00.0000] <smaug>
DOM2Events doesn't define default actions properly

[07:50:01.0000] <smaug>
it doesn't say whether or not they are happening in node level or document level

[07:51:00.0000] <annevk>
http://www.w3.org/TR/DOM-Level-2-Events/events.html#Events-flow-cancelation

[07:51:01.0000] <smaug>
nope

[07:52:00.0000] <smaug>
that just says that event listeners must fire before default action

[07:52:01.0000] <annevk>
hmm, they are happening at node-level I'd say

[07:52:02.0000] <annevk>
not sure how you can test that and I'm also not sure how that's relevant here

[07:53:00.0000] <smaug>
but for all elements there is the default action for "click" event to dispatch DOMActivate if the target of the "click" is the current node

[07:53:01.0000] <smaug>
?

[07:53:02.0000] <annevk>
(well, I see that in Gecko you could get Multiple DOMActivate events now because of how <a>.onclick is handled right now...)

[07:54:00.0000] <annevk>
smaug, yes, the default action of a click event being dispatched is to dispatch a DOMActivate event

[07:54:01.0000] <annevk>
although not for synthetic click events aiui

[07:54:02.0000] <smaug>
but that happens only if the target of the element is the "current" element

[07:54:03.0000] <smaug>
there is after all some sort of posthandleevent phase

[07:55:00.0000] <smaug>
apparently in opera too

[07:55:01.0000] <smaug>
if default actions are node level

[07:55:02.0000] <annevk>
it doesn't happen during the bubbling phase, no

[07:55:03.0000] <annevk>
only when it's dispatched

[07:56:00.0000] <annevk>
I suppose it might be worthwhile to be a bit more explicit about such details in the draft; either in HTML 5 or DOM3Events

[08:00:00.0000] <smaug>
annevk: hmm, but does that meaning of default actions apply to all cases

[08:00:01.0000] <smaug>
how about xbl2?

[08:00:02.0000] <smaug>
at least it used to be possible to define own default actions

[08:01:00.0000] <smaug>
node level, and not depending on whether or not the current node is the target of the event

[08:02:00.0000] <annevk>
I suppose HTML 5 should just be clarified that the click event only does that when the current node is also the target node

[08:03:00.0000] <smaug>
yes, because that is a special case for a default action

[08:03:01.0000] <smaug>
xbl2 still has the way to define default actions

[08:03:02.0000] <smaug>
node level and triggered reversed tree order

[08:03:03.0000] <annevk>
Yeah, much like <html:form> has a default action for 'submit' too...

[08:04:00.0000] <smaug>
so html5 is not at all clear

[08:04:01.0000] <smaug>
it has special meaning for default action in one case

[08:05:00.0000] <smaug>
in gecko default actions happen always pretty much like in xbl2

[08:05:01.0000] <annevk>
what it says sort of implies the special case though

[08:05:02.0000] <smaug>
"implies" :)

[08:06:00.0000] <annevk>
by using "the element" etc. but it can be better :)

[08:07:00.0000] <annevk>
I'd suggest raising it on the list and implementing it as if it was more clear :)

[08:08:00.0000] <smaug>
or maybe I should come up some more reasonable solution for the problem

[08:08:01.0000] <smaug>
special casing default action handling sounds like a hack

[08:09:00.0000] <annevk>
the default action of the click event, if the target node is the same as the current node, is to dispatch a DOMActivate at the current node

[08:09:01.0000] <annevk>
seems reasonable to me

[08:10:00.0000] <smaug>
hmm, yes. that way

[08:10:01.0000] <smaug>
and it should be clear that it is not about interactive elements

[08:10:02.0000] <smaug>
but all elements

[08:10:03.0000] <smaug>
er, all nodes

[08:11:00.0000] <smaug>
(click may happen also on document, I think)

[08:11:01.0000] <annevk>
and for DOMActive you just have event listeners set up at the activation behavior elements and a special flag that tells you whether some element has its behavior triggered already

[08:11:02.0000] <annevk>
or some such

[08:11:03.0000] <smaug>
activation behavior is clear, sure

[08:12:00.0000] <annevk>
oh, ok, I thought there was some confusion about first in reverse tree order, but that might have been because of the multiple DOMActivate events...

[08:19:00.0000] <smaug>
(need to still think about how the click event's default action work with XBL, when there is retargeting happening...)

[08:20:00.0000] <annevk>
hah, have fun with that, while I take a break from this stuff :)

[13:06:00.0000] <hsivonen>
http://htmlpurifier.org/


2007-08-06
[04:22:00.0000] <ROBOd>
Hixie: you there?

[04:27:00.0000] <ROBOd>
i'd have a question, regarding the procedure on working on HTML 5. can proposals and reviews be made, in any order? obviously, dave hyatt and hixie do not work simultaneously on all the aspects of the spec. thus, i'd like to know if it's not a problem to send proposal/reviews for things that are not currently discussed on the mailing list. i suppose the editors "store" the emails in a (huge) "TODO folder", such that when they get to the sections of the spec,

[04:27:01.0000] <ROBOd>
 they also review all/most of the emails sent on the matter at hand. is that correct?

[04:31:00.0000] <hsivonen>
Hixie: btw, I think Karl is right when he calls two-tiered conformance levels "a form of versioning is coming back by the backdoor".

[04:31:01.0000] <hsivonen>
Hixie: I'd prefer to have one-tiered conformance with the one notion of conformance permissive enough to allow what authors want to do (including style='').

[04:33:00.0000] <ROBOd>
karl nailed that very well, conformance levels are a form of versioning . i think it's wrong to have different conformance levels

[04:45:00.0000] <gsnedders>
ROBOd: emails are sent about any and every section of the spec. there's no reason not to.

[04:47:00.0000] <ROBOd>
gsnedders: since i reviewed the "media element" section ... i'm again "into the video element", and i am a bit concerned i might "ressurect" hot debates about the video element (like when it was proposed)

[04:48:00.0000] <gsnedders>
ROBOd: you just have to accept that such a debate may come up, and that any constructive comments made will be taken into account

[04:49:00.0000] <ROBOd>
yeah, but still generally i do not want to cause/contribute hot debates ;)

[04:49:01.0000] <gsnedders>
they'll be caused whenever you mention such things :P

[04:49:02.0000] <ROBOd>
if you noticed or not, i seldom reply to other threads than mine

[04:50:00.0000] <gsnedders>
I just stay out of hot debates

[04:50:01.0000] <gsnedders>
:)

[04:51:00.0000] <ROBOd>
me too. but, but but ... what if *you want* to say something "hotly debateable"? :)

[04:52:00.0000] <gsnedders>
then just say it.

[04:52:01.0000] <gsnedders>
:P

[04:53:00.0000] <ROBOd>
i shall see


2007-08-07
[18:11:00.0000] <Hixie>
/me discovers Net::IMAP::Simple

[18:11:01.0000] <Hixie>
/me pokes at his IMAP mail box

[18:13:00.0000] <Hixie>
3741 e-mails in my whatwg folders

[18:31:00.0000] <Hixie>
the top five names still in my pile of e-mails to deal with are:

[18:31:01.0000] <Hixie>
Henri Sivonen (245)

[18:31:02.0000] <Hixie>
Jim Ley (144)

[18:31:03.0000] <Hixie>
Matthew Raymond (128)

[18:31:04.0000] <Hixie>
Anne van Kesteren (110)

[18:31:05.0000] <Hixie>
Lachlan Hunt (109)

[18:32:00.0000] <jruderman>
too bad you can't just off those people. kill one person and everyone thinks you're a murderer!

[18:33:00.0000] <Hixie>
what's amusing to me is that two of those five people haven't really posted much recently

[18:38:00.0000] <Hixie>
two most common subject lines:

[18:38:01.0000] <Hixie>
Re: [whatwg] Mathematics in HTML5 : 138

[18:38:02.0000] <Hixie>
Re: [whatwg] on codecs in a 'video' tag. : 74

[18:38:03.0000] <Hixie>
(remember that these are as yet unanswered e-mails, so they're heavily biased by what i've already replied to)

[02:46:00.0000] <zcorpan>
Philip`: did you implement microsyntaxes?

[02:52:00.0000] <hsivonen>
hmm. some Java APIs are like namespaces: can't remember them

[03:23:00.0000] <krijnh>
Why do spambots keep POSTing to 404 URIs :S

[03:32:00.0000] <virtuelv_>
krijnh: because they also keep POSTing to any URL they have decided

[03:32:01.0000] <virtuelv_>
I got 10000 POST requests to a document only strictly gettable

[03:35:00.0000] <krijnh>
Bah

[05:39:00.0000] <hsivonen>
hmm. I connected the Xalan serializer manually to Xalan and got ill-formed output (duplicate xmlns)

[06:00:00.0000] <Philip`>
zcorpan: I didn't - are you thinking of http://geoffers.no-ip.com/svn/php-html-5-direct/src/trunk/ ?

[06:04:00.0000] <zcorpan>
Philip`: hmm, perhaps

[06:48:00.0000] <zcorpan>
interesting that type="" is authorative on <embed> but ignored on <object>

[06:51:00.0000] <annevk>
I'd agree that the Web is interesting

[06:51:01.0000] <zcorpan>
is it required for compat?

[06:58:00.0000] <annevk>
zcorpan, maybe, at least when src= is not present

[06:59:00.0000] <annevk>
not sure it makes much sense trying to risk it for a useless element like <embed>

[06:59:01.0000] <zcorpan>
annevk: you lost me there

[07:05:00.0000] <zcorpan>
what happens with <param name="" value="foo">

[07:07:00.0000] <hsivonen>
I wonder what XHTML-to-XHTML XSLT transformations people tend to do

[07:07:01.0000] <hsivonen>
I now have only one real-world test case (DanC's task survey transformation)

[07:07:02.0000] <zcorpan>
(which is broken)

[07:08:00.0000] <hsivonen>
how?

[07:08:01.0000] <zcorpan>
it outputs elements in the null namespace

[07:08:02.0000] <zcorpan>
but works because it's served as text/html

[07:08:03.0000] <annevk>
zcorpan, <embed type> handling is required for compat when src= is not present (at least)

[07:08:04.0000] <hsivonen>
ooh. perhaps that's why I saw weirdness with Xalan

[07:08:05.0000] <annevk>
zcorpan, I'm not sure if changing browser behavior for <embed src type> is worth it

[07:09:00.0000] <hsivonen>
I already hacked a workaround that trashes xmlns attributes 

[07:09:01.0000] <hsivonen>
but

[07:09:02.0000] <hsivonen>
with the hack, Xalan gives me sane namespaces scopes.

[07:09:03.0000] <hsivonen>
hmm.

[07:09:04.0000] <zcorpan>
annevk: ok. (i'm not suggesting to change it)

[07:09:05.0000] <hsivonen>
anyway, I am still interested in examples of non-broken XHTML-to-XHTML XSLT transformations

[07:14:00.0000] <hsivonen>
zcorpan: are you sure that DanC's transformation is broken?

[07:14:01.0000] <zcorpan>
yes

[07:14:02.0000] <zcorpan>
<xsl:template match='h:div[h:h3/h:a/@id="xbgbio"]/h:table'>

[07:14:03.0000] <zcorpan>
  <ul>

[07:14:04.0000] <zcorpan>
no default namespace on the ul

[07:14:05.0000] <hsivonen>
oh, right that part

[07:15:00.0000] <hsivonen>
/me notes the non-use of id()

[07:15:01.0000] <hsivonen>
probably related to the IDness establishment probs

[07:18:00.0000] <hsivonen>
http://www.mnot.net/blog/2007/08/07/etags

[07:57:00.0000] <zcorpan>
<map ...><area><area .../> doesn't work in firefox

[08:03:00.0000] <annevk>
seems that only Firefox implies </address> at <table> and <ol>/<ul> ...

[08:03:01.0000] <zcorpan>
firefox implies </address>?

[08:04:00.0000] <zcorpan>
/me didn't know

[08:04:01.0000] <annevk>
I was wondering whether the <address> content model had legacy reasons

[08:04:02.0000] <annevk>
doesn't seem like it has in IE7 and O9

[08:15:00.0000] <zcorpan>
also div, address, dl, center...probably others too

[08:15:01.0000] <zcorpan>
blockquote

[08:16:00.0000] <zcorpan>
h1-h6

[08:18:00.0000] <zcorpan>
hr, li, menu, pre, fieldset, form

[08:19:00.0000] <zcorpan>
annevk: address does allow tables in address though

[08:19:01.0000] <zcorpan>
er

[08:19:02.0000] <zcorpan>
html5

[08:19:03.0000] <annevk>
sure

[08:19:04.0000] <annevk>
it would be silly otherwise

[08:20:00.0000] <zcorpan>
seems to be that Ben Boyle thought it didn't

[08:21:00.0000] <annevk>
oh, allowing it in the parser I meant

[08:21:01.0000] <annevk>
/me didn't check conformance

[08:23:00.0000] <Philip`>
Aha, I was wondering if Firefox treated anything else the same as <unknown> in terms of implying end tags, but couldn't think of any, but it looks like <address> is one

[08:26:00.0000] <zcorpan>
we should preserve wordings that makes readers smile! :)

[08:33:00.0000] <zcorpan>
hmm, ie doesn't support coords=default

[08:34:00.0000] <annevk>
yeah, known

[08:35:00.0000] <annevk>
I think IE parsing was not reverse engineered fully or something

[08:36:00.0000] <zcorpan>
parsing of what?

[08:37:00.0000] <zcorpan>
s/coords/shape/

[08:38:00.0000] <annevk>
coords, I believe it was not really a problem that IE didn't do shape=default

[08:39:00.0000] <zcorpan>
ok

[08:39:01.0000] <zcorpan>
/me needs to play more with image maps

[08:39:02.0000] <annevk>
I'd spend time on stuff that causes actual problems :)

[08:39:03.0000] <zcorpan>
such as

[08:40:00.0000] <annevk>
good question

[08:40:01.0000] <zcorpan>
seems we apply the areas in the wrong order

[08:40:02.0000] <annevk>
guess most of those sections are sort of covered, just awaiting implementation experience

[08:40:03.0000] <zcorpan>
yeah

[08:40:04.0000] <zcorpan>
i could write test cases for video

[08:41:00.0000] <zcorpan>
i've avoided that section so far

[08:42:00.0000] <annevk>
parsing of <meta http-equiv=refresh> ? :)

[08:45:00.0000] <zcorpan>
http://simon.html5.org/test/html/semantics/area/001.xht

[08:45:01.0000] <zcorpan>
yeah

[08:45:02.0000] <zcorpan>
or actually

[08:45:03.0000] <zcorpan>
http-equiv in general

[08:45:04.0000] <zcorpan>
it is implemented differently from what html5 says i think

[08:45:05.0000] <zcorpan>
browsers really treat the values as if they were http headers

[08:46:00.0000] <zcorpan>
and Refresh is supported as a real header

[08:46:01.0000] <annevk>
in some browsers, yes

[08:46:02.0000] <annevk>
the question is whether they can drop such support

[08:46:03.0000] <zcorpan>
in which browsers is that not so?

[08:47:00.0000] <annevk>
I thought Refresh was only supported by Gecko, actually

[08:47:01.0000] <zcorpan>
oh no

[08:47:02.0000] <zcorpan>
last time i checked i didn't find any browser not supporting it

[08:47:03.0000] <zcorpan>
but i might have done something wrong by then

[08:47:04.0000] <annevk>
interesting

[08:48:00.0000] <Philip`>
/me sees <meta http-equiv> used on more pages than <table>, but doesn't have any way to determine the attribute's values :-(

[08:49:00.0000] <zcorpan>
/me assumes content-type and content-language are the 2 top

[09:42:00.0000] <krijnh>
Perhaps set-cookie

[09:48:00.0000] <met_>
hsivonen: http://about.validator.nu/author/ from the footer gives 404

[14:14:00.0000] <gsnedders>
is there any content that relies on browsers treating HTTP headers ending in "CRLF as the value including the quote?

[14:25:00.0000] <gsnedders>
this is something browsers do unusually (for HTTP) consistently

[15:45:00.0000] <Hixie>
here's an example of a page that blatently shouldn't validate: http://gallery.mac.com/emily_parker#100335&bgcolor=black

[15:46:00.0000] <Hixie>
and not for the reasons the validator currently gives

[15:46:01.0000] <Hixie>
disable stylesheets on that page to see what i mean

[15:47:00.0000] <Hixie>
hsivonen: http://html5.validator.nu/?doc=http%3A%2F%2Fgallery.mac.com%2Femily_parker%23100335%26bgcolor%3Dblack ?

[15:49:00.0000] <gavin_>
wow

[15:50:00.0000] <Hixie>
hsivonen: btw i've added a form to validator.whatwg.org so that people can submit URIs from there instead of having one more page to go through

[15:50:01.0000] <Hixie>
hsivonen: let me know if that's a problem

[15:54:00.0000] <hsivonen>
Hixie: apparently & is not allowed in the fragment identifier so the IRI parser barfs

[15:54:01.0000] <hsivonen>
Hixie: the form on validator.whatwg.org is ok

[15:55:00.0000] <Hixie>
interesting about &

[15:56:00.0000] <Philip`>
Is <base target="_self"> meant to do anything at all useful? (If it does, I can't work out what it is, but TinyMCE appears to use it quite frequently)

[16:01:00.0000] <hsivonen>
Hixie: hmm. perhaps it isn't the IRI parser at all. but the HTTP lib doesn't like getting any fragment ids

[16:01:01.0000] <hsivonen>
I guess I should use the IRI lib to zap the fragment id

[16:03:00.0000] <Hixie>
yeah you shouldn't send fragids over the wire

[16:03:01.0000] <Hixie>
maybe the http lib just doesn't know to drop them :-)

[16:04:00.0000] <Hixie>
Philip`: it probably creates a new window called _self and then targets that a lot - unless something first did window.name = "_self_ ?


2007-08-08
[20:20:00.0000] <Hixie>
anyone have examples of pages that have custom context menus?

[20:20:01.0000] <Hixie>
like google maps or live search maps

[21:29:00.0000] <gavin_>
the zimbra web ui uses custom context menus

[22:15:00.0000] <Hixie>
so anyone know what's up with anne's site?

[22:38:00.0000] <Hixie>
/me is more and more convinced that we should only allow integers for <img height/width> and they should match the image coordinates

[23:10:00.0000] <Hixie>
othermaciej: (in case you have any influence over the web gallery / iphoto / .mac folks, see http://gallery.mac.com/emily_parker#100335&bgcolor=black with stylesheets disabled)

[23:10:01.0000] <Hixie>
othermaciej: (that page is an example of why i don't like inlines-in-<div>s and style="" being considered 1st class citizens)

[23:11:00.0000] <othermaciej>
Hixie: I don't like the div soup design antipattern either

[23:11:01.0000] <othermaciej>
what iWeb generates is even worse

[23:11:02.0000] <Hixie>
in other news, <input type=image usemap="#..."> only seems to work in opera and firefox, not in safari and IE

[23:11:03.0000] <Hixie>
i wonder whether we should support it or not

[23:11:04.0000] <Hixie>
othermaciej: yikes

[23:12:00.0000] <othermaciej>
there output is full of, I kid you not, <div class="paragraph">

[23:12:01.0000] <Hixie>
hah

[23:17:00.0000] <Hixie>
according to some old data I have, 17% of usemap="" attributes on <img> elements have the value "#Map"

[23:18:00.0000] <Hixie>
not original, but it works, i guess

[23:19:00.0000] <Hixie>
roughly 0.008% of <input type=image> elements have usemap="" attributes

[23:19:01.0000] <Hixie>
i wonder how many have useful values

[23:20:00.0000] <Hixie>
(it's less common than <input price="">, <input heigth="">, <input typ="">, <input enable="">, <input 1px="">, etc)

[23:22:00.0000] <othermaciej>
price=""?

[23:22:01.0000] <Hixie>
dunno

[23:22:02.0000] <Hixie>
(see topic)

[02:56:00.0000] <zcorpan>
"[ac] (1) Make id='' required on <map>. Let me know if that's a problem."

[02:56:01.0000] <zcorpan>
Hixie: i've suggested to use name="" instead

[02:58:00.0000] <zcorpan>
http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-April/010975.html

[04:07:00.0000] <hsivonen>
Hixie: fwiw, any time a common attribute becomes required on a particular element, it is an inconvenience for me. (though this probably isn't a reason not to require them, but anyway)

[04:07:01.0000] <annevk>
you shouldn't say that :)

[04:08:00.0000] <hsivonen>
annevk: say what? the part in parentheses or before?

[04:08:01.0000] <annevk>
the whole sentence

[04:09:00.0000] <annevk>
anyway, isn't it quite easy to add required attributes using RNG?

[04:10:00.0000] <hsivonen>
annevk: it sucks when you can't use common attributes but have to rebuild them on a per-element basis without the one you need to require and so that you can make the one required

[04:10:01.0000] <hsivonen>
annevk: this is why the grammar for bdo is ugly

[04:11:00.0000] <annevk>
oh, you can't override global stuff on a per element basis?

[04:11:01.0000] <annevk>
annoying

[04:11:02.0000] <hsivonen>
annevk: very

[04:12:00.0000] <annevk>
yeah, that actually sucks a lot given the amount of attributes we have now

[04:12:01.0000] <zcorpan>
hsivonen: would it help you if the attribute was called name instead of id? :)

[04:12:02.0000] <annevk>
:p

[04:12:03.0000] <hsivonen>
zcorpan: yes :-)

[04:12:04.0000] <zcorpan>
well then :)

[04:12:05.0000] <annevk>
terrible

[04:13:00.0000] <hsivonen>
annevk: at least this is my recollection. there's a small chance that I am misremembering something or that my mind is clouded by the DTD compat extension

[04:19:00.0000] <hsivonen>
annevk: my recollection was correct. 

[04:19:01.0000] <hsivonen>
http://validator.nu/?doc=http%3A%2F%2Fabout.validator.nu%2F&schema=http%3A%2F%2Fhsivonen.iki.fi%2Ftest%2Fduplicate-attr.rnc

[04:20:00.0000] <hsivonen>
that is, schemas that can derive duplicate attributes are forbidden. 

[04:20:01.0000] <hsivonen>
this sucks pretty badly compared to embracing grammar ambiguity

[04:21:00.0000] <annevk>
you should raise this (and other issues with RNG) on some list so it can be improved

[04:22:00.0000] <hsivonen>
I suppose

[04:23:00.0000] <annevk>
i bet some geeks will find it interesting and fix it :)

[04:24:00.0000] <hsivonen>
actually, now that James Clark has moved on from hacking on Jing and RNG is an ISO standard, I'm pretty pessimistic when it comes to improving RNG

[04:30:00.0000] <annevk>
whoa, Elliotte Harold sort of turned around it seems :)

[04:30:01.0000] <hsivonen>
eww. the feedback mechanism for RNG is overlawyered

[04:30:02.0000] <annevk>
http://www-128.ibm.com/developerworks/library/x-html5/

[04:32:00.0000] <hsivonen>
It totally sucks that these days the patent system is so sick that people don't want to hear from you until you jump through the hoops of a patent covenant

[04:33:00.0000] <hsivonen>
too much legal stuff to read for today if I wanted to use the official channel for saying something

[04:34:00.0000] <hsivonen>
I'm going to ping an unofficial channel first to see if there's a reason why the suggestion wouldn't work anyway

[04:47:00.0000] <hsivonen>
Hixie: btw, the same rationale that was used for allowing xmlns on root and /> could be used for allowing old-style meta charset syntax

[04:53:00.0000] <hsivonen>
whoa. Elliotte Harold (or a copy editor) takes the risk of taking a stand on what RSS abbreviates :-)

[05:27:00.0000] <zcorpan>
http://niquimerret.com/?p=94

[05:32:00.0000] <zcorpan>
what's the use case for HTMLMapElement.images?

[05:33:00.0000] <annevk>
good question

[05:33:01.0000] <annevk>
I had one

[06:57:00.0000] <zcorpan>
"Each remaining area element in areas represents a hyperlink. Those hyperlinks should all be made available to the user in a manner associated with the text of the img or input element." -- why is this not a MUST?

[07:24:00.0000] <zcorpan>
wonder if we should support showing the <area> links when <object> is used and images are disabled

[07:25:00.0000] <met_>
https://bugzilla.mozilla.org/show_bug.cgi?id=382267#c15

[07:28:00.0000] <zcorpan>
<video> doesn't have fallback content in html5-aware UAs

[07:28:01.0000] <zcorpan>
alternative video formats are covered with <source>

[07:28:02.0000] <annevk>
the videostream itself needs to be accessible

[07:29:00.0000] <zcorpan>
yeah

[07:29:01.0000] <zcorpan>
non-video alternatives would have to be presented alongside the video

[07:29:02.0000] <zcorpan>
("alternatives")

[07:33:00.0000] <annevk>
can someone tell me what the effect of <input type=image usemap> is?

[07:34:00.0000] <annevk>
/me goes to test it

[07:35:00.0000] <annevk>
it seems harmful

[07:35:01.0000] <zcorpan>
same as <img usemap> afaict (as long as it is actually associated with a map)

[07:36:00.0000] <annevk>
yeah, which makes no sense at all for a control

[07:36:01.0000] <zcorpan>
agreed

[07:36:02.0000] <annevk>
maybe to make particular areas unclickable?

[07:38:00.0000] <annevk>
i think i rather make <input> less bloated than catering for that silly usecase

[07:38:01.0000] <zcorpan>
ah, clicking outside an area will submit the form as normal

[08:03:00.0000] <annevk>
"I agree with Anne for once!"

[08:03:01.0000] <annevk>
/me ponders

[13:05:00.0000] <Wolfman2000>
I know that HTML 5 is only in a working draft right now, but does anyone know which browsers already support it?

[13:06:00.0000] <annevk>
http://wiki.whatwg.org/wiki/Implementations_in_Web_browsers

[13:07:00.0000] <annevk>
maybe the WHATWG FAQ should point to that...

[13:07:01.0000] <Wolfman2000>
In other words, not a lot.

[13:10:00.0000] <annevk>
depending on how you look at it, sure

[13:11:00.0000] <annevk>
(that page doesn't list all the features from HTML4 that are also part of HTML5 it seems which are obviously implemented too)

[13:11:01.0000] <Wolfman2000>
The only tag that I see as implemented on that page is <canvas>

[13:18:00.0000] <Philip`>
If you're just counting tags, there's <event-source> too

[13:18:01.0000] <Philip`>
(and <video> if you count experimental implementations)

[14:38:00.0000] <Hixie>
hmm

[14:43:00.0000] <hsivonen>
Hixie: did you notice the accessibility comment on the Mozilla <video> bug? the spec is silent on captions

[14:45:00.0000] <Hixie>
uri?

[14:45:01.0000] <hsivonen>
Hixie: https://bugzilla.mozilla.org/show_bug.cgi?id=382267#c15

[14:46:00.0000] <Hixie>
/me grumbles about people who think of accessibility as a separate concern from feature design

[14:46:01.0000] <Hixie>
it's like people who bolt on security afterwards

[14:46:02.0000] <hsivonen>
what browsers do people use for browsing the spec when they complain about locking? WFM in Firefox on OS X and in Opera on Maemo

[14:46:03.0000] <Hixie>
or who think "performance" is a separate module

[14:47:00.0000] <hsivonen>
Hixie: I'd expect Ogg <video> to use in-Ogg Writ captions.

[14:47:01.0000] <Hixie>
yeah, same with any video stream

[14:47:02.0000] <Hixie>
it's not an all-or-nothing deal

[14:47:03.0000] <hsivonen>
Hixie: but the spec is silent on how to enable/disable captions of pick a language in case of multiple subtitles, etc.

[14:53:00.0000] <Hixie>
hsivonen: isn't that a UA issue?

[14:53:01.0000] <Hixie>
if we rely on the page providing that, i predict many "inaccessible" pages

[14:54:00.0000] <hsivonen>
Hixie: good point.

[14:54:01.0000] <hsivonen>
Hixie: perhaps the spec should say this

[14:54:02.0000] <Hixie>
it will, in due course

[15:01:00.0000] <Hixie>
can anyone work out what chris meant in today's "RE: [whatwg] Where did the "rev" attribute go?"

[15:02:00.0000] <annevk>
/me skipped those e-mails

[15:03:00.0000] <annevk>
I guess he means that a hyphen "means" a space, but I'm not sure what the point is

[15:03:01.0000] <Hixie>
wow, usemap on <input type=image> is used

[15:03:02.0000] <hsivonen>
Hixie: looks like a case of getting semantics from reading the class name in English

[15:04:00.0000] <annevk>
Hixie, really? What's used for Internet Explorer?

[15:04:01.0000] <hsivonen>
someone really should write an article that explains why that is a fallacy

[15:04:02.0000] <Hixie>
i guess it just doesn't do anything in IE

[15:04:03.0000] <Hixie>
hsivonen: i didn't understand the e-mail at all

[15:07:00.0000] <hsivonen>
Hixie: I had a flashback of http://tantek.com/log/2002/08.html#L20020821t2114 again

[15:08:00.0000] <Hixie>
annevk: http://shop.divebooty.com/78.html is one example

[15:09:00.0000] <annevk>
Hixie, there is nothing matching "searchmap" (the usemap value) in that page though

[15:09:01.0000] <tantek>
hsivonen - heh

[15:10:00.0000] <Hixie>
annevk: it's right at the top

[15:12:00.0000] <annevk>
hmm, doesn't work in Opera if you click with a mouse, if you tab to it and hit enter it does...

[15:12:01.0000] <annevk>
removing the <map> makes the page work better in Opera...

[15:12:02.0000] <Hixie>
i love my new imap-fu

[15:13:00.0000] <Hixie>
annevk: i'm gonna get a more complete list of pages that actually have differences between IE and other browsers

[15:13:01.0000] <annevk>
this has differences

[15:13:02.0000] <annevk>
it works less good in Opera and Firefox

[15:13:03.0000] <Hixie>
yeah, but others in the list i got don't

[15:13:04.0000] <Hixie>
(don't have differences, i mean)

[15:14:00.0000] <Hixie>
so i need to get a better sample list

[15:14:01.0000] <annevk>
ok

[15:14:02.0000] <Hixie>
e.g. some have usemap="" pointing to non-existent maps

[15:14:03.0000] <annevk>
why does that not surprise me :)

[15:14:04.0000] <Hixie>
what surprises me is that some have usemap=""s pointing to actual maps! :-)

[15:20:00.0000] <annevk>
seems IE might be Unicode case-insensitive for ID values (related to <map> anyway, and at least for the character ë)

[15:20:01.0000] <Hixie>
ew

[15:21:00.0000] <annevk>
(same for name=)

[15:22:00.0000] <Hixie>
noted

[15:25:00.0000] <annevk>
seems the developerworks article initiated a few new blogposts

[15:28:00.0000] <Hixie>
hey anne what's the story with your site?

[15:28:01.0000] <Hixie>
did it change url?

[15:34:00.0000] <annevk>
not really

[15:34:01.0000] <annevk>
oh, you mean this XML5 thing?

[15:34:02.0000] <Hixie>
http://annevankesteren.nl/ hasn't responded for me for some time

[15:34:03.0000] <annevk>
oh ok

[15:35:00.0000] <annevk>
dunno, I've yet to switch it to dreamhost, which might not be better, but at least I get ssh access

[15:35:01.0000] <Hixie>
oh so it's just down?

[15:40:00.0000] <Philip`>
Is http://www.whatwg.org/specs/web-apps/current-work/#audio0 likely to become more specific about numbers of channels and sample rates and bits per sample, or is it expected that people can support any crazy wave file?

[15:40:01.0000] <Hixie>
probably neither

[15:41:00.0000] <Hixie>
though once we have more implementation experience it might describe the common supported subset

[15:42:00.0000] <Hixie>
feel free to mail suggestions to the list for what it should require, though

[15:42:01.0000] <Hixie>
(the minimum supported subset)


2007-08-09
[18:01:00.0000] <Hixie>
any opera people here?

[18:03:00.0000] <karlUshi>
which nationality?

[18:03:01.0000] <karlUshi>
http://en.wikipedia.org/wiki/Category:Opera_singers_by_nationality

[18:04:00.0000] <Hixie>
...

[18:04:01.0000] <Hixie>
any opera software employees here?

[18:12:00.0000] <Dashiva>
Hixie: Temporarily, at least

[18:13:00.0000] <Hixie>
hey

[18:14:00.0000] <Hixie>
do you know what the offical Opera opinion is of moving postMessage from Document to Window?

[18:14:01.0000] <Hixie>
is it something you guys can do without breaking too much content? would you have to support both for a while? is that ok?

[18:17:00.0000] <Dashiva>
Let's see...

[18:18:00.0000] <Dashiva>
Last I heard it was "window might be nicer, but not breaking the API is probably better" 

[18:20:00.0000] <Dashiva>
You'll have to poke anne (I think) for anything official

[18:20:01.0000] <Hixie>
ok

[18:20:02.0000] <Hixie>
thanks

[18:20:03.0000] <Hixie>
i think i'm gonna break the API, in case you can poke people internally about this to make them aware of it

[18:22:00.0000] <Dashiva>
I'll see what I can do. Not sure what the official Opera policy on shooting messengers is ;)

[18:22:01.0000] <Hixie>
:-)

[19:01:00.0000] <Hixie>
<!DOCTYPE html> a<ruby> b <rt>rt1</rt> <rt>rt2</rt> </ruby>c

[19:01:01.0000] <Hixie>
and

[19:01:02.0000] <Hixie>
<!DOCTYPE html> a<ruby> b <rt>rt1</rt><rt>rt2</rt> </ruby>c

[19:01:03.0000] <Hixie>
render very differently in IE

[19:01:04.0000] <Hixie>
grr

[19:11:00.0000] <karlUshi>
judt because of the space?

[19:12:00.0000] <karlUshi>
I think it would be a good opportunity to include Richard Ishida in the discussion. He's on the list. 

[19:14:00.0000] <Hixie>
it's not really a ruby problem, ironically

[19:14:01.0000] <Hixie>
it's an IE parsing problem

[19:15:00.0000] <Hixie>
knowledge of ironically ruby doesn't really help with speccing what to do with parsing ruby :-(

[19:15:01.0000] <Hixie>
er

[19:15:02.0000] <Hixie>
knowledge of ruby ironically doesn't really help, even

[19:18:00.0000] <Hixie>
i wonder if we should make <ruby> a formatting element

[19:19:00.0000] <Hixie>
it's the only way i can see to handle <ruby>a<rt>b<p>c</ruby>d

[19:19:01.0000] <Hixie>
in a way that resembles IE

[19:19:02.0000] <Hixie>
though maybe the solution is to not resemble IE

[19:44:00.0000] <Lachy>
http://www.ibm.com/developerworks/library/x-html5/?ca=dgr-lnxw01NewHTML

[19:44:01.0000] <Lachy>
http://developers.slashdot.org/article.pl?sid=07/08/08/1558225

[20:14:00.0000] <karlUshi>
first comment "And here I was thinking that solved all of my web design problems. Now I might have to learn a second type of tag!"

[20:14:01.0000] <karlUshi>
:)

[20:29:00.0000] <deltab>
that comment as intended: "And here I was thinking that <pre> solved all of my web design problems. Now I might have to learn a second type of tag!"

[23:29:00.0000] <Hixie>
http://junkyard.damowmow.com/287 comes from the new study

[23:29:01.0000] <Hixie>
(about usemap)

[23:31:00.0000] <Hixie>
i think we can throw it out safely

[23:32:00.0000] <Hixie>
but if someone wants to look at how those pages use usemap="" on <input> to see if any of them use it in a way that matters, that'd be great

[01:11:00.0000] <Lachy>
Hixie, the first 6 results from that study don't use the map for anything useful at all. I don't expect to find any that do if I keep looking

[01:20:00.0000] <Hixie>
all the pages have an <input> with type=image with a usemap="" that points to a <map> with a name or ID that matches

[01:20:01.0000] <Hixie>
i didn't check if the <map> had <area> elements

[01:21:00.0000] <Hixie>
lachy: in what way were they useless?

[01:22:00.0000] <Hixie>
hm, gotta go (i'm amused that people randomly twitter single words on whatwg.org btw)

[01:22:01.0000] <Lachy>
some had no <area> elements, some had them without href attributes and others had <area ... href=""> or <area ... href="#">

[01:39:00.0000] <annevk>
Hixie, we don't want to break the API I think

[01:40:00.0000] <annevk>
Hixie, there was this bug report you commented on relevant to that

[04:08:00.0000] <zcorpan>
http://simon.html5.org/test/html/semantics/image-maps/ 003..004 confirms that id/name with usemap is unicode case insensitive

[04:09:00.0000] <zcorpan>
in all 4 browsers

[04:09:01.0000] <zcorpan>
(except in xhtml in fx and saf, where it is case sensitive)

[04:11:00.0000] <zcorpan>
Hixie: ^

[10:24:00.0000] <zcorpan>
Lachy: is http://lachy.id.au/dev/presentation/developing-with-html5/ announced anywhere? can it be added to the wiki?

[10:59:00.0000] <annevk>
http://www.prochoiceforum.org.uk/comm47.asp is another page which is worse in browsers that support <input usemap>

[11:00:00.0000] <Hixie>
hmm

[11:01:00.0000] <Hixie>
guess we'll remove it

[11:01:01.0000] <annevk>
http://www.soe.uwm.edu/pages/welcome/Certification_and_Degrees too

[11:01:02.0000] <annevk>
(if you click advanced search you go to the normal search page except that it doesn't submit any data you typed in)

[11:02:00.0000] <annevk>
although I think that site actually had the intention of making it useful

[11:03:00.0000] <annevk>
http://www.universaldizayn.com/irtibat.php login breaks in Opera and Firefox

[11:04:00.0000] <Hixie>
i'm gonna do the study again but only get pages that have <map>s with <area>s

[11:04:01.0000] <Hixie>
which themselves have both coords="" and href="" and their href="" isn't "" or "#"

[11:04:02.0000] <annevk>
these all had <area>

[11:04:03.0000] <annevk>
oh

[11:04:04.0000] <Hixie>
anything else i should filter out?

[11:05:00.0000] <annevk>
I just looked at those three so far

[11:05:01.0000] <annevk>
I already filed a bug for removal...

[11:05:02.0000] <Hixie>
yeah i'm pretty sure we want to remove it

[11:05:03.0000] <Hixie>
i just want to make sure

[11:16:00.0000] <annevk>
http://www.ljagri.gov.cn/nykj_syjs.asp also has the # problem

[11:17:00.0000] <Hixie>
yeah that one wouldn't be caught in the new survey i just launched

[11:20:00.0000] <annevk>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%0A%3Ciframe%20src%3Dhttp%3A//www.opera.com/%3E%3C/iframe%3E%0Axx%3Cscript%3E%20document.getElementsByTagName%28%22iframe%22%29%5B0%5D.contentDocument.location%20%3D%20%22http%3A//google.com%22%20%3C/script%3E

[11:20:01.0000] <annevk>
is one reason why postMessage might just as well stay on Document

[11:20:02.0000] <annevk>
oh, interesting, it only seems to work in Firefox

[11:20:03.0000] <annevk>
/me ponders

[11:21:00.0000] <Hixie>
yeah document shouldn't be accessible cross-origin

[11:21:01.0000] <annevk>
I thought they were the ones having issues with this document being accessible?

[11:21:02.0000] <annevk>
We'd like to keep postMessage() as is, if possible...

[11:21:03.0000] <Hixie>
no they were the ones who said that they had too many security problems

[11:21:04.0000] <Hixie>
oh?

[11:21:05.0000] <Hixie>
why?

[11:22:00.0000] <annevk>
see pm with link to internal bug report

[11:22:01.0000] <annevk>
basically, jl doesn't really see a good reason to break the API

[11:25:00.0000] <annevk>
btw, would be nice if Live DOM Viewer had something like $(id) and $$(tagname) in the global scope or something similarly named

[11:41:00.0000] <annevk>
oops, made a mistake

[11:41:01.0000] <annevk>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3Ciframe%20src%3Dhttp%3A//www.opera.com/%3E%3C/iframe%3E%3Cscript%3Eonload%3Dfunction%28%29%7Bdocument.getElementsByTagName%28%22iframe%22%29%5B0%5D.contentDocument.location%20%3D%20%22http%3A//google.com/%22%7D%3C/script%3E%20

[11:41:02.0000] <annevk>
does make the redirect in Opera

[11:43:00.0000] <annevk>
lol, that variant throws an exception in Firefox

[11:44:00.0000] <zcorpan>
ie7 as well (when you change contentDocument to contentWindow.document)

[11:46:00.0000] <Lachy>
zcorpan: sure, you can add it to the wiki. I will be blogging about it shortly too

[11:56:00.0000] <Hixie>
/me slowly builds the most elaborate cross-domain attempt he has ever made

[11:57:00.0000] <zcorpan>
Lachy: done

[12:03:00.0000] <Hixie>
http://software.hixie.ch/utilities/js/live-dom-viewer/?<!DOCTYPE html><body onload%3D"setTimeout(test1%2C 100)%3B setTimeout(test2%2C 200)%3B">...<iframe src%3D"http%3A//labs.google.com"></iframe><script>function test1() {w('contentDocument%3A')%3Bvar y %3D document.getElementsByTagName('iframe')[0]%3Bw(y)%3Bw(y.contentDocument)%3Bw(y.contentDocument.write)%3By.contentDocument.write('test')%3B}</script><script>function test2() {w('contentWindow%3A')%3Bvar x %3D do

[12:03:01.0000] <Hixie>
(did that truncate?)

[12:03:02.0000] <Hixie>
anyway my conclusion is that only Opera lets you iterate over the Document, and only Firefox lets you access document.write()

[12:06:00.0000] <Hixie>
actually i can't even call document.write() in firefox

[12:08:00.0000] <Hixie>
wait that wasn't the right test

[12:10:00.0000] <Hixie>
annevk: i commented on the bug

[12:10:01.0000] <Hixie>
http://junkyard.damowmow.com/288 is my "testcase"

[12:25:00.0000] <Hixie>
i can't work out how to make <ruby> work

[12:26:00.0000] <Hixie>
specifically, the parsing of <ruby> in IE is not compatible with a non-tree DOM

[12:26:01.0000] <Hixie>
i wonder how much ruby is out there

[12:27:00.0000] <Hixie>
hm, not much

[12:27:01.0000] <Hixie>
it's not in the top 200

[13:09:00.0000] <Hixie>
woot, my first official google blog post - http://google-code-updates.blogspot.com/2007/08/optimisation-data-for-html5-parser.html

[13:09:01.0000] <Hixie>
(also on code.google.com)

[14:16:00.0000] <Hixie>
/me launches a study of <ruby> parsing

[14:17:00.0000] <Hixie>
i don't understand how people wrote specs before

[14:18:00.0000] <hsivonen>
Hixie: have you taken a look at the ODF spec?

[14:18:01.0000] <Hixie>
no

[14:18:02.0000] <Hixie>
should i?

[14:18:03.0000] <hsivonen>
well, it certainly is very different from HTML 5.

[14:18:04.0000] <Hixie>
heh

[14:19:00.0000] <Hixie>
/me starts looking at adding URIs and titles to pushState()

[14:19:01.0000] <hsivonen>
it is basically a commentary on the schema

[14:19:02.0000] <Hixie>
eh, that's a pretty common way of writing specs

[14:19:03.0000] <Hixie>
HTML4 is pretty much that

[14:20:00.0000] <hsivonen>
not on that level

[14:20:01.0000] <Hixie>
scary

[14:22:00.0000] <hsivonen>
anyway, as far as spec writing goes, this processing model thing as well as error handling seem to be novel :-/

[14:23:00.0000] <hsivonen>
OOXML has detail, but more like a DoS of detail. I haven't figured out yet, if it is useful and sufficient detail

[14:23:01.0000] <Hixie>
but how do you write a "version 2" without being able to study the existing content and examine the cowpaths?

[14:25:00.0000] <hsivonen>
you take the next version of MS Office and figure out what new data structures there are? :-)

[14:25:01.0000] <Hixie>
on another note, i have this small slice of the web that i use for testing my scripts before i run them on the real data, and every time i do a test run over that data my parser skips over these same pages, which it outputs the url of to the console

[14:25:02.0000] <Hixie>
one of them is http://www.online-dating-facts.net/

[14:26:00.0000] <Hixie>
which has so many copies of "<b>Warning</b>:  feof(): supplied argument is not a valid stream resource in <b>/home/online/public_html/inc/articles.inc</b> on line <b>4</b><br />" that my script runs out of bits to store elements in

[14:27:00.0000] <Hixie>
it may in fact be an infinite page, i can't quite tell

[14:27:01.0000] <Hixie>
i don't think i've ever reached the end of it

[14:27:02.0000] <Hixie>
(my test data file has a truncated copy of it)

[14:29:00.0000] <Hixie>
anyone have an opinion on whether pushState() should always require a URI or whether we should continue to allow state to be included without a URI?

[14:45:00.0000] <KevinMarks>
Afternoon

[14:46:00.0000] <KevinMarks>
having a bit fo trouble following the rel="bookmark" stuff 

[14:46:01.0000] <KevinMarks>
http://www.whatwg.org/specs/web-apps/current-work/#link-type2

[14:47:00.0000] <Hixie>
what don't you follow?

[14:50:00.0000] <KevinMarks>
the implications of sections - if there are no <article> elements, the rel="bookmark"  only applies to a subset fo the document?

[14:51:00.0000] <KevinMarks>
or can the bookmark link apply to the whole page?

[14:52:00.0000] <KevinMarks>
concrete example, go to google maps, and do a search. The 'link to this page' url has rel="bookmark" on it, which is intended to mean 'use this for the page permalink, not the top url as the content changed"

[14:52:01.0000] <Hixie>
it applies to the nearest <article>, or if there are no such ancestors, to the nearest <body>, <section>, <nav>, <blockquote>, or <aside> element

[14:53:00.0000] <Hixie>
whichever comes first

[14:55:00.0000] <KevinMarks>
that 'section' discovery is tricky

[14:55:01.0000] <Hixie>
you mean working out which element is a section?

[14:57:00.0000] <KevinMarks>
what seems 'natural' to me is to flood-fill outwards from each rel="bookmark" link  until you hit another's boundary

[14:58:00.0000] <KevinMarks>
is the intent of that code to not 'leak' out from the subsections to the filler?

[14:59:00.0000] <Hixie>
i don't recall what the thinking was

[14:59:01.0000] <Hixie>
i think it was to make it possible to mark blog posts primarily

[14:59:02.0000] <KevinMarks>
right

[14:59:03.0000] <Hixie>
and then when you're not doing a post, to mark sections and stuff

[15:00:00.0000] <Hixie>
but maybe we should drop that altogether

[15:00:01.0000] <Hixie>
and make it the nearest <article> or <body>

[15:00:02.0000] <KevinMarks>
no, broadly it makes sense

[15:00:03.0000] <Hixie>
i'm not sure i really see the point of permalinking to a section really

[15:00:04.0000] <Hixie>
i recommend sending feedback to the list (whatwg⊙wo) or me (ian⊙hc) so i look at it more closely, if you want it changed at all

[15:00:05.0000] <KevinMarks>
it is a huge usecase in blogs

[15:01:00.0000] <Hixie>
certainly <article> should definitely scope permalinks

[15:01:01.0000] <Hixie>
i'm talking mainly about <section> et al

[15:01:02.0000] <KevinMarks>
it's the independence of the sction parsing and the bookmark linking that confuses me a bit

[15:01:03.0000] <Hixie>
i don't see a real use case to scope to them

[15:01:04.0000] <Hixie>
i don't understand what you mean by "section parsing"

[15:02:00.0000] <KevinMarks>
http://www.whatwg.org/specs/web-apps/current-work/#outlines

[15:02:01.0000] <Hixie>
oh outlining

[15:03:00.0000] <KevinMarks>
wondering if anyone has implemented that yet - is that part of Sam's html5lib?

[15:03:01.0000] <Hixie>
i don't know of any implementation work short of the small script in the spec itself

[15:04:00.0000] <KevinMarks>
OK, thanks

[16:20:00.0000] <Hixie>
preliminary results suggest we can get away with making <ruby> parse like phrasing elements! sweet

[16:20:01.0000] <Hixie>
let's hope it's representative

[16:29:00.0000] <Hixie>
Ok I can only find about 500 sites (about 12000 pages) that use usemap="" on <input> in a way involving <area> elements that have actual href=""s with possibly useful values

[16:43:00.0000] <Hixie>
continuing the usemap=" saga -- http://junkyard.damowmow.com/289 is the new survey data

[16:43:01.0000] <Hixie>
in case people want to look at them


2007-08-10
[17:14:00.0000] <Hixie>
anyone have an opinion on whether pushState() should take an absolute URI which the UA must then verify, or whether we should require that it be frefixed either with ? or with # and that it then replaces either the query part or the hash part?

[17:17:00.0000] <othermaciej>
Hixie: I think it should accept either relative or absolute and apply appropriate constraints to the resulting resolved URI

[17:17:01.0000] <othermaciej>
Hixie: a lot of "REST" style sites put the interesting info in path segments of the URI, not in a query or fragment ID

[17:18:00.0000] <Hixie>
we could allow the path part to change too, i guess

[17:19:00.0000] <Hixie>
so you're saying take the URI, resolve it relative to whatever it is you resolve URIs in scripts to, and then check the protocol/server/port are the same

[17:20:00.0000] <othermaciej>
yes, that would be my suggestion

[17:20:01.0000] <Hixie>
k

[17:21:00.0000] <Hixie>
sweet, my ilife '08 package arrived

[17:21:01.0000] <Hixie>
now go go el wire package!

[17:34:00.0000] <kingryan>
/me finds it interesting that http://www.whatwg.org/specs/web-apps/current-work/#outlines includes actual code

[18:11:00.0000] <Hixie>
hm

[18:11:01.0000] <Hixie>
an event to inform people that the fragment identifier has changed

[18:12:00.0000] <Hixie>
on...hashchanged?

[18:17:00.0000] <kingryan>
isn't there already an event for that?

[18:17:01.0000] <kingryan>
or maybe its just onscroll that gets emitted

[18:21:00.0000] <Hixie>
i don't think there's an event other than onscroll, no

[18:21:01.0000] <Hixie>
and onscroll doesn't always fire

[18:25:00.0000] <kingryan>
right

[19:47:00.0000] <deltab>
Hixie: would it be when navigation starts, or when the target fragment is shown?

[20:22:00.0000] <Hixie>
deltab: it fires when you're on a page and you do something to scroll the page to a new fragid

[21:08:00.0000] <mpt>
Hixie, how about when you load a page with a fragment in the first place?

[21:09:00.0000] <mpt>
ah, maybe that's already catered for

[05:08:00.0000] <annevk>
Hixie, http://annevankesteren.nl/2007/08/ajax-history people ask for replaceState()

[05:09:00.0000] <Philip`>
Hixie: About the parser performance data: I got significant speedups by measuring the expected lengths of strings (tag names, attributes names/values) and preallocating enough space for ~99% of those, so that's quite useful data to have too

[05:10:00.0000] <Philip`>
(I suppose data about attribute-value-length per attribute-name could be useful too - src and href and onclick etc are usually really long, but maybe you could easily preallocate space for all the shorter attributes)

[05:13:00.0000] <annevk>
Philip`, I mentioned this in a pm, but I suppose it can be discussed here as well, how about merging the <canvas> testsuites? It would be great if your testsuite became accessible to more people so they can contribute I think

[05:18:00.0000] <Philip`>
annevk: That sounds like a good thing to do

[05:19:00.0000] <annevk>
seems that you cover most, if not everything, of the tests I covered and more so using your stuff as base makes more sense now :)

[05:19:01.0000] <Philip`>
I can't think of any immediately obvious ways to merge all the existing ones, since mine is weird and puts everything in tiny boxes and tries to avoid ever having humans look at the results

[05:20:00.0000] <annevk>
it does mark the boxes either red or green which is enough for humans

[05:25:00.0000] <Philip`>
annevk: I think the problem is more with other tests like http://tc.labs.opera.com/html/canvas/getContext/2d/arc/004.htm and http://hixie.ch/tests/adhoc/html/canvas/005.html whose pass/fail outcome can't really be determined automatically

[05:29:00.0000] <annevk>
hmm, yeah, I suppose it makes sense that the testsuite consists of an automated part an a non-automated one

[05:29:01.0000] <Philip`>
(I think I've only got three which can't be done automatically in a perfect browser, plus two that can't be done automatically in Firefox/Opera because of bugs)

[05:30:00.0000] <Philip`>
((Bugs related to security of data: and drawImage and getImageData, in particular))

[05:32:00.0000] <Philip`>
(But IE/Safari can hardly do any of them automatically, since they lack getImageData or equivalent)

[05:34:00.0000] <annevk>
it makes some sense to require getImageData for the testsuite... it's a feature of <canvas> after all :)

[08:10:00.0000] <annevk>
someone suggested that hashchanged shouldn't be past tense to be more in line with other comments

[13:01:00.0000] <Hixie>
http://www.chewontech.com/2007/08/more-things-added-in-html-5.html is surprisingly well-written

[13:15:00.0000] <Philip`>
Hixie: Probably because they nicked it from http://www.ibm.com/developerworks/xml/library/x-html5/

[13:16:00.0000] <Hixie>
huh

[13:16:01.0000] <Hixie>
interesting

[13:17:00.0000] <Hixie>
the style of the chewontech page make it much easier to read than the ibm one

[13:17:01.0000] <Hixie>
to the point where i read the chewontech page but couldn't read the ibm one

[16:13:00.0000] <gsnedders>
how many times have I defended IE from people trying to make MS look bad?

[16:14:00.0000] <othermaciej>
is that a trick qustion?

[16:14:01.0000] <gsnedders>
othermaciej: a rhetorical one

[16:14:02.0000] <gsnedders>
complaining filter goes against the option in CSS2.1 of using hyphens or underscore this time, despite it predating CSS2.1

[16:15:00.0000] <othermaciej>
I think the CSS vendor prefix convention also predates CSS 2.1

[16:15:01.0000] <gsnedders>
othermaciej: I just looked it up. Couldn't find it in CSS2

[16:15:02.0000] <othermaciej>
although I don't know if it predates the IE filter property

[16:16:00.0000] <gsnedders>
IE filter was introduced in IE4

[16:16:01.0000] <gsnedders>
(which also predates CSS2)

[16:16:02.0000] <othermaciej>
the worse thing about filter is that SVG introduced a filter CSS property with an incompatible syntax

[16:17:00.0000] <othermaciej>
but I blame SVG for that, not Microsoft

[16:19:00.0000] <gsnedders>
it's often on a forum which I go to less and less, as I've been called an idiot so many times for believing that XHTML isn't the future, and that MS isn't totally shit

[16:19:01.0000] <othermaciej>
well, IE is really not that great in terms of standards complliance and has not advanced much in a long time

[16:20:00.0000] <othermaciej>
so I'm not sure all criticism of MS is off-base

[16:21:00.0000] <gsnedders>
a lot of it is

[16:24:00.0000] <Hixie>
there are a lot of people who criticise IE unfairly and incorrectly

[16:24:01.0000] <Hixie>
but there's a lot of criticism that IE could get fairly and accurately that it doesn't

[16:24:02.0000] <Hixie>
it's possible the two balance out :-)

[16:24:03.0000] <gsnedders>
also a lot who criticise the devs, cwilso especially

[16:24:04.0000] <Hixie>
chris is a fine man

[16:24:05.0000] <Hixie>
not the most active of chairmen, but that's another story

[16:25:00.0000] <Hixie>
someone give me a random number

[16:25:01.0000] <gsnedders>
42

[16:25:02.0000] <Hixie>
dom-events it is

[16:25:03.0000] <gsnedders>
(not really random, but hey)

[16:26:00.0000] <gsnedders>
in the case of some people in the WG they'll complain if he isn't active as chair, and if he is, they'll complain he isn't improving Trident

[16:26:01.0000] <Hixie>
he could be more active than he is without really eating into his Trident productivity much

[16:26:02.0000] <gsnedders>
I know, but they'll find reasons to complain at him someway or another

[16:27:00.0000] <gsnedders>
I remember one reaction I got around a year ago when I told someone Tantek  was lead of IE5/Mac. Then my saying how old IE5/Mac was. They had assumed it was recent, and it was terrible because it was MS.

[16:28:00.0000] <gsnedders>
Hixie: why did you want a random number, though?

[16:29:00.0000] <Hixie>
i'm picking things to work on in the spec

[16:29:01.0000] <gsnedders>
then how do numbers map to things in the spec?

[16:29:02.0000] <Hixie>
42 was the 42nd folder on my IMAP server

[16:29:03.0000] <gsnedders>
ah.

[16:29:04.0000] <Hixie>
INBOX.input-for-whatwg-dom-events

[16:30:00.0000] <hendry>
:)

[16:30:01.0000] <Hixie>
ok

[16:30:02.0000] <Hixie>
<img alt>

[16:30:03.0000] <Hixie>
it seems there are three kinds of images

[16:30:04.0000] <gsnedders>
I wonder if Douglas Adams' ever knew what the consequences of trying to think of what a normal number would be

[16:30:05.0000] <Hixie>
images that are really just fancy alternatives to actual text

[16:31:00.0000] <Hixie>
images that are purely decorative

[16:31:01.0000] <Hixie>
and images that are intrinsic to the content but have no textual alternative

[16:32:00.0000] <hendry>
/me would agree with that

[16:32:01.0000] <gsnedders>
what about graphs?

[16:32:02.0000] <hendry>
gsnedders: vector graphics?

[16:32:03.0000] <gsnedders>
they have vital to the content, but can have a textual alternative

[16:32:04.0000] <gsnedders>
hendry: no, just images that contain graphs

[16:32:05.0000] <gsnedders>
anyhow, I must go sleep

[16:32:06.0000] <Hixie>
examples being a simple diagram or graph that can be described textually, a snapshot from a movie in a movie review or a diagram or graph that just repeats what the prose says, and a photo on flickr respectively


2007-08-11
[17:13:00.0000] <Hixie>
any

[17:13:01.0000] <Hixie>
er

[17:13:02.0000] <Hixie>
http://www.whatwg.org/specs/web-apps/current-work/#the-img is the starting point for my <img>/alt="" redesign

[17:13:03.0000] <Hixie>
work in progress

[17:35:00.0000] <jruderman>
interesting, the alt attribute is no longer required and having no alt now has a specific meaning

[17:37:00.0000] <jruderman>
"The download of the image must delay the load event." might put one variant of the intranet-port-scanning attack into spec, but i'm not sure if you want to be trying to prevent that attack

[17:37:01.0000] <jruderman>
speaking of specs and security

[17:38:00.0000] <jruderman>
are you interested in possible changes to the same-origin policy?

[17:38:01.0000] <jruderman>
some people think IP addresses should become part of origin (along with hostname, protocol, and port)

[17:40:00.0000] <h3h>
my guess is that would break sites (mostly intranet) and Microsoft would be reluctant at best

[17:40:01.0000] <h3h>
but that's just a guess

[17:42:00.0000] <jruderman>
won't TCPConnection allow malicious web pages to get around the "servers should check the Host header" solution to DNS rebinding attacks?

[17:42:01.0000] <othermaciej>
the TCPConnection from HTML5?

[17:43:00.0000] <othermaciej>
I think it would be hard to use it to connect to an http server

[17:43:01.0000] <Hixie>
jruderman: (i'm not really here but if you could mail me your feedback ian⊙hc -- or to one of the lists -- that'd be great)

[17:43:02.0000] <jruderman>
oh, it's not allowed to connect to port 80? i guess that helps a little

[17:43:03.0000] <othermaciej>
but yes, the protocol itself would itself be vulnerable to DNS rebinding

[17:43:04.0000] <othermaciej>
jruderman: no, it uses a wacky custom protocol and requires a particular handshake

[17:44:00.0000] <othermaciej>
jruderman: you're not supposed to be able to actually use it to connect to any existing network service

[17:44:01.0000] <jruderman>
oh, neat

[17:44:02.0000] <othermaciej>
although the way it enforces this is pretty weak

[17:44:03.0000] <othermaciej>
the protocol itself could be subject to DNS rebinding, like I said

[17:46:00.0000] <jruderman>
it looks like it tries not to be, by making both sides send the hostname(?)

[17:48:00.0000] <Hixie>
that's the source domain, but yes, that's intended to mitigate the problem partly

[17:48:01.0000] <Hixie>
the idea of having the IP in the origin might work

[17:48:02.0000] <Hixie>
though it would cause all kinds of issues in multi-homed sites like, say, google

[17:50:00.0000] <jruderman>
why are ports below 1024 and ports >= 1024 treated differently?

[17:50:01.0000] <Hixie>
i recommend starting a thread on whatwg⊙wo about the origin idea

[17:50:02.0000] <Hixie>
/me looks at the spec

[17:51:00.0000] <Hixie>
oh it's to try and avoid people abusing services by connecting to services and faking the protocol

[17:51:01.0000] <Hixie>
in case the handshake isn't good enough

[17:51:02.0000] <Hixie>
to stop that

[17:52:00.0000] <jruderman>
i think it would be good to specify wraparound behavior for port numbers (or lack of wraparound behavior), since many apps have wraparound behavior without intending to have it

[17:52:01.0000] <Hixie>
what is "wraparound behaviour"?

[17:52:02.0000] <jruderman>
port 65616 = port 80

[17:52:03.0000] <jruderman>
wouldn't want someone to be able to get around the "no port 80 connections" rule that way

[17:53:00.0000] <Hixie>
the spec already defines that

[17:53:01.0000] <Hixie>
it says to raise an exception

[17:54:00.0000] <jruderman>
oh, i read the whole "the port argument is neither equal to 80, nor equal to 443, nor greater than or equal to 1024 and less than or equal to 65535," sentence backwards

[17:55:00.0000] <jruderman>
it *only* allows connections on port 80, 443, and 1024..65535?

[17:55:01.0000] <jruderman>
why allow ports 80 and 443? the TCPConnection protocol isn't http or https...

[17:56:00.0000] <Hixie>
to tunnel out of fascist regimes

[17:56:01.0000] <Hixie>
basically

[17:56:02.0000] <jruderman>
hehe, ok

[17:56:03.0000] <Hixie>
(many sites proxy port 80, and block everything else, so the only way out is 443)

[17:57:00.0000] <Hixie>
note that othermaciej and others have quite low opinions of this section

[17:57:01.0000] <jruderman>
ok

[17:57:02.0000] <Hixie>
personally i'm not sure what to replace it with if we remove it -- the usual suggestion (embedding the whole http stack into this) is not one i feel is plausible

[17:58:00.0000] <Hixie>
but i guess i need to draw up the requirements/problem space to see what else could address it (and to show if/why http doesn't handle it)

[17:58:01.0000] <Hixie>
i think notwithstanding the abuse of ports 80 and 443, and not withstanding the fact that it's Yet Another custom protocol, it actually is pretty neat

[17:59:00.0000] <Hixie>
but those two things are the main things people have against it, and i don't have good answers to those concerns

[18:02:00.0000] <Hixie>
i've added notes about the two security attacks you mentioned -- rebinding and ip in origin, and port scanning through load events -- to the spec and will look at them in due course.

[18:03:00.0000] <Hixie>
now i'm really going offline -- bbl. feel free to e-mail the list on things you see, especially security issues.

[06:27:00.0000] <Philip`>
HTML5 parsing is incompatible with IE and Opera on http://www.gerv.net/security/content-restrictions/ - it doesn't think half the page is blue

[06:31:00.0000] <Dashiva>
Philip`: A little bit more detailed?

[06:33:00.0000] <Dashiva>
Do you mean the unclosed span causing long runs of blue text?

[06:33:01.0000] <Philip`>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20HTML%3E%3Cp%3E%3Cspan%20style%3D%22color%3Ablue%22%3EA%3C/p%3E%3Cp%3EB is about what that page does

[06:33:02.0000] <Philip`>
Yes

[06:36:00.0000] <Philip`>
Firefox/Safari/HTML5 put the B outside the span, Opera/IE put it inside

[06:36:01.0000] <Dashiva>
I'm more concerned about P being a descendant of another P

[07:29:00.0000] <hendry>
does anything use <link rel='archives' title='July 2007' type html? googlebot?

[08:35:00.0000] <zcorpan>
/me is now back in sweden

[08:58:00.0000] <zcorpan>
anyone who has written an html parser: is it faster to parse quoted attributes or unquoted attributes, or indifferent?

[08:59:00.0000] <zcorpan>
Hixie: ^

[08:59:01.0000] <zcorpan>
hsivonen: ^

[09:30:00.0000] <mpt>
hendry, iCab? :-)

[09:40:00.0000] <Philip`>
zcorpan: Quoted should be theoretically a tiny bit faster for long attribute values, since you don't need to keep checking for whitespace, but maybe a tiny bit slower for short values since the quotes mean there are more characters to process, but I think that's all trivial and irrelevant compared to all the other costs (of getting input characters in the first place, of constructing attribute value strings, etc)

[09:45:00.0000] <Philip`>
The extra 2 bytes for a quoted attribute, if you're downloading a web page at 100KB/sec, would take about 20 microseconds, which is about 40,000 CPU cycles, so a few extra comparison instructions won't be at all significant

[09:45:01.0000] <Philip`>
((unless you're writing in Python or something where a comparison might be closer to being that expensive))

[10:56:00.0000] <zcorpan>
Philip`: what's the difference between checking for whitespace and checking for the closing quote character? whitespace can be several different characters?

[10:58:00.0000] <Philip`>
It's just a few more values to check for - unquoted needs to detect 9/10/11/12/32/38/62/EOF, double-quoted only needs to look for 34/38/EOF, so you'd usually need a few more comparisons in the unquoted case

[10:59:00.0000] <zcorpan>
yeah, ok

[11:02:00.0000] <Philip`>
I don't think it's an especially compelling reason to use HTML instead of XHTML :-)

[11:16:00.0000] <zcorpan>
Philip`: oh, i was just surprised that mskinner's implementation experience suggested that parsing quoted attributes would be "MUST faster" than unquoted

[11:16:01.0000] <zcorpan>
http://forums.whatwg.org/viewtopic.php?t=93#395

[11:36:00.0000] <annevk>
/me is in Spain without luggage

[11:44:00.0000] <annevk>
/me looks at <img> redesign

[11:45:00.0000] <annevk>
/me notes that <audio> prolly needs a note about its constructor too

[11:48:00.0000] <annevk>
/me reads molly's rant

[11:48:01.0000] <annevk>
/me notes that #whatwg is the new twitter

[11:50:00.0000] <annevk>
/me reads the comments about people being frustrated by the WHATWG now it's suddenly in the W3C

[11:51:00.0000] <annevk>
/me thinks the comments at http://www.molly.com/2007/08/11/dear-w3c-dear-wasp/ are quite hilarious given what has been said so far about headers= etc. on the public-html list

[11:52:00.0000] <tantek>
/me reads about annevk reading molly's rant and goes to read molly's rant.

[11:52:01.0000] <takkaria>
/me wonders what all the fuss is about

[11:53:00.0000] <tantek>
takkaria, a disconnect between those that hold meetings, their topics/agendas, and those that work in the field day-to-day, their real-world challenges and desired solutions.

[11:56:00.0000] <takkaria>
heh

[11:57:00.0000] <takkaria>
since when has WHATWG been a semi-secret society?

[11:59:00.0000] <tantek>
takkaria, it is a general trolling technique

[12:01:00.0000] <tantek>
often used by bureaucracy trolls: http://tantek.pbwiki.com/TrollTaxonomy#Bureaucraytroll

[12:02:00.0000] <takkaria>
semi-secret in the sense of "public". :)

[12:02:01.0000] <takkaria>
I wonder when MS will send feedback on the current spec

[12:05:00.0000] <takkaria>
nice taxomony, there

[12:23:00.0000] <Philip`>
zcorpan: Maybe quoted vs unquoted matters more if you're writing an actual SGML-like HTML4 parser, since that's much more particular than HTML5 about what characters are allowed in unquoted values

[12:24:00.0000] <zcorpan>
Philip`: so what would you do when you hit a character that is not allowed?

[12:27:00.0000] <Philip`>
Given what the W3C validator does, you sometimes start reading a new attribute name that begins with the non-allowed character

[12:27:01.0000] <Philip`>
(at least for <div class=foo'bar></div> - but it seems different for <div class=foo#bar></div> since that just complains about an invalid character inside the value)

[12:28:00.0000] <zcorpan>
crazy

[12:28:01.0000] <zcorpan>
/me assumes that sgml doesn't define what to do

[12:28:02.0000] <hasather_>
zcorpan: SGML has no defined error handling

[12:29:00.0000] <zcorpan>
hasather_: it has some

[12:29:01.0000] <hasather_>
like?

[12:29:02.0000] <zcorpan>
hasather_: e.g., </foo in a cdata element closes the element but is an error if the end tag doesn't match the start tag

[12:30:00.0000] <zcorpan>
not sure if there are others

[12:45:00.0000] <hasather_>
zcorpan: yes, so the error there is that is that the end tag doesn't match the start tag. But SGML doesn't define what to do in that case

[12:50:00.0000] <zcorpan>
hasather_: i thought it did (namely close the element anyway)

[12:54:00.0000] <hasather_>
zcorpan: it says that the first end-tag delimiter-in-context (i.e. ETAGO followed by a name start character) closes the element.

[12:55:00.0000] <zcorpan>
hasather_: yes, exactly

[12:58:00.0000] <zcorpan>
but yeah, i guess you could argue that it doesn't say if you should do something different in case it is an error

[13:12:00.0000] <tantek>
takkaria, thanks.  Also, I have attempted to add a rational response to the comments on molly's post: http://www.molly.com/2007/08/11/dear-w3c-dear-wasp/#comment-610009

[13:15:00.0000] <takkaria>
tantek: I tried that a while ago.  never seems to do much, but I suppose one has to try

[13:16:00.0000] <tantek>
rationality is worth persistence

[13:18:00.0000] <tantek>
i'm actually optimistic that accessibility "advocates" can and will adopt more purely rational/scientific approaches to their advocacy, thus build stronger, more appealing cases for accessibility, which more will then listen to, which will hopefully result in more accessibility in technologies, tools, etc.

[13:21:00.0000] <tantek>
here is an example of such a rational/scientific approach in progress, thanks largely to James Craig and a few other accessibility advocates: http://microformats.org/wiki/assistive-technology-abbr-results

[13:21:01.0000] <takkaria>
/me takes a look

[13:21:02.0000] <takkaria>
I was optimistic for a while, maybe it'll return now public-html has calmed down a little

[13:21:03.0000] <tantek>
it's not perfect, but it's a huge step in the right direction in terms of approach etc.

[13:24:00.0000] <takkaria>
yeah, that looks pretty good

[13:25:00.0000] <tantek>
we have to work on reducing the use of negative/personal rhetoric as a method of discourse - it doesn't help anybody

[16:39:00.0000] <othermaciej>
Molly's post is confusing

[16:39:01.0000] <othermaciej>
I'm not sure which things she thinks are a problem or why

[16:40:00.0000] <takkaria>
it seemed a bit of an undirected rant to me

[16:43:00.0000] <othermaciej>
actually it sounds like a Microsoft enemies ist

[16:43:01.0000] <othermaciej>
*list

[16:43:02.0000] <othermaciej>
this might be an accident

[16:44:00.0000] <othermaciej>
but I know that Microsoft is against many advanced features in HTML5, against the new ECMAScript standards process, and against AIR

[16:44:01.0000] <takkaria>
AIR?

[16:44:02.0000] <takkaria>
/me is new around these parts

[16:45:00.0000] <othermaciej>
it's the Adobe Flash+HTML for local apps thing

[16:45:01.0000] <mgdm>
http://en.wikipedia.org/wiki/Adobe_AIR

[16:47:00.0000] <takkaria>
ah, right.  last I heard, it was still Apollo

[16:48:00.0000] <takkaria>
is that MS as a company that's against them, or the IE team against them?

[16:48:01.0000] <mgdm>
/me is very new around these parts, hello everyone

[16:52:00.0000] <webben>
I think folks would benefit from waiting for Molly to clarify her post (or asking her too) rather than trying to guess what she means.

[16:53:00.0000] <webben>
othermaciej: Is there some document somewhere that explains how the standards process has changed? I wasn't aware it had.

[16:53:01.0000] <othermaciej>
takkaria: MS is against them because it encroaches on (a) native windows app development with win32 or .NET and (b) Silverlight

[16:54:00.0000] <takkaria>
/me nods

[16:54:01.0000] <takkaria>
that's pretty much why they're being added, so I'm not suprised, really


2007-08-12
[17:02:00.0000] <othermaciej>
well, I commented on Molly's post

[17:02:01.0000] <othermaciej>
although mostly responding to her, not her commenters

[17:02:02.0000] <othermaciej>
who seem to be off on another tangent altogether

[18:17:00.0000] <annevk>
the comments on molly.com latest post are just sad

[18:25:00.0000] <annevk>
xml parser error: http://tommorris.org/blog/

[18:25:01.0000] <annevk>
/me hits "Reparse document as HTML"

[18:37:00.0000] <Philip`>
XHTML isn't great at marketing itself, since the only time you ever notice that a site uses XHTML is when it gives you an XML parse error

[23:29:00.0000] <Hixie>
i don't really understand molly

[23:29:01.0000] <Hixie>
didn't she used to be pro progress?

[23:34:00.0000] <othermaciej>
I'm not sure if she's as against progress as she sounds

[23:35:00.0000] <othermaciej>
I thought it was funny that she picked on AIR but not Silverlight

[23:41:00.0000] <Lachy>
it looks like JF and others are still complaining about pave the cowpaths (see his comment on molly's blog). I think he just misunderstands it, misrepresents it and then objects to it

[23:41:01.0000] <Lachy>
but I think we need to reword it to make it a little less vague

[23:42:00.0000] <Lachy>
this is an alternative wording I've come up with:

[23:42:01.0000] <Hixie>
i still don't really understand why this was never a problem before the w3c htmlwg

[23:42:02.0000] <Lachy>
Where possible, evaluate the success and failure of existing practices or solutions and consider adopting and/or improving upon them in preference to forbidding them and/or inventing new features.

[23:43:00.0000] <tantek>
o hai hixie, othermaciej & lachy

[23:43:01.0000] <Hixie>
lachy: what's the current wording?

[23:43:02.0000] <Hixie>
hey tantek

[23:43:03.0000] <Lachy>
hi tantek 

[23:43:04.0000] <tantek>
/me hopes he didn't speak too out of line on molly's post.

[23:43:05.0000] <Lachy>
Hixie: "When a practice is already widespread among authors, consider adopting it rather than forbidding it or inventing something new."

[23:44:00.0000] <Hixie>
Lachy: i prefer your new wording, yeah

[23:44:01.0000] <othermaciej>
hello tantek 

[23:44:02.0000] <othermaciej>
Lachy: I agree it should be reworded, I like your wording

[23:44:03.0000] <tantek>
Hixie, this *was* a problem before (the new) w3c htmlwg

[23:44:04.0000] <othermaciej>
Lachy: I do think a lot of people are misunderstanding it

[23:45:00.0000] <Lachy>
IIRC, a few people have compared pave the cowpaths with adopting tables for layout, since it's so common, but such comparisons miss the point about evaluation the solutions

[23:45:01.0000] <othermaciej>
Lachy: but some misunderstandings seem to be wildly out of line with even the current wording

[23:45:02.0000] <Hixie>
lachy: we might want to look at the usemap="" thing as another example for that btw, since it's not accessiblity-related and doesn't involve xml syntax...

[23:45:03.0000] <othermaciej>
for example, claiming that "pave the cowpaths" would calll for removing particular html features

[23:45:04.0000] <tantek>
pave the cowpaths is not about adopting markup usage literally - it's about analyzing people's *data* *publication* behaviors

[23:45:05.0000] <Hixie>
Lachy: it highlights the focus on research

[23:46:00.0000] <tantek>
at least in the context of microformats

[23:46:01.0000] <tantek>
people are starting to use "pave the cowpaths" out of context

[23:46:02.0000] <Hixie>
though as othermaciej says, that's probably best given as a separate principle

[23:46:03.0000] <Hixie>
to avoid confusing people

[23:46:04.0000] <othermaciej>
I think the principle might need a new name as well as new text along lines proposed by Lachy

[23:47:00.0000] <Lachy>
one issue with that new wording is that it sounds a lot like Don't Reinvent the Wheel

[23:47:01.0000] <tantek>
i'm ok with confused people as long as they keep their dialog civil

[23:47:02.0000] <Hixie>
tantek: what i meant about the problem is that before the w3c htmlwg started, we had very few people argue from a position of authority rather than using rational discussion and basing their opinions on research

[23:48:00.0000] <tantek>
interesting

[23:48:01.0000] <Hixie>
tantek: still now, indeed, the discussions in whatwg's list are more rational than many on public-html (though there has been much progress on that front in public-html, i must admit)

[23:48:02.0000] <tantek>
well that's hopeful

[23:48:03.0000] <Hixie>
it might just be that the w3c community has fostered the development of "experts" whose entire careers are positioned around being on working groups and having their way

[23:50:00.0000] <Lachy>
yeah, Ben Millard on public-html has done some good work on finding table examples for doing research on headers.  I wish that had happened when the headers debate started and we were asking for it

[23:50:01.0000] <Lachy>
but at least it is improving now

[23:50:02.0000] <Hixie>
yeah

[23:50:03.0000] <Hixie>
i'm gonna have to look at that issue soon, i have a ton of data i collected in early july about it

[23:51:00.0000] <Hixie>
i guess i can do that after <img alt>

[23:51:01.0000] <Lachy>
what needs to be done for alt?

[23:56:00.0000] <Hixie>
see the spec for my wip revamp of <img alt>

[00:00:00.0000] <Lachy>
Hixie, another example for a piece of text with an alternate graphical representation would be some company logos

[00:00:01.0000] <Hixie>
yeah

[00:01:00.0000] <tantek>
Hixie, yeah I need to write up the table headers usage examples for hCalendar more properly

[00:01:01.0000] <Lachy>
You're also missing "An iconic representation of the surrounding content" - it's similar to "graphical representation of some of the surrounding text", but not quite the same

[00:02:00.0000] <tantek>
wow: http://twitter.com/rogerjohansson/statuses/200225882

[00:03:00.0000] <Lachy>
see, for example, http://www.webnauts.net/ - the wheelchair icon represents the concept of accessibility, the magnifying glass represents SEO, etc.

[00:03:01.0000] <Lachy>
(although, the alt text used on that site is ironically inappropriate)

[00:07:00.0000] <Lachy>
woah, this is surprising too http://twitter.com/rogerjohansson/statuses/132542642

[00:09:00.0000] <tantek>
clearly Roger felt bullied whether or not he actually was

[00:10:00.0000] <othermaciej>
I'm not sure why he felt bullied

[00:10:01.0000] <tantek>
i think there is a bit of a culture transition clash going on

[00:11:00.0000] <othermaciej>
I don't see that many messages from him, and none that are in my public-html folder seem to have engendered particularly heated responses

[00:11:01.0000] <Lachy>
some people, particularly the accessibility advocates, just have difficulty seeing our point of view and object whenever someone questions the conventional wisdom

[00:12:00.0000] <Lachy>
I'll speak to Roger later and find out what he really thinks

[00:12:01.0000] <tantek>
standards organizations, and w3c is no exception, often designed things "a priori", depending on the opinion of "experts" to just "pick the right thing" a priori, based on their expertise.

[00:13:00.0000] <tantek>
the approach that microformats (and apparently WHATWG) is taking with scientifically challenging features and demanding justification through research/evidence is quite new

[00:13:01.0000] <tantek>
and quite threatening to those that are used to simply getting their way because they say so

[00:13:02.0000] <tantek>
as in, how dare anyone question their expertise?

[00:13:03.0000] <tantek>
shouldn't their expertise be sufficient to justify things?

[00:15:00.0000] <Lachy>
one thing I've also noticed is that some people refuse to look at the issues from the authoring point of view, and only consider the benefits for those with accessibility-related needs

[00:15:01.0000] <tantek>
yes, i've noticed that tooo

[00:16:00.0000] <tantek>
it's like, just because you ask, people aren't necessarily going to do it.  econ 101, psych 101 etc.

[00:16:01.0000] <Lachy>
i.e. they advocate features regardless of how easy they are for authors, and refuse to consider authoring incentives beyond legal and moral arguments

[00:16:02.0000] <othermaciej>
it really depends on what the standards body in question is, as far as past experience goes

[00:16:03.0000] <tantek>
Lachy, to be fair, *some* advocate features that way

[00:17:00.0000] <tantek>
painting all accessibility advocates with the same brush is inaccurate

[00:17:01.0000] <othermaciej>
my understanding is that, for instance, things like the C99 or H.264 had quite a bit of empirical work go into changes

[00:17:02.0000] <Lachy>
yeah, I realise it's not all, but it is at least a few of the more vocal ones

[00:17:03.0000] <othermaciej>
go into their design rather

[00:17:04.0000] <tantek>
yes on both counts

[00:18:00.0000] <othermaciej>
I have seen a number of people appear to take the stance that disagreeing with their preferred technical approach means you are against human rights

[00:18:01.0000] <Lachy>
indeed, though I think some recent discussions clarified that a bit

[00:19:00.0000] <othermaciej>
voicing opinions like that tends to lead to communication breakdown

[00:19:01.0000] <othermaciej>
John Foliot still seems to stand by this stance

[00:19:02.0000] <othermaciej>
I say this based on both his comment post on Molly's blog and some private communication after I raised that issue

[00:19:03.0000] <tantek>
indeed, labeling someone "against human rights" is quite close to Godwin's law.

[00:20:00.0000] <tantek>
and of course, one should simply call out such ad hominem attacks for what they are

[00:21:00.0000] <Lachy>
/me google's Godwin's law

[00:21:01.0000] <Lachy>
ah http://en.wikipedia.org/wiki/Godwin's_Law

[00:21:02.0000] <tantek>
ah yes John Foliot. he's the one that provided the most ad hominem attack fodder

[00:22:00.0000] <Lachy>
tantek: also the one to most vocally complain about the attitude problems

[00:22:01.0000] <tantek>
keep pointing out peoples' ad hominem attacks, for those are the true attitude problems

[00:22:02.0000] <tantek>
if you can't be civil, then you have an attitude problem and don't deserve any sympathy

[00:23:00.0000] <tantek>
or put more economically, IMHO, uncivil comments are lower priority than civil comments

[03:06:00.0000] <takkaria>
http://twitter.com/rogerjohansson/statuses/200225882

[03:09:00.0000] <takkaria>
oh.  wrong window there

[04:08:00.0000] <hsivonen>
zcorpan: Re: attribute parsing, I don't have anything to add to what Philip` said

[04:08:01.0000] <zcorpan>
hsivonen: ok

[04:09:00.0000] <hsivonen>
/me suffers from a bad case of sudden summer flu and should probably go offline instead of reading irc logs

[04:10:00.0000] <hsivonen>
hmm. looks like Roger Johansson removed the twitters mentioned in the IRC log today

[04:25:00.0000] <zcorpan>
btw, ben millard and me have discussed possible ways to improve the table headers algorithm

[04:26:00.0000] <zcorpan>
it will probably be forwarded to the list at some point

[04:27:00.0000] <zcorpan>
the current algorithm is inadequate for many tables

[05:01:00.0000] <Lachy>
I had a long, rational and very productive discussion with Roger on ICQ a few hours ago. I think we were able to clarify a lot of issues and understand each others perspectives a lot better.

[05:01:01.0000] <Lachy>
I suspect that's why he removed the twitters

[06:51:00.0000] <Whiskey_M>
'lo

[06:52:00.0000] <zcorpan>
yo

[06:52:01.0000] <Whiskey_M>
hi Zcorpan, how goes?

[06:53:00.0000] <zcorpan>
goes well thank you, you?

[06:53:01.0000] <Whiskey_M>
none too shabby :-)

[08:57:00.0000] <met_>
http://manwithnoblog.com/2007/08/12/are-we-becoming-complacent/

[09:00:00.0000] <Whiskey_M>
read it earlier today

[09:00:01.0000] <Lachy>
wow, see Joe's comment on molly's article. http://www.molly.com/2007/08/11/dear-w3c-dear-wasp/#comment-611988

[09:02:00.0000] <Whiskey_M>
isn't that normal Joe ;-)

[09:10:00.0000] <Lachy>
possibly, but it's ironic that he said that after my comment explaining that we are not against accessibility features.  I just don't understand why he'd come back with a strawman argument about us dropping the alt attribute

[09:17:00.0000] <virtuelv>
/me notes that he respectfully disagrees with fantasai on browser vendor influence over CSS

[09:18:00.0000] <virtuelv>
from http://fantasai.inkedblade.net/weblog/2007/css-wg-opening-up/

[09:41:00.0000] <sirPalook>
Noob (just read the specs).  So in <p><tt><b>foo</p>, 'tt' and 'b' are *not* closed, but cloned?

[09:45:00.0000] <sirPalook>
In the DOM that is

[09:46:00.0000] <Philip`>
If you have something after that (other than a </b> or </tt> or EOF), then yes

[09:46:01.0000] <Philip`>
http://james.html5.org/cgi-bin/parsetree/parsetree.py?source=%3Cp%3E%3Ctt%3E%3Cb%3Efoo%3C%2Fp%3Ebar

[09:46:02.0000] <Lachy>
sirPalook, yes, I believe they are reopened after the P element, or inside any following block level element (the adoption agency algorithm is confusing)

[09:47:00.0000] <sirPalook>
I'm trying to figure out the differences with SGML, which theoretically sais that every closing tag closes all open tags...

[09:47:01.0000] <sirPalook>
Yep, thanks.

[09:48:00.0000] <Lachy>
sirPalook, yes, that's the theory and that's how OpenSP implemented it. Not sure if that's actually defined in the SGML spec, though

[09:48:01.0000] <sirPalook>
If i understand correctly, cloning happens for "Formatting tags"

[09:48:02.0000] <Lachy>
yes

[09:48:03.0000] <sirPalook>
Lachy. Yep I never invested any precious brain cells on SGML either :)

[09:51:00.0000] <Philip`>
/me wonders if it's possible to get worse than quadratic behaviour

[09:51:01.0000] <Philip`>
(i.e. worse than http://james.html5.org/cgi-bin/parsetree/parsetree.py?source=%3Cp%3E%3Cb%3E%3Cb%3E%3Cb%3E%3Cb%3E%3Cb%3Ex%3Cp%3Ex%3Cp%3Ex%3Cp%3Ex%3Cp%3Ex )

[09:53:00.0000] <sirPalook>
nice one. Although I think that DOM sais the "clones" point to the same thing.

[09:54:00.0000] <sirPalook>
It's interesting what happens with <A NAME=> though

[10:00:00.0000] <Philip`>
Not quite sure what you mean about <A NAME=>

[10:01:00.0000] <hendry>
hsivonen: do you have an idea with the element isn't used by UAs anymore really for aiding navigation?

[10:01:01.0000] <hendry>
hsivonen: do you have an idea with the LINK element isn't used by UAs anymore really for aiding navigation?

[10:03:00.0000] <hasather_>
hendry: at least Opera uses it

[10:03:01.0000] <Lachy>
/me responded to Joe

[10:04:00.0000] <Lachy>
... on molly's blog

[10:04:01.0000] <Lachy>
I hope this clears up the issue http://www.molly.com/2007/08/11/dear-w3c-dear-wasp/#comment-612166

[10:04:02.0000] <Philip`>
Does Opera use anything other than <link rel=next>?

[10:05:00.0000] <hendry>
i have loads <links on my blog natalian.org. I think I should remove them.

[10:05:01.0000] <hasather_>
Philip`: yes

[10:06:00.0000] <hasather_>
Philip`: activate the navigation bar (Appearance > Toolbar >Navigation bar)

[10:07:00.0000] <Philip`>
hasather_: Aha, thanks - I never noticed that before

[10:11:00.0000] <hendry>
hasather_: it's useless on natalian.org :) /me wonders if i can get that view on Opera mobile

[10:12:00.0000] <virtuelv>
hendry: afaik, no

[10:13:00.0000] <virtuelv>
besides, it only uses known values for rel

[10:14:00.0000] <Philip`>
/me wonders if it's possible to use some kind of heuristics to decide whether a longdesc string is an attempt at writing a URL or is (as seems far more common) a descriptive phrase

[10:15:00.0000] <Lachy>
I think Hixie was going to attempt something like that, there are ways you could test it

[10:15:01.0000] <Philip`>
(By "far more common", I mean I looked at about six pages which use longdesc, and most were text instead of links)

[10:16:00.0000] <Philip`>
If people are going to the effort of writing long descriptions, it seems unhelpful to just ignore them because they never knew it should have been a link instead

[10:16:01.0000] <Lachy>
check if it begins with http://, search the string for spaces (which would indicate a phrase), attempt to resolve and retrieve the URL, etc

[10:18:00.0000] <Lachy>
could you just produce a page that lists all the values found in longdesc attributes, and check it manually?

[10:18:01.0000] <Lachy>
it shouldn't be too hard for someone to visually inspect the list and recognise phrases vs. URLs

[10:19:00.0000] <takkaria>
/me volunteers to do such inspection

[10:22:00.0000] <Lachy>
Philip`, you could also compare the value with the alt attribute and list the values for alt and longdesc side by side on the result page

[10:22:01.0000] <Philip`>
I don't have any data about attribute values at the moment - all I've got is the list at http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/pages/attr/longdesc

[10:23:00.0000] <Lachy>
well, we can rule out wikipedia, since they use it incorrectly and also redundantly

[10:23:01.0000] <Philip`>
...but it shouldn't require significant effort to get some data about alt/longdesc/etc in particular, since they're not that rare

[10:24:00.0000] <Philip`>
(though I don't have code that handles non-ASCII character encodings at all correctly, which is a pain if you're trying to actually read the text on a page)

[10:26:00.0000] <Lachy>
schirmer.com has empty longdesc=""

[10:27:00.0000] <Lachy>
I think Hixie was also going to compare the longdesc attribute with the value of the surrounding <a href="">, if any

[10:27:01.0000] <Lachy>
there should be some discussion of this research in the IRC logs somewhere

[10:28:00.0000] <Lachy>
the 4th result in your list, for instance, links to the same page, as does wikipedia

[10:29:00.0000] <Lachy>
tv.com has <iframe longdesc="">!

[10:30:00.0000] <takkaria>
shakedown.fi seems to replicate the title text as longdesc

[10:31:00.0000] <Philip`>
<IMG align="left" alt="Aurajoen soudut" longdesc="Aurajoen soudut" src="images/aurajoki.jpg" title="Aurajoen soudut">

[10:31:01.0000] <Lachy>
oh, HTML4 allows longdesc on iframe. I didn't realise that

[10:32:00.0000] <takkaria>
Philip`: but a fair few of them don't have alt text at all

[10:32:01.0000] <Philip`>
(See http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/pages/tagattr/iframe/longdesc vs http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/pages/tagattr/frame/longdesc vs http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/pages/tagattr/img/longdesc if you're interested in a specific tag)

[10:33:00.0000] <takkaria>
heh, a longdesc with the same value as src

[10:34:00.0000] <takkaria>
stopitnow.com: <img src="images/logo-homer.gif" alt="stopitnowlogo" ... longdesc="images/logo-homer.gif">

[10:38:00.0000] <takkaria>
puntoluce.tv uses longdesc on a product image to link to that product's page; most of the others I've tried have either been attempts at keyword spam, or have been links to the image or to the page they're on

[10:42:00.0000] <takkaria>
ACLU in Iowa seem to use it on a "get adobe reader!" image to point to the download page; I wonder if that's copy-and-pasted from an adobe webpage?

[10:50:00.0000] <met_>
Lachy, isn't the story about dropping alt based on http://www.whatwg.org/specs/web-apps/current-work/multipage/section-embedded.html#the-img where "alt" isn't marked as required, although bellow is written "The text is given by the alt attribute, which must be present and must have a non-empty value"?

[10:51:00.0000] <takkaria>
based on those links (on img), longdesc is generally misused and even when it, there's little useful information there

[10:52:00.0000] <takkaria>
if some more links could be dredged up from somewhere, I'd go through and put together a wiki page on longdesc usage

[10:53:00.0000] <Lachy>
takkaria, you can use either the whatwg wiki or the HTML WG section of the W3C wiki

[10:53:01.0000] <takkaria>
I think I prefer the whatwg one, mediawiki's friendlier

[10:54:00.0000] <takkaria>
I'll go through the links there after a bite to eat

[10:54:01.0000] <takkaria>
/me bbl

[10:54:02.0000] <Lachy>
met_, look at last alternative titled "A key part of the content that has no textual alternative", that says "no alt" below

[10:54:03.0000] <Lachy>
that appears to be why it's no longer required

[10:56:00.0000] <Lachy>
I wasn't aware of that change until tonight, but it's going to make the validation issue interesting. It's useful to know when an alt has been omitted by mistake, so maybe validators would have to starrt issuing warnings instead.

[10:57:00.0000] <met_>
ah I see

[10:58:00.0000] <Lachy>
I'm not sure why there is a distinction between no alt and alt="" in the spec

[10:59:00.0000] <met_>
but I think that examples mentioned there are not good example, I would prefer even alt="screenshot" before blank alt (personal opinion) - maybe find better examples where alt realle has not sense?

[11:01:00.0000] <Lachy>
hmm. Maybe, I'm not sure what the best approach is with screenshots and alt text.

[11:02:00.0000] <Lachy>
it might be good to look up some guidelines about alt usage (like Jukka's pages, htmlhelp.com, etc.)

[11:02:01.0000] <met_>
just looking in czech accessibility book and there is good example: when has image only decorative meaning e.g. decorative illistration there is preferer blank alt, maybe this

[11:05:00.0000] <met_>
like this http://www.cbsnews.com/sections/eveningnews/main3420.shtml - every article has some small  picture, which often has no information and is mainly for decorative purposes, to draw readers attention

[11:07:00.0000] <Lachy>
yeah, I call those iconic representations

[11:08:00.0000] <met_>
or here the man and 2 women http://en.navrcholu.cz/

[11:08:01.0000] <Lachy>
though, on News sites like that, the pictures often have associated captions

[11:08:02.0000] <met_>
ok its not in 100%

[11:09:00.0000] <Lachy>
yeah, but you're right. I already mentioned this exact issue to Hixie earlier today

[11:09:01.0000] <Lachy>
or yesterday, actually

[11:49:00.0000] <zcorpan>
hmm. why is the distributed spec out of sync with svn?

[11:52:00.0000] <zcorpan>
some issue with revisions after r999?

[11:53:00.0000] <zcorpan>
r999 was 2007-08-10, the spec header says 11 August 2007

[11:53:01.0000] <zcorpan>
Hixie: ^

[11:54:00.0000] <takkaria>
/me starts doing some research on longdesc properly

[12:08:00.0000] <zcorpan_>
http://methisto.blogspot.com/2007/08/html5-trailer-find-your-hero.html

[12:10:00.0000] <zcorpan_>
krijnh: in the logs, perhaps you could have a default right border or padding so that the text doesn't jump around when you hover it?

[12:29:00.0000] <Lachy>
cool trailer :-)

[12:36:00.0000] <Hixie>
zcorpan_: the draft on whatwg.org is my working copy -- i updated it to let people see my wip on alt="", but it's not done yet

[13:35:00.0000] <takkaria>
http://wiki.whatwg.org/wiki/Longdesc_usage

[13:35:01.0000] <takkaria>
if anyone's interested

[13:36:00.0000] <takkaria>
of the 62 pages that philip gave with <img longdesc="">, I reckon two of them are valid uses according to the spec

[13:43:00.0000] <Philip`>
Two? That's probably better than td@headers :-)

[13:43:01.0000] <takkaria>
one of them is explaining what an acessibility icon means

[13:44:00.0000] <takkaria>
the other one at one point links to a text/plain document telling the user they can download speech software to hear the website

[13:45:00.0000] <takkaria>
overall, I'd have to say that longdesc is horribly broken

[13:45:01.0000] <Philip`>
The data does appear to suggest that invisible metadata is bad - authors get no feedback when they do it wrong, and so they almost never get it right

[13:46:00.0000] <Philip`>
"3 of which look like SEO attempts (13, 18, 48, 61)" - that's four, not three

[13:47:00.0000] <takkaria>
true

[13:47:01.0000] <takkaria>
/me fixes

[13:47:02.0000] <takkaria>
if I ever do this again, I'm going to use something other than pen and paper

[13:48:00.0000] <Philip`>
Perhaps it'd be possible to use html5lib to extract the relevant data from the pages

[13:49:00.0000] <takkaria>
well, I used a system of curl'ing all the pages to numerically numbered files, then grepping them for "longdesc", then noting down in a column what category they fit into

[13:49:01.0000] <takkaria>
hooking up html5lib to a "is this right? yes/no" UI would simplify it a lot in future, I imagine

[13:49:02.0000] <takkaria>
I don't know python or ruby, though, so I'm no help there

[13:50:00.0000] <Philip`>
An html5grep might be nice

[13:50:01.0000] <Philip`>
(so you can find elements/attributes/etc without worrying about case-sensitivity or extraneous whitespace or whatever)

[13:55:00.0000] <Philip`>
http://search.cpan.org/dist/XML-Twig/tools/xml_grep/xml_grep - ah, there's already at least one XML one - all it needs is an HTML5 parser stuck on the front

[14:07:00.0000] <takkaria>
what should I do with that page?  should I post it to public-html or whatwg?

[14:17:00.0000] <Philip`>
/me doesn't know whether the chance of starting a new longdesc discussion on public-html should be considered an opportunity or a danger

[14:18:00.0000] <takkaria>
I reckon it's probably a danger at the moment

[14:18:01.0000] <takkaria>
I just wonder what to do with the page now I've made it. :)

[14:19:00.0000] <Hixie>
i recommend waiting until i post my results and start looking at the topic, and then forwarding it to the relevant list :-)

[14:19:01.0000] <Philip`>
Maybe you could mention it on the WHATWG list so people know it exists, and when some longdesc discussion starts again on public-html then point to the page since it's useful data

[14:20:00.0000] <takkaria>
Hixie: I can do that, assuming I'm still subscribed in a couple years' time ^_^

[14:21:00.0000] <Hixie>
that's a pretty cool study

[14:21:01.0000] <Hixie>
were _any_ of them correctly used?

[14:22:00.0000] <takkaria>
two, sort=of.  but while they might have been correct, they weren't useful

[14:22:01.0000] <takkaria>
I was greatly amused by the number of sites trying to do SEO though

[14:23:00.0000] <Philip`>
How many were non-correct but would still be useful if a UA acted in a certain way?

[14:23:01.0000] <Hixie>
/me logs in to his work computer to see what his own longdesc study's results were like and finds that his ruby parser study, which he was running over the weekend, completed

[14:23:02.0000] <Hixie>
sweet

[14:26:00.0000] <Hixie>
let's see

[14:26:01.0000] <Hixie>
longdesc

[14:26:02.0000] <takkaria>
Philip`: er, I'm not sure any

[14:26:03.0000] <hsivonen>
w many were non-correct but would still be useful if a UA  acted in a certain way?

[14:26:04.0000] <hsivonen>
oops. sorry

[14:27:00.0000] <Hixie>
so my script found some hundred billion or so longdescs

[14:27:01.0000] <Hixie>
er

[14:27:02.0000] <Hixie>
hundred billion or so <img>s rather

[14:27:03.0000] <Hixie>
of which about a hundred billion had no longdesc=""

[14:28:00.0000] <Hixie>
of the hundred million or so that _did_ have longdesc and src and in which neither was blank:

[14:29:00.0000] <Hixie>
about 800 thousand had longdesc = src

[14:29:01.0000] <Hixie>
about 85 million had longdesc = an ancestor <a href>

[14:29:02.0000] <Hixie>
about 7 million pointed to the root of another domain (!)

[14:30:00.0000] <takkaria>
/me nods, I got a fair few pointing to the root of another domain

[14:30:01.0000] <Hixie>
about 12 million had a space in the longdesc

[14:30:02.0000] <Philip`>
/me wonders if the 85 million includes the two million pages of Wikipedia * average number of images per page

[14:30:03.0000] <Hixie>
basically i got it down to about 8 million potentially useful longdesc=""s that need further study

[14:31:00.0000] <Hixie>
Philip`: might well

[14:31:01.0000] <Hixie>
Philip`: wikimedia in general

[14:31:02.0000] <takkaria>
nice stats

[14:32:00.0000] <Hixie>
looking at the sample of those 8 million, most seem to be URIs to images

[14:32:01.0000] <Philip`>
http://en.wikipedia.org/wiki/Image:STS-118_launch_M-cropped.jpg - that one page has five longdescs which match ancestor hrefs

[14:33:00.0000] <hsivonen>
Hixie: are you going to post a summary of the results to public-html?

[14:33:01.0000] <Hixie>
the result, i think, though, is that longdesc="" data is so widely abused and corrupt that an accessibility tool would get better usability without supporting it than with supporting it

[14:33:02.0000] <Hixie>
hsivonen: yeah, when i look at longdesc=""

[14:34:00.0000] <takkaria>
/me nods

[14:34:01.0000] <takkaria>
ach, I best be off, bbl

[14:35:00.0000] <Philip`>
Can an accessibility tool get better usability by automatically determining (/guessing) whether a longdesc is valid or not?

[14:35:01.0000] <Hixie>
hard to say

[14:36:00.0000] <Hixie>
i don't know how the heuristic would work

[14:36:01.0000] <hsivonen>
Hixie: comparing with ancestor href='' and doing a HEAD request to see if the resource is another image

[14:37:00.0000] <Hixie>
look at, e.g.: http://photofile.ru/users/marusyan/?page=2

[14:38:00.0000] <Hixie>
several of the longdesc=""s on that page look useful to that heuristic

[14:40:00.0000] <zcorpan_>
Hixie: ah, ok

[14:44:00.0000] <Hixie>
http://wiki.whatwg.org/wiki/Talk:Longdesc_usage

[14:44:01.0000] <Philip`>
The exact heuristic could be left as an area for UA developers to innovate in

[14:48:00.0000] <Hixie>
i think before we give ATs things to innovate on, we should wait for their products to become usable in the first place.

[16:35:00.0000] <takkaria>
/me loks at those urls

[16:58:00.0000] <takkaria>
Hixie: I'll have a bash at those urls tomorrow (UK time)


2007-08-13
[17:05:00.0000] <Hixie>
cool

[17:27:00.0000] <Philip`>
http://molly.com is interesting in Links, since it has fifty lines of "Link: favicon", "Link: pingback", "Link: archives: August 2007", ... at the top of the screen, followed by the "skip to the content" link at the top of the third page

[17:29:00.0000] <Philip`>
(Perhaps that counts as an indication that it'd be nice if pages identified where their content was and let the UAs add a skip-to-content link in an appropriate part of their UI)

[18:53:00.0000] <othermaciej>
Hixie: what would an implementation have to do to support the new semantic block-level elements?

[18:54:00.0000] <othermaciej>
Hixie: would it be appropriate to give them default style of display: block, or no?

[18:54:01.0000] <othermaciej>
Hixie: is it necessary to do some special rendering of <h1> - <h6> in the presence of sectioning elements?

[18:54:02.0000] <othermaciej>
Hixie: I'm wondering because those seem like some of the least controversial additions in the spec, but I'm not sure what, if anything, implementations should do for those elements

[18:58:00.0000] <Lachy>
othermaciej, display: block; for them would be very useful

[18:59:00.0000] <Lachy>
I'm not sure about heading rendering though

[18:59:01.0000] <othermaciej>
Lachy: I could see arguments either way - if they don't have display: block, content authors will be more likely to make documents that degrade gracefully by explicitly setting display: block

[18:59:02.0000] <othermaciej>
I'm not sure what Hixie's intent was

[19:02:00.0000] <Lachy>
I think content authors will still set display: block for them to be compatible with legacy UAs during the transitional period until most UAs support them, though I think having the default has display: block; would be the most logical

[19:02:01.0000] <Lachy>
I think the bigger issue of backwards compatibility is the differences in parsing for known vs. unknown elements

[19:03:00.0000] <othermaciej>
is there a parsing difference per spec?

[19:06:00.0000] <Lachy>
no, I meant the way in when legacy UAs will handle them as unknown elements and new UAs will parse them more sensibly

[19:06:01.0000] <Lachy>
although they're not in the parsing algorithm yet (last time I checked), I assume the sectioning elements will end up being parsed more like div

[22:28:00.0000] <Hixie>
othermaciej: my intent (unsure whether it's workable in reality) is for <h2>-<h6> to continue as is, for all the new block-level elements to have display:block as a style, and for <h1> to automatically style appropriately according to the section depth as defined by the outline section

[22:29:00.0000] <Hixie>
othermaciej: but i have no idea if that is practical back-compat wise, let alone if it is implementable with good perf

[22:30:00.0000] <G0k>
hixie: did you get a chance to look at that event-source stuff yet?

[22:30:01.0000] <G0k>
oh wait i got an email back from you

[22:53:00.0000] <G0k>
oh heh actually i have another question about that

[22:55:00.0000] <G0k>
if you have >1 RemoteEventTarget node within a single document, and they all get the same event source URI added to them, should the UA open one connection for each object, or just open one and then dispatch events from that connection to all the listeners?

[23:12:00.0000] <Hixie>
G0k: i believe the spec unambiguously requires the former

[23:14:00.0000] <G0k>
hm. isn't there like a 2 HTTP connections per UA rule per server?

[23:14:01.0000] <Hixie>
yes, i wouldn't recommend that people have more than one of these things

[23:15:00.0000] <G0k>
yeah i suppose that makes sense

[23:17:00.0000] <G0k>
and is there like a pre-draft of what the Access-Control HTTP header might look like?

[23:27:00.0000] <Hixie>
yeah

[23:27:01.0000] <Hixie>
www.w3.org/tr/access-control or some such

[23:30:00.0000] <G0k>
ah

[23:42:00.0000] <G0k>
i guess this is the wrong place to ask this, but i don't understand what this is supposed to accomplish

[23:47:00.0000] <G0k>
so i put a document on server-a.com, then another resource on server-b.com, and i want the server-a document to access the server-b resource....so now i have to make server-b.com provide an HTTP header that says it's ok for that to happen?

[23:48:00.0000] <Hixie>
yeah

[23:48:01.0000] <G0k>
how is this better than just having server-b check the Referer field?

[00:10:00.0000] <Hixie>
most servers don't check the referer field

[00:11:00.0000] <Hixie>
so if we relied on that, most servers would be vulnerable today

[00:11:01.0000] <G0k>
but...zero servers check the Content-Access-Control fiel

[00:11:02.0000] <G0k>
*d

[00:11:03.0000] <Hixie>
no the server has to _send_ the field

[00:11:04.0000] <G0k>
i guess i'm not clear who's being protected here

[00:12:00.0000] <Hixie>
the user is being protected from server-a pretending to be the user when talking to server-b

[00:15:00.0000] <G0k>
ah ok so this could prevent the resource from server-b, which is accessed by the user directly, from being sent back to server-a?

[00:16:00.0000] <Hixie>
right

[00:16:01.0000] <Hixie>
(on an unrelated note, http://www.w3.org/mid/20070808115703.GC5388⊙sc summarises where i'm coming from regarding the wysiwyg stuff)

[00:17:00.0000] <G0k>
though it requires server-b to implement it, and the UA to implement it, and also requires breaking all other current cross-domain communication techniques?

[00:18:00.0000] <Hixie>
no

[00:18:01.0000] <Hixie>
it requires server-b to implement it only if server-b wants to allow access

[00:19:00.0000] <Hixie>
it requires the UA to implement it only if the UA is to support cross-domain stuff

[00:19:01.0000] <Hixie>
and it doesn't affect any existing cross-domain features

[00:22:00.0000] <G0k>
ok but tricks like dom-inserting a <script> element could still get around this

[00:23:00.0000] <Hixie>
dom-inserting a script only works if you're inserting a script

[00:23:01.0000] <Hixie>
so yes, scripts aren't protected

[00:30:00.0000] <G0k>
ok but...let's say server-b doesn't want server-a's web pages from loading its images

[00:30:01.0000] <G0k>
it could say Content-Access-Control: deny server-a.com

[00:31:00.0000] <G0k>
but that would only be honored by new UAs

[00:32:00.0000] <G0k>
i mean i guess that's somewhat outside the scope of access control's purpose

[00:33:00.0000] <G0k>
but it's something of an broken-at-conception default behavior

[00:42:00.0000] <Hixie>
G0k: images are also not affected by this

[00:43:00.0000] <Hixie>
G0k: this doesn't affect anything that is allowed now

[00:43:01.0000] <Hixie>
G0k: it only affects new cross-domain stuff

[00:43:02.0000] <G0k>
so it only applies in situations where the same origin policy is currently being used?

[00:43:03.0000] <Hixie>
it only applies in situations where the specs say it applies :-)

[00:44:00.0000] <G0k>
yeah right so "To facilitate clear and controlled read access to resources, this specification defines a read access control mechanism that enables a web resource to permit access to its content from external domains when such access would otherwise be prohibited by a same origin policy."

[00:44:01.0000] <G0k>
i guess my question is....is there a formal definition of when the same origin policy is used?

[00:45:00.0000] <G0k>
i mean clearly it seems to always apply to XMLHttpRequest stuff

[00:46:00.0000] <G0k>
and presumably the event-source gunk

[00:47:00.0000] <G0k>
or is it just always left up the implementation?

[00:49:00.0000] <Hixie>
access-control only applies in situations where the specs explicitly say it applies

[00:49:01.0000] <Hixie>
so e.g. xmlhttprequest

[00:49:02.0000] <Hixie>
or rather xmlhttprequest2

[00:50:00.0000] <G0k>
ooo the sequel

[00:50:01.0000] <Hixie>
because that spec says it applies

[00:50:02.0000] <Hixie>
same with xbl2

[00:50:03.0000] <Hixie>
xbl2 says it applies

[00:52:00.0000] <G0k>
alright

[00:53:00.0000] <G0k>
well i guess that was a key point i was missing

[00:53:01.0000] <G0k>
i thought it was more of a replacement for previous access control methods

[00:56:00.0000] <Hixie>
nope

[01:02:00.0000] <Hixie>
oops, replied to a thread that already got replies

[01:02:01.0000] <Hixie>
i need to remember to read the whole thread before replying

[01:10:00.0000] <zcorpan_>
Hixie: what is the legacy parsing behaviour for <address> you're referring to? firefox's behavior?

[01:19:00.0000] <Hixie>
"address" is mentioned in all kinds of places in the parser

[01:20:00.0000] <zcorpan_>
i only find two "address" in the parsing section

[01:22:00.0000] <Whiskey_M>
'lo

[01:23:00.0000] <zcorpan_>
morning Whiskey_M

[01:23:01.0000] <Whiskey_M>
Good morning, how goes today?

[01:24:00.0000] <zcorpan_>
tired... woke up 5am this morning

[01:24:01.0000] <Whiskey_M>
ouch, lots of coffee today then

[01:25:00.0000] <zcorpan_>
nah

[02:17:00.0000] <met_>
http://therealcrisp.xs4all.nl/blog/2007/08/13/getelementsbyclassname-re-re-re-visited/ see also crisp comment

[03:31:00.0000] <virtuelv>
exactly what is Molly talking about here? http://www.molly.com/2007/08/11/dear-w3c-dear-wasp/

[03:32:00.0000] <virtuelv>
I simply don't get what she's trying to point out

[03:37:00.0000] <Dashiva>
/me considers a humorous comment, but decides it's just going to end up in another angry mail some time in the future

[04:01:00.0000] <gsnedders>
virtuelv: the impression I got is W3C's long-term vision for the web (XML)

[04:01:01.0000] <gsnedders>
and that HTML5 is undermining that

[04:02:00.0000] <virtuelv>
I _still_ don't get it

[04:03:00.0000] <gsnedders>
Why is HTML5 being developed at all? It's undermining everything (XML)! Why isn't WaSP and the like doing anything about it?

[04:04:00.0000] <virtuelv>
xml pretty much undermined itself as a language for end-user display on the web when it defined draconian error handling anyway

[04:05:00.0000] <gsnedders>
It's obvious that from Molly's POV it _must_ be the future of the web

[04:05:01.0000] <Whiskey_M>
just a quick one - it's not XML you are talking about, rather XHTML (although the error handling does come from XML parsers)

[04:06:00.0000] <gsnedders>
Whiskey_M: The W3C's vision is to use XML for everything though, not just XHTML

[04:06:01.0000] <virtuelv>
Whiskey_M: note that XHTML as text/html is not XHTML

[04:08:00.0000] <Whiskey_M>
gsnedders, most things yes (and for a lot it works well), virtuelv - if you're on about serving, although transitional format does allow text/html - you're right 

[04:08:01.0000] <Whiskey_M>
heading for tea

[04:08:02.0000] <gsnedders>
Whiskey_M: all XHTML 1.0 DOCTYPEs are allowed under text/html

[04:09:00.0000] <virtuelv>
Whiskey_M: note that no: xhtml as text/html is never parsed using an xml parser

[04:09:01.0000] <virtuelv>
/me points to http://www.hixie.ch/advocacy/xhtml'

[04:09:02.0000] <virtuelv>
err, http://www.hixie.ch/advocacy/xhtml

[04:10:00.0000] <gsnedders>
virtuelv: that's not what he's saying.

[04:10:01.0000] <gsnedders>
virtuelv: he's saying you're allowed to serve it as text/html.

[04:10:02.0000] <gsnedders>
virtuelv: what it is parsed as is a whole different matter

[04:10:03.0000] <virtuelv>
yes, I am not denying that. I'm just saying that you SHOULD NOT

[04:11:00.0000] <virtuelv>
(I do it on my site, but will stop doing so once I get time to rework the templates)

[04:11:01.0000] <gsnedders>
virtuelv: then why the "no"?

[04:12:00.0000] <virtuelv>
gsnedders: because by the time you're doing it, it may have an "xhtml" doctype, but the browser doesn't treat it as such

[04:12:01.0000] <gsnedders>
virtuelv: but that still doesn't contradict the statement that you can serve it as text/html

[04:12:02.0000] <zcorpan_>
virtuelv: with what expected benefit except perhaps for a few saved bytes and a pat on the back? re rework the templates

[04:14:00.0000] <virtuelv>
zcorpan_: just changing the doctype has no real-world benefit as such

[04:15:00.0000] <zcorpan_>
virtuelv: indeed

[04:15:01.0000] <virtuelv>
but that isn't the point of redoing the templates. It's rather to get rid of cruft, and incorporating a few html5 features

[04:15:02.0000] <zcorpan_>
ok

[04:16:00.0000] <virtuelv>
but it's more of a project utopia, given that I never have the time to do it anyway :-P

[04:16:01.0000] <gsnedders>
but in the real-world you don't need to redo it to add HTML 5 features

[04:16:02.0000] <zcorpan_>
removing cruft might need redoing :)

[04:17:00.0000] <zcorpan_>
/me -> lunch

[04:17:01.0000] <gsnedders>
my redesign has gone through so many phases I have no idea what sort of markup it has

[04:17:02.0000] <gsnedders>
probably HTML 4.01 Strict

[04:18:00.0000] <gsnedders>
an earlier concept of the redesign was XHTML 1.1, but that was all scrapped anyway

[04:28:00.0000] <Whiskey_M>
gs / virt:  On reading the piece and without wanting to put words in Molly's mouth - it would seem to say that although a lot of people, companies and organisations are putting together a lot of good work there does seem to be some pulling in different directions, some of the stuff  (AIR), doesn't appear to sit comfortably in standards (also between AIR and Silverlight are we going to see a new "browser wars"), and the normal bodies tha

[04:33:00.0000] <mpt>
Whiskey_M, "...and the normal bodies th"

[04:33:01.0000] <Whiskey_M>
mpt?

[04:33:02.0000] <mpt>
you got cut off

[04:33:03.0000] <Whiskey_M>
ahhh, sorry

[04:34:00.0000] <Whiskey_M>
and the normal bodies that you'd expect to try to bring people / companies / organisations together aren't really saying / doing anything about it

[04:34:01.0000] <Whiskey_M>
btw.  I'm neither advocating or disagreeing with the opinion - just how I understood it

[04:35:00.0000] <hsivonen>
Silverlight is a rather big elephant in the room on Molly's blog. :-(

[05:34:00.0000] <krijnh>
"GET /irc-logs/html-wg/ HTTP/1.1" 302 5 "-" "Mozilla/4.0 (compatible; MSIE 5.5; Windows NT 5.0)"

[05:34:01.0000] <krijnh>
:O

[05:36:00.0000] <zcorpan_>
/me fires up Mosaic

[06:00:00.0000] <met_>
heh, we has even one IE4 hit today http://en.navrcholu.cz/Statistics/58501/browsers-versions/?s=13.8.2007&e=13.8.2007&p=h&pi=1

[06:01:00.0000] <met_>
/me found also one hit from Mozilla Suite 1.1

[06:02:00.0000] <krijnh>
38% Fx ? :|

[06:08:00.0000] <Philip`>
/me probably leaves an unusual log trail of IE6+Win98 when looking around in IE in Wine

[06:15:00.0000] <krijnh>
"...demeaning comments regarding accessibility issues and "smell-o-vision". [http://krijnhoetmer.nl/irc-logs/whatwg/20070701#l-217]"

[06:15:01.0000] <krijnh>
And thank you guys for putting those comments on my site! ;)

[06:17:00.0000] <krijnh>
I should have a disclaimer somewhere..

[06:24:00.0000] <met_>
krijnh, of course it site for Mozilla/FF users 8-)

[06:24:01.0000] <krijnh>
met_: Ah :)

[06:24:02.0000] <met_>
but we have MUCH better if you want http://en.navrcholu.cz/Statistics/81644/browsers-families/?s=13.8.2007&e=13.8.2007&p=h&pi=1

[06:25:00.0000] <krijnh>
You call that much better? Opera has only 1 visit :(

[06:25:01.0000] <zcorpan_>
yeah :(

[06:26:00.0000] <met_>
99.8% of Gecko, I like it 8-)

[06:27:00.0000] <met_>
wait Opera has 1/3 or IE usage ther, it's cool, isn't?

[06:27:01.0000] <met_>
and if you wisit the page once you can increase is twice 8-)

[06:29:00.0000] <zcorpan_>
done already

[06:30:00.0000] <met_>
btw http://xhtml.com/en/future/fixing-the-web-1/

[06:35:00.0000] <mpt>
I don't understand what's so demeaning about smell-o-vision, honestly

[06:36:00.0000] <mpt>
It was the best example I could think of, because smell is a medium that is very poorly supported

[06:36:01.0000] <mpt>
actually, that's not quite right

[06:37:00.0000] <mpt>
Smell is a sense that is supported in very few media.

[06:37:01.0000] <mpt>
So whenever you watch someone's holiday video, for example, you're missing out on all the wonderful smells

[06:38:00.0000] <mpt>
but the narrators of such videos *usually* don't mention them, unless they're particularly noteworthy.

[06:38:01.0000] <mpt>
Similarly, when you listen to radio documentaries, you're missing out on the appearance of all the things being discussed

[06:39:00.0000] <mpt>
but the narrator usually doesn't mention the appearance of things, unless it's relevant to the topic at hand.

[06:40:00.0000] <mpt>
(Podcast hosts who say "I'll put the picture of X in the show notes" are the clumsy exception, rather than the rule.)

[06:42:00.0000] <mpt>
So, I don't think that HTML's accessibility features for people with visual impairments, for example, should take the attitude of "And here's what you would be able to see if you weren't so blind".

[06:42:01.0000] <mpt>
(Which is what longdesc= does.)

[06:43:00.0000] <takkaria>
s/does/tries to do/ ;)

[06:43:01.0000] <mpt>
yeah

[06:43:02.0000] <krijnh>
Again, I need a disclaimer ;p

[06:44:00.0000] <mpt>
Rather, I think that they should concentrate on producing the best self-contained text-only version possible

[06:44:01.0000] <Whiskey_M>
mpt - thinking of an example where longdesc would be appropriate ... describing a chart / graph, just thinking of one of hand

[06:46:00.0000] <mpt>
Whiskey_M, perhaps, though I think a long alt= could get 90% of the effect.

[06:46:01.0000] <Whiskey_M>
off hand, sorry

[06:47:00.0000] <takkaria>
talking of longdesc, I should go over some more URLs and see how many use longdesc correctly

[06:47:01.0000] <mpt>
Sometimes the best text equivalent to a graph would be a <table>, though

[06:47:02.0000] <hsivonen>
mpt: out of curiosity, do you work on usability for the general population these days or does your dayjob also involve accessibility issues?

[06:47:03.0000] <mpt>
which is why the <object> situation is so sad

[06:48:00.0000] <Whiskey_M>
don't know - would perhaps be overkill for alt.  Say we can a graph of user agent usage over the last 12 months - the alt would be "graph of user agent usage over the last 12 months" - the longdesc could give a rough breakdown of the numbers etc. apparent through the chart

[06:48:01.0000] <gsnedders>
how much content actually relies on IE's treatment of |object|?

[06:48:02.0000] <zcorpan_>
Whiskey_M: that wouldn't be very good alt text

[06:48:03.0000] <mpt>
hsivonen, the general population, but a few days ago we had someone on our mailing list who is blind and has trouble with a few pages

[06:49:00.0000] <Whiskey_M>
zcor: yeah I know

[06:49:01.0000] <Whiskey_M>
but at the same time you wouldn't want the full breakdown of the data thrown at you in an alt

[06:49:02.0000] <zcorpan_>
why not?

[06:50:00.0000] <mpt>
Whiskey_M, perhaps minimums and maximums, or starting and ending values, for each series in the graph

[06:51:00.0000] <mpt>
for those at least, you have a fair shot at being able to auto-generate them

[06:51:01.0000] <mpt>
(we have bar charts with auto-generated alt= on our site)

[06:51:02.0000] <Whiskey_M>
because on something like that it would take too long, especially if it were only suplimental information backing up the rest of the information on the page.  Better alt contains an overview and longdesc (if there were one), had the additional information for anyone that wanted to delve in

[06:51:03.0000] <hsivonen>
mpt: any user feedback on the autogenerated alts?

[06:52:00.0000] <mpt>
hsivonen, no

[06:53:00.0000] <zcorpan_>
Whiskey_M: how would you present the information in a text-only context? have a brief overview and a link to a separate page for more detail?

[06:53:01.0000] <mpt>
I suppose if a graph accompanies a news story where the salient changes are already described in text, the alt= should go into deeper detail, because anyone listening to the graph has already listened to the article

[06:54:00.0000] <Whiskey_M>
thats what throws me zcorpan - I'm aware longdesc should point to a text resource of more info, but must admit to never seeing it done (not even as an example on accessify forum)

[07:08:00.0000] <Whiskey_M>
Zcor: thinking about how I'd present that info - probably in a table, user agent / month as the two axis 

[07:11:00.0000] <Whiskey_M>
my bad, look and you will find...  http://www.accessifyforum.com/viewtopic.php?t=1807

[07:21:00.0000] <gsnedders>
Hixie: how come the WHATWG WD is 11th Aug, and SVN is 10th?

[07:22:00.0000] <Philip`>
gsnedders: http://krijnhoetmer.nl/irc-logs/whatwg/20070812#l-324

[07:23:00.0000] <gsnedders>
ah

[07:28:00.0000] <hsivonen>
http://www.sitepoint.com/blogs/2007/08/13/nihilism-accessibility-and-the-preponderence-of-amazing-co-incidences/

[07:28:01.0000] <hsivonen>
"the microformats community and WHATWG are behaving like cabals in their self-interested refusal to acknowledge the accessibility issues with that they’re doing"

[07:29:00.0000] <takkaria>
sigh

[07:35:00.0000] <Lachy>
wow, the misconceptions spreading througout the web are getting out of hand

[07:42:00.0000] <takkaria>
that comment seems to be a throwaway line before a somewhat existentialist rant about nothing in particular

[07:42:01.0000] <Lachy>
takkaria, my comment?

[07:43:00.0000] <takkaria>
no, the "the microformats community and WHATWG are behaving..." quote mentioned above

[07:43:01.0000] <Lachy>
oh, I thought mine because I am actually writing a rant right now :-)

[07:43:02.0000] <sirPalook>
Hi.  I believe that end tag </li> should be completely ignored.

[07:44:00.0000] <sirPalook>
Here is why:

[07:44:01.0000] <Lachy>
sirPalook, yes, IIRC, that idea has been mentioned before

[07:45:00.0000] <sirPalook>
In <ul><ul><li></li></li></ul></ul>, there is an extra </li> which has the effect of closing the <ul>

[07:45:01.0000] <sirPalook>
This *does* happen in some pages on the internet (gcc archives)

[07:45:02.0000] <sirPalook>
And neither mozilla nor any other browser act like that.

[07:46:00.0000] <zcorpan_>
sirPalook: http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2007-June/011884.html

[07:46:01.0000] <Lachy>
why would the second </li> close the <ul>? Is that in the spec's algorithm?

[07:46:02.0000] <sirPalook>
html5lib closes it.

[07:47:00.0000] <sirPalook>
um, sorry, maybe there is another <li> in the outer list...

[07:47:01.0000] <sirPalook>
revised example <ul><li><ul><li></li></li></ul></li></ul>

[07:48:00.0000] <zcorpan_>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%0D%0A%3Cul%3E%3Cli%3E%3Cul%3E%3Cli%3E%3C/li%3E%3C/li%3E%3C/ul%3E%3C/li%3E%3C/ul%3E

[07:49:00.0000] <zcorpan_>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%0D%0A%3Cul%3E1%3Cli%3E2%3Cul%3E3%3Cli%3E4%3C/li%3E5%3C/li%3E6%3C/ul%3E7%3C/li%3E8%3C/ul%3E might be more useful

[07:49:01.0000] <sirPalook>
ops! maybe i have an old version.

[07:50:00.0000] <sirPalook>
sorry for the noise.

[07:50:01.0000] <zcorpan_>
no worries :)

[09:21:00.0000] <met_>
see for the first time, that google suggest you other results http://www.google.com/search?q=html%205 - after three "html 5" results google suggest me "html 4" results

[09:23:00.0000] <zcorpan_>
met_: that has been so for some time i think

[10:21:00.0000] <Philip`>
Hmm, does IE not support iframe.onload?

[10:34:00.0000] <Philip`>
http://canvex.lazyilluminati.com/misc/dom-viewer/ -- half live + half dead = zombie?

[10:34:01.0000] <Philip`>
http://canvex.lazyilluminati.com/misc/dom-viewer/?%3C%21DOCTYPE%20html%3E%0D%0A%3Co%3Ap%3E%3Ca%20b%3D - IE, FF and Opera all do different things on each side

[10:36:00.0000] <Lachy>
I've done some more thinking about the design principles and revised my suggestions a bit (see yesterday's discussion http://krijnhoetmer.nl/irc-logs/whatwg/20070812#l-66 )

[10:36:01.0000] <Lachy>
I think we need to reword Don't Reinvent the Wheel and Pave the Cowpaths

[10:36:02.0000] <Lachy>
these are my new suggestions: ...

[10:37:00.0000] <Lachy>
Don't Reinvent the Wheel: Evaluate the success and failure of existing solutions. For those that have proven reasonably successful in terms of benefits, usage and implementation, consider adopting, retaining and/or improving upon them in preference to dismissing and inventing new features.

[10:37:01.0000] <Lachy>
Pave the Cowpaths: Investigate existing practices and design or adopt features that meet the desires of authors.  Where possible, solutions should leverage the existing techniques and skill sets of authors which will help promote the adoption of new features.

[10:39:00.0000] <Lachy>
I'm going to mail them to the list shortly, any suggestions before I do?

[10:39:01.0000] <othermaciej>
Lachy: please do mail the list

[10:44:00.0000] <Whiskey_M>
both admirable, but I wouldn't rule out fresh invention where appropriate.  Take for example the DVD, if cow paths were followed we'd just have video recorders with faster rewinds / skips and greater media protection.  I know there is a time and place for innovation thus I guess the most important word is *appropriate* - just my 2p worth ;-)

[10:45:00.0000] <Lachy>
Whiskey_M: that's a perfect example of misunderstanding and inappropriately applying the principle

[10:46:00.0000] <Whiskey_M>
fair enough, although I might not be the only one to get the wrong end of the stick then

[10:46:01.0000] <Lachy>
DVDs solve a significant problem that video tapes had with their quality

[10:46:02.0000] <othermaciej>
Whiskey_M: the original intent of the principles was:

[10:48:00.0000] <othermaciej>
Don't Reinvent the Wheel -- if there's a deployed solution in existing implementations (i.e. web browsers) to some problem, then even if it has some quirks and minor problems it is likely to be a better choice than something brand new, and effort is probably better spent on improving it than making up something new

[10:48:01.0000] <Lachy>
but if you want to apply pave the cowpaths to DVD and VHS, then you would look at how users are already familiar with inserting media into a player and pressing play, seeking, etc. e.g. tapes, CDs, records, etc. So DVDs to leverage that existing skill set

[10:49:00.0000] <othermaciej>
Pave the Cowpaths -- if authors are doing something that is harmless but either forbidden or not explicitly encouraged, it might be a good idea to allow or even encourage it, because if conformance imposes pointless requirements then authors won't do it

[10:50:00.0000] <othermaciej>
neither of these principles would be very applicable to inventing a brand new format

[10:50:01.0000] <othermaciej>
as opposed to making a new version of an existing one

[10:50:02.0000] <othermaciej>
DVD is not VHS5

[10:50:03.0000] <Whiskey_M>
thanks macie :-)  Lachy ok, bad analogy

[10:51:00.0000] <othermaciej>
in both cases you need to do some investigation to see what problems are being addressed, whether the de facto solution is a good one, etc

[10:52:00.0000] <othermaciej>
this is why all the principles are written in a hedged way ("might", "consider", etc) and not as absolute commandments

[10:52:01.0000] <othermaciej>
it is a balancing act

[10:53:00.0000] <Lachy>
yes, that's why I've tried to emphasize the necessity of research in the proposed wordings

[10:55:00.0000] <Philip`>
What are the alternative viewpoints that "don't reinvent the wheel" is opposed to? Is the problem when people just ignore existing implementations entirely, or is it when they actively want to avoid any relation with existing implementations (i.e. implementations of a feature are seen as a negative point)?

[10:55:01.0000] <Philip`>
(or both?)

[10:56:00.0000] <Philip`>
(or something else? but presumably there's at least some kind of alternative viewpoint, because otherwise it's just a pointless obvious statement rather than a design principle that's helpful when making decisions)

[10:58:00.0000] <othermaciej>
Philip`: there's been some active opposition voiced to "Don't Reinvent the Wheel"; the idea is that we can invent better solutions if we start with a blank slate, than if we have to accept the legacy issues of an existing solution

[10:58:01.0000] <Lachy>
Philip`: pave the cowpaths tends to get conflated with reinventing the wheel and the objection is that it's seen as a way to block the introduction of new features.  I think the best example of this would be the in the discussion of the role attribute

[10:59:00.0000] <othermaciej>
for example, some might propose that instead of <canvas> we should have a whole different approach to immediate-mode graphics that is perhaps somehow part of SVG

[10:59:01.0000] <othermaciej>
or that instead of contentEditable we should invent a whole new editing API that seems more elegant

[11:00:00.0000] <othermaciej>
obviously it is a matter of balance (no one has proposed adopting IE's support for namespaces in HTML for instance), but some thing that existing de facto implementations should be given no weight at all

[11:01:00.0000] <Lachy>
IIRC, some people have even compared it with introducing CSS for layout when tables were already in use for that purpose, and then using that as a strawman to object to it

[11:01:01.0000] <othermaciej>
a similar line of argument has been used to argue against "Support Existing Content" - that we can make something better and "fix the web" if we break compatibility

[11:01:02.0000] <hsivonen>
othermaciej: actually, discussion Re: namespaces has come pretty close to adopting what IE does

[11:02:00.0000] <Lachy>
crap! I don't want to adopt what IE does for namespaces :-(

[11:03:00.0000] <hsivonen>
I'm on a PPC Mac, so I'm not personally in a good position research IE's behavior. I'm hoping that MS or Opera would share their experiences on this.

[11:03:01.0000] <Philip`>
I thought IE looked almost sensible, but that's just because I was using the Live DOM Viewer which gives incorrect answers

[11:04:00.0000] <Philip`>
(like http://canvex.lazyilluminati.com/misc/dom-viewer/?%3C%21DOCTYPE%20html%3E%0D%0A%3Cbody%3E%0D%0A%3Co%3Ap%3E%0D%0A%3Chtml%20xmlns%3Ao%3E%0D%0A%3Co%3Ap%3E has "p > p" on the left, "O:P + p" on the right)

[11:05:00.0000] <Lachy>
in IE, <foo:bar>...</foo:bar> is parsed as pseudo-XML, rather than as ordinary unknown tags.  The major problem with it is that it also creates it's typical broken DOM with badly nested elements

[11:06:00.0000] <Lachy>
so some special variation of the adoption agency algorithm would need to be developed for it

[11:06:01.0000] <othermaciej>
is there a lot of legacy content depending on details of IE's parsing for namespaces in HTML?

[11:07:00.0000] <Lachy>
sadly, there's a lot of XForms in tag-soup on the web

[11:07:01.0000] <Lachy>
and probably MathML too

[11:07:02.0000] <Lachy>
thanks in part to applications like FormsPlayer and MathPlayer (or whatever they're called)

[11:08:00.0000] <Philip`>
It's (generally) only parsed as pseudo-XML if you have a <html xmlns:foo> before it, and not if you use document.write, and people do use document.write with IE namespaces, which probably makes it painful to reproduce

[11:08:01.0000] <hsivonen>
Lachy: aren't you taking the purist attitude instead of paving the cowpaths there? ;-)

[11:09:00.0000] <Philip`>
There's loads of <o:p> and quite a bit of VML, but I don't know if people care how that's parsed or how it interacts with the DOM

[11:09:01.0000] <othermaciej>
if IE's implementation would be hard to emulate enough to be interoperable then I guess it is better to have syntax which won't collide with it instead

[11:10:00.0000] <Lachy>
hsivonen, possibly :-)

[11:10:01.0000] <Philip`>
("loads" = more than 1% of pages)

[11:11:00.0000] <Lachy>
well, actually, I don't object to adding support for MathML and SVG in HTML, but I just the existing solution of using pseudo-XML syntax in HTML is more harmful, and thus it is a case where a better solution should be developed

[11:13:00.0000] <hsivonen>
Lachy: just adding SVG and MathML isn't the kind of distributed extensibility Sam called for

[11:13:01.0000] <Lachy>
I don't understand Sam's use cases for adding it.

[11:14:00.0000] <Lachy>
although I haven't read that whole thread yet.

[11:14:01.0000] <hsivonen>
Lachy: the cases cited thus far have been about non-browser server-side templating languages leveraging HTML5 toolchains

[11:14:02.0000] <Lachy>
But I think the facebook example he gave is actually a case where the authoring environment should be XML, and then serialised as HTML after processing

[11:15:00.0000] <hsivonen>
Lachy: well, yeah, except that XML is too hard

[11:16:00.0000] <Lachy>
do you really think it's too hard for the developers that FBML is targetted at?

[11:17:00.0000] <othermaciej>
There's really multiple reasons people have for wanting namespaces in HTML, and they don't necessarily all point to the same solution

[11:17:01.0000] <hsivonen>
othermaciej: agreed

[11:17:02.0000] <othermaciej>
1) support for specific well-known XML vocabularies (perhaps best handled via special-casing of the root elements, but needs a good fallback story)

[11:18:00.0000] <hsivonen>
Lachy: did you notice Sam's PHP spaghetti example? it has all the same problems regarding well-formedness as PHP spaghetti spewing XHTML

[11:18:01.0000] <Philip`>
https://secure.mysociety.org/cvstrac/fileview?f=mysociety/pb/phplib/pbfacebook.php&v=1.1

[11:18:02.0000] <othermaciej>
2) support for well-known custom tags without regard to XML compatibility but with the ability for any third party to define an extension that is not expected to be understood natively by browsers

[11:19:00.0000] <othermaciej>
3) same as 2 but with the additional requirement of being able to correspond to a "namespaces in XML" infoset

[11:19:01.0000] <othermaciej>
(and to generate an XML namespace DOM)

[11:19:02.0000] <hsivonen>
othermaciej: 4) same as 3) with non-XHTML nodes

[11:20:00.0000] <othermaciej>
hsivonen: non-XHTML nodes?

[11:20:01.0000] <hsivonen>
(#3 is satisfied with all elements being in the XHTML namespace)

[11:20:02.0000] <othermaciej>
ah, I see, you could make custom elements in the XHTML namespace, yes

[11:21:00.0000] <othermaciej>
I guess my point is that the solution for #2 that has the best backward compatibility and best behavior in pre-HTML5 browsers is probably something that doesn't use XML namespaces at all

[11:23:00.0000] <Lachy>
hsivonen: I hadn't noticed the PHP example, but that's a problem with PHP and a good illustration of the danger of developing XML under tag soup conditions

[11:23:01.0000] <Philip`>
Does it matter how #2 is handled in pre-HTML5 browsers, since the extensions are (if they're like FBML) meant to be processed by the HTML5 tools on the server and are no longer present in what's served to the browser?

[11:24:00.0000] <hsivonen>
Philip`: Sam said he was targeting both browsers and non-browsers

[11:25:00.0000] <othermaciej>
it does matter if you want to serve it on the public World Wide Web

[11:26:00.0000] <hsivonen>
I'm not so convinced that distributed extensibility on the Web is a good idea. as opposed to extensibility by centralized clearing houses like the microformats community

[11:27:00.0000] <Philip`>
hsivonen: othermaciej's #2 was for things not understood by browsers, so perhaps there's room for separate browser (#1) vs non-browser (#2/#3) solutions

[11:28:00.0000] <othermaciej>
Philip`: I meant not natively understood by browsers with special handling; not necessarily not served to browsers

[11:28:01.0000] <othermaciej>
/me hopes that distinction is clear

[11:29:00.0000] <othermaciej>
I think IE's weird handling of unknown tags may break any hope of a backwards-compatible extension mechanism

[11:29:01.0000] <hsivonen>
whoa, http://hsivonen.iki.fi/producing-xml/ is almost 2 years old. (speaking of PHP and XML)

[11:30:00.0000] <Lachy>
I don't object to the principle of extensibility on the web, but I really think it needs to be done in a way that doesn't mess too much with the parsing. i.e. I don't think letting authors make up their own elements is a good idea

[11:31:00.0000] <othermaciej>
there does not seem to be a huge need for custom elements that are unknown to browsers, in publicly served content

[11:31:01.0000] <Lachy>
I think extension mechanisms using attributes can be successful, like microformats have used class and rel.

[11:32:00.0000] <othermaciej>
you would probably get better results by styling an existing semantic-free element like <span> or <div> with a class

[11:32:01.0000] <Lachy>
also something like role="", if it were introduced

[11:33:00.0000] <Philip`>
You could perhaps have UAs treat <x:y> the same way as HTML5 currently does (i.e. an element named "x:y"), except say UAs may support specific namespace prefixes (svg, math, etc) and parse them under different rules into an XML-like DOM. Then HTML5 browsers wouldn't break old content (<o:p>, <v:rect>, etc), but they could support new content (<svg:svg>), and the new content would degrade no worse than any other tag soup in IE, and non-browser tools cou

[11:33:01.0000] <Philip`>
...could support their own extensions

[11:33:02.0000] <othermaciej>
custom attributes seem more potentially useful

[11:33:03.0000] <othermaciej>
for cases where you want key/value data and would rather not parse it out by hand

[11:33:04.0000] <gsnedders>
Philip`: having an <svg> element has already been suggested without namespaces at an HTML document level, by simply changing the namespace that the elements are put within

[11:34:00.0000] <othermaciej>
what's the o namespace referred to in <o:p>?

[11:34:01.0000] <Lachy>
I supposed we should also consider which use cases are solved by things like XBL, which can give all the practical benfits of extending the language, but without all the mess

[11:34:02.0000] <othermaciej>
is that a real commonly used namespace or just a popular random example?

[11:34:03.0000] <Philip`>
othermaciej: <o:p> is from Microsoft Office

[11:35:00.0000] <othermaciej>
Philip`: ewww

[11:35:01.0000] <Philip`>
(and it was in 1.7% of pages I looked at a while ago)

[11:35:02.0000] <Lachy>
Philip`: that seems a little low

[11:36:00.0000] <othermaciej>
Lachy: XBL+CSS lets you add new behavior and presentation, but it doesn't help you represent the custom semantics that you want to attach this styling and behavior to

[11:36:01.0000] <othermaciej>
Lachy: so it's kind of orthogonal

[11:37:00.0000] <othermaciej>
using XBL with <div class="foobar"> doesn't address any of the reasons people might want a <foobar> tag in their markup, whatever those might be

[11:38:00.0000] <othermaciej>
XBL also potentially creates a greater need for custom attributes

[11:38:01.0000] <Lachy>
hmm, I suppose, though it's not really clear what the use cases are for such extensibility on the client side

[11:39:00.0000] <Lachy>
if it's extensibility like microformats are doing, then their solution already works.

[11:40:00.0000] <Lachy>
maybe it's something like these JavaScript triggers http://www.alistapart.com/articles/scripttriggers/

[11:40:01.0000] <Lachy>
basically, custom attributes to extend existing elements

[11:41:00.0000] <othermaciej>
it seems like XBL is more likely to increase rather than reduce the desire to send custom elements and attributes

[11:42:00.0000] <Lachy>
really?

[11:42:01.0000] <Lachy>
why?

[11:42:02.0000] <othermaciej>
if you make some kind of complex custom control using XBL, you'll tend to think it should be its own tag

[11:42:03.0000] <othermaciej>
rather than binding it to <div class="something">

[12:10:00.0000] <othermaciej>
so was Molly actually just annoyed about that article anointing Zeldman as "King of Web Standards"?

[12:10:01.0000] <othermaciej>
is that the personal agenda she was referring to?

[12:20:00.0000] <Whiskey_M>
'lo

[12:21:00.0000] <Lachy>
what does 'lo mean?

[12:21:01.0000] <Whiskey_M>
as in hello

[12:22:00.0000] <Lachy>
ah

[12:25:00.0000] <Philip`>
It's the opposite of 'hi' and therefore means the same

[13:01:00.0000] <met_>
http://www.molly.com/2007/08/13/dear-what-wg-and-html-5-wg/

[13:09:00.0000] <othermaciej>
wow, molly sure likes all-caps boldface

[13:14:00.0000] <hsivonen>
hmm. having the spec on too servers seems to be perceived as a problem. but XBL is published by 2 orgs. as is RELAX NG and ODF

[13:15:00.0000] <hsivonen>
at least in the case of HTML 5 and XBL, the content is verbatim as opposed to ISO RELAX NG having gone through the ISO obfuscation process

[13:15:01.0000] <hsivonen>
s/too/two/

[13:53:00.0000] <molly>
Hi all

[13:53:01.0000] <molly>
I've posted some clarifications as many WHAT WG members were concerned about conversation on my blog, perhaps this will help http://www.molly.com/2007/08/13/dear-what-wg-and-html-5-wg/

[13:55:00.0000] <molly>
Hixie

[13:55:01.0000] <molly>
hsivonen

[13:55:02.0000] <molly>
kevinmarks

[13:55:03.0000] <molly>
lachy

[13:55:04.0000] <hasather_>
molly: s/Sevonin/Sivonen/

[13:56:00.0000] <molly>
hello?

[13:56:01.0000] <jgraham>
molly: hello (although I am not on your list :))

[13:57:00.0000] <molly>
hi jgraham

[13:57:01.0000] <molly>
hasather, thanks, I always do that to Henri!

[13:57:02.0000] <molly>
sorry Henri

[13:57:03.0000] <molly>
fixed no

[13:57:04.0000] <molly>
now

[13:58:00.0000] <jgraham>
One reason there are two lists dealing with HTML5 stuff is that the W3C policy prevents members of W3C-Member organisations who are not nominated representatives from providing feedback

[13:58:01.0000] <hasather_>
molly: now it's Sevonen which is still wrong :)

[13:59:00.0000] <molly>
ha! ok, fixed again.

[13:59:01.0000] <molly>
jgraham, that's a problem then that W3C policy needs to repair

[13:59:02.0000] <molly>
having two lists at two trajectories is a fragmentation

[14:00:00.0000] <jgraham>
I agree that the dual-list situation is not ideal. But I don't think it's more harmful than the alterntaives

[14:00:01.0000] <gsnedders>
then we have the issue of the time taken to change the process document

[14:00:02.0000] <molly>
I think it's very harmful. 

[14:00:03.0000] <jgraham>
Why?

[14:01:00.0000] <jgraham>
Most discussion now happens on public-html

[14:01:01.0000] <molly>
which is authorotative? How is that expressed to the wider public?

[14:01:02.0000] <molly>
gsnedders: agree that the time issue with w3c is awful.  But that's why my initial comments were explicitly not to the WHAT WG

[14:01:03.0000] <molly>
but to the W3C and to WaSP

[14:02:00.0000] <jgraham>
Neither. The spec is authoritative. public-html is authoritative in producing deliverables insofar as the W3C brand carries much more weight than the WHATWG brand

[14:02:01.0000] <gsnedders>
molly: but it's why we can't do anything about the issue with the W3C mailing list situation. It'd be nice if we were just allowed to use the existing mailing list.

[14:03:00.0000] <gsnedders>
(which to this day still has more subscribers)

[14:03:01.0000] <molly>
so how do we make that allowable? The W3C bears enormous responsibility in accommodating WHAT WG process as it relates to HTML 5. 

[14:03:02.0000] <jgraham>
(or rather the HTMLWG, which conducts its business via public-html, is authoritative)

[14:03:03.0000] <hsivonen>
molly: hi

[14:03:04.0000] <molly>
Hi Henri!

[14:03:05.0000] <molly>
sorry to mis-spell your name so many times, I think I corrected it now :D

[14:03:06.0000] <hsivonen>
molly: no problem

[14:04:00.0000] <gsnedders>
molly: we'd have to get the W3C group re-chartered

[14:05:00.0000] <molly>
so what's the point of HTML 5 serialization under W3C then?

[14:05:01.0000] <hsivonen>
molly: I still stand by what I said earlier (not in the comments of your previous post but even earlier): HTML 5 fixes many things in HTML 4.01, but releasing merely the fixes is not a good idea as it would slow down launching new features

[14:05:02.0000] <jgraham>
/me doesn't really understand that question

[14:05:03.0000] <gsnedders>
/me doesn't either

[14:05:04.0000] <hsivonen>
which need to be launched to keep HTML competitive

[14:06:00.0000] <molly>
I understand the need and desire to keep HTML moving forward. I have no problem with that

[14:06:01.0000] <molly>
but I think there's an audience that people are missing

[14:06:02.0000] <molly>
and that's why I keep digging around this issue

[14:06:03.0000] <jgraham>
The point of the HTML 5 serialization compared to what? An XML serialisation? 

[14:06:04.0000] <gsnedders>
I, in many ways, would've preferred we developed the spec separate from the W3C and just submit it when completed, thereby avoiding the entire over-bureaucracy of the W3C

[14:06:05.0000] <jgraham>
Compared to leaving HTML stagnant?

[14:06:06.0000] <molly>
and that is the fact that we don't even have stable specs implemented as is

[14:07:00.0000] <hsivonen>
molly: merely releasing bug fixes is not interesting enough to justify a new thing to people who don't know how important the fixes are

[14:07:01.0000] <molly>
no, jgraham, i didn't mean that

[14:07:02.0000] <molly>
I meant under w3c

[14:07:03.0000] <molly>
not that WHAT WG wouldn't continue its work

[14:07:04.0000] <molly>
but if the W3C is an impediment

[14:07:05.0000] <hsivonen>
molly: if anything, new exciting features are needed for marketing and the bug fixes are delivered on the side

[14:07:06.0000] <molly>
a detriment

[14:07:07.0000] <jgraham>
From who's point of view?

[14:07:08.0000] <molly>
well, this is all well and good but what do you say to the poor suckers of the world who still don't know what a DOCTYPE is much less why they should care?

[14:07:09.0000] <jgraham>
Rejoice!

[14:08:00.0000] <molly>
we forget how we see this versus how the rest of the world sees it

[14:08:01.0000] <hsivonen>
molly: I'd give them a link to my doctype page :-)

[14:08:02.0000] <gsnedders>
your markup will be parsed consistently!

[14:08:03.0000] <jgraham>
For HTML5 is designed to make your life easier

[14:08:04.0000] <molly>
MY life?

[14:08:05.0000] <jgraham>
(at least, if I can help it at all... ;))

[14:08:06.0000] <hsivonen>
/me has been educating people about doctypes since 2000

[14:08:07.0000] <molly>
yes, Henri, and I often to send people to you

[14:08:08.0000] <jgraham>
The life of people who know little about doctypes + etc.

[14:08:09.0000] <gsnedders>
/me once ended up doing HTML in computing at school.

[14:08:10.0000] <molly>
however, perhaps the magnitude of "them"

[14:08:11.0000] <gsnedders>
I ended up teaching the class, not the teacher

[14:08:12.0000] <molly>
um, these are people who work the Web today

[14:08:13.0000] <molly>
yes, exactly

[14:08:14.0000] <molly>
thank you

[14:09:00.0000] <molly>
now you see my pov

[14:09:01.0000] <molly>
I'm the educator

[14:09:02.0000] <molly>
that's the role I play

[14:09:03.0000] <molly>
I have to take this and make it make sense to people

[14:09:04.0000] <molly>
not only who are doing daily work

[14:09:05.0000] <hsivonen>
molly: the hardest part in that is this: http://annevankesteren.nl/2007/04/html-red-pill

[14:09:06.0000] <molly>
but who can't figure out how the fuck to implement anything as it is much less as we all see it through our various colored glasses

[14:09:07.0000] <jgraham>
Right, so one thing that has been happening is to simplify needness complexity

[14:10:00.0000] <jgraham>
like now we have <!DOCTYPE html> rather than the HTML4 cruft

[14:10:01.0000] <molly>
I know

[14:10:02.0000] <jgraham>
so it's easier to learn what a doctype is

[14:10:03.0000] <molly>
I'm not so sure that's true

[14:10:04.0000] <jgraham>
Why not?

[14:10:05.0000] <gsnedders>
jgraham: maybe not what it is

[14:10:06.0000] <jgraham>
It's a magic string...

[14:11:00.0000] <molly>
most of the people working on specs seem to not realize that the majority of people working in the front lines of web design and dev

[14:11:01.0000] <gsnedders>
a doctype is a far from simple concept

[14:11:02.0000] <molly>
are not trained, do not understand

[14:11:03.0000] <molly>
do not get exposed to the ideas

[14:11:04.0000] <molly>
and just do "what works"

[14:11:05.0000] <gsnedders>
jgraham: what's magic about it?

[14:11:06.0000] <hsivonen>
molly: how can we do anything if we assume no learning at all?

[14:11:07.0000] <jgraham>
That depends what you mean by "what"

[14:11:08.0000] <molly>
in fact, one of the most common questions in XHTML classes?  "if elements are supposed to be lower case, why is DOCTYPE upper case?"

[14:11:09.0000] <jgraham>
:)

[14:11:10.0000] <molly>
:: bangs head on desk ::

[14:12:00.0000] <molly>
back to Anne's "no sgml" point

[14:12:01.0000] <othermaciej>
hi molly, thanks for visiting

[14:12:02.0000] <molly>
hi! other maciej

[14:12:03.0000] <molly>
are you really another maciej?

[14:12:04.0000] <molly>
or the one I'm familiar with? :D

[14:12:05.0000] <othermaciej>
probably the one you are familiar with

[14:12:06.0000] <hsivonen>
molly: HTML5 solves that. no more explaining XHTML-as-text/html that doesn't have an explanation that makes sense to students

[14:12:07.0000] <othermaciej>
but there is a different maciej on this server

[14:12:08.0000] <gsnedders>
/me ponders where the other comes from

[14:12:09.0000] <othermaciej>
damn him for stealing my name

[14:12:10.0000] <gsnedders>
I'm too slow, dangnamit!

[14:12:11.0000] <molly>
hehe

[14:13:00.0000] <Hixie>
molly: commented on your blog

[14:13:01.0000] <molly>
Hixie, hi there

[14:13:02.0000] <molly>
thank you!

[14:13:03.0000] <jgraham>
gsnedders: If the class were sufficiently competent you could explain that it's needed to turn off some odd behaviour needed for back-compat

[14:14:00.0000] <gsnedders>
jgraham: what's wrong with the odd behaviour?

[14:14:01.0000] <hsivonen>
gsnedders: odd behavior is counter-intuitive

[14:14:02.0000] <jgraham>
It's odd :)

[14:14:03.0000] <hsivonen>
gsnedders: counter-intuitive is hard

[14:14:04.0000] <gsnedders>
/me is trying to remember how he explained it

[14:14:05.0000] <Hixie>
molly: to answer your question "so what's the point of HTML 5 serialization under W3C then?", the answer is that we had to move the spec development to have it covered by the w3c's patent policy in order to get microsoft to review the spec.

[14:14:06.0000] <gsnedders>
hsivonen: HTML is odd regardless, though

[14:15:00.0000] <Philip`>
People seem to be very good at remembering <html> and <head> and <body> - what makes <!doctype html> so much harder?

[14:15:01.0000] <Hixie>
molly: (there were other minor benefits of having it under the w3c, but that was by far the main reason)

[14:15:02.0000] <molly>
Hixie, I am following the lists but part of my frustration is the volume. I will be present at the working group during the tech plenary in november

[14:15:03.0000] <jgraham>
Anyway molly, things like the parsing algorithm will, eventually, make things better for authors and they won't even know about it

[14:15:04.0000] <molly>
yes, Chris filled me in on that part

[14:15:05.0000] <othermaciej>
having it covered by the w3c patent policy is a benefit to all interested parties

[14:15:06.0000] <molly>
but there is still expressed concern over that

[14:16:00.0000] <Hixie>
othermaciej: indeed

[14:16:01.0000] <molly>
I'm not well-versed on IP/Patent issues so I really wouldn't want to comment on that

[14:16:02.0000] <othermaciej>
the mail volume is indeed hard to follow

[14:16:03.0000] <molly>
but I understand the general points

[14:16:04.0000] <othermaciej>
it would be good if it was easier to follow only a subset of issues, for those who have strong interest and expertise in particular areas but limited time

[14:16:05.0000] <othermaciej>
I'm not sure how to make that happen though

[14:16:06.0000] <molly>
that's a really good point maciej

[14:17:00.0000] <othermaciej>
having the whatwg list is good as a forum for those who are for whatever reason unwilling or unable to join the HTML WG

[14:17:01.0000] <Hixie>
you don't really have to follow the lists to contribute, though

[14:17:02.0000] <othermaciej>
but the balance of useful work continues to shift more towards the HTML WG list

[14:17:03.0000] <molly>
Hixie: I know, and as I said, I'll be there in person in November

[14:17:04.0000] <Hixie>
in the public-html list you need but send out your feedback, then write it up in the wiki according to that e-mail i sent

[14:17:05.0000] <Hixie>
and then it'll be taken into account

[14:18:00.0000] <molly>
sot hat's a good starting point for me in terms of any actual participation

[14:18:01.0000] <molly>
I have lots of reasons to be hesitant, as you can see

[14:18:02.0000] <jgraham>
The flip side is public-html probably has a poorer signal/noise than the WHATWG list still

[14:18:03.0000] <Hixie>
molly: i'm not sure i will, we'll see (i don't see much point to face-to-face meetings, they cost a lot in carbon offsets and they don't tend to bring the spec forward much)

[14:18:04.0000] <molly>
Hixie: Funny, i find it the opposite. The effort and commitment to show up, pay the way, and face each other can be very useful IMO

[14:19:00.0000] <gsnedders>
molly: there are plenty of people, like myself, who simply cannot afford to go to meetings for whatever reason

[14:19:01.0000] <molly>
I understand that gsnedders

[14:19:02.0000] <molly>
that's a realistic barrier that's an unfortunate one

[14:19:03.0000] <molly>
but we all work in different ways

[14:20:00.0000] <othermaciej>
face-to-face meetings can be useful on a human and community level, but are rarely an efficient way to make progress on the spec

[14:20:01.0000] <gsnedders>
molly: which thereby has the affect that everything has to go through the mailing lists after a real meeting

[14:20:02.0000] <molly>
Those here who have met me know that I'm a very personable human and enjoy the F2F work a great deal

[14:20:03.0000] <molly>
and benefit most from it

[14:20:04.0000] <othermaciej>
meeting people in person aids mutual understanding and helps soften future online disputes

[14:20:05.0000] <molly>
this is my feeling too, maciej

[14:20:06.0000] <Hixie>
othermaciej: yeah, i would be interested in an unconference style meeting, if we could get a good chunk of the group to attend (like, 50-100)

[14:20:07.0000] <molly>
that's what works for me

[14:20:08.0000] <molly>
how could something like that work? Be affordable for people and actually happen?

[14:20:09.0000] <molly>
I'd love to see something of that nature

[14:21:00.0000] <molly>
"unconference style meeting"

[14:21:01.0000] <Hixie>
othermaciej: i just don't want to sit around a table for 3 days while people talk about how their pet feature is a good idea but without even having explained what they're trying to solve and without any evidence

[14:21:02.0000] <hsivonen>
molly: I guess the big problem is where to have it: Paris, Boston or the Bay Area

[14:21:03.0000] <Hixie>
to support their claims

[14:21:04.0000] <jgraham>
hsivonen: Why one of those

[14:21:05.0000] <molly>
Hixie: I surely can't disagree with you on THAT! :D

[14:21:06.0000] <jgraham>
?

[14:22:00.0000] <Hixie>
molly: i don't think i can think of a single f2f meeting i've been to that hasn't involved hours of that :-(

[14:22:01.0000] <othermaciej>
Hixie: I would echo your loathing of the standard w3c format

[14:22:02.0000] <hsivonen>
jgraham: the hot spots of Europe, East Coast and West Coast

[14:22:03.0000] <molly>
within the W3C format yesw

[14:22:04.0000] <molly>
well, Henri - we had that informal day in Paris

[14:22:05.0000] <molly>
something could grow from that sort of thing

[14:22:06.0000] <Hixie>
gotta go, bbiab

[14:22:07.0000] <molly>
that approach, let's say

[14:23:00.0000] <molly>
round table discussion external from W3C, corporate agendas

[14:23:01.0000] <molly>
etc.

[14:23:02.0000] <molly>
no one's saying any of this is easy or has a fast solution

[14:23:03.0000] <molly>
my real concern is what happened to WaSP

[14:23:04.0000] <molly>
and the failings of the W3C

[14:23:05.0000] <gsnedders>
less and less appears to be happening at WaSP

[14:23:06.0000] <molly>
to provide a fair forum and process

[14:24:00.0000] <jgraham>
In any case the problem is not where to have it, it's getting enough people to attend

[14:24:01.0000] <molly>
gsnedders: As its retired mom, you know it breaks my heart

[14:24:02.0000] <molly>
jgraham: maybe it has to take place in several locations and we synch up somehow

[14:24:03.0000] <molly>
this way it can be global and reduce the carbon concerns that Hixie pointed to

[14:25:00.0000] <gsnedders>
and also, most likely, get more attendees

[14:25:01.0000] <gsnedders>
(due to lower costs of travel, etc)

[14:25:02.0000] <molly>
yes

[14:25:03.0000] <molly>
absolutely

[14:25:04.0000] <hsivonen>
jgraham: exactly, the three places I mentioned probably are the best for attendance, but still people are reluctant to travel across the Atlantic or from coast to coast. so no matter what the place is, it is always wrong from some point of view

[14:25:05.0000] <molly>
that's why maybe several locations

[14:25:06.0000] <molly>
where there's onsite, linked up

[14:25:07.0000] <molly>
Paris, SF, etc.

[14:25:08.0000] <gsnedders>
(though if there's one in Paris, and it's in late Oct, maybe I'll be there :P)

[14:25:09.0000] <jgraham>
molly: Even if you have three locations, most people would have expenses of >£100 ($200) to attend

[14:26:00.0000] <jgraham>
which is a lot for people not being paid for this stuff

[14:26:01.0000] <molly>
hey, you're preaching to the soprano in the choir here

[14:26:02.0000] <molly>
I've spent a fortune in my own money

[14:26:03.0000] <molly>
on standards-related efforts

[14:26:04.0000] <molly>
believe me, I know

[14:26:05.0000] <gsnedders>
what about students, who simply don't have their own money to spend?

[14:27:00.0000] <molly>
there has to be some kind of relay

[14:27:01.0000] <molly>
clearly we'd be synched up

[14:27:02.0000] <molly>
and on IRC

[14:27:03.0000] <othermaciej>
maybe we could get some of the big corporations involved to provide travel stipends for at least some volunteer contributors

[14:27:04.0000] <molly>
and there can be some kind of moderated approach where people who could not do it F2F

[14:27:05.0000] <molly>
and maciej - I would be willing to knock on a lot of doors to try and get some resources for that

[14:28:00.0000] <jgraham>
/me should mention he is very much in favour of a f2f since it would help sort out some of the most persistent culture problems on public-html

[14:28:01.0000] <gsnedders>
othermaciej: end up with people screaming "I DESERVE IT MORE THAN HIM!"

[14:28:02.0000] <molly>
gsnedders: yes but there could be specifics put into place

[14:28:03.0000] <jgraham>
gsnedders: Maybe that's not a problem

[14:28:04.0000] <molly>
such as agreements that a sponsored participation requires an equal action (ie, involvement for a year with the group, etc.)

[14:28:05.0000] <othermaciej>
gsnedders: I'm not too worried about that

[14:28:06.0000] <molly>
neither am I to be honest

[14:29:00.0000] <molly>
there are ways to manage that

[14:29:01.0000] <jgraham>
People have to compete for funding all the time

[14:29:02.0000] <molly>
but I think we're on to something here

[14:29:03.0000] <gsnedders>
wasn't it brought up before when we last discussed a f2f on public-html?

[14:29:04.0000] <molly>
this would be third party

[14:29:05.0000] <molly>
or non party

[14:29:06.0000] <molly>
I should say

[14:29:07.0000] <jgraham>
gsnedders: Yes. I think some people didn't like it

[14:30:00.0000] <jgraham>
but maybe they were wrong

[14:30:01.0000] <molly>
in other words, this is not W3C, not WHAT WG, not WaSP, not Microsoft, Mozilla, etc.

[14:30:02.0000] <hsivonen>
the problem with the mountain view f2f was that it lacked any stated purpose

[14:30:03.0000] <jgraham>
or it was pitched in the wrong way

[14:30:04.0000] <molly>
well, a big plate of food for thought, I Think

[14:31:00.0000] <gsnedders>
arguably the state of the politeness of some is a purpose on its own, now

[14:31:01.0000] <molly>
yes

[14:31:02.0000] <molly>
yes yes yes

[14:32:00.0000] <molly>
but I wonder if some F2F time can help that. I mean, not just sitting around the table arguing, but gaining real insight into perspectives. That, at least for me, comes through human discourse 

[14:32:01.0000] <molly>
that is not guided by process

[14:32:02.0000] <gsnedders>
molly: I think just being dumped F2F might help on its own. Plenty of people are acting far worse than you ever would offline.

[14:32:03.0000] <molly>
I know. I think this is why I'm so upset

[14:33:00.0000] <molly>
I've got long-term friends and colleagues fighting with each other

[14:33:01.0000] <molly>
when clearly they have common goals as well as uncommon ones

[14:33:02.0000] <molly>
it disturbs me very much

[14:33:03.0000] <molly>
to see people getting that frustrated

[14:33:04.0000] <jgraham>
hmm. have to go for a bit.

[14:33:05.0000] <molly>
I don't want to add to the frustration, I didn't realize what I was stepping in honestly

[14:33:06.0000] <hsivonen>
molly: tantek had insightful things to say about the culture clash

[14:34:00.0000] <molly>
so now I'd like to help reduce the frustration

[14:34:01.0000] <hsivonen>
/me looks up the URL

[14:34:02.0000] <molly>
yes, he and I had a long talk about it last night Henri

[14:34:03.0000] <molly>
I thought he made some excellent points about the various factions

[14:34:04.0000] <hsivonen>
http://krijnhoetmer.nl/irc-logs/whatwg/20070812#l-123

[14:34:05.0000] <molly>
but that's still no excuse for what Tantek himself points out as extremely negative behavior

[14:35:00.0000] <molly>
not from anyone, including myself. I get frustrated too

[14:35:01.0000] <molly>
but the accessibility guys, for one

[14:35:02.0000] <molly>
that stuff has to stop

[14:35:03.0000] <molly>
I mean, you'll never get Joe Clark to stop

[14:35:04.0000] <molly>
but that's Joe

[14:35:05.0000] <molly>
however, there have to be some human connections made here between those factions

[14:36:00.0000] <molly>
or more accurately, those fractions (badumpump) ;-)

[14:36:01.0000] <gsnedders>
I think one thing that people need to get past is that evangelism is not enough on its own. Too many people are pretty much saying that any accessibility feature MUST NOT be removed, without any justification.

[14:36:02.0000] <molly>
I think removing things is part of the fear and concern over HTML 5 in general

[14:37:00.0000] <molly>
why do things have to be "removed" per se

[14:37:01.0000] <hsivonen>
it seems to me that not all "accessibility guys" are the same. and there are some who don't think HTML 5 is all bad. but I understand that they don't want to engage in the smear discussions

[14:37:02.0000] <gsnedders>
my posting has gone down and down the more and more this has all taken over

[14:37:03.0000] <gsnedders>
because there's very little real discussion happening

[14:37:04.0000] <molly>
yes, and that's a problem!

[14:37:05.0000] <molly>
signal / noise

[14:38:00.0000] <gsnedders>
not just that, it's that there is barely any signal at all, now

[14:38:01.0000] <molly>
Everyone who cares, I think, feels that

[14:38:02.0000] <gsnedders>
I've been posting everything bar HTML WG spec review to the WHATWG list

[14:40:00.0000] <hsivonen>
molly: part of the problem is that public-html is so unpleasant that the signal get routed around it (e.g. to here)

[14:40:01.0000] <molly>
I think I'd also like to see some status stuff

[14:40:02.0000] <molly>
like "these are the key points"

[14:40:03.0000] <hsivonen>
s/get/gets/

[14:40:04.0000] <molly>
I'd help with that - but writtten for the masses

[14:40:05.0000] <molly>
not for the groups

[14:41:00.0000] <molly>
that might help cause less concern in the broader community?

[14:41:01.0000] <molly>
right now, that noise is frightening people

[14:41:02.0000] <molly>
or overwhelming them, even our best and brightest

[14:43:00.0000] <gsnedders>
the other interesting factoid is that as the volume of email has increased, the time taken for me to read them has gone down, simply as more and more of it is pointless and takes scarcely any time to deal with

[14:43:01.0000] <gsnedders>
molly: what sort of status things?

[14:43:02.0000] <molly>
some way of sorting through the noise

[14:43:03.0000] <molly>
to get to the signal

[14:44:00.0000] <molly>
and clarify those points on a regular basis to the broader public

[14:44:01.0000] <gsnedders>
so some sort of summary about what is going on?

[14:44:02.0000] <molly>
exactly

[14:44:03.0000] <gsnedders>
and do it how frequently? when needed?

[14:45:00.0000] <molly>
I think an update now would be great, something to provide the community at large with more insight into the signal and get the frustration level that the noise is creating down a bit

[14:45:01.0000] <molly>
I think another update around the Tech Plenary time would also be good

[14:45:02.0000] <molly>
and if needed in between

[14:45:03.0000] <gsnedders>
when is the Tech Plenary?

[14:45:04.0000] <molly>
November

[14:46:00.0000] <hsivonen>
molly: there's no clear WG level going on. but watching Hixie here and watching the svn commit messages gives an idea of the goings on with the spec

[14:46:01.0000] <molly>
http://www.w3.org/2007/11/TPAC/overview.html

[14:47:00.0000] <molly>
like the Twitter updates? Those are cool. But how about annotated every few weeks

[14:47:01.0000] <molly>
that sort of thing

[14:47:02.0000] <hsivonen>
molly: btw, one PR problem is that since Hixie doesn't really do firefighting in spec writing, the time it takes from an issue flaring up to the spec changing can be in the order of months

[14:48:00.0000] <hsivonen>
molly: so even if Hixie says here that a bad part of the spec sucks, people still keep thinking that the spec is destined to remain sucky on that point 

[14:49:00.0000] <Hixie>
can even take years

[14:49:01.0000] <hsivonen>
(I think not doing firefighting is the reasonable way, but it still creates the wrong impressions sometimes)

[14:50:00.0000] <hsivonen>
molly: so people who don't follow along here get outraged about <font> again and again

[14:50:01.0000] <hsivonen>
molly: I'm here, so I just leave <font> unimplemented in the conformance checker and trust that the part will go away in due course

[14:50:02.0000] <Hixie>
i should go around adding red boxes to the most controversial parts

[14:50:03.0000] <Hixie>
that might help

[14:50:04.0000] <hsivonen>
Hixie: yeah

[14:51:00.0000] <molly>
Something of that nature, yes!

[14:51:01.0000] <molly>
good idea

[14:51:02.0000] <Hixie>
can you send a mail to me or one of the lists listing the controversial parts and suggesting i add red boxes? i'll try to do it after my next commit (the alt text revamp)

[14:51:03.0000] <Hixie>
i have to go get my bike fixed first

[14:53:00.0000] <gsnedders>
g'nite

[14:53:01.0000] <gsnedders>
but, molly, if you want any help to do such an update, I'm more than welcome to do so

[14:54:00.0000] <molly>
absolutely!

[14:54:01.0000] <molly>
I'd totally like to do that

[14:54:02.0000] <molly>
HTML 5 news capsules, like Hixie does on Twitter but with annotations

[14:55:00.0000] <gsnedders>
molly: I'm on here pretty much all the time (though not always physically here, but I'll read anything addressed to me), and there are other contact details listed on my blog <http://geoffers.uni.cc/the-author/>

[14:56:00.0000] <gsnedders>
(that design is almost two years old. wow.)

[14:56:01.0000] <Hixie>
anyone should feel free to post on the WHATWG blog if they want to post regular updates

[14:56:02.0000] <Hixie>
in fact i highly encourage it

[14:57:00.0000] <Hixie>
just sign up to the blog and then post (and ask Lachy to set you up as an editor so you can actually post the entries once you've drafted them, we had to turn that off because spammers were using it)

[14:57:01.0000] <molly>
but see, that goes back to the issue of fractioning

[14:57:02.0000] <molly>
Shouldn't that really be also on the W3C?

[14:57:03.0000] <Hixie>
well, feel free to post on the w3c blog too :-)

[14:57:04.0000] <molly>
hahaha

[14:57:05.0000] <Hixie>
or instead :-)

[14:57:06.0000] <Hixie>
but you'll have to speak to DanC about that, i don't know how to post to that blog

[14:58:00.0000] <Hixie>
it's not an open blog like the whatwg's

[14:58:01.0000] <molly>
yep

[14:58:02.0000] <Hixie>
my point is just that you should just go ahead and do things

[14:58:03.0000] <molly>
Hixie, I think my track record proves I learned that one a long time ago ;-)

[14:59:00.0000] <molly>
however, battles must be picked carefully

[14:59:01.0000] <molly>
that's my concern.

[14:59:02.0000] <molly>
for me, personally. That's why I think WaSP has a responsibility here.

[14:59:03.0000] <Hixie>
well, i don't care who does the updates or where

[14:59:04.0000] <Hixie>
i just don't have time to edit the spec _and_ do the updates

[14:59:05.0000] <Hixie>
so someone else has to do it

[14:59:06.0000] <molly>
no of course not

[14:59:07.0000] <Hixie>
it can be you, or it can be someone else, that's not my problem :-)

[15:00:00.0000] <molly>
getting the highlighting done would be an awesome thing to help out too

[15:00:01.0000] <molly>
so you committed to that

[15:00:02.0000] <molly>
I'll commit to working with whoever here wants to work on

[15:00:03.0000] <molly>
summaries

[15:00:04.0000] <molly>
and we can cross blog 'em

[15:00:05.0000] <hsivonen>
also http://people.w3.org/mike/planet/html5/

[15:00:06.0000] <molly>
which is the best idea to get the word out to broad communities anyway

[15:01:00.0000] <molly>
Henri: thanks!

[15:01:01.0000] <Hixie>
(i recommend starting by assuming no-one will help out, because usually people only help out once it is proven that something is happening)

[15:03:00.0000] <molly>
Hixie: I appreciate your kind guidance. 

[15:03:01.0000] <molly>
I'm pretty used to herding cats.

[15:04:00.0000] <molly>
and occasionally effective at it, too :D

[15:04:01.0000] <Hixie>
:-)

[15:04:02.0000] <Hixie>
bbiab

[15:06:00.0000] <gsnedders>
molly, Hixie: IMO such a blog should be separate to both WHATWG and the W3C

[15:06:01.0000] <molly>
gsnedders: I think if it's just updates 

[15:06:02.0000] <molly>
it is best in as many places as possible

[15:07:00.0000] <gsnedders>
I think it's both or separate. People will always conclude something wrong if it is just in one place.

[15:16:00.0000] <hsivonen>
molly: thanks for following up to my question

[15:16:01.0000] <hsivonen>
bed time now

[15:16:02.0000] <hsivonen>
nn

[15:18:00.0000] <molly>
night Henri!

[15:18:01.0000] <molly>
thank you 

[15:20:00.0000] <gsnedders>
time for me to go sleep now, too (I really mean it, this time)

[15:32:00.0000] <molly>
night :D

[15:32:01.0000] <molly>
time for me to get ready for a meeting - thanks everyone. I'll be interested as to where this goes, and I shall visit again very soon


2007-08-14
[21:10:00.0000] <Hixie>
Lachy: yt?

[21:10:01.0000] <Lachy>
yeah

[21:11:00.0000] <Hixie>
someone submitted a blog entry

[21:11:01.0000] <Hixie>
(just fyi, i'm sure you'll see the e-mail in due course!)

[21:13:00.0000] <Lachy>
I haven't yet. Who sent you the mail?

[21:13:01.0000] <Hixie>
it was sent to the list

[21:15:00.0000] <Lachy>
oh, I thanks for pointing it out.  It would have been a while before I read it

[21:16:00.0000] <Lachy>
there's actually 2 waiting for publishing.  One is a short description and a link to http://blog.to-be.pl/2007/08/11/nowe-elementy-w-html-5/

[21:16:01.0000] <Lachy>
the other is about ruby

[21:16:02.0000] <Hixie>
cool

[21:16:03.0000] <Lachy>
not sure what language that is, though

[21:17:00.0000] <Lachy>
I like to include the language in the title, if the post isn't in english

[21:17:01.0000] <Hixie>
.pl is poland

[21:17:02.0000] <Hixie>
i think

[21:17:03.0000] <othermaciej>
that's Polish

[21:17:04.0000] <Lachy>
oh, I didn't notice the domain

[21:17:05.0000] <Hixie>
(btw, i recommend not reviewing the posts -- so long as they're not spam, i'd just post them)

[21:18:00.0000] <Hixie>
(after all, we only added the review step because of spammers)

[21:18:01.0000] <othermaciej>
it's about "new elements in html5"

[21:18:02.0000] <Lachy>
ah, that's a translation of that IBM article I read recently

[21:18:03.0000] <Lachy>
ah, there's a link to it at the bottom http://www-128.ibm.com/developerworks/library/x-html5/

[21:19:00.0000] <othermaciej>
I can read Polish

[21:19:01.0000] <othermaciej>
although technical terms sometimes confuse me

[21:26:00.0000] <Lachy>
oops! I should have set the post slug before publishing - it's a bit long. http://blog.whatwg.org/xhtml-5-will-have-the-only-usable-implementation-of-ruby-markup

[21:26:01.0000] <Lachy>
oh well, too late

[21:26:02.0000] <Hixie>
heh

[23:40:00.0000] <othermaciej>
which is up to #3 on my todo queue

[02:46:00.0000] <Hixie>
sigh

[02:46:01.0000] <Whiskey_M>
what's up?

[02:46:02.0000] <Hixie>
i scan several billion documents, and robert burns isn't happy with the quality of my research

[02:46:03.0000] <zcorpan_>
Hixie: you're not trying hard enough ;)

[02:47:00.0000] <zcorpan_>
/me hides

[02:47:01.0000] <Whiskey_M>
I hope you did them by hand ;-)

[02:47:02.0000] <krijnh>
Yeah, we should get ourselves a new editor ;)

[02:47:03.0000] <Hixie>
and the other reply to my last public-html mail tells me that i shouldn't be making decisions

[02:47:04.0000] <krijnh>
Robert Burns likes writing..

[02:47:05.0000] <Hixie>
oh well

[02:59:00.0000] <hendry>
/me sighs at the weather

[03:07:00.0000] <Hixie>
/me updates the <img alt> section

[03:11:00.0000] <karlUshi>
I think what robert burns is saying

[03:11:01.0000] <karlUshi>
is that stats are not enough

[03:12:00.0000] <karlUshi>
stats give us a rough evaluation of quantity, but not the quality. 

[03:12:01.0000] <Hixie>
hah, the great alt="" revamp got the auspicious revision number r1000.

[03:12:02.0000] <karlUshi>
quality being what are the processes which have led to this stats. How authoring tools operate etc

[03:12:03.0000] <Hixie>
karlUshi: except in the case he's talking about, the stats were specifically about determining the quality of the content.

[03:13:00.0000] <gsnedders>
Hixie: scan the whole google cache. kthxbai.

[03:13:01.0000] <Hixie>
that's not far from basically what i'm doing actually (i work for google)

[03:13:02.0000] <karlUshi>
quality (in my sentence above) = using authoring tool by hand and sees how the markup is produced in an user interaction.

[03:14:00.0000] <karlUshi>
s/an user/a user/

[03:14:01.0000] <gsnedders>
Hixie: I'm aware of what you're doing. I'm just saying _whole_ :)

[03:14:02.0000] <Hixie>
(though i scan a subset of the whole thing usually for reasons of not wanting to wait three months to get the results)

[03:14:03.0000] <Hixie>
i've done the whole thing sometimes :-)

[03:14:04.0000] <gsnedders>
Hixie: I'm just wondering how you can satisfy Robert Burns better with the data you have access to

[03:14:05.0000] <karlUshi>
gsnedders: that would not solve the problems ;)

[03:14:06.0000] <gsnedders>
karlUshi: I know

[03:15:00.0000] <Hixie>
gsnedders: i don't think bigger numbers would make the slightest difference

[03:15:01.0000] <Hixie>
gsnedders: i'm really not sure what would

[03:15:02.0000] <Hixie>
ok, any immediate comments on the alt text revamp?

[03:15:03.0000] <gsnedders>
Hixie: can the entire web.

[03:15:04.0000] <karlUshi>
exactly because bigger numbers don't do any differences

[03:15:05.0000] <gsnedders>
*scan

[03:15:06.0000] <gsnedders>
even if that is technically impossible!

[03:15:07.0000] <karlUshi>
quantity versus quality. 

[03:15:08.0000] <Hixie>
statistically, what i've scanned is close enough to the entire web as makes no difference, to be honest

[03:16:00.0000] <gsnedders>
there is realistically no point in doing anything bigger, though

[03:16:01.0000] <Hixie>
right

[03:16:02.0000] <gsnedders>
you'd just get so similar results

[03:17:00.0000] <Hixie>
it's actually amasing how similar results others have been getting with just samples of a few thousand

[03:17:01.0000] <karlUshi>
Hixie: If you want different results, imho, you have to categorize

[03:18:00.0000] <Hixie>
and that's like 0.001% of what i've scanned

[03:18:01.0000] <karlUshi>
small random sample = big random sample with less confidence

[03:19:00.0000] <karlUshi>
what is interesting is what is hidden in the data when we just do samples

[03:19:01.0000] <karlUshi>
http://roslingsblogger.blogspot.com/

[03:22:00.0000] <virtuelv>
Hixie: re usemap: I'd like to see the opposite stats for what you did here, http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-August/012334.html

[03:22:01.0000] <virtuelv>
tell me which percentage breaks in which way

[03:23:00.0000] <virtuelv>
for some weird reason, that seems to work better as a justification for people

[03:24:00.0000] <Hixie>
virtuelv: well, go through the list of URLs there and categorise them into the various reasons they fail (or pass), and you'll have that data :-)

[03:24:01.0000] <karlUshi>
"the problem is not ignorance but preconceived ideas"

[03:25:00.0000] <Lachy>
Hixie: which UAs support <input useemap>?

[03:25:01.0000] <Hixie>
firefox and opera, for some definition of "support"

[03:26:00.0000] <Hixie>
(i mean, the html4 spec was no more detailed about what it should do than it is for anything else)

[03:26:01.0000] <Lachy>
ok, so the fact that IE doesn't support it should be a good reason for not bothering to include it

[03:26:02.0000] <virtuelv>
Hixie: neh, just mentioning for future reference.   I only follow certain parts of html5 work

[03:27:00.0000] <Lachy>
though I'm really confused by Robert's statement "Dropping client-side image maps for forms in a recommendation that is focussed on web applications seems completely counter-intuitive to me."

[03:27:01.0000] <Hixie>
yeah i didn't follow his e-mail at all

[03:27:02.0000] <Lachy>
I have no idea what use cases he's imaginign for it

[03:27:03.0000] <Whiskey_M>
is there a URL where his e-mail is stored?

[03:28:00.0000] <Lachy>
http://lists.w3.org/Archives/Public/public-html/2007Aug/0462.html

[03:28:01.0000] <Hixie>
http://lists.w3.org/Archives/Public/public-html/2007Aug/0462.html

[03:28:02.0000] <Hixie>
d'oh

[03:28:03.0000] <Whiskey_M>
'ta :)

[03:32:00.0000] <zcorpan_>
Hixie: i love the examples on <img alt> :D

[03:32:01.0000] <Hixie>
:-D

[03:35:00.0000] <Hixie>
there are about 10 in-jokes in those examples

[03:35:01.0000] <Hixie>
some obvious

[03:35:02.0000] <Hixie>
some very obscure

[03:36:00.0000] <gsnedders>
Hixie: I'm now going to get no work done today, just looking for them :P

[03:36:01.0000] <Hixie>
hah

[03:37:00.0000] <gsnedders>
at first glance I thought "XYZ.gif" read "YYZ.gif"

[03:37:01.0000] <Hixie>
XYZ.gif is one of the less obvious references

[03:37:02.0000] <gsnedders>
/me ponders

[03:37:03.0000] <gsnedders>
(and starts playing YYZ while doing so)

[03:39:00.0000] <Lachy>
heh, I can't wait till I get an iPhone so I can compare the reflectiveness of my apple logo :-)

[03:39:01.0000] <gsnedders>
Hixie: please don't say 1100670787_6a7c664aef is a joke.

[03:39:02.0000] <Hixie>
hah

[03:39:03.0000] <Hixie>
gsnedders: yup :-)

[03:39:04.0000] <gsnedders>
/me stabs Hixie 

[03:39:05.0000] <gsnedders>
sorry, but that's just evil.

[03:39:06.0000] <Hixie>
actually that one is one of the more funny ones

[03:39:07.0000] <Hixie>
and one of the easiest to track down

[03:39:08.0000] <gsnedders>
I can make several guesses about what it is

[03:40:00.0000] <gsnedders>
/me wonders if he can search such things for such an ID easily

[03:41:00.0000] <Hixie>
actually i don't know how you would track it down, come to think of it

[03:41:01.0000] <Hixie>
oh, i know

[03:42:00.0000] <Hixie>
aha, yes

[03:42:01.0000] <Lachy>
oh, don't reveal the answer yet, I want to try and figure them all out !

[03:42:02.0000] <Hixie>
i can do one search for a very specific, obvious string, on one site, to find the reference that's to

[03:42:03.0000] <gsnedders>
OBVIOUS!?

[03:43:00.0000] <gsnedders>
I really should work :\

[03:43:01.0000] <zcorpan_>
/me too

[03:43:02.0000] <krijnh>
/me too

[03:43:03.0000] <Hixie>
hah

[03:43:04.0000] <Lachy>
I'm trying to work out what "sequence of rounded squares with varying shades of green and bold white outlines" refers to.  That doesn't sound like the old WHATWG logo

[03:43:05.0000] <zcorpan_>
Hixie: stop wasting our time! ;)

[03:43:06.0000] <gsnedders>
we're so hard working here

[03:43:07.0000] <Hixie>
tee hee

[03:43:08.0000] <gsnedders>
just on the wrong things

[03:43:09.0000] <Hixie>
Lachy: that one is one of the other pretty funny ones

[03:44:00.0000] <Lachy>
oh, got it!

[03:45:00.0000] <Hixie>
hah

[03:45:01.0000] <gsnedders>
/me still ponders

[03:45:02.0000] <Lachy>
I missed the bit about the blue book, and then that reminded me of John's book

[03:46:00.0000] <gsnedders>
Lachy: shhhh

[03:46:01.0000] <Hixie>
assuming you all got the reference in the first example (and the other later one with the same text), then there are only three that are really possible to get

[03:46:02.0000] <gsnedders>
Hixie: so kind

[03:46:03.0000] <Hixie>
that's the XYZ one, the "consider for a moment" one, and the 1100670787_6a7c664aef one

[03:47:00.0000] <hsivonen>
/me found 1100670787_6a7c664aef

[03:47:01.0000] <Hixie>
teehee

[03:47:02.0000] <Hixie>
you're good

[03:47:03.0000] <gsnedders>
hmmm…

[03:47:04.0000] <gsnedders>
it'll be some ID specific to some site

[03:47:05.0000] <gsnedders>
(probably)

[03:48:00.0000] <Hixie>
(the other examples are either references that aren't funny, or just made up things based on the other in-jokes)

[03:48:01.0000] <gsnedders>
Hixie: but no, I didn't get that reference in the first example

[03:48:02.0000] <Hixie>
the first example should be immediately obvious to anyone here :-P

[03:49:00.0000] <gsnedders>
where it comes from, or what?

[03:49:01.0000] <Hixie>
yeah

[03:49:02.0000] <Hixie>
it's not funny

[03:49:03.0000] <gsnedders>
yeah, I got that straight away

[03:49:04.0000] <Hixie>
right

[03:49:05.0000] <gsnedders>
I simply glanced over that, looking too deeply

[03:49:06.0000] <Hixie>
hehe

[03:49:07.0000] <gsnedders>
/me needs to implement that, actually

[03:49:08.0000] <Lachy>
how is the XYZ company a joke?

[03:50:00.0000] <hsivonen>
I like the /images/logo example

[03:50:01.0000] <Hixie>
lachy: it's a reference to something quite old, and the third example from that series is based on it

[03:50:02.0000] <Hixie>
not saying it's especially funny mind you

[03:50:03.0000] <gsnedders>
/me laughs at what iTunes randomly plays

[03:50:04.0000] <gsnedders>
/me is listening to I Still Haven't Found What I'm Looking For by U2 from Rattle And Hum

[03:50:05.0000] <Hixie>
hah

[03:51:00.0000] <gsnedders>
myself and a friend have a running joke about iTunes being psychic. it _always_ does such things, but _only_ on OS X.

[03:51:01.0000] <Lachy>
Hixie: the 3rd example in the whole spec from the same series?

[03:52:00.0000] <Lachy>
or, actually, I'm not sure what you meant by "the third example from that series is based on it"

[03:53:00.0000] <gsnedders>
I can't get the ΑΒΓ joke at all

[03:53:01.0000] <Hixie>
Lachy: the third example in the xyz/alpha-beta-gamma series is a reference to something specific

[03:54:00.0000] <Hixie>
gsnedders: which one?

[03:54:01.0000] <gsnedders>
Hixie: both

[03:54:02.0000] <gsnedders>
Hixie: can't get the XYZ one either

[03:54:03.0000] <Hixie>
ah :-)

[03:54:04.0000] <gsnedders>
am I just too young? :)

[03:55:00.0000] <Hixie>
could be, it's reference to something from the late 90s

[03:55:01.0000] <Whiskey_M>
am I right in thinking (sorry doing other stuff so slow), that the <input type="image" usemap="..." >, functionality would trigger a get user agent request?  (sorry for being dim, obviously need more tea)

[03:55:02.0000] <Hixie>
though i don't expect anyone here to get it without searching for it

[03:55:03.0000] <gsnedders>
/me feels young again

[03:55:04.0000] <Lachy>
oh, I see.  I thought "series" was referring to some TV series or something

[03:56:00.0000] <Hixie>
Whiskey_M: how do you mean? when clicked?

[03:56:01.0000] <Whiskey_M>
yup

[03:57:00.0000] <gsnedders>
"sequence of rounded squares with varying shades of green and bold white outlines"

[03:57:01.0000] <gsnedders>
hmmm…

[03:57:02.0000] <Whiskey_M>
sorry, into a meeting now 

[03:57:03.0000] <Hixie>
Whiskey_M: yeah, it would work much the same way as <img usemap>

[03:57:04.0000] <gsnedders>
I'd expect it to have been published on a blue book

[03:58:00.0000] <gsnedders>
11:06+01 and I haven't started work yet. :\

[04:02:00.0000] <Lachy>
Hixie: "... but if the image makes it a lot easier for users of visual browsers to understand the concept." - either remove "if" or finish the statement.

[04:03:00.0000] <Hixie>
oops, missing "including"

[04:04:00.0000] <Lachy>
I think you should s/if/including/

[04:05:00.0000] <Hixie>
i was going to do s/if/if including/

[04:05:01.0000] <Hixie>
but yeah, i'll remove the second if

[04:05:02.0000] <Lachy>
nah, that doesn't work

[04:05:03.0000] <Hixie>
(and replace it with including)

[04:06:00.0000] <hsivonen>
/me doesn't get the specific reference of the alpha-beta-gamma pie chart

[04:06:01.0000] <Lachy>
it sounds like it could be some Dilbert reference or something

[04:06:02.0000] <Hixie>
much more obscure than that

[04:06:03.0000] <Hixie>
(much closer to home than that, too)

[04:07:00.0000] <hsivonen>
I thought it was a reference to the latest Stevenote but the colors are wrong

[04:07:01.0000] <Hixie>
oooh, i didn't think of having examples to keynotes

[04:07:02.0000] <Lachy>
Hixie: where's the obligatory Stargate reference in this section?

[04:08:00.0000] <Hixie>
heh

[04:08:01.0000] <Hixie>
haven't seen stargate in months, it's not on the mind

[04:09:00.0000] <Hixie>
talking about not understanding references, can anyone work out wtf this e-mail is talking about? http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-August/012351.html

[04:10:00.0000] <Hixie>
"My sense is that a consensus is building in the group to use a generalized XML-like syntax for the examples and illustrations in the semantics chapter."

[04:10:01.0000] <Hixie>
it is?

[04:11:00.0000] <hsivonen>
Hixie: no

[04:11:01.0000] <Hixie>
no to the e-mail, or to the examples' syntax? :-)

[04:11:02.0000] <Hixie>
or both? :-)

[04:11:03.0000] <hsivonen>
Hixie: the latter

[04:12:00.0000] <hsivonen>
Hixie: syntax that is

[04:12:01.0000] <Hixie>
ah ok

[04:12:02.0000] <Hixie>
well good

[04:12:03.0000] <Hixie>
though someone should probably let him know

[04:12:04.0000] <Lachy>
I found 1100670787_6a7c664aef, but I don't know who the person in the photo is

[04:12:05.0000] <Hixie>
not it!

[04:12:06.0000] <Hixie>
Lachy: me

[04:12:07.0000] <Lachy>
oh

[04:14:00.0000] <gsnedders>
hah

[04:15:00.0000] <Hixie>
right

[04:15:01.0000] <Hixie>
bed time

[04:15:02.0000] <Hixie>
hsivonen: i added the red boxes like you suggested

[04:16:00.0000] <gsnedders>
Lachy: mind PMing me the link?

[04:16:01.0000] <gsnedders>
Lachy: to 1100670787_6a7c664aef?

[04:16:02.0000] <Lachy>
no

[04:16:03.0000] <gsnedders>
why was I expecting that? :)

[04:16:04.0000] <hsivonen>
Hixie: thanks. the div one is rather opinionated ;-)

[04:16:05.0000] <Lachy>
I thought it would be obvious which site it came from based on the filename. After that, it's not too hard to figure out what to search for

[04:16:06.0000] <Hixie>
hsivonen: :-D

[04:17:00.0000] <Lachy>
the div one? Which one is that?

[04:17:01.0000] <gsnedders>
Lachy: but surely such a filename could be any photo gallery?

[04:17:02.0000] <Hixie>
hsivonen: i have but one thing to say about <div>: http://gallery.mac.com/emily_parker

[04:18:00.0000] <hsivonen>
Lachy: see rev 1001

[04:19:00.0000] <Hixie>
that url also serves well as an argument for irrelevant=""

[04:19:01.0000] <colione>
/me found the 1100670787_6a7c664aef too

[04:19:02.0000] <zcorpan_>
/me also, btw

[04:19:03.0000] <gsnedders>
/me feels dumb

[04:20:00.0000] <colione>
/me wonders how many really famous photo sharing sites there is.. :)

[04:20:01.0000] <Hixie>
colione: well, i just posted a link to one above :-D

[04:21:00.0000] <Hixie>
so at least 2 ;-)

[04:21:01.0000] <gsnedders>
/me can't get what to search for

[04:21:02.0000] <colione>
 :D

[04:22:00.0000] <gsnedders>
I'm probably on the right site, at least

[04:23:00.0000] <gsnedders>
ah. searching for a substring of what is needed doesn't help :D

[04:23:01.0000] <gsnedders>
hmm… it only has two views :\

[04:26:00.0000] <hsivonen>
gsnedders: try the API

[04:27:00.0000] <gsnedders>
hsivonen: multiple have confirmed I got it right, though

[04:27:01.0000] <zcorpan_>
/me got the logo joke

[04:27:02.0000] <gsnedders>
*multiple people

[04:29:00.0000] <Hixie>
zcorpan_: woot

[04:31:00.0000] <Hixie>
right, *gone*

[04:35:00.0000] <zcorpan_>
xml doesn't define error handling??

[04:37:00.0000] <gsnedders>
WHAT!?

[04:37:01.0000] <zcorpan_>
http://www.w3.org/mid/54DFD15E-C8EA-445C-A674-8611405F77CD⊙rc

[04:37:02.0000] <gsnedders>
actually, I think XML says it's an error, but not fatal

[04:38:00.0000] <gsnedders>
no, that must be something else

[04:39:00.0000] <gsnedders>
here we go.

[04:39:01.0000] <gsnedders>
"Note that non-validating processors are not obligated to to read and process entity declarations occurring in parameter entities or in the external subset; for such documents, the rule that an entity must be declared is a well-formedness constraint only if standalone='yes'."

[04:39:02.0000] <gsnedders>
http://w3.org/TR/xml/#wf-entdeclared

[04:40:00.0000] <zcorpan_>
right, so if there's an external dtd and you're non-validating, entity references you don't know about is an error but not fatal

[04:40:01.0000] <gsnedders>
it doesn't specify what to do, though

[04:40:02.0000] <Lachy>
zcorpan_: yes

[04:41:00.0000] <Lachy>
but I have no idea where Rob is getting his ideas about stray & and < characters

[04:41:01.0000] <zcorpan_>
/me neither

[04:41:02.0000] <Lachy>
they're defined as fatal well formedness errors

[04:42:00.0000] <gsnedders>
an entity to be treated as above still has to match the requirements for an entity ('&' Name ';')

[04:43:00.0000] <zcorpan_>
indeed

[04:43:01.0000] <gsnedders>
which does mean there will be some where that is the case

[04:44:00.0000] <gsnedders>
surely we have to assume that we have a standard XML parser that creates a DOM, so we can't do anything to it?

[04:44:01.0000] <hsivonen>
fwiw, the reason why unrecognized entities are fatal in Gecko is that as far as expat is concerned, it hasn't skipped the DTD but parsed a zero-length DTD

[04:45:00.0000] <hsivonen>
the Gecko entity resolver resolves unknown external entities to a zero-length stream

[04:45:01.0000] <hsivonen>
unknown: those that don't have a mapping in the built-in catalog

[04:49:00.0000] <hsivonen>
Rob is right that by configuring the XML processor to skip extenal entities and by using a reference to an external entity in the document, one can open up a loop hole that allows entity references to be expanded on the application level

[04:50:00.0000] <zcorpan_>
hsivonen: but he says that stray & and < are not well-formedness errors

[04:50:01.0000] <hsivonen>
however, the question that one must ask what good is using XML if you do your utmost to exploit loop holes to escape the nature of XML

[04:50:02.0000] <hsivonen>
zcorpan_: yeah, that part is bogus

[05:02:00.0000] <takkaria>
Hixie: typo in new <img> text, grep for "Standadrs" in one of the examples

[05:18:00.0000] <hendry>
anyknow what what WICD is about?

[05:18:01.0000] <hendry>
anyone know what WICD is about? http://www.w3.org/TR/WICD/

[05:22:00.0000] <zcorpan_>
hendry: it's wicked

[05:22:01.0000] <hendry>
zcorpan_: not crazy?

[05:23:00.0000] <zcorpan_>
</pun>

[06:23:00.0000] <Lachy>
I wrote a script to mark the issues in the spec with the text "Big Issue:" http://lachy.id.au/dev/utilities/#mark-issues

[06:23:01.0000] <Lachy>
I made a bookmarklet to do it manually, though we could probably get the script added to the spec

[06:28:00.0000] <hsivonen>
Lachy: wouldn't CSS generated content work fine?

[06:28:01.0000] <Lachy>
how well does that work with assistive technology? e.g. JAWS running on IE

[06:29:00.0000] <Lachy>
I wrote it based on Steve Faulkner's post to public-html 

[06:37:00.0000] <hsivonen>
Lachy: oh. IE...

[06:37:01.0000] <Lachy>
yeah, I just fixed that bug, reload the script

[06:38:00.0000] <Lachy>
now I can't figure out why it's not working in Opera

[06:39:00.0000] <Philip`>
http://www.twaynet.com/trade/t.asp?ID=1100670787 - aha, that alt text example is a reference to Peek Pi Pps Engineering Plastics Bearings!

[06:41:00.0000] <hsivonen>
Philip`: huh?

[06:42:00.0000] <Philip`>
It's the same number as in 1100670787_6a7c664aef.jpg and there's only a one-in-a-billion chance of it being a coincidence :-)

[06:42:01.0000] <hsivonen>
http://diveintomark.org/archives/2007/08/13/cc-elephants accessibility guy uses alt for title & authorship data

[06:42:02.0000] <Lachy>
ah, I had uploaded the old version of the bookmarklet.  Now it works! :-)

[06:42:03.0000] <hsivonen>
Philip`: oh

[06:43:00.0000] <virtuelv>
Lachy: got a URL I can test that script on?

[06:43:01.0000] <Lachy>
yeah, the spec

[06:44:00.0000] <Philip`>
Hixie: Other typos in the <img> thing: "greek" -> "Greek"; "aquisition" -> "acquisition"; "shallot.jpeg" -> "shalott.jpeg"

[06:47:00.0000] <Lachy>
hmm. It doesn't work in Safari :-(

[06:47:01.0000] <Lachy>
is there some trick to getting bookmarklets to work in it?

[06:52:00.0000] <Lachy>
hmm. mark-issues.js works fine in Safari, but the bookmarklet is broken.  Oh well.

[07:02:00.0000] <virtuelv>
Lachy: it's the bookmarklet itself that breaks

[07:02:01.0000] <Lachy>
yeah, I know. Do you know why?

[07:03:00.0000] <zcorpan_>
http://simon.html5.org/temp/zon/ working prototype parser for a custom tests format i came up with to be able to write lots of tests for getting innerHTML

[07:04:00.0000] <virtuelv>
Lachy: looking into it, because it should not break

[07:08:00.0000] <virtuelv>
Lachy: dunno exactly what happened, but this: javascript:(function(){var d=document.documentElement.appendChild(document.createElement('script'));d.src='http://lachy.id.au/dev/utilities/mark-issues.js'})() works in my version of Opera

[07:09:00.0000] <Lachy>
yeah, it works in Opera fine now.  I fixed the bug that was causing Opera to break already

[07:09:01.0000] <Lachy>
oh, did you mean Safari?

[07:10:00.0000] <virtuelv>
don't have Safari here

[07:11:00.0000] <Lachy>
oh, then why was my version breaking in Opera?  Were you using the older version?

[07:11:01.0000] <virtuelv>
no, I tested in an internal version, in which your bookmarklet broke

[07:12:00.0000] <colione>
Nice work zcorpan_

[07:15:00.0000] <zcorpan_>
thanks colione

[07:26:00.0000] <Lachy>
virtuelv: ok, you'd better file a bug report for it then :-)

[07:26:01.0000] <Lachy>
virtuelv: I didn't realise you worked at Opera

[07:27:00.0000] <virtuelv>
Lachy: but you changed the bookmarklet somewhat?   I retested it, and now it seems to work

[07:28:00.0000] <Lachy>
the last change I made to it was to add the outer (function() { ... })()

[07:28:01.0000] <Lachy>
the original one I had uploaded omitted it by mistake and, although it worked in Firefox, it broke in Opera

[07:29:00.0000] <virtuelv>
I tested the one that didn't wrap it in a function

[07:29:01.0000] <Lachy>
ok

[07:39:00.0000] <zcorpan_>
"User agents may adjust prefixes and namespace declarations in the serialisation (and indeed might be forced to do so in some cases to obtain namespace-well-formed XML)." -- should this perhaps be fleshed out and be specced as a set of requirements so that it can be tested?

[07:39:01.0000] <zcorpan_>
getting innerHTML in XML that is

[08:01:00.0000] <gsnedders>
or even speccing it out as an algorithm, so everything must implement it consistently?

[08:03:00.0000] <zcorpan_>
yeah

[08:03:01.0000] <zcorpan_>
that would be good

[08:26:00.0000] <Lachy>
charlvn added the script to the annotate-web-apps.js script, so now the issues get annotated automatically :-)

[08:44:00.0000] <Lachy>
ah... how should I respond to this? http://www.w3.org/mid/46C1BF25.9060808⊙RAU

[08:45:00.0000] <gsnedders>
Lachy: that phrasing confuses me slightly…

[08:46:00.0000] <Lachy>
in what way?

[08:46:01.0000] <gsnedders>
Lachy: "Many sites that used usemap="" on <input type="image">  cease to work in browsers that implement usemap, therefore it has been dropped from HTML 5."

[08:46:02.0000] <Philip`>
Try giving an example where it breaks

[08:46:03.0000] <gsnedders>
Lachy: I dunno. It just is confusing.

[08:46:04.0000] <Philip`>
(preferably of a real site)

[08:47:00.0000] <Lachy>
yeah, I was just making a test case

[08:47:01.0000] <Lachy>
to see how it works

[08:47:02.0000] <Philip`>
and then say that UAs which don't implement <img usemap> work better than those which do

[08:47:03.0000] <Philip`>
and hopefully that'd be clear enough :-)

[08:48:00.0000] <Lachy>
ok.

[08:48:01.0000] <gsnedders>
and list UAs

[08:48:02.0000] <Whiskey_M>
<img usermap, or <input type="image" usermap ?

[08:48:03.0000] <gsnedders>
Whiskey_M: the latter

[08:48:04.0000] <Philip`>
/me wonders if this is the kind of place where the HTML 5 test suite should explicitly test that UAs don't support usemap

[08:48:05.0000] <takkaria>
Lachy: I was very tempted to ask if he could just read it again and save us all bandwidth ;)

[08:49:00.0000] <Philip`>
(and how you can do that kind of testing nicely when the spec doesn't talk about usemap at all)

[08:49:01.0000] <Lachy>
ha!

[08:49:02.0000] <gsnedders>
Philip`: does anything prohibit UAs that implement HTML5 from supporting further elements?

[08:50:00.0000] <Philip`>
Oops, yes, I didn't mean <img usemap>

[08:50:01.0000] <Lachy>
perhaps, in cases like this, the spec should give explicit requirements for certain attributes to be ignored

[08:52:00.0000] <Philip`>
gsnedders: Don't know - it just seems good for UA quality assurance if the HTML test suite can test that UAs work as well as possible, and ignoring usemap appears to be some that is needed for a UA to work as well as possible

[08:53:00.0000] <Philip`>
and I don't know how that situation would be best handled

[08:58:00.0000] <takkaria>
I think the spec requring a certain attribute not be handled that was defined in a previous version of the spec would be a good solution; or at least saying that for compatiblity, UAs should not implement the attribute

[09:02:00.0000] <Philip`>
I posted http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-June/012063.html a while ago about supporting non-standard features - keeping non-standardised support for compatibility reasons seems like a bad idea in terms of having a relevant-in-the-real-world spec

[09:02:01.0000] <Philip`>
(because if it's needed for compatibility, it should be in the spec, and if it's not needed, it shouldn't be in the implementations)

[09:03:00.0000] <Philip`>
It feels kind of cruel to look in people's source code and create test cases specifically for them to fail, though

[09:06:00.0000] <gsnedders>
Philip`: keeping? what about adding?

[10:11:00.0000] <hsivonen>
Lachy: I don't understand what difference between Safari and Firefox http://alforag.com/donate/ illustrates

[10:13:00.0000] <Lachy>
it's the Sign up for updates form that breaks, were you testing the other one below (which I didn't notice originally)

[10:14:00.0000] <Lachy>
In Firefox, when you click the go button, it's treated as a link to a file:// URI and goes nowhere.  In Safari, it submits the form.

[10:15:00.0000] <hsivonen>
Lachy: oh, right. I clicked the image outside the red rectangle, so I didn't notice

[10:16:00.0000] <Lachy>
oh, I didn't notice one could click outside the red

[10:29:00.0000] <gsnedders>
Lachy: being very observant today? :)

[10:30:00.0000] <Lachy>
gsnedders: me being observerant? I don't think so!

[10:31:00.0000] <gsnedders>
maybe this is why I try and do backend development. less observation needed.

[11:11:00.0000] <takkaria>
there does seem to be a strong expert-bias in public-html

[11:15:00.0000] <Lachy>
takkaria: what do you mean by expert-bias?

[11:17:00.0000] <takkaria>
the comment that brought it about was Philip Taylor's "allowing the craziness of a small number of individuals to outweigh the combined wisdom of all those who formulated the specification for HTML 4.01"

[11:19:00.0000] <takkaria>
the idea being that an expert's opinion is worth far more than anyone else's, in short

[11:19:01.0000] <hsivonen>
takkaria: curiously, those who formulated HTML 4.01 are automatically assumed to have had wisdom whereas those who formulate HTML 5 aren't

[11:19:02.0000] <hsivonen>
I guess that happens when a spec is old enough

[11:19:03.0000] <takkaria>
interesting how people acquire mythical status over time. :)

[11:20:00.0000] <Lachy>
yeah, despite the fact that we have a whole 10 years worth of real world evidence that the HTML4 WG didn't have

[11:23:00.0000] <gsnedders>
why don't we just continue using SGML? fuck the real world, it keeps us consistent with the wisdom of HTML 4.01

[11:25:00.0000] <Lachy>
gsnedders, XML is the future, SGML failed

[11:25:01.0000] <gsnedders>
Lachy: but that goes against the wisdom of the old HTML WG!

[11:26:00.0000] <gsnedders>
Lachy: and they must be right!

[11:26:01.0000] <gsnedders>
Lachy: and you can't say anything to change that!

[11:26:02.0000] <Lachy>
the old HTMLWG already began the migration to XML as soon as it became available

[11:26:03.0000] <Lachy>
they ditched HTML4 after XHTML1

[11:26:04.0000] <takkaria>
I imagine who you define as an expert is a function of how much your views and theirs match up. :)

[11:27:00.0000] <gsnedders>
Lachy: were they still called the HTML WG?

[11:27:01.0000] <Lachy>
takkaria, I define you as an expert :-)

[11:27:02.0000] <takkaria>
/me grins

[11:27:03.0000] <gsnedders>
/me defines self as idiot

[11:27:04.0000] <Philip`>
Did they actually care about SGML when writing HTML4, or did they just keep using it because that's what the previous versions of HTML used?

[11:27:05.0000] <gsnedders>
Philip`: IIRC it's that they didn't want to reverse engineer everything

[11:27:06.0000] <Lachy>
gsnedders, yes, the status of XHTML1 still talks about the "HTML Working Group"

[11:28:00.0000] <gsnedders>
Lachy: when did the name change?

[11:28:01.0000] <gsnedders>
Lachy: the rechartering this year?

[11:28:02.0000] <Lachy>
it changed from HTML WG to XHTML2 WG earlier this year

[11:28:03.0000] <gsnedders>
so yeah

[11:29:00.0000] <Lachy>
I'm sure they just kept using it based on the wisdom of those who wrote HTML 3.2 using SGML

[11:29:01.0000] <Lachy>
which was in turn based on the wisdom of those who wrote HTML2

[11:29:02.0000] <Lachy>
which was actually not based on the wisdom of TBL who didn't originally use SGML when he first created it

[11:30:00.0000] <gsnedders>
which was based on the criticism of the IIIR-HTML draft?

[11:30:01.0000] <Philip`>
So they were following the SGML cow path? :-)

[11:30:02.0000] <gsnedders>
was I even born when the IIIR-HTML draft was written?

[11:30:03.0000] <gsnedders>
1993. ah good. I was alive!

[11:31:00.0000] <Whiskey_M>
gs: lol

[11:31:01.0000] <takkaria>
did they have cowpaths back then? :)

[11:31:02.0000] <Lachy>
what does IIIR stand for?

[11:32:00.0000] <Lachy>
ah, Integration of Internet Information Resources Working Group

[11:32:01.0000] <Philip`>
They should have called it the I3R

[11:32:02.0000] <gsnedders>
looking it up on Wikipedia gives a search page, results including "Transsexual sexuality" :\

[11:33:00.0000] <takkaria>
sounds like grounds for a libel claim to me

[11:33:01.0000] <gsnedders>
note to self: wikipedia does not have the answers to everything. google does.

[13:43:00.0000] <Jero>
hey, was reading http://pearlcrescent.com/products/pagesaver/ and read their extension uses the <canvas> element to somehow make a screenshot of the page.

[13:43:01.0000] <Jero>
"Page Saver uses the canvas feature that was introduced in Firefox 1.5."

[13:43:02.0000] <Jero>
does anyone know how they pull that off?

[13:44:00.0000] <Philip`>
There's a (non-standard) drawWindow method, which is usable only by extensions and not normal web content

[13:45:00.0000] <Jero>
i see...

[13:46:00.0000] <Philip`>
ctx.drawWindow(window, x, y, w, h, bgcolor)

[13:46:01.0000] <Philip`>
(I think)

[13:46:02.0000] <Jero>
interesting

[13:47:00.0000] <Philip`>
(It's disabled for web content because of security concerns)

[13:47:01.0000] <Philip`>
( http://lxr.mozilla.org/mozilla/source/content/canvas/src/nsCanvasRenderingContext2D.cpp#2305 )

[13:50:00.0000] <Jero>
Thanks!

[13:52:00.0000] <Jero>
the thing is, i was checking out the possibilities of creating a web application that takes a URL, feeds it to Gecko, and outputs a PNG image

[13:53:00.0000] <Jero>
sort of like khtml2png (http://khtml2png.sourceforge.net/)

[13:53:01.0000] <met_>
Philip`, this drawWindow is only for Mozilla XPCOM or is it accessible from JavaScript?

[13:54:00.0000] <met_>
Jero, check http://www.screengrab.org/ extension

[13:54:01.0000] <Philip`>
http://weblogs.mozillazine.org/roc/archives/2005/05/rendering_web_p.html

[13:55:00.0000] <Philip`>
met_: It works in JS too

[13:55:01.0000] <met_>
Philip`, just trying but it gives me some security error

[13:56:00.0000] <met_>
ah there is example, going to try it, thx

[13:56:01.0000] <Jero>
met_, thanks

[13:56:02.0000] <Philip`>
met_: You need UniversalBrowserRead privileges, either by being chrome (like an extension) or by doing some magic to ask the user for permission

[13:57:00.0000] <met_>
Philip`, ah this 8-(

[13:57:01.0000] <met_>
so it is not primary designed for web but for XUL apps

[13:57:02.0000] <Philip`>
netscape.security.PrivilegeManager.enablePrivilege('UniversalBrowserRead')

[13:57:03.0000] <Philip`>
(Hooray for netscape)

[13:58:00.0000] <Philip`>
met_: Yes, since the security issues of exposing it to the web are seemingly non-trivial

[13:59:00.0000] <Philip`>
(I think it would be quite a useful thing to have web access to, but I don't know how it'd be implemented)

[14:02:00.0000] <met_>
/me is not sure if there are still some security problems if you can draw only window from your domain

[14:02:01.0000] <met_>
you can screenshot filled form, but you have access to it anyway

[14:02:02.0000] <colione>
There will still be som XSS concerns

[14:03:00.0000] <met_>
colione: iframes and frames from other domains can remain blank

[14:03:01.0000] <Philip`>
The window from your domain can contain content from other domains (images, iframes, etc) and it's hard to filter all that stuff out correctly

[14:03:02.0000] <met_>
in rendered view

[14:03:03.0000] <met_>
not sure if images are the real problem too

[14:04:00.0000] <colione>
Imagine some badly coded application that takes arguments in a querystring and don't sanitize it...you can put in any arbitrary script there

[14:05:00.0000] <met_>
colione, but what is problem with canvas2d.drawWindow ?

[14:07:00.0000] <colione>
that page might lead to a registration page and you can for example hook a function to the submit button that takes a screenshot of the page and send in over the web

[14:07:01.0000] <colione>
prehaps not the best example, but still possible

[14:09:00.0000] <met_>
if you have access to this page, frame, you can do it anyway - making screenshot is only better comfort

[14:11:00.0000] <Jero>
Okay, so lets say if i wanted to make a web application that takes a URL to create a screenshot of it using Gecko. Do you think it would be possible to make the app call Firefox with an extension that uses the drawWindow method on the server to save the image on the server? Or would there be a better way?

[14:14:00.0000] <Philip`>
Jero: http://www.hackdiary.com/archives/000055.html sounds like what you're doing

[14:16:00.0000] <Jero>
Philip`, oh wow, awesome, thanks!

[14:16:01.0000] <Philip`>
(Using an embedded Mozilla is probably much easier than trying to automate the Firefox UI)

[14:17:00.0000] <Jero>
yeah

[14:27:00.0000] <Hixie>
Philip`: there's more than one-in-a-billion chance, it's a birthday attack ;-)

[14:31:00.0000] <Hixie>
fixed the other typos, thanks

[14:31:01.0000] <Philip`>
Someone should do a survey of all the numbers that are on the web, limited to ten digits or so - it could be interesting to see if any particular numbers are unrepresented...

[14:32:00.0000] <Philip`>
(And then publish the results somewhere, hence filling in all the gaps)

[14:34:00.0000] <Hixie>
it would probably show a propensity towards numbers that start with 1

[14:35:00.0000] <Hixie>
there's a theory regarding the likelihood of that in any number sample

[14:36:00.0000] <gsnedders>
/me wonders how high up 42 would be

[14:36:01.0000] <Philip`>
Hixie: Something I noticed earlier today: Why add <nav> instead of reusing <map>? HTML4 says <map> can be used for navigation stuff, and http://www.noraradcliffe.co.uk/ uses it for that (though that's the only one I could find in ~8000 pages), and you can style it properly in current versions of IE and Firefox (unlike <nav>)

[14:41:00.0000] <zcorpan_>
Philip`: then <map> would be sectioning element or not depending on if it was used as an image map or not... and afaict using <map> meaning navigation isn't widely understood by authors

[14:42:00.0000] <Hixie>
Philip`: i prefer not to overload things where it isn't necessary

[14:48:00.0000] <met_>
/me is lost while reading http://webkit.org/blog/115/webcore-rendering-ii-blocks-and-inlines/ Blocks inside Inline Flows section

[14:49:00.0000] <met_>
/me didn't undestund why inlines  around  block are splitted

[14:49:01.0000] <zcorpan_>
jgraham__: that would be awsome (re table algorithm prototyping)

[14:50:00.0000] <zcorpan_>
/me has some ideas for possible algorithms, but not really written down

[14:52:00.0000] <zcorpan_>
trying to figure out header cells from TD-only tables could be interesting, but might be taking it too far...? i mean, there are data tables where all cells are just data

[15:00:00.0000] <Philip`>
zcorpan_, Hixie: Okay, makes sense

[15:11:00.0000] <Hixie>
/me sighs as he contributes yet another unhelpful e-mail to public-html

[15:11:01.0000] <Hixie>
maybe i should just unsubscribe

[15:14:00.0000] <gsnedders>
Hixie: yeah, just leave the WG

[15:14:01.0000] <Hixie>
the funny thing is, i have no trouble keeping up with the volume of e-mail

[15:14:02.0000] <Hixie>
my problem is that i sometimes can't help replying to e-mails

[15:14:03.0000] <gsnedders>
ah. I have no problem with either.

[15:15:00.0000] <gsnedders>
The amount of junk makes it easier to do the first though, as there is less in-depth reading needed

[15:31:00.0000] <takkaria>
I've drafted several replies to things on public-html, but they go in the "drafts" folder and I never remember about things that go there

[15:34:00.0000] <zcorpan_>
takkaria: yep, having public-html mail in the drafts folder for a while might be a good thing... :)

[15:35:00.0000] <takkaria>
well, when I come back to it, I don't remember writing it, so it gets deleted

[15:35:01.0000] <takkaria>
also, thunderbird has mysteriously hidden one so that it only shows up in the folder view when I look at only-unread stuff

[15:35:02.0000] <takkaria>
despite it not being unread

[15:40:00.0000] <zcorpan_>
my point being that it results in less mail to the list. :)

[15:40:01.0000] <takkaria>
/me nods

[15:41:00.0000] <zcorpan_>
if one doesn't remember what a mail was about, it probably wasn't important anyway ;)

[15:41:01.0000] <takkaria>
no.  not on the grand scale of things

[15:41:02.0000] <colione>
tell the person with amnesia...

[15:42:00.0000] <takkaria>
normally someone else comes along and makes the same point, more or less eloquently, and I can be happy that I didn't waste bandwidth

[16:17:00.0000] <zcorpan_>
i'm sure i've read the logo joke somewhere before. where is it from, Hixie?

[16:58:00.0000] <mpt>
Philip`, the numbers survey has been done. <http://www.turbulence.org/Works/nums/prepare.html>

[16:58:01.0000] <mpt>
Unfortunately it hasn't been updated since 2002


2007-08-15
[17:02:00.0000] <Philip`>
mpt: Oh, thanks - that looks like it would be interesting if I had Java installed :-)

[17:20:00.0000] <tndH_>
http://blogoscoped.com/archive/2006-07-28-n51.html is a lot less thorough, but it's colourful :)

[17:23:00.0000] <mpt>
I wonder how much the frequencies of "9" and "11" have increased in the past decade

[20:25:00.0000] <Hixie>
why i do believe Leif just asked me to shut up

[20:33:00.0000] <karlUshi>
paranoia. 

[20:33:01.0000] <karlUshi>
I think Leif said that 80%-20% was a poor argument in a discussion.

[20:33:02.0000] <Hixie>
"Save it up for later." seems pretty clear to me

[20:34:00.0000] <karlUshi>
yes :) but I do not read it the same way than you

[20:34:01.0000] <Hixie>
how do you read it?

[20:34:02.0000] <Hixie>
i'm curious as to what he meant

[20:34:03.0000] <Hixie>
i'm confused because he immediately sent another e-mail asking questions

[20:34:04.0000] <Hixie>
and i don't know if he wants me to reply or not

[20:35:00.0000] <Hixie>
probably safest not to

[20:35:01.0000] <karlUshi>
I always try to take a step back for any comments on the mailing-list :/ because I would end up at leaving my job every day ;)

[20:35:02.0000] <karlUshi>
So the way I read it is

[20:35:03.0000] <Lachy>
hmm. I read it as shut up too

[20:36:00.0000] <karlUshi>
"80% is an argument that you always use, I'm not convinced by it, save it for later."

[20:36:01.0000] <Hixie>
it wasn't intended to be an argument

[20:36:02.0000] <karlUshi>
ooops

[20:36:03.0000] <karlUshi>
French

[20:37:00.0000] <karlUshi>
argument = point of discussion

[20:37:01.0000] <Hixie>
and i don't understand how saving it for later would make it any more acceptable

[20:37:02.0000] <karlUshi>
not dispute

[20:37:03.0000] <Hixie>
right, it wasn't intended to be a point of discussion -- it's just how languages have to be designed

[20:37:04.0000] <Hixie>
you always have to draw a line as to how much to include

[20:37:05.0000] <karlUshi>
in your perspective 

[20:37:06.0000] <Hixie>
it's not clear to me that there can be useful perspectives where you don't draw a line

[20:39:00.0000] <Hixie>
though i'd love to hear of one if there are some

[20:39:01.0000] <Lachy>
of course the line has to be drawn somewhere, it's just a matter of finding out wheere

[20:39:02.0000] <karlUshi>
imagine a line based on the inclusion of diversity more than the democracy (democracy being the power of the majority as opposed to the Republics without entering too much into details)

[20:39:03.0000] <Hixie>
what do you mean by "diversity"?

[20:40:00.0000] <karlUshi>
we are drifting... but let's say. For example, the republics guarantee the rights of minority because the respect of this diversity is important for the design of a society. 

[20:40:01.0000] <karlUshi>
Democracies at the opposite impose the power of big numbers on minority

[20:40:02.0000] <karlUshi>
usually a mix of the two is useful.

[20:40:03.0000] <Hixie>
ok... (looking for how this applies to language design)

[20:41:00.0000] <karlUshi>
accessibility for example, or internationalisation.

[20:42:00.0000] <karlUshi>
a feature can be useful for a limited number of persons, but without the feature, the language will not be useful in this precise context.

[20:42:01.0000] <karlUshi>
Same for the inclusion of some rare characters in unicode.

[20:43:00.0000] <karlUshi>
/me suddenly remembers how much usenet sucks when we could not type accents typing French. 

[20:44:00.0000] <karlUshi>
When some people started to mention that it would be cool to modify the implementation. Some people said it was folly.

[20:45:00.0000] <Lachy>
karlUshi, I don't understand the point you are trying to make. I thought you were trying to explain a perspective where a line doesn't have to be drawn, but then you gave an example where it was drawn in a different place.

[20:46:00.0000] <karlUshi>
Lachy: I just said there is not one unique line, I said it was a question of perspective, point of view. :) 

[20:46:01.0000] <karlUshi>
being binary doesn't help

[20:47:00.0000] <Hixie>
well sure, my point was just that you do have to draw a line

[20:47:01.0000] <Hixie>
which you seem to agree with, despite saying earlier that that was just "in my perspective"

[20:48:00.0000] <Hixie>
/me is now confused as to both karl's opinion _and_ leif's opinion!

[20:49:00.0000] <Lachy>
I understand Lief's opinion, he just objects to the 80%-20% argument because he sees it as a way to ignore the minority, particularly on accessibility

[20:49:01.0000] <Hixie>
ooo, yet another mailing list

[20:50:00.0000] <karlUshi>
your line is 80/20 rationale. what I said being your perspective :)

[20:50:01.0000] <Hixie>
Lachy: ah, then he's probably misunderstanding it. i thought i was pretty careful about saying it was 80% of _tables_, not 80% of _users_.

[20:50:02.0000] <karlUshi>
you decided to draw the line on this principle

[20:50:03.0000] <Lachy>
Hixie: yes, a lot of people do misunderstand it

[20:51:00.0000] <karlUshi>
Lachy: because it is often used not only to categorize data, but to remove options to people. 

[20:51:01.0000] <Lachy>
some don't realise that it's not a fixed ratio either, it needs to be evaluated on a case by case basis depending on numerous other factors

[20:52:00.0000] <Hixie>
what's the status of the design proposals? can i add one?

[20:52:01.0000] <Hixie>
principles, rather

[20:52:02.0000] <Lachy>
maybe, what are you thinking of adding?

[20:52:03.0000] <karlUshi>
So I guess each time we categorize data like this, we have to propose something that will not exclude people falling into the 20%.

[20:52:04.0000] <Hixie>
people don't fall into percentages

[20:53:00.0000] <Hixie>
it's 80% of _content_, not 80% of _users_

[20:53:01.0000] <Hixie>
Lachy: the line

[20:53:02.0000] <karlUshi>
Hixie: unfortunately managerial decisions show every days that people fall into percentages. I'm pretty I just have to read Yahoo News right now

[20:53:03.0000] <karlUshi>
pretty sure

[20:54:00.0000] <Hixie>
i'm just talking about language design here

[20:54:01.0000] <Hixie>
nothing else

[20:54:02.0000] <Hixie>
and in designing languages for the web, we should aim to cater for 100% of users, and n% of use cases, where n < 100.

[20:54:03.0000] <karlUshi>
yes but people are "humans" not machines. When they read something, they charge it with emotions, etc. The same way you read leif comment with emotions.

[20:54:04.0000] <Hixie>
lachy: (that's the principle i was going to add)

[20:55:00.0000] <karlUshi>
so when you say 80% of something, many people will think 20% of people

[20:55:01.0000] <Hixie>
if people aren't going to read what i write, that's not really my problem

[20:55:02.0000] <Hixie>
:-)

[20:55:03.0000] <Hixie>
at least not in this context

[20:56:00.0000] <karlUshi>
:) unfortunately it is ;) or you have missed everything about social relationship :p

[20:59:00.0000] <Hixie>
not in this context -- if people are going to ignore what i write and argue against what they think i'm writing, then i'll just ignore them, as they aren't being productive.

[20:59:01.0000] <Hixie>
no big loss to me, since someone who doesn't slow down long enough to actually read what they're talking about probably isn't going to be slowing down long enough to think about what they're writing either.

[21:00:00.0000] <Lachy>
exactly! It's more productive to not respond in some cases

[21:01:00.0000] <Lachy>
and just hope that, eventually, people get the idea that unproductive, emotionally charged responses don't get responses, and then they stop sending them

[21:05:00.0000] <Lachy>
oh, nice, Jason White gave a usecase for <input usemap>

[21:05:01.0000] <Lachy>
yet he's ignoring that virtually *nobody* uses it for that purpose anyway

[21:15:00.0000] <Hixie>
ok, added Baby Steps to the principles

[21:19:00.0000] <Lachy>
that's quite well written

[21:19:01.0000] <Hixie>
why thank you

[21:19:02.0000] <Lachy>
hmm. I wonder if it's at all possible to search for sites that use <input type=image> as a server side image map

[21:20:00.0000] <Hixie>
as opposed to a button?

[21:20:01.0000] <Hixie>
i.e. those that care about the .x and .y coordinates?

[21:20:02.0000] <Lachy>
yes

[21:20:03.0000] <Lachy>
but there'd be a lot that use type=image, but don't care about x,y

[21:20:04.0000] <Hixie>
right

[21:22:00.0000] <Hixie>
i don't understand why people have such a gut reaction to us removing unused or abused features

[21:22:01.0000] <Hixie>
(like usemap)

[21:22:02.0000] <Hixie>
(on input)

[21:23:00.0000] <Hixie>
the philip in http://www.w3.org/mid/46C1DE89.1080305⊙RAU apparently missed the list of URIs that has been cited several times

[21:23:01.0000] <Lachy>
some people just seem to think the designers of HTML4 had more wisdom then we do, and think they must have included it for a good reason

[21:50:00.0000] <Lachy>
I think I know how we could search for type=image used as a server side image map...

[21:50:01.0000] <Hixie>
oh?

[21:51:00.0000] <Lachy>
my hypothesis is that if the only visible control in a form is type=image (may also include type=hidden or readonly controls), then it's more likely that the site would be interested in the x,y coords

[21:52:00.0000] <Hixie>
/me replies to 58 e-mails about alt=""

[21:52:01.0000] <Lachy>
also, larger images are more likely to be image maps, than smaller images which are likely to be simple buttons

[21:53:00.0000] <Lachy>
so searching for pages that meet those criteria should reveal the pages that use it

[21:55:00.0000] <Lachy>
should I send you a mail about that for you to put on your todo list to do the search?

[21:55:01.0000] <Hixie>
it would be hard to do in my current framework, because my parser doesn't really do the form association stuff

[21:55:02.0000] <Lachy>
ok

[21:56:00.0000] <Hixie>
and i'm not at all convinced that there aren't many pages with large <input type=image>s :-)

[21:56:01.0000] <Lachy>
yeah, and the only way to know how big the image is, is to retrieve it, which would be quite a slow process

[21:57:00.0000] <Hixie>
my current framework can't handle looking at external files at all

[21:57:01.0000] <Hixie>
and making it do so would be far more work than i'm interested in doing

[21:58:00.0000] <Hixie>
(FAR more work)

[21:58:01.0000] <Lachy>
eyah

[21:58:02.0000] <Lachy>
yeah

[21:58:03.0000] <Hixie>
though height/width might help

[22:00:00.0000] <Lachy>
oh look, Robert has described hypothetical use cases for <input usemap>

[22:00:01.0000] <Hixie>
i hope he makes sure to create a wiki page for each of those use cases / problems

[22:00:02.0000] <Hixie>
so i can make sure to address them

[22:01:00.0000] <Lachy>
I'll get them started

[22:02:00.0000] <Hixie>
(don't bother unless the use cases are compelling)

[22:02:01.0000] <Hixie>
ok, home time. bbl.

[22:02:02.0000] <Lachy>
ok, fair enough

[22:02:03.0000] <Lachy>
cya

[01:02:00.0000] <Lachy>
ah, I'm not looking forward to the backlash against Hixie's comment "I don't really have much time for process"

[01:03:00.0000] <Lachy>
(although I actually agree with it)

[01:06:00.0000] <Hixie>
the guy told me i shouldn't have edited the wiki

[01:06:01.0000] <Hixie>
fuck that

[01:06:02.0000] <Lachy>
yeah, I know, that's what I thought too

[01:06:03.0000] <Hixie>
on a more practical note, we need to work out a story for height and width on <img>

[01:06:04.0000] <Lachy>
what about them? Are they being dropped?

[01:07:00.0000] <Hixie>
working out what they mean, what their conformance criteria are, etc

[01:07:01.0000] <Hixie>
i think it was hsivonen who pointed out that it's getting quite common for pages to include extremely high res pictures, and then they are using height=/width= to make them size sanely even in the absence of CSS

[01:08:00.0000] <Hixie>
i'm a big fan of making sure pages can render sanely even without CSS (or with difference CSS)

[01:08:01.0000] <Lachy>
BTW, I just reworded cowpaths and reinventing the wheel principles in the wiki too

[01:08:02.0000] <Hixie>
different, even

[01:08:03.0000] <Hixie>
but it seems weird for the page to give a particular size

[01:08:04.0000] <Hixie>
since that's very media-specific

[01:08:05.0000] <Hixie>
what if my device is bigger? smaller?

[01:09:00.0000] <Lachy>
some sites that use high res images and then scale them down seem to do so because they don't know how to operate an image editor

[01:09:01.0000] <Hixie>
when i've done it it's been because i wanted the print version to have access to more pixels

[01:09:02.0000] <Hixie>
these days it's probably also to let higher res displays have access to more pixels

[01:12:00.0000] <Hixie>
i guess i can pretend that the height/width are encoding the "real" dimensions of the image in CSS pixels

[01:12:01.0000] <Lachy>
wouldn't that be a use case for media queries on <img>?

[01:12:02.0000] <Lachy>
although, I"m not sure how practical that would be to implement, it's what XHTML2 have done, I think

[01:13:00.0000] <Hixie>
i don't think there's anything wrong with simply providing a very high res image

[01:13:01.0000] <karlUshi>
Hixie: replied

[01:13:02.0000] <Hixie>
and expecting the UAs to scale appropriately

[01:13:03.0000] <Lachy>
ok

[01:14:00.0000] <Hixie>
karlUshi: thanks

[01:15:00.0000] <arve__>
<Hixie> i guess i can pretend that the height/width are encoding the "real" dimensions of the image in CSS pixels <-- care to elaborate?

[01:15:01.0000] <arve__>
It's the word "real" I don't quite get

[01:15:02.0000] <Hixie>
i think i will pretend that the height/width are encoding the "real" dimensions of the image in CSS pixels, where those dimensions might differ from the actual dimensions of the image because we're expecting UAs to ignore DPI info in the image, but the ratio has to match the real ratio

[01:16:00.0000] <Hixie>
so if the image is actually 1000dpx by 500dpx, but the author puts width=100 height=50, then that's ok

[01:16:01.0000] <Hixie>
but if they have an image that's 100dpx by 50dpx and they put height=100 width=100 that's bad

[01:16:02.0000] <Hixie>
dpx = device pixel

[01:16:03.0000] <Hixie>
as opposed to CSS pixel

[01:16:04.0000] <Lachy>
ok, so they have to keep aspect ratios to be conforming

[01:17:00.0000] <arve__>
what should happen in the latter case, then?

[01:17:01.0000] <arve__>
I think there are legitimate uses of that for image stretching purposes

[01:17:02.0000] <Hixie>
arve__: oh i'm not changing anything that happens, it'll stay as per today

[01:17:03.0000] <Hixie>
arve__: like?

[01:18:00.0000] <arve__>
for instance, screencaps from some video source may be aspect-ratio-distorted, and you can use width/height to carefully correct it

[01:19:00.0000] <arve__>
or the other case, which I concede is abuse, to create progress bars and similar using images

[01:19:01.0000] <Lachy>
arve__: wouldn't it be better to correct such distortions in an image editor, rather than letting browsers scale it?

[01:19:02.0000] <arve__>
Lachy: surely, but people do stuff like that anyway

[01:19:03.0000] <Lachy>
browsers tend to scale images quote poorly compared with image editors, which use more complex algorithms

[01:19:04.0000] <Lachy>
really? Do you have any examples?

[01:20:00.0000] <Lachy>
but just because people do that, should we allow it?  People do a lot of other invalid stuff

[01:20:01.0000] <arve__>
Lachy: other than "from memory", no?

[01:21:00.0000] <Lachy>
ok

[01:21:01.0000] <arve__>
I've distorted aspect ratio on purpose in scripting, though

[01:23:00.0000] <Lachy>
Hixie: I've found several sites that do actually depend on <input usemap> working, and which break in IE

[01:23:01.0000] <Lachy>
But all of them would be solved by using <a><img></a> or <img usemap>

[01:38:00.0000] <Lachy>
Hixie, these sites were listed in your results, but they don't contain <input usemap> http://gctinc.com/index.html http://globulus-ural.com/links.php

[02:06:00.0000] <Hixie>
virtuelv_: if the images are aspect-ratio distorted (that is, anamorphic) then they should be fixed. there's no reason to have anamorphic static images.

[02:07:00.0000] <virtuelv_>
Hixie: not that I disagree

[02:07:01.0000] <Hixie>
Lachy: they might have changed since i looked at them

[02:07:02.0000] <Lachy>
ok, how long ago did you look at them?

[02:07:03.0000] <Hixie>
Lachy: and yes, some pages might require <input usemap>, but unless it's a majority...

[02:08:00.0000] <Lachy>
is it possible that the usemap actually occured on other pages of the site and you listed the wrong page?

[02:08:01.0000] <Hixie>
no

[02:08:02.0000] <Hixie>
but the data can be several weeks out of date depending on exactly what i was doing

[02:09:00.0000] <Lachy>
oh right, since it's from Google's cache

[02:22:00.0000] <Whiskey_M>
'lo

[02:22:01.0000] <Hixie>
hey

[02:23:00.0000] <hsivonen>
Hixie: do you expect conformance checkers to download images and check if they are fully transparent?

[02:24:00.0000] <Whiskey_M>
Hixie, just saw the e-mail about web forms,  I know it's at the backburner, but (and sorry if I should have RTFMed from the website and missed it), is there any effort in to write a server side forms processor + front end JS for new controls?

[02:36:00.0000] <Hixie>
hsivonen: i dunno

[02:37:00.0000] <Hixie>
hsivonen: i'm starting to think that maybe we should have a level of conformance for conformance checkers that does involve checking external resources, because i keep coming across features that would be nice to add that require that

[02:41:00.0000] <hsivonen>
Hixie: anything else besides transparent images and image dimensions?

[02:42:00.0000] <hsivonen>
(I'm not yet convinced that tying conformance to external resources is a good idea)

[02:43:00.0000] <Whiskey_M>
/me guesses his was a silly question and goes back to the site to look for details

[02:44:00.0000] <Lachy>
I found one site that would be a valid use case for <input usemap>, except that due to the way it's implemented in Firefox, it's actually better in browsers that don't support it http://realestate.bannisterdesign.com/grostick.com/search.php

[02:45:00.0000] <Lachy>
the problem is the <area> links don't submit the coordinates where the user clicked, but the site uses them for the search

[02:51:00.0000] <Hixie>
hsivonen: type="" on <link> and <script> and <object> matching the resource, all kinds of stuff

[02:51:01.0000] <Hixie>
hsivonen: not having links to 404s

[02:52:00.0000] <Hixie>
though i guess that should be a should

[02:52:01.0000] <Hixie>
i'm not sure how we should phrase it

[02:52:02.0000] <Hixie>
maybe document conformance and document context conformance or some such

[02:53:00.0000] <hsivonen>
Hixie: the former is theoretically hard, because GET doesn't allow you to see what someone else GETs

[02:54:00.0000] <hsivonen>
Hixie: this has already been a problem with krijn's and anne's Apache configs

[02:54:01.0000] <hsivonen>
which makes it a practical problem, too

[02:54:02.0000] <Hixie>
well, the same can be said of any URI, even the one the validator gets

[02:54:03.0000] <Hixie>
unless you're uploading the file

[02:54:04.0000] <hsivonen>
yes

[02:55:00.0000] <hsivonen>
but at least when the validator has received something from the main URI, it can say definite things about that stream of bytes plus content-type

[02:55:01.0000] <Hixie>
it's possible that conformance criteria dependent on external resources are only useful in a dynamic environment, same as conformance criteria that apply to script-driven scenarios

[02:55:02.0000] <Hixie>
e.g. i don't expect a conformance checker to say if the scripts are conforming, but we have requirements on those already

[03:24:00.0000] <Whiskey_M>
outside of on-going discussions here does anyone have a link to the usemap thread 

[03:26:00.0000] <zcorpan_>
"If either attribute is specified, the other must be as well" [img width/height] -- why? specifying only one is a good way to get the right ratio, especially when you use percentages

[03:29:00.0000] <Lachy>
I hope my detailed explanation of the usemap issue I just sent puts a conclusive end to the issue.  I've wasted too much time on it already today. :-(

[03:44:00.0000] <hsivonen>
I wonder if the people talking about "education" have ever done upgrade evangelism on behalf of a browser

[03:45:00.0000] <hsivonen>
that is, contacted sites to explain that the sites are doing something wrong and need to be fixed in order to work with a new browser

[03:52:00.0000] <Hixie>
zcorpan_: because the point is to allow for authors to prevent flickering page load, not for them to be able to get the layout they want

[03:54:00.0000] <zcorpan_>
Hixie: right. i thought percentages were allowed but looking again it seems they aren't. ok

[03:57:00.0000] <zcorpan_>
Hixie: still, having images with more pixels than the size you want it in order to get better quality on high resolution screens, and not caring about page load flickering, it is easier to just specify one if you're authoring by hand

[03:57:01.0000] <Hixie>
mmm

[03:57:02.0000] <Hixie>
fair enough

[04:07:00.0000] <Hixie>
woah, robert's latest e-mail is weird

[04:07:01.0000] <Hixie>
he's arguing in favour of <input usemap> in response to an e-mail where Lachy showed that <input usemap> was a net accessibility loss

[04:14:00.0000] <Hixie>
(and citing accessibility reasons for it)

[04:15:00.0000] <Hixie>
/me decides to let others reply to the bulk of that e-mail, but jumps in on the last paragraph

[04:16:00.0000] <Hixie>
zcorpan_: spec updated

[04:19:00.0000] <zcorpan_>
Hixie: looks good

[04:29:00.0000] <Hixie>
hsivonen: lachlan (writing in nov 2006) said, about src="":

[04:29:01.0000] <Hixie>
> And, as I mentioned in IRC, I think it should be defined that the value

[04:29:02.0000] <Hixie>
> should resolve to a valid URI for an image

[04:29:03.0000] <Hixie>
what should i do? say something like "the resource specified by the src attribute must be [an image]"?

[04:29:04.0000] <Hixie>
or say nothing?

[04:30:00.0000] <hsivonen>
hmm. perhaps there indeeds to be a class of tests for external resources

[04:30:01.0000] <hsivonen>
Hixie: what's an image?

[04:31:00.0000] <Hixie>
that's a separate rat hole that i don't want to get into tonight

[04:31:01.0000] <hsivonen>
Hixie: is an application/pdf file an image in the Safari context?

[04:31:02.0000] <Hixie>
but i'll probably define it as image/* resources that correspond to bitmap data

[04:31:03.0000] <Hixie>
thus excluding pdfs, svg, and wmfs

[04:31:04.0000] <hsivonen>
Hixie: what about SVG and Opera?

[04:34:00.0000] <zcorpan_>
who are we helping by having requirements for what <img src> points to?

[04:35:00.0000] <Hixie>
the original reason for Lachy's request was to be able to ban <img src=""> (pointing to the HTML file itself)

[04:35:01.0000] <Hixie>
which it would be vaguely useful to discourage

[04:40:00.0000] <Hixie>
ok i'm gonna sleep now

[04:40:01.0000] <Hixie>
more <img> nonsense tomorrow

[04:44:00.0000] <Lachy>
oh wow, this is going around in circles already: "Unless you can layout what problems you think we will solve by eliminating <input type=image usemap= > from HTML, why are you even posing it."

[04:44:01.0000] <Lachy>
I thought we already outlined the problems that removing it solves

[06:19:00.0000] <krijnh>
Ping

[06:21:00.0000] <Philip`>
Ooh, IE actually parses <blink>...</blink> properly and not like an unrecognised tag

[06:22:00.0000] <Philip`>
Could be quite useful if someone wanted to add a Backward-LINK element in a future version of HTML, since it already has the correct parsing in IE

[06:22:01.0000] <takkaria>
heh

[06:23:00.0000] <zcorpan_>
wonder if the serializing html fragments algorithm should drop PIs on the floor instead of bailing out

[06:23:01.0000] <zcorpan_>
or emit them as "<?" + target + " " + data + ">"

[06:24:00.0000] <takkaria>
Philip`: if that's not repurposing of elements, I don't know what is

[06:24:01.0000] <hsivonen>
Philip`: the new rev=''! :-)

[10:31:00.0000] <Philip`>
I did http://zaynar.demon.co.uk/misc2/font4.png a while ago with Latin, Latin-Extended, Greek, Cyrillic and Hebrew, and it fits alright, but it doesn't seem very scalable to things like CJK

[12:51:00.0000] <guardian>
hello

[12:52:00.0000] <guardian>
please, where could i find the canvas spec ?

[12:52:01.0000] <guardian>
http://www.whatwg.org/specs/web-apps/current-work/#the-canvas remains silent

[12:52:02.0000] <gavin_>
"remains silent"?

[12:52:03.0000] <guardian>
empty page :)

[12:52:04.0000] <gavin_>
not for me

[12:53:00.0000] <guardian>
really ?

[12:53:01.0000] <Philip`>
http://www.whatwg.org/specs/web-apps/current-work/multipage/section-the-canvas.html#the-canvas might work better if your browser doesn't like really big pages

[12:53:02.0000] <grimboy>
Yeah, works for me too.

[12:53:03.0000] <guardian>
i just get an html page with an empty body

[12:53:04.0000] <guardian>
i'm using firefox

[12:53:05.0000] <Philip`>
(Works for me too :-) )

[12:54:00.0000] <Philip`>
(but the multipage one works quicker)

[12:54:01.0000] <guardian>
duh

[12:54:02.0000] <guardian>
very strange

[12:54:03.0000] <guardian>
surprisingly ie7 displays it 

[12:54:04.0000] <guardian>
well thanks for the other link 

[12:55:00.0000] <guardian>
while i'm at it, is there a way i could make a canvas element have the same size as the client area of the browser window ?

[12:56:00.0000] <Philip`>
Is it adequate to rescale it with CSS, or do you want the bitmap dimensions to change too?

[12:58:00.0000] <guardian>
well i'm trying to make it have the same size as the client area, then i want to draw something inside

[12:59:00.0000] <zcorpan_>
btw, do sites rely on image maps being css pixels rather than following the size of the image? i've found 1 site that doesn't work correctly because of this behavior

[12:59:01.0000] <guardian>
i tried <canvas style="width: 100%"> but firefox keeps giving me a 300px wide canvas

[12:59:02.0000] <zcorpan_>
http://sodertalje.se/ when you change the text size

[12:59:03.0000] <zcorpan_>
the banner at the top

[13:00:00.0000] <Philip`>
guardian: I'd expect that to give a 300x150 bitmap which is rescaled to 100% of the width of whatever

[13:00:01.0000] <Philip`>
If you want to change the bitmap's size, you need <canvas width=... height=...>

[13:00:02.0000] <Philip`>
or set canvas.width=... etc with JavaScript

[13:00:03.0000] <zcorpan_>
or rather, Isac Lagerblad found it

[13:01:00.0000] <guardian>
you mean drawing happens in the 300x150 space then it's interpolated ?

[13:01:01.0000] <Philip`>
guardian: Yes (since the bitmap's default size is 300x150 if you don't specify width= anywhere)

[13:03:00.0000] <guardian>
ok i did not know bitmap's default size was 300x150

[13:03:01.0000] <guardian>
i'm new to this

[13:03:02.0000] <guardian>
(to canvas)

[13:04:00.0000] <Philip`>
http://www.whatwg.org/specs/web-apps/current-work/multipage/section-the-canvas.html#width3 is where the default sizes are defined

[13:04:01.0000] <guardian>
gonna try to play a bit with javascript to adjust the size then

[13:05:00.0000] <Philip`>
It might be useful to know that the canvas gets reset to its default empty state whenever you resize it, which might be a pain if you're making it dynamically resize to fit some other object

[13:06:00.0000] <guardian>
thx

[13:06:01.0000] <Philip`>
/me wonders [in a completely unrelated direction] if anybody actually likes/uses server-side image maps nowadays

[13:07:00.0000] <Philip`>
I get the impression that they were much more popular when HTML4 was being developed, so it was more worthwhile trying to handle them accessibly; but now anybody who cares about accessibility won't be using server-side image maps at all

[13:11:00.0000] <hsivonen>
Philip`: I use quite frequently a service that uses server side image maps without using ismap

[13:11:01.0000] <hsivonen>
dunno how they do it

[13:11:02.0000] <hsivonen>
http://aikataulut.ytv.fi/reittiopas/fi/?mm=trip&sm=&m=&adv=&map=1&n=&a=8411&b=8127&c=&keya=Kuusitie%252525252525252010&keyb=Karvaamokuja%25252525252525204&keyc=&an=10&bn=4&cn=&date=20070815&time=2216&params=1l3l1l3l2l2ll1l1l1l1l1l0&temp_a=2550006%2C6676237&temp_b=2549061%2C6678598&temp_c=&time=2216&date=20070815&which=&px=2547000&py=6679334&ppx=&ppy=&zx=2545666&zy=6670268&zoom=1&width=600&height=400&route=t0a2550006a6676237t0a0a0t9a183a90t1a35b20t9b

[13:11:03.0000] <hsivonen>
that's an example

[13:13:00.0000] <Lachy>
hsivonen: I don't see the example image map

[13:13:01.0000] <Philip`>
That mostly just says "Valitettavasti paikkaa/reittiä ei voitu näyttää kartalla." which I guess translates to "sorry, there isn't an image map on this page" or something similar :-(

[13:14:00.0000] <Lachy>
do we have to be logged in or something?  Or have some cookie?

[13:14:01.0000] <hsivonen>
hmm. wfm. cookies or something, I guess

[13:15:00.0000] <hsivonen>
try http://aikataulut.ytv.fi/reittiopas/en/?keya=pasila&a=17816&an=&bb=17838%3At1a2551837a6673309%3AKamppi%3A740%3A&bn=&keyb=kamppi&hour=22&min=23&vm=1&day=15&month=08&year=2007&va=2&adv= and one of the show route links

[13:16:00.0000] <Philip`>
It's odd how <a href=.><img usemap=#m ismap></a> in IE appears to send 'click' to the 'a' (in areas not covered by <area>s), but then doesn't actually load the page after you've clicked on it

[13:16:01.0000] <Lachy>
hsivonen: I clicked show route from that page and got the map

[13:17:00.0000] <hsivonen>
Lachy: so the map is in practice a server-side image map

[13:17:01.0000] <hsivonen>
Lachy: but in HTML terms, it is not

[13:18:00.0000] <Philip`>
That's just an <input type=image>

[13:18:01.0000] <Lachy>
hsivonen: yes it is.  It uses <input type=image src=...>, which sends the x,y coords when click

[13:19:00.0000] <hsivonen>
whoa. I'm really bad at viewing source then

[13:19:01.0000] <Philip`>
/me used Firebug -> Inspect :-)

[13:20:00.0000] <Lachy>
checking the HTTP request headers revealed the &x=00&y=00

[13:20:01.0000] <Lachy>
XRAY would have helped too

[13:26:00.0000] <Philip`>
Bah, I can't find anything crazy or inconsistent about how Firefox and Opera handle <input usemap> :-(

[13:27:00.0000] <Philip`>
Oh, except that Opera applies it to type!=image inputs too

[13:27:01.0000] <Lachy>
lol

[13:28:00.0000] <Lachy>
I tested non image inputs yesterday in Firefox hoping to find a bug like that.  I should have checked opera

[13:38:00.0000] <zcorpan_>
Philip`: have you filed a bug report?

[13:40:00.0000] <Philip`>
Hmm, I'm not actually sure it is a bug

[13:41:00.0000] <Philip`>
HTML4 appears to say that INPUTs with 'ismap' must have type "image", but it says nothing about usemap

[13:42:00.0000] <Philip`>
(<input usemap> seems to just be covered by "The MAP element specifies a client-side image map (or other navigation mechanism) that may be associated with another elements (IMG, OBJECT, or INPUT)." which doesn't even make grammatical sense)

[13:51:00.0000] <Lachy>
if I were to narrow down the selection for the 3 forms task forms members, I'd have to go with at least 1 or 2 (if not 3) of these people: Hixie, Maciej, Anne, Matthew Raymond and maybe Ben Boyle

[14:24:00.0000] <Philip`>
http://yuiblog.com/blog/2007/08/13/rte-notes/ - "My biggest problem with the native execCommand method (in all browsers) is that the browser doesn?t tell you what it applied the command to. ... The world of JavaScript editors would be so much more civilized if this would happen (hint, hint, nudge, nudge)."

[14:24:01.0000] <Philip`>
That sounds like a feature request for http://www.whatwg.org/specs/web-apps/current-work/#command

[14:27:00.0000] <jgraham__>
http://blog.ianbicking.org/2007/08/14/reflection-and-description-of-meaning/

[14:58:00.0000] <Hixie>
someone moved universal access to the disputed section

[14:58:01.0000] <Hixie>
who the hell is disputing universal access

[15:02:00.0000] <tndH>
a slight disagreement over wording apparently makes it disputed

[15:22:00.0000] <hsivonen>
some of my photos on flickr from May still lack proper *titles* even... let alone descriptions

[15:24:00.0000] <Philip`>
Would you mind if the validator said those pages were in error, or would it be a useful reminder that you should go back and fix them some time?

[15:25:00.0000] <hsivonen>
Philip`: I already stuff approximate filler data in there, so a validator wouldn't know

[15:25:01.0000] <hsivonen>
(I have a sensible title for each photo set and I duplicate the title of the set on untitled photos)

[15:27:00.0000] <hsivonen>
I don't expect to give proper titles to all my photos in my lifetime

[15:27:01.0000] <hsivonen>
I don't expect to describe them all, either.

[15:31:00.0000] <Philip`>
If there hadn't been a decade of alt being required and everybody telling everybody they must use it, would you (or, more particularly, other people) ever put in alt descriptions of images even when they were appropriate?

[15:31:01.0000] <Philip`>
(I don't know the answer to that, so I can only guess)

[15:32:00.0000] <hsivonen>
Philip`: on flickr photos, definitely no. on diagrams in my thesis, probably

[15:32:01.0000] <hsivonen>
Philip`: it between, hard to say

[15:34:00.0000] <Philip`>
I'd probably guess that if validators couldn't complain about missing alts, and if the conformance rules for it were as complicated as they are now (i.e. you must use alt except when it's too hard and then you mustn't use it) so you couldn't easily explain it to someone, there would be less usage of alt text (both correct and incorrect)

[15:56:00.0000] <zcorpan_>
http://simon.html5.org/temp/zon/001.htm -- the framework should be pretty good now, so now i can start actually writing tests using it... :)

[16:15:00.0000] <jgraham__>
/me is supremely unconcerned with whether his flickr page validates

[16:17:00.0000] <jgraham>
But perhaps I should be. I mean I have never considered it as an issue when choosing where to put my photos

[16:17:01.0000] <jgraham>
I guess valid markup would be easier to scrape

[16:18:00.0000] <jgraham>
if I wanted to retrieve data and the API was taken away

[16:18:01.0000] <Hixie>
man if http://my.opera.com/oedipus/albums/ is supposed to be an example of good alt text, my standards are too high

[16:18:02.0000] <Hixie>
(i tried switching to no CSS and turned off images and that page and its subpages became unusable)

[16:24:00.0000] <jgraham>
Yeah, those images seem to have no useful alt text

[16:26:00.0000] <Philip`>
Hixie: Maybe it indicates that your perspective is not the same as that of the target audience for alt text

[16:29:00.0000] <Hixie>
Philip`: i'd love to see that perspective

[16:40:00.0000] <Hixie>
Philip`: so... i'm basically ignorant of the statistics stuff, which i should have learnt at university but it seems i've forgotten all of it

[16:40:01.0000] <Hixie>
Philip`: how can i determine what the accuracy of my numbers is?

[16:40:02.0000] <Hixie>
that is, how can i work out the error margin?

[16:40:03.0000] <Hixie>
the way that you do

[16:44:00.0000] <Hixie>
(feel free to reply here, i'll be back online in about 25 minutes)

[16:44:01.0000] <Philip`>
I'm basically ignorant of it too, but I think I can remember just enough to find the parts of Wikipedia that look relevant and non-vandalised :-)

[16:46:00.0000] <Philip`>
When looking at a random sample of pages and getting a binary value for each (e.g. page does/doesn't have <blink>), the measured mean should follow a binomial distribution (i.e. each value is 'yes' with a probability p, independent of all other values)

[16:47:00.0000] <Philip`>
and, as in <http://en.wikipedia.org/wiki/Binomial_distribution#Normal_approximation>, that can be approximated as a normal distribution with mean n*p and variance n*p*(1-p) (as long as n*p and n*(1-p) are large enough, like about 10)

[16:47:01.0000] <Philip`>
and then standard deviation = sqrt(variance) = sqrt(n*p*(1-p)) (by definition)

[16:48:00.0000] <Philip`>
and, as in <http://en.wikipedia.org/wiki/Normal_distribution#Standard_deviation>, about 95% of values are within two standard deviations of the mean

[16:50:00.0000] <Philip`>
so there is a 95% probability that the real population mean (from which the sample was taken) is within 2 std devs of the measured sample mean

[16:51:00.0000] <Philip`>
p is the real population mean, which you can't know since you're just looking at a sample, but there's probably some theory somewhere that says it's valid to do this with p = the measured sample mean

[16:52:00.0000] <Philip`>
which means you can count n (sample size), calculate p (proportion of sample which were true), then with 95% confidence the real population mean is n*p +/- 2*sqrt(n*p*(1-p))

[16:53:00.0000] <Philip`>
But I wouldn't mind checking with someone who really knows how to do statistics to make sure that's all valid and not just cheating :-)

[16:57:00.0000] <Philip`>
(At least that final equation has roughly the right behaviour - the proportional error decreases with sqrt(n), so using a million times larger sample only gives an extra three decimal places of accuracy)

[16:57:01.0000] <Philip`>
(which is why I'm happy looking at eight thousand web pages instead of eight billion :-) )

[16:58:00.0000] <Philip`>
(...at least for things which are >1%, because otherwise n*p gets really small and the data becomes useless if the sample isn't huge)


2007-08-16
[17:04:00.0000] <jgraham>
Philip`: That final answer looks familiar

[17:10:00.0000] <Philip`>
Oops, I think I gave Robert Burns the wrong idea about how <input usemap> is currently supported

[17:11:00.0000] <Philip`>
(I'm not sure why he doesn't just try the example in Firefox and see that it works like it's meant to...)

[17:18:00.0000] <Hixie>
ok so if p=0.0001878 (fraction of pages with a <td<> element [sic] in my sample)

[17:19:00.0000] <Hixie>
and n=3.5e9

[17:20:00.0000] <Hixie>
then the error is 6e6e-7

[17:21:00.0000] <Hixie>
which if n is, say, 650000, then that's about +/- 2000

[17:21:01.0000] <Hixie>
which is actually smaller than the error margin that my methodology introduces in and of itself, heh

[17:22:00.0000] <Hixie>
interesting

[17:22:01.0000] <Hixie>
(my numbers are obtained using some statistical estimators that have, in the case of this particular sample, an error margin of +/- 4000 or so)

[17:36:00.0000] <Philip`>
I'm not sure how you're going from n=3.5e9 to n=650000

[17:37:00.0000] <Hixie>
er, s/n/p/ and s/650000/650000 div 3.5e9/

[17:41:00.0000] <Philip`>
Did you miss the sqrt?

[17:42:00.0000] <Philip`>
With n=3.5e9, p=0.0001878, then n*p*(1-p) =~ 650000, and 2*sqrt(n*p*(1-p)) =~ 1600

[17:42:01.0000] <Philip`>
Ah, I think that about makes sense with what you said

[17:43:00.0000] <Philip`>
(There is an assumption somewhere that the population is infinite, so it might be necessary to be a bit careful if the sample size is a significant fraction of the actual population, but I have no idea how relevant that is)

[17:46:00.0000] <mpt>
/me is confused by the aspect ratio restriction for <img>

[17:47:00.0000] <mpt>
"If both [width and height] attributes are specified, then the ratio of the specified width to the specified height must be the same as the ratio of the logical width to the logical height in the image file."

[17:48:00.0000] <mpt>
Does that mean in <https://translations.launchpad.net/ubuntu> we're doing something wrong? (pretending for the moment that that page is HTML 5)

[17:51:00.0000] <Hixie>
Philip`: sorry i wasn't clear

[17:51:01.0000] <Hixie>
Philip`: let me try again

[17:51:02.0000] <Hixie>
Philip`: if i have N=3.5e9 (number of pages examined)

[17:51:03.0000] <Hixie>
Philip`: and r=650000 (number of pages with an element <tr<>

[17:52:00.0000] <Hixie>
Philip`: then p=0.000186 ish

[17:54:00.0000] <Hixie>
Philip`: and the error (95% confidence the real population mean r_real is within this much of r) is 2*sqrt(n*(r/n)*(1-r/n))

[17:54:01.0000] <Hixie>
right?

[17:54:02.0000] <Hixie>
which is about 2000 (to 1sf)

[18:40:00.0000] <Philip`>
Hixie: That sounds highly plausible to me

[18:41:00.0000] <Hixie>
ok

[18:41:01.0000] <Hixie>
so that error is about half of the error that i get just from the statistical estimators that i use in getting my data

[18:44:00.0000] <Hixie>
btw i think one thing that's being missed in this input/usemap discussion is that the majority (all but three of the 50000+ that have been examined) of uses of input/usemap are either no-ops or actually result in _worse_ user experience in browsers that support usemap

[18:45:00.0000] <Hixie>
and that therefore browsers are better off not implementing it if they want to compete over their handling of existing pages

[18:46:00.0000] <Philip`>
They're not better off to a sufficiently significant extent that Mozilla or Opera have bothered changing their implementation because of it, so that doesn't seem like an especially important issue

[18:47:00.0000] <Hixie>
fair enough

[18:48:00.0000] <Philip`>
Oops, seems I responded to an old message of Rob Burns after he already said it was wrong...

[18:49:00.0000] <Philip`>
But I still don't understand "the image map areas should instead submit the form with the coordinates (again, there doesn't have to be a server-side to this; some use-cases do not require a server-side round-trip)" - how do you submit a form without a server side?

[18:49:01.0000] <Hixie>
don't ask me

[18:49:02.0000] <Hixie>
i don't have any idea what he's talking about

[18:54:00.0000] <Philip`>
Maybe you could do an accessible email-based version of Where's Wally, so you have to find and click on him in the image and then it submits the coordinates to a mailto: link, and to be accessible it also has a client-side image map with an <area> per person in the image where the alt text is the name of the person and it submits the email form with the person's coordinates when you choose one

[18:55:00.0000] <Philip`>
/me struggles to think of more useful examples

[19:01:00.0000] <Hixie>
i have written a script that reads the mbox files for public-html, and scores each participant by giving them 1 point per e-mail sent and 1 point per e-mail that was a reply to their e-mail.

[19:02:00.0000] <Philip`>
Does that only count direct replies?

[19:02:01.0000] <Hixie>
yes

[19:02:02.0000] <Hixie>
it counts them via In-Reply-To headers

[19:03:00.0000] <mjs>
I wonder4 who scores the highest?

[19:04:00.0000] <Philip`>
I'd tend to assume most messages only get one or two replies, because all the annoyingly large conversations are annoyingly long rather than wide

[19:05:00.0000] <Philip`>
in which case the score would be mainly dependent on how many emails the particular individual sends, and not affected much by the replies

[19:18:00.0000] <Hixie>
so i'm actually relatively surprised as to who scores the worst on this thing

[19:19:00.0000] <Philip`>
Are you assuming high scores are bad? :-)

[19:19:01.0000] <Hixie>
yes :-)

[19:20:00.0000] <Philip`>
Depending on who's at the top, you could say it shows who has blessed us most with their knowledge, experience and ideas

[19:21:00.0000] <Hixie>
the person at the top is mostly at the top because they have triggered so many replies

[19:23:00.0000] <mjs>
http://www.zeldman.com/2007/08/15/what-crisis/

[19:23:01.0000] <mjs>
who's got the high score? I gotta know

[19:24:00.0000] <Hixie>
mjs: you

[19:25:00.0000] <mjs>
I'm not all that surprised, if it's mainly through triggering replies

[19:25:01.0000] <Hixie>
i think the results would be very different if i only did the last month

[19:26:00.0000] <mjs>
well that's true

[19:27:00.0000] <mjs>
I've hardly sent any mail, let alone had lots of replies

[19:28:00.0000] <Hixie>
bbl

[20:09:00.0000] <mpt>
Hixie, the definition of alt= for <area> is much vaguer than the definition of alt= for <img>

[20:13:00.0000] <mpt>
Perhaps the former should reference the latter somehow

[20:30:00.0000] <Lachy>
Hixie, what's my score in the email count?

[20:31:00.0000] <Lachy>
and what position do I get?

[20:44:00.0000] <Hixie>
http://junkyard.damowmow.com/290

[20:46:00.0000] <Lachy>
7th isn't too bad.  I correct guessed who would get second, though I would have expected him to get first if mjs hadn't

[20:47:00.0000] <mjs>
wow, I'm the only one who elicited more responses than emails sent?

[20:47:01.0000] <mjs>
no, I guess not

[20:47:02.0000] <mjs>
misread chart at first

[20:47:03.0000] <Lachy>
it's a shame it doesn't distinguish between productive discussion and useless trolling

[20:49:00.0000] <mjs>
regarding contribution value of the #2 entry on the list, I present this: http://bugs.webkit.org/show_bug.cgi?id=14945

[20:51:00.0000] <mjs>
(but maybe it's just a different guy with the same name?)

[20:51:01.0000] <Lachy>
Hixie, for alt text in "A key part of the content that has no textual alternative", the spec should probably try to emphasise more that alt text should be provided if possible

[20:52:00.0000] <Lachy>
"In such cases, the alt attribute must be omitted." should be reprhased to say "... may be omitted, but should be provided if at all possible" or something like that

[20:54:00.0000] <Lachy>
mjs: it's the same person, he posted about the same topic to public-html

[20:54:01.0000] <Lachy>
suggesting that XHTML5 should somehow more clearly specify XML parsing

[20:55:00.0000] <Lachy>
(though I didn't bother paying too much attention to it)

[20:57:00.0000] <Lachy>
oh, I understand Gregory's point about his photo album now!  http://my.opera.com/oedipus/albums/showpic.dml?album=212490&picture=3302055

[20:58:00.0000] <Lachy>
he wasn't suggesting that "perception - photography - image interpretation - blindness" is good alt text, he was suggesting that the comments people have provided for him are good descriptions

[21:01:00.0000] <mjs>
really the relevant alternative isn't good alt text vs no alt text

[21:01:01.0000] <mjs>
alt="" has always been allowed

[21:01:02.0000] <mjs>
the question is just whether leaving alt out entirely is in some cases better than alt=""

[21:02:00.0000] <mpt>
It's better than authoring tools producing automated alt="" for sticking-plaster compliance

[21:02:01.0000] <mjs>
yes, I think so too

[21:03:00.0000] <mjs>
but there's incentive effects both ways

[21:03:01.0000] <mjs>
the HTML4 way made it likely that images with no provided alternative text would be treated the same as decorative images, which is harmful to blind users or users of text-only browsers

[21:03:02.0000] <mpt>
For the same reason that good transcribers include "[inaudible]" in transcripts, instead of just omitting the words/phrases they didn't hear

[21:04:00.0000] <mjs>
the (current draft) HTML5 way might do less to encourage thoughtful authors writing by hand to add good alt text

[21:04:01.0000] <mjs>
since the conformance checker won't remind them

[21:04:02.0000] <mjs>
that was one potential advantage of noalt

[21:04:03.0000] <mjs>
you could verify that all your images that lack alt text are doing so on purpose

[21:05:00.0000] <mpt>
ooh, noalt

[21:05:01.0000] <mpt>
that's a nifty idea

[21:05:02.0000] <mpt>
but then would authoring tools produce automated noalt...

[21:06:00.0000] <mjs>
presumably, yes

[21:06:01.0000] <mjs>
but that is better than them adding alt=""

[21:06:02.0000] <mjs>
and it makes it harder to forget to add alt accidentally when authoring by hand

[21:07:00.0000] <mjs>
I could see the argument either way

[21:07:01.0000] <mpt>
Maybe there could be an authoring-tool exception, like there is/was for <font>

[21:07:02.0000] <mjs>
ironically I think Gregory's image gallery shows a valid use case for omitted alt

[21:08:00.0000] <mpt>
Apologies if I'm unwittingly rehashing suggestions made in public-html@

[21:08:01.0000] <mjs>
I think this was mostly discussed on the whatwg list actually

[21:08:02.0000] <mjs>
I don't think the authoring tool exception is a good concept

[21:08:03.0000] <mjs>
and doesn't really work well for user-generated content embedded in a hand-authored dynamic page

[21:09:00.0000] <mpt>
Whenever anyone says "user-generated content" I want to throw something

[21:09:01.0000] <mpt>
but anyway

[21:10:00.0000] <mjs>
if you have a better phrase to use for what I'm referring to I would be happy to use that instead

[21:10:01.0000] <mjs>
I mean to distinguish content created by casual end-users who are doing it incidentally, not professionally or as their primary goal

[21:11:00.0000] <mpt>
Well, part of the problem with it is that I don't know what you're referring to

[21:11:01.0000] <mpt>
Photos?

[21:11:02.0000] <mpt>
Weblog comments?

[21:11:03.0000] <mjs>
photos posted to a photo sharing site, weblog posts, comments on a review site like yelp...

[21:11:04.0000] <mpt>
Forums drizzled with emoticons?

[21:11:05.0000] <mjs>
email messages

[21:11:06.0000] <mjs>
web forums

[21:11:07.0000] <mjs>
IM messages

[21:11:08.0000] <mpt>
So what are you excluding?

[21:12:00.0000] <mjs>
profesionally developed web sites

[21:12:01.0000] <mjs>
I should have said "web forum comments"

[21:12:02.0000] <mpt>
If it's a hand-authored page, I'm not sure how the authoring tool exception kicks in at all

[21:12:03.0000] <mpt>
oh, unless it's a Weblog comment system that allows images?

[21:12:04.0000] <mjs>
I would distinguish the web forum itself ("profesionally developed" in some sense), from comments posted by users

[21:12:05.0000] <mjs>
right

[21:12:06.0000] <mpt>
but they usually don't anyway, to prevent CSRF

[21:13:00.0000] <mjs>
but photo sharing sites allow embedding images

[21:13:01.0000] <mpt>
Maybe "usually" is overstating it

[21:13:02.0000] <mjs>
and email programs

[21:13:03.0000] <mpt>
ok, so e-mail programs would use the authoring-tool exception

[21:13:04.0000] <mpt>
while photo sharing sites would use noalt

[21:14:00.0000] <mjs>
now I'm not sure what your proposal is

[21:14:01.0000] <mpt>
That the authoring-tool exception save authoring tools from having to use alt="" or noalt or whatever they think the best default is

[21:16:00.0000] <mpt>
so that the UA can tell whether any thought has been put into the alternate text

[21:17:00.0000] <mjs>
if noalt existed I think the right thing would be to treat it the same as missing alt

[21:17:01.0000] <mjs>
so I don't think that would buy anything

[21:17:02.0000] <mjs>
the question is just whether noalt is worth it

[21:17:03.0000] <mpt>
While a photo sharing site knows that the photos are the point of the site, and can therefore apply noalt to all the images.

[00:25:00.0000] <G0k>
oy all

[00:36:00.0000] <G0k>
so er...more questions about the server-side events thing

[00:38:00.0000] <G0k>
when dealing with boolean event attributes, should I follow the pattern specified for Cancelable and Bubbles (i.e. "No" is false but everything else is true) or should it treat "false" and "0" as false also?

[00:46:00.0000] <Lachy>
G0k: IIRC, the default cancelable and bubbles values are defined for each event in the DOM Events spec, and I think HTML5 gives default values for custom events (though I'd have to check the spec)

[00:46:01.0000] <G0k>
yeah i got that

[00:47:00.0000] <G0k>
i just mean how should i parse the fields

[00:47:01.0000] <G0k>
for example, for the MouseEvent interface

[00:47:02.0000] <G0k>
if i get, say

[00:47:03.0000] <G0k>
ctrlKey: false

[00:47:04.0000] <Lachy>
doesn't the spec define that?

[00:47:05.0000] <G0k>
well it says

[00:47:06.0000] <G0k>
"If the specified interface has an attribute that exactly matches the name of the field, and the value of the field can be converted (using the type conversions defined in ECMAScript) to the type of the attribute, then it must be used."

[00:47:07.0000] <G0k>
but that doesn't really make sense for booleans

[00:47:08.0000] <G0k>
i don't think any string in ECMAScript evaluates to false

[00:48:00.0000] <Lachy>
hmm. interesting

[00:48:01.0000] <G0k>
plus, that's inconsistent

[00:48:02.0000] <G0k>
"No" doesn't evaluate to false in ECMAScript

[00:49:00.0000] <Lachy>
you could send an email to one of the lists and ask

[00:49:01.0000] <Lachy>
it may be a bug in the spec

[00:50:00.0000] <G0k>
oki

[01:58:00.0000] <takkaria>
/me considers unsubscribing from public-html :/

[02:05:00.0000] <G0k>
what'd they do?

[02:06:00.0000] <takkaria>
just constant sniping and personal attacks and seemingly knee-jerk reactions

[02:06:01.0000] <takkaria>
to /everything/

[02:07:00.0000] <G0k>
yeah so fuck em. revolution up in here

[03:16:00.0000] <hsivonen>
Hixie: I think the score would illustrate the issue better if you divided it by # of days the person has been on the list. (yeah, I realize that developing the score script isn't your main activity)

[03:20:00.0000] <Hixie>
hsivonen: hehe, not a bad idea. time from first post to last post?

[03:21:00.0000] <Hixie>
mpt: see the recent large alt text revamp e-mail i bcc'ed you on

[03:21:01.0000] <Hixie>
so i'm thinking maybe the lack of an alt="" attribute should also be a sign of a "low quality" page, though maybe that would just make people set it to alt="". (empty)

[03:21:02.0000] <hsivonen>
Hixie: yes

[03:22:00.0000] <hsivonen>
Hixie: or time from first post to today

[03:25:00.0000] <G0k>
should specify that not having an alt attribute means that the UA must not load any banner ads on that page

[03:26:00.0000] <G0k>
designers never fix anything until you mess with their money

[03:28:00.0000] <G0k>
i suppose it would just lead to people setting alt attributes like alt="asz89aj9g" but still

[03:41:00.0000] <Whiskey_M>
G0K, trust me that would be even worse - also what if you have a page with no images on that aren't purely for display.  Therefore they shouldn't have any alt

[03:41:01.0000] <G0k>
see I almost wonder if there should really be like two different tags

[03:42:00.0000] <Whiskey_M>
/me would be intrigued by the stats of pages using primarily alt="" which is AFAIK pretty much a kick back against the measurement of automated accessibility checkers

[03:42:01.0000] <G0k>
an element for images with semantic meaning, then another element for those without

[03:42:02.0000] <G0k>
thus people who don't care could go on using img-no-alt for their purely display images

[03:43:00.0000] <G0k>
then people who cared could give use a new image element which absolutely needed the alt tag

[03:45:00.0000] <G0k>
er attribute

[03:45:01.0000] <G0k>
thing

[03:46:00.0000] <Whiskey_M>
/me doesn't see what would necessarily be gained - off to a meeting now, but will be thinking about it

[04:21:00.0000] <Hixie>
http://junkyard.damowmow.com/291

[04:22:00.0000] <Hixie>
lachy and i don't even appear on that chart

[04:24:00.0000] <Lachy>
that aligns with my expectations better than the last one - it has expected winner!

[04:24:01.0000] <Hixie>
we come somewhat below the top 20

[04:24:02.0000] <Hixie>
 2.59    218    199  0.91    161  Dan Connolly <connolly⊙wo>

[04:24:03.0000] <Hixie>
 2.50     23     24  1.04     19  John Boyer <boyerj⊙cic>

[04:24:04.0000] <Hixie>
 2.49    171    211  1.23    154  Lachlan Hunt <lachlan.hunt⊙lia>

[04:24:05.0000] <Hixie>
 2.44      2      3  1.50      3  Shane McCarron <shane⊙ac>

[04:24:06.0000] <Hixie>
 2.36    188    191  1.02    161  Ian Hickson <ian⊙hc>

[04:24:07.0000] <Hixie>
 2.35      2      3  1.50      3  Geoffrey Sneddon <geoffers⊙gc>

[04:25:00.0000] <Lachy>
how is the score calculated?

[04:26:00.0000] <Hixie>
(sent+recvd)/days

[04:26:01.0000] <Hixie>
more or less

[04:26:02.0000] <Hixie>
it's actually done by seconds

[04:27:00.0000] <Hixie>
if i merge tina's two accounts she drops off the top 20 too

[04:28:00.0000] <Lachy>
ok, so the score is the average number of emails sent per day

[04:29:00.0000] <Hixie>
right

[04:29:01.0000] <Hixie>
sent and received

[04:29:02.0000] <Lachy>
yeah

[04:30:00.0000] <Lachy>
so Robert sends 6.63 emails per day

[04:31:00.0000] <Lachy>
I wonder if there's some correlation between the number of posts sent+received per day and the quality of the post

[04:34:00.0000] <hsivonen>
Hixie: thanks for the stats

[04:34:01.0000] <Hixie>
are philip⊙zdcu and excors⊙gc the same person?

[04:34:02.0000] <Hixie>
or are those two different philips?

[04:35:00.0000] <Lachy>
Hixie: I think they're the same person, we're fairly sure there's only 2 Philip Taylors

[04:36:00.0000] <Hixie>
really? i thought there were three.

[04:38:00.0000] <zcorpan_>
http://www.w3.org/2000/09/dbwg/details?group=40318&public=1 lists 2

[04:38:01.0000] <Lachy>
excors⊙gc is Philip`

[04:38:02.0000] <Hixie>
ok

[04:38:03.0000] <hsivonen>
zaynar is Philip` too

[04:38:04.0000] <Hixie>
ok

[04:38:05.0000] <Hixie>
excellent

[04:38:06.0000] <Lachy>
@zaynor is the same as P.Taylor⊙RAU

[04:38:07.0000] <Hixie>
no need to special case the philips in my script!

[04:38:08.0000] <hsivonen>
Hixie: I believe rhul and turnbridge (sp?) are one as well

[04:39:00.0000] <hsivonen>
Lachy: huh?

[04:39:01.0000] <Hixie>
wait, wait, you both just contradicted yourselves

[04:39:02.0000] <Lachy>
oh, maybe I'm mistaken

[04:39:03.0000] <Lachy>
let me check...

[04:39:04.0000] <Hixie>
i have:

[04:39:05.0000] <Hixie>
Philip TAYLOR <Philip-and-LeKhanh⊙RO>

[04:39:06.0000] <Hixie>
Philip Taylor (Webmaster) <P.Taylor⊙RAU>

[04:39:07.0000] <Hixie>
Philip Taylor <excors⊙gc>

[04:39:08.0000] <Hixie>
Philip Taylor <philip⊙zdcu>

[04:40:00.0000] <hsivonen>
Hixie: my belief is the the first two are one person and the second two are another person

[04:40:01.0000] <Hixie>
my script will turn those into three, collapsing the bottom two

[04:40:02.0000] <Hixie>
ok

[04:40:03.0000] <Hixie>
so i do need to special case them

[04:41:00.0000] <Lachy>
I thought there were only 3 email addresses for the Philips, so I assumed the @zaynar one had to be the same as the other

[04:42:00.0000] <Lachy>
Philip`: yt? Care to settle this mystery about the last 2 email addresses?

[04:43:00.0000] <Philip`>
The last two are me, and the first two are not me

[04:43:01.0000] <Philip`>
(I don't remember having sent to the list with any other email address too)

[04:48:00.0000] <Lachy>
Philip`: you have about 22 times by my count of my local archive

[04:49:00.0000] <Lachy>
oh, wait, I misread your last statement.

[04:49:01.0000] <Philip`>
I meant I don't remember ever using any other than those two

[04:50:00.0000] <Lachy>
yeah, I realise that now, I thought it said you didn't remember using the latter email address

[04:56:00.0000] <Hixie>
wow, a lot of people use mutliple addresses

[04:57:00.0000] <Hixie>
/me adds generic name collapsing to this script

[04:58:00.0000] <Hixie>
http://junkyard.damowmow.com/291 updated

[04:58:01.0000] <zcorpan_>
/me has used zcorpan@hotmail and zcorpan@gmail on whatwg, and zcorpan@gmail and simonp⊙oc on public-html

[04:59:00.0000] <Hixie>
yeah my script collapses those all together now

[05:01:00.0000] <hsivonen>
zcorpan_: you have an admirable skill of staying out of the ratholes on public-html

[05:04:00.0000] <Lachy>
Hixie, can you plot the volume of email from each user over time on a graph?

[05:05:00.0000] <Hixie>
i could

[05:05:01.0000] <Hixie>
not gonna :-)

[05:05:02.0000] <zcorpan_>
hsivonen: thanks :)

[05:05:03.0000] <Hixie>
if you know perl i'm happy to send you my script and you can repurpose it

[05:06:00.0000] <Lachy>
dammit Hixie! I demand that you do my research for me! :-)

[05:06:01.0000] <Lachy>
unfortunately I don't know perl

[05:06:02.0000] <Hixie>
:-)

[05:07:00.0000] <Philip`>
Perl is easy to learn - it's just like Python without all the words

[05:07:01.0000] <hsivonen>
perl gives you a hangover. python doesn't

[05:08:00.0000] <Hixie>
perl is just misunderstood

[05:08:01.0000] <Hixie>
mostly because it's hard to understand, but still

[05:09:00.0000] <hsivonen>
by hangover I mean that the day after when one reads one's own perl code, one goes wtf

[05:20:00.0000] <Hixie>
btw spec is updated

[05:25:00.0000] <Hixie>
with new alt text advice

[05:30:00.0000] <Lachy>
Hixie: I think the new text balances the issue nicely now

[05:33:00.0000] <Philip`>
"Note that the following would be a very bad use of alternative text:" - why "very bad"? It doesn't really hurt much, except wasting a couple of seconds of someone's time when they listen to the image, so it only seems a little bit bad

[05:35:00.0000] <Lachy>
/me sets up message filter to clearly identify the sensible Philip from the other one

[05:35:01.0000] <zcorpan_>
Hixie: how does missing alt now integrate with <figure> semantics in the fallback case?

[05:36:00.0000] <Lachy>
(Philip`: now messages from you get marked with a gold star in my inbox :-))

[05:37:00.0000] <Lachy>
Hixie: maybe s/very bad/redundant/ in that paragraph

[05:37:01.0000] <Philip`>
(Lachy: I feel very special)

[05:38:00.0000] <Philip`>
(though I remember I used to like shiny silver stars more than gold stars, at least when I was 5)

[05:39:00.0000] <gsnedders>
/me tries to remember who Philip` is

[05:40:00.0000] <gsnedders>
which Philip is he?

[05:41:00.0000] <Philip`>
I'm hope that I'm the sensible one

[05:41:01.0000] <Philip`>
s/e/ing/

[05:41:02.0000] <zcorpan_>
the one who wrote the spec splitter, the 8000 tokenization tests, the tokenization diagrams, the canvas test suite

[05:41:03.0000] <zcorpan_>
and canvex

[05:42:00.0000] <gsnedders>
zcorpan_: yes, but on the mailing list, which? :P

[05:42:01.0000] <gsnedders>
the title case one?

[05:42:02.0000] <Philip`>
and three tokenisers :-)

[05:42:03.0000] <zcorpan_>
gsnedders: the other than Philip TAYLOR / Philip Taylor (Webmaster)

[05:42:04.0000] <gsnedders>
zcorpan_: yeah. that's what I thought.

[05:43:00.0000] <gsnedders>
silly common names…

[05:43:01.0000] <hsivonen>
use case for namespaces!

[05:43:02.0000] <Lachy>
Philip`: sorry, Thunderbird only gives me gold stars

[05:43:03.0000] <Hixie>
Lachy/zcorpan_: send mail, it's way past my bed time

[05:43:04.0000] <Hixie>
nn

[05:44:00.0000] <Lachy>
Hixie:  you gonna wake up for the telcon?

[05:48:00.0000] <Lachy>
hsivonen: they're email addresses are the namespaces.  The problem is they both break compatibility by using 2 each, and thus expect each reader to be aware of both!

[05:49:00.0000] <Philip`>
Readers should just apply message analysis heuristics to determine which addresses are equivalent

[05:50:00.0000] <Lachy>
that's what I did, it's just confusing because it takes reading the first few sentences to know

[05:50:01.0000] <hsivonen>
Philip`: that's how I figured there are two persons

[05:50:02.0000] <zcorpan_>
oops, perhaps i should have changed the subject line for that email... oh well

[05:50:03.0000] <Philip`>
Also, if you see two of the email addresses replying to each other, they're either distinct people or are one person who is crazy

[05:51:00.0000] <Philip`>
Anyway, I don't have any problem telling who each message came from :-p

[05:52:00.0000] <Lachy>
what about in the case of Robert Burns, who frequently responds to himself from the same email address?

[05:52:01.0000] <Lachy>
zcorpan_: I don't understand the question in your email

[05:53:00.0000] <zcorpan_>
Lachy: "If the embedded content cannot be used, then, for the purposes of establishing what the figure element represents:"

[05:54:00.0000] <Philip`>
Hmm, I meant (but didn't say) "replying" as in "responding and discussing an issue from different sides", rather than as in just having Reply-To headers that might be followup messages instead of responses

[05:56:00.0000] <zcorpan_>
Lachy: the figure section assumes that there will be fallback content (which might be empty), but no alt means that there is no fallback content

[05:56:01.0000] <Lachy>
yeah, I just read that section

[05:58:00.0000] <Philip`>
(That reduces to the message analysis heuristics case again, so I guess it's not much help)

[06:00:00.0000] <zcorpan_>
thinking about it, <embed>, <video> and <audio> don't have fallback content either

[07:14:00.0000] <zcorpan_>
Philip`: do you have any suggestions for how to create lots of testcases for serializing html fragments? i already have a framework to run test cases, now i just need to create them

[07:18:00.0000] <zcorpan_>
Philip`: the format looks like http://simon.html5.org/temp/zon/tests.dat

[07:19:00.0000] <Philip`>
The way I did it with the tokeniser was to create a load of short (mostly 1 character) strings, run the tokeniser algorithm on all of them, then remove the ones that were redundant (i.e. resulted in exactly the same tokeniser state), then made a load more slightly-longer strings by appending short strings to each of the ones from the previous stage

[07:19:01.0000] <Philip`>
...and then repeated until the strings were sufficiently long that they covered every tokeniser state

[07:20:00.0000] <Philip`>
(and then did another half-step so that it (almost) covered every transition out of every state)

[07:20:01.0000] <zcorpan_>
aha. interesting

[07:21:00.0000] <Philip`>
That requires the ability to run the tokeniser algorithm and see what state it ends up in - I have no idea if you can do the same with HTML serialisation

[07:25:00.0000] <Philip`>
(Also, it's only really testing that specific implementation of the algorithm - that works alright for the tokeniser since everyone implements it the same way, but it doesn't work in more general cases)

[07:28:00.0000] <zcorpan_>
hmm, the algorithm doesn't require any specific order for attributes

[07:28:01.0000] <zcorpan_>
which might be good for performance optimazation but not good for writing test cases

[07:56:00.0000] <hsivonen>
we are developing a *Web* spec, right? how are local drives part of the Web?

[07:58:00.0000] <Philip`>
I've seen quite a few file:/// links on web pages, though I'm not sure if that counts

[07:58:01.0000] <Lachy>
hsivonen: Google should just index file:// URIs too, and then we could survey local hard drives too ;-)

[07:59:00.0000] <Lachy>
we could survey intranets too. I'm sure there are plenty that don't have enough security and would let a bot in to fetch the pages

[07:59:01.0000] <Philip`>
<a href="file:///etc/passwd">SomeUniqueWord</a> then search Google and view the cached copy to see Googlebot's passwords

[08:00:00.0000] <Lachy>
LOL

[08:03:00.0000] <zcorpan_>
Philip`: actually, Support Existing Content applies to <input usemap>, it's just that in order to support existing content you have to remove support for the feature... :)

[08:06:00.0000] <Philip`>
I'm not really convinced by that as a reason, since the number of affected pages seems to be small enough that Mozilla and Opera haven't bothered removing the feature

[08:07:00.0000] <zcorpan_>
yet

[08:07:01.0000] <zcorpan_>
we didn't know it was a problem in supporting it

[08:07:02.0000] <zcorpan_>
:)

[08:08:00.0000] <zcorpan_>
but it's not critical

[08:09:00.0000] <Lachy>
I like Rob's claim that there were no examples presented that break! :-)

[08:10:00.0000] <Lachy>
in fact, if it were to be implemented as he imagines, it would break any of the sites that use <input usemap> as a hyperlink image map (though, they should use <img> anyway)

[08:12:00.0000] <zcorpan_>
i thought Hixie specifically tried to find valid uses of <input usemap>, and filtered out obvious misuse

[08:23:00.0000] <Lachy>
zcorpan_: it depends on your definition of misuse.  Hixie filtered out areas with no href on the grounds that it's completely useless in all existing UAs, whereas Rob was to include those and exlude those with href, because that's how his imaginary implementation works

[08:24:00.0000] <Lachy>
s/was/wants/

[08:25:00.0000] <Lachy>
but what Rob should be looking for, is pages that fulfil the use case using alternative techniques.  But he's refusing to do any work

[08:26:00.0000] <zcorpan_>
right

[08:41:00.0000] <Philip`>
/me wonders when to give up

[08:51:00.0000] <Lachy>
Philip`: just give up now.  I have.  He's clearly not interested in resolving the use case issue, and there's no point discussing it any further

[09:43:00.0000] <Lachy>
can anyone make sense of William Loughborough's recent post in the baby steps thread?

[09:48:00.0000] <hober>
It didn't make the slightest bit of sense to me.

[09:52:00.0000] <Philip`>
I have problems parsing his first sentence-paragraph

[09:52:01.0000] <Lachy>
I just carefully re-read it, looking up obscure words I'd never seen before like 'engendered' and 'deleterious', he seems to be talking about the harm  of eliminating the alt attribute and replacing it with something else

[09:52:02.0000] <Philip`>
(I can't find any way to read it that makes grammatical sense)

[09:53:00.0000] <Lachy>
nor can I.  I mostly did a sort of heuristic analysis and constructed the meaning from a few key words

[09:54:00.0000] <hsivonen>
it is totally counter-productive that people dramaticize making the alt attribute omissible in some cases as the removal of the possibility to use the attribute

[09:54:01.0000] <Philip`>
Seems easier to just ignore it :-)

[09:54:02.0000] <Philip`>
(though maybe less fun, depending on what kind of fun you like having)

[09:56:00.0000] <Lachy>
it's too stressful to deal with misunderstandings.  If someone wants to jump to conclusions about us dropping the alt attribute, good for them.  We know we're not, we're trying to improve it as a whole

[09:58:00.0000] <Philip`>
It seems the intention is to remove bad features, which nobody really minds (like all the presentational features, and frames, and @name, and script@charset, etc), but some accessibility features are bad features, and people focus on the fact that they're accessibility features rather than that they're bad features

[10:00:00.0000] <Lachy>
that's because all accessibility features are automatically *good*, regardless of how ineffective they are

[10:11:00.0000] <Lachy>
I've been thinking it might be a good idea to explain in the spec that alt text is dependent upon the context in which it's used, rather than just the image itself.

[10:11:01.0000] <Lachy>
so it's possible for the same image to be used in different contexts, and have completely different alt text

[10:12:00.0000] <mpt>
oh yes

[10:12:01.0000] <Lachy>
/me will send mail

[10:12:02.0000] <mpt>
I tried to explain that to the Wikipedia developers once

[10:12:03.0000] <Lachy>
/me wonders whether I should send to public-html or whatwg?

[10:12:04.0000] <mpt>
They were wondering if the alt text for an image could be kept on the image's page

[10:12:05.0000] <zcorpan_>
http://simon.html5.org/temp/zon/001.htm -- safari does surprisingly well

[10:14:00.0000] <Lachy>
it would be clearer if you listed the tests that pass as well.  I looked and thought it had failed

[10:14:01.0000] <Lachy>
or gave some indication about how many tests there are and how many it passed, listing the failed ones below

[10:17:00.0000] <Lachy>
mpt: do you know of an example I could list in my email?

[10:17:01.0000] <Philip`>
http://som.uthscsa.edu/ is weird - it looks like it's really trying to sensibly use <td usemap>, but that doesn't work at all in any browser I've tried

[10:18:00.0000] <Philip`>
I guess they must have started with a <img usemap> and then changed it to a <td background=... usemap>

[10:21:00.0000] <zcorpan_>
Lachy: hmm... is that good enough? :)

[10:23:00.0000] <Lachy>
oh I see it now.  I didn't notice it at first, cause it was just below the bottom of the viewport in Firefox

[10:24:00.0000] <mpt>
Lachy, unfortunately not, I couldn't find one at the time

[10:24:01.0000] <Lachy>
could you add a summary that says "Passed: 20, Failed 5" (or equivalent)

[10:24:02.0000] <mpt>
Lachy, however, Wikipedia's "This image is used in these articles:" feature might be useful

[10:25:00.0000] <zcorpan_>
Lachy: now have "Finished processing. 3 of 25 tests passed."

[10:25:01.0000] <Lachy>
yeah, that's good

[10:29:00.0000] <Philip`>
I can't find any real examples, but I can imagine a page that has a fancy logo in the header (alt text being the textual part of it, like the organisation name), where the page is about the history of the organisation and includes images of various logos that it has had (including the same one it's currently using in its header, with alt text being a description of its appearance)

[10:34:00.0000] <Lachy>
like Google's list of special event logos?

[10:34:01.0000] <Lachy>
good example!

[10:34:02.0000] <Lachy>
http://www.google.com/holidaylogos.html

[10:35:00.0000] <gsnedders>
what's worse to come across: the HTML WG, or two hunters on legendary when the rest of the UNSC forces are dead?

[10:35:01.0000] <gsnedders>
(and when you're out of practice)

[10:35:02.0000] <Lachy>
what's UNSC?

[10:36:00.0000] <Lachy>
and what does "two hunters on legendary" mean?

[10:36:01.0000] <gsnedders>
Halo.

[10:36:02.0000] <Philip`>
http://www.google.com/stickers.html - top-left logo should presumably have alt="Google", other logos should perhaps have "Small Google logo on white background" so people who want to use it on their own site can find the appropriate one

[10:37:00.0000] <Lachy>
I'll include both examples

[10:37:01.0000] <gsnedders>
Lachy: http://en.wikipedia.org/wiki/United_Nations_Space_Command and http://en.wikipedia.org/wiki/The_Covenant_%28Halo%29#Hunters

[10:37:02.0000] <Lachy>
although, perrhaps not, if the table included row headers that describeed the white, grey and black alternatives

[10:37:03.0000] <gsnedders>
Lachy: and legendary is the hardest difficulty in Halo (and Marathon, IIRC)

[10:37:04.0000] <Philip`>
The holiday logos might be interesting to consider when they're used on the main search page - the alt text should probably still just say "Google", but it'd be nice to have a longdesc or something so people can find out more about the particular image if they're interested

[10:38:00.0000] <Lachy>
indeed, it should say google, but they use the alt text given in that page

[10:38:01.0000] <Lachy>
it should have the extra info in the title attributwe

[11:08:00.0000] <Philip`>
Hmm, dangerously close to getting a "+1" on public-html...

[12:28:00.0000] <Philip`>
/me wonders what the OOXML namespaces are

[12:29:00.0000] <Philip`>
The five hundred page Primer document is full of examples like '<revisions xmlns="..." xmlns:r="...">' and I can't find anywhere saying what you actually write in place of '...'

[12:42:00.0000] <Philip`>
Oh, the five thousand page specification does mention the namespaces in various places, so that's alright

[12:43:00.0000] <Lachy>
Philip`: did you download the entire 5000 page spec?

[12:43:01.0000] <G0k>
well is "..." a valid namespace?

[12:44:00.0000] <Lachy>
G0k: no, it needs to be a URI

[12:44:01.0000] <Lachy>
(though, in practice, "..." would work since consumers simply do string comparison.  They never need to resolve the URI

[12:45:00.0000] <Philip`>
Lachy: It's only 33MB

[12:46:00.0000] <Lachy>
ok

[12:46:01.0000] <hsivonen>
Lachy: true for just about any API except XOM

[12:46:02.0000] <Lachy>
what does XOM do specially?

[12:46:03.0000] <hsivonen>
Lachy: XOM is designed to barf if the string does not look like a URI

[12:47:00.0000] <Philip`>
"<w:r><w:t>This sentence needs to be long enough to cause some kind of line br</w:t><w:softHyphen/><w:t>eaking.</w:t></w:r>" - isn't there, like, a Unicode character for soft hyphens?

[12:47:01.0000] <Lachy>
what's its criteria for looking like a URI?

[12:47:02.0000] <hsivonen>
Lachy: dunno. I haven't checked

[12:48:00.0000] <Philip`>
(Their examples of line-breaking are all rather broken because the PDF page width is wider than their "long enough" sentences so it doesn't actually demonstrate any line-breaking)

[12:51:00.0000] <zcorpan_>
Philip`: there is, U+00AD

[12:52:00.0000] <hsivonen>
Lachy: https://xom.dev.java.net/source/browse/xom/src/nu/xom/URIUtil.java?rev=1.26&view=markup

[12:52:01.0000] <G0k>
http://www.oreillynet.com/xml/blog/images/OpenXMLSchemas3.pdf

[12:52:02.0000] <Philip`>
/me wonders why they use the term "high ANSI" for all Unicode characters that are not ASCII, "complex script" or "East Asian"

[12:52:03.0000] <G0k>
is that list of namespaces at the end what you're looking for?

[12:52:04.0000] <hsivonen>
Lachy: doesn't appear to be comprehensive

[12:52:05.0000] <Lachy>
hsivonen: thanks

[12:53:00.0000] <Philip`>
zcorpan_: I guess the more relevant question is, why <w:softHyphen/> instead of &#xAD;?

[12:54:00.0000] <Philip`>
G0k: Ah, thanks, looks like it is

[12:56:00.0000] <zcorpan_>
Philip`: &#xAD; doesn't have enough angle brackets? :)

[12:57:00.0000] <G0k>
think we should just remove all non-tag stuff from our formats

[12:58:00.0000] <G0k>
<word value="hello"/><comma /><space /> <word value="world"/>

[12:59:00.0000] <Philip`>
It's not like they're that fond of angle brackets - they use one tag with a comma-separated lists of font names in an attribute, instead of something properly XMLish like <fonts><font name="1"/><font name="2"/></fonts>

[13:01:00.0000] <hsivonen>
http://about.validator.nu/htmlparser/

[13:01:01.0000] <hsivonen>
does that look like something that could be announced?

[13:02:00.0000] <hsivonen>
nope

[13:02:01.0000] <Philip`>
ODF says to use Unicode characters for soft/non-breaking hyphens

[13:02:02.0000] <Philip`>
but it does have a <text:tab/> element

[13:02:03.0000] <Philip`>
Oh, it also has <text:s/> for a space character

[13:02:04.0000] <Philip`>
(and <text:s text:c="42"/> for multiple space characters)

[13:03:00.0000] <hsivonen>
/me fixed the download link

[13:04:00.0000] <Philip`>
I guess XML isn't so great at marking up text documents when you want to have complex typography like whitespace

[13:05:00.0000] <hendry>
is there some spec that specifies img {display none;} shouldn't download the image?

[13:06:00.0000] <Philip`>
hsivonen: http://about.validator.nu/htmlparser/apidocs/nu/validator/htmlparser/sax/HtmlParser.html says "Version: $Id$" which doesn't sound terribly useful

[13:06:01.0000] <Philip`>
hendry: I believe HTML5 says they must download the image

[13:07:00.0000] <hsivonen>
Philip`: hmm. template dating from CVS days. how do I do -kkv with Subclipse?

[13:07:01.0000] <Philip`>
(or at least it says they must download the image when the image element is created, regardless of whether it's invisible via CSS)

[13:07:02.0000] <Philip`>
(but Opera doesn't do that, if I remember correctly)

[13:08:00.0000] <Philip`>
hsivonen: I have no idea with Subclipse, but you'd have to set the file's "svn:keywords" property to "Id" (for each affected file)

[13:08:01.0000] <hsivonen>
Philip`: thanks

[13:09:00.0000] <hendry>
Philip`: {display none;} seems to be a strategy for not downloading imgs on network expensive environments like handhelds/mobiles

[13:09:01.0000] <Philip`>
hsivonen: Is htmlparser-1.0.zip meant to contain a __MACOSX directory containing only hidden files?

[13:09:02.0000] <hsivonen>
Philip`: artifact of OSX

[13:10:00.0000] <Philip`>
I guessed that from the name ;-)

[13:10:01.0000] <Philip`>
(but have no idea what it's containing)

[13:11:00.0000] <hsivonen>
Philip`: HFS resource fork and metadata emulation cruft

[13:11:01.0000] <Philip`>
hendry: Do you know if browser other than Opera acts that way?

[13:11:02.0000] <Philip`>
*if any browser

[13:15:00.0000] <hendry>
Philip`: no, but I think it's a good idea if certain imgs can be marked not to download for the "@media handheld"

[13:15:01.0000] <Philip`>
hsivonen: The XSLT4HTML5 example appears to run correctly, so that seems good :-)

[13:15:02.0000] <hsivonen>
Philip`: good

[13:16:00.0000] <hsivonen>
proof by implementation that XSLT can be used with HTML5

[13:19:00.0000] <Philip`>
/me tries sorting the HTML5 spec

[13:19:01.0000] <Philip`>
"Exception in thread "main" org.xml.sax.SAXParseException: Start tag ?h1? has a non-NCName name." - oh, that didn't work :-(

[13:20:00.0000] <hsivonen>
whoa

[13:21:00.0000] <hsivonen>
Philip`: thanks

[13:21:01.0000] <hendry>
"""When the src attribute is set, the user agent must immediately begin to download the specified resource, unless the user agent cannot support images, or its support for images has been disabled."""

[13:23:00.0000] <hendry>
its support for images has  been disabled.  ==  img {display none;} ??

[13:24:00.0000] <hsivonen>
forgot the Digit production...

[13:24:01.0000] <takkaria>
hendry: no, as in the user has told the browser to not display images

[13:24:02.0000] <takkaria>
setting an image to display:none does not amount to diabling image support

[13:24:03.0000] <takkaria>
since e.g. alt text isn't displayed

[13:24:04.0000] <hendry>
that's really binary when it comes to images.

[13:25:00.0000] <hendry>
though I guess it could prevent odd things happening.

[13:30:00.0000] <hsivonen>
Philip`: fixed

[13:31:00.0000] <Philip`>
"Exception in thread "main" org.xml.sax.SAXParseException: This document is not mappable to XML 1.0 without data loss to ?--? in a comment."

[13:31:01.0000] <Philip`>
(I assume that's intentional)

[13:32:00.0000] <hsivonen>
Philip`: that is correct, isn't it

[13:35:00.0000] <Philip`>
Hmm... Your validator complains about it on line 41550, column 5

[13:35:01.0000] <Philip`>
which is

[13:35:02.0000] <Philip`>
<!-- XXX alt="": Define that either the src="" is shown (as an image)

[13:35:03.0000] <Philip`>
or the alt="" is shown (inline) but should not ever have both at

[13:35:04.0000] <Philip`>
once. -->

[13:35:05.0000] <Philip`>
I'm not quite sure where the "Consecutive hyphens did not terminate a comment." comes from in that section

[13:35:06.0000] <Philip`>
Oh

[13:35:07.0000] <Philip`>
There's a <!-- just before that

[13:36:00.0000] <hsivonen>
whew

[13:37:00.0000] <Philip`>
If I fix that, it all seems to work correctly

[13:38:00.0000] <hsivonen>
Philip`: thanks

[13:41:00.0000] <Philip`>
If I pass a load of arbitrary HTML through it, I get "Exception in thread "main" org.xml.sax.SAXParseException: Attribute name ?type"block"? is not an NCName." which seems entirely true

[13:45:00.0000] <Philip`>
/me can't think of any other easy ways to try to break it

[15:51:00.0000] <tndH>
/me is really fed up of hearing the phrase "fight the good fight" in a web standards context

[15:51:01.0000] <G0k>
yeah we gotta fight dirty

[15:51:02.0000] <gavin_>
we need to fight the good fight and get rid of that phrase

[15:53:00.0000] <Philip`>
We could all be friends and stop fighting

[15:53:01.0000] <G0k>
what's the fun in that?

[15:53:02.0000] <Philip`>
Well, maybe we could just act like we're friends, which leaves open the opportunity for backstabbing people when they're not careful

[15:54:00.0000] <G0k>
aha excellent

[16:03:00.0000] <karlUshi>
1. You don't talk about fight club.

[16:03:01.0000] <karlUshi>
2. You don't talk about fight club.

[16:04:00.0000] <Philip`>
/me has probably been playing Defcon too much, where the only reason you ever ally with someone is so you can drop nukes on them from behind when they are busy elsewhere

[16:05:00.0000] <Hixie>
i'm amused that none of the people discussing the KDE example in the spec have noticed that i didn't make it up

[16:05:01.0000] <Hixie>
it's lifted straight from wikipedia

[16:06:00.0000] <takkaria>
Hixie: you're missing an "attribute" in the new img text for the no-alt section: "In such cases, the alt attribute may be omitted, but the alt should be included"

[16:07:00.0000] <G0k>
uh. "may be omitted, but should be include" isn't that like...repetitive?

[16:09:00.0000] <G0k>
not that there's anything wrong with that

[16:09:01.0000] <Hixie>
takkaria: fixed, thanks

[16:10:00.0000] <Hixie>
G0k: in this case, the words "may" and "should" have very specific meaning that make this contrived phrasing useful

[16:10:01.0000] <G0k>
fair enough

[16:11:00.0000] <G0k>
Hixie: not to bug but did you get my other note on event-source stuff?

[16:13:00.0000] <Hixie>
yeah, my mail server is having issues and i can't send my reply

[16:13:01.0000] <Hixie>
but it's written

[16:13:02.0000] <G0k>
neato

[16:15:00.0000] <G0k>
sorry to keep bugging about that...i have an implementation that i don't want to commit until it's pretty

[16:15:01.0000] <Hixie>
ah

[16:15:02.0000] <Hixie>
ok

[16:15:03.0000] <Hixie>
coo

[16:15:04.0000] <Hixie>
l

[16:15:05.0000] <G0k>
don't want to start of broken

[16:15:06.0000] <G0k>
*off

[16:17:00.0000] <zcorpan_>
was it good ideas and bad code that was successful in building communities? :)

[16:27:00.0000] <Hixie>
the people working on the html5 status thing are charlvn and zcorpan_ right?

[16:30:00.0000] <zcorpan_>
yeah

[16:30:01.0000] <zcorpan_>
though i haven't done anything on that for quite some time now

[16:31:00.0000] <zcorpan_>
charlvn had set up some php backend iirc

[16:31:01.0000] <gsnedders>
ergh. PHP is horrid.

[16:31:02.0000] <zcorpan_>
but i don't know what is supposed to be passed on to the php script

[16:32:00.0000] <gsnedders>
(sorry, I've just been paid to do two weeks of work in PHP)

[16:32:01.0000] <G0k>
php is the VB of the web generation

[16:32:02.0000] <Philip`>
Writing CGI scripts in C is fun

[16:33:00.0000] <Philip`>
http://canvex.lazyilluminati.com/misc/dom-viewer/reflect.cgi?%3Cblink%3EHello%20world - I don't know if I should worry that it's the ultimate XSS hole

[16:34:00.0000] <G0k>
heh

[16:34:01.0000] <Hixie>
zcorpan_: yeah i need to find someone to take the ball on that and run with it

[16:35:00.0000] <zcorpan_>
yep, would be good

[16:35:01.0000] <G0k>
what's the status thing supposed to do?

[16:36:00.0000] <zcorpan_>
mark sections as being "work in progress", etc

[16:37:00.0000] <Philip`>
Could any of it be rolled into the spec-splitter script, or does it all need to work in the single-page version too?

[16:38:00.0000] <zcorpan_>
dunno

[16:39:00.0000] <zcorpan_>
i think right now only the interface to update the markers is actually missing

[16:41:00.0000] <Hixie>
yeah i really want something that i can just have as a floating toolbar over the spec itself that i can just click on a section and then say "this is stable" or "known issue here" and include links to feedback (like blogs) that i have to deal with, as well as linking to test cases and listing what implementations support a particular section, etc

[16:43:00.0000] <G0k>
would be really cool if it like..scanned the mailing lists and automagically added links to messages which quoted sections

[16:43:01.0000] <G0k>
how crrrazy would that be?

[16:45:00.0000] <Philip`>
That doesn't sound incredibly crazy

[16:45:01.0000] <Philip`>
at least if "quoted sections" means "posted a link to a section", rather than quoting the actual content

[16:45:02.0000] <G0k>
sure that

[16:46:00.0000] <Philip`>
/me already has a script that matches up quoted bits of the spec within a paragraph, though

[16:46:01.0000] <Philip`>
(to generate the annotations for http://canvex.lazyilluminati.com/tests/tests/spec.html )

[16:48:00.0000] <Hixie>
i'd be happy to have cray cool stuff too, but i recommend starting small :-)

[16:49:00.0000] <Philip`>
/me needs to get around to finishing rejigging the spec-splitter at some point...


2007-08-17
[17:59:00.0000] <G0k>
check it out all http://thefridgeowl.com/stuff/domeventtest/test01.html

[17:59:01.0000] <G0k>
server side event unit test thing

[18:00:00.0000] <G0k>
anyone with a recent copy of opera?

[18:04:00.0000] <Philip`>
"Recent" like 9.22, or like 9.5?

[18:04:01.0000] <G0k>
like..last 6 months i guess

[18:05:00.0000] <Philip`>
"Statement on line 50: Type mismatch (usually a non-object value used where an object is required)"

[18:05:01.0000] <G0k>
uh. of which file?

[18:06:00.0000] <Philip`>
(in 9.22, which is almost the most recent public one)

[18:06:01.0000] <Philip`>
  Line 50 of linked script http://thefridgeowl.com/stuff/domeventtest/test01.js

[18:06:02.0000] <Philip`>
    listener.addEventSource("test01.php");

[18:06:03.0000] <G0k>
so do any of the Waiting fields turn to "Passed" ?

[18:06:04.0000] <colione>
three

[18:06:05.0000] <colione>
mouseover MouseEvent Passed

[18:06:06.0000] <colione>
mousemove MouseEvent Passed

[18:06:07.0000] <colione>
mouseout MouseEvent Passed

[18:07:00.0000] <Philip`>
They all say "Waiting" for me

[18:07:01.0000] <Philip`>
Oh, now those three say "Passed"

[18:07:02.0000] <colione>
Opera 9.22.3687 Mac Os X

[18:08:00.0000] <G0k>
oh heh

[18:08:01.0000] <Philip`>
They seem to go to "Passed" when I move my mouse over a bit just off the top-right corner of the table

[18:08:02.0000] <G0k>
because when you move your mouse over the thing

[18:08:03.0000] <G0k>
yeah

[18:08:04.0000] <Philip`>
Oh, I mean top rather than top-right

[18:08:05.0000] <G0k>
hm. so they don't seem to support the addEventListener method

[18:09:00.0000] <G0k>
lemme try the event-source tag

[18:10:00.0000] <colione>
same result in latest webkit

[18:10:01.0000] <G0k>
yeah i haven't checked my webkit implementation in yet. :)

[18:24:00.0000] <G0k>
wow well

[18:24:01.0000] <G0k>
http://thefridgeowl.com/stuff/domeventtest/test02.html

[18:24:02.0000] <G0k>
it does do a great job making opera leak memory anyway

[18:31:00.0000] <G0k>
finally got opera working...seems to dislike the divx plugin a lot

[18:31:01.0000] <G0k>
well if anyone wants to play with it http://thefridgeowl.com/stuff/domeventtest/test02.html now kinda works on opera 9.2.whatever

[18:32:00.0000] <G0k>
some things work some don't

[18:32:01.0000] <G0k>
dunno who's at fault for the nonworking ones

[18:32:02.0000] <Philip`>
I get some combination of "Passed", "Waiting" and "Not a (...)Event" on testf02.html

[18:33:00.0000] <Philip`>
Uh, test02.html

[18:33:01.0000] <G0k>
well. they seem to support some and not support others

[18:33:02.0000] <G0k>
and/or my tests are bad

[18:35:00.0000] <Philip`>
and/or the spec is bad

[18:36:00.0000] <G0k>
well that's almost undeniable. :)

[18:37:00.0000] <G0k>
my tests are particularly biased, since i kinda invented them to test my webkit implementation

[18:38:00.0000] <Philip`>
Opera's implementation is fairly old and untested, so I wouldn't expect it to be a good approximation of correctness. I've got no idea if they've made any improvements to this stuff for v9.5, though

[18:39:00.0000] <G0k>
is there a 9.5 public beta?

[18:39:01.0000] <Philip`>
Not yet

[18:39:02.0000] <Philip`>
The only date I've heard is "soon", for the past few months :-)

[18:40:00.0000] <G0k>
hm. i know there was a patch for an implementation in mozilla at some point too

[18:40:01.0000] <G0k>
but it was never accepted

[18:40:02.0000] <G0k>
kinda curious how that was implemented too

[18:42:00.0000] <colione>
i know that zcorpan_ has a copy of opera 9.5 *tries to draw his attention to irc*

[18:42:01.0000] <Philip`>
https://bugzilla.mozilla.org/show_bug.cgi?id=338583

[18:43:00.0000] <zcorpan_>
someone want me to test something in kestrel? :)

[18:44:00.0000] <G0k>
http://thefridgeowl.com/stuff/domeventtest/test02.html

[18:44:01.0000] <colione>
 :D

[18:44:02.0000] <Philip`>
zcorpan_: You could save a lot of effort if you accidentally sent an email to the WHATWG list with the Kestrel installer attached

[18:44:03.0000] <zcorpan_>
10 passed, 7 failed, the rest are waiting

[18:44:04.0000] <zcorpan_>
Philip`: ha

[18:45:00.0000] <G0k>
k so doesn't look like any progress from 9.2

[18:45:01.0000] <G0k>
zcorpan_:  what about http://thefridgeowl.com/stuff/domeventtest/test01.html ?

[18:45:02.0000] <colione>
hmm 11 past in 9.22

[18:45:03.0000] <colione>
passed*

[18:45:04.0000] <G0k>
heh uhoh

[18:46:00.0000] <G0k>
to be honest, my mechanism for type checking an event is really pathetic

[18:46:01.0000] <Philip`>
In what way?

[18:46:02.0000] <G0k>
i check to see if "event" is a XXXEvent by seeing if event.initXXXEvent is defined

[18:46:03.0000] <Philip`>
Ah

[18:46:04.0000] <zcorpan_>
G0k: 9 passed, 7 failed

[18:47:00.0000] <G0k>
zcorpan_: for test01?

[18:47:01.0000] <zcorpan_>
G0k: yes

[18:47:02.0000] <G0k>
ah neato so the addEventSource thing is in kestrel. thanks.

[18:47:03.0000] <Philip`>
Maybe you could do something like Object.prototype.toString(event) and see if it's called "[object XXXEvent]"

[18:50:00.0000] <G0k>
ah

[18:51:00.0000] <G0k>
alright well all those that were previously failing in opera now aren't

[18:52:00.0000] <G0k>
but they're not getting called either

[18:52:01.0000] <G0k>
i mean

[18:52:02.0000] <G0k>
the ones that weren't getting called

[18:52:03.0000] <G0k>
still are not

[18:54:00.0000] <Philip`>
"if (Object.prototype.toString(event) == interfaceName) { passed = false; ..." - that doesn't make sense

[18:55:00.0000] <Philip`>
It should be "if (... != "[object "+interfaceName+"]") ...", I think

[18:55:01.0000] <G0k>
well. it's returning just the class name

[18:55:02.0000] <G0k>
in both webkit and opera

[18:56:00.0000] <Philip`>
But you're setting passed to false if the value is equal to the expected value

[18:56:01.0000] <Philip`>
when it should be the other way around

[18:56:02.0000] <G0k>
oh craps

[18:56:03.0000] <G0k>
wow

[18:57:00.0000] <Philip`>
Also, I was wrong anyway

[18:57:01.0000] <Philip`>
It should be Object.prototype.toString.call(event)

[18:57:02.0000] <Philip`>
else you'll always get "[object Object]"

[18:57:03.0000] <Philip`>
Also this assumes the objects have the same type as the interface, rather than a subclass, and I don't know if that's valid

[19:00:00.0000] <G0k>
well i think the server side event spec currently explicitly says which class the events need to be

[19:01:00.0000] <G0k>
but yeah i mean i suppose it would still be valid if it were a subclass of that class

[19:03:00.0000] <G0k>
man javascript is really in need of "does this implement some interface" function

[19:06:00.0000] <zcorpan_>
foo.constructor ?

[19:07:00.0000] <G0k>
well for instance

[19:08:00.0000] <G0k>
Opera seems to use the KeyEvent prototype everywhere i would expect a KeyboardEvent prototype

[19:08:01.0000] <zcorpan_>
hm

[19:08:02.0000] <G0k>
which...i now notice isn't in dom 3. heh

[19:09:00.0000] <zcorpan_>
anyway... well beond bedtime for me

[22:30:00.0000] <Hixie>
"Niether Hixe nor Philip looked for use of <input usemap> . Hixie looked for misuse (or esoteric use) of <input usemap>."

[22:30:01.0000] <Hixie>
i thought i had specifically looked for _valid_ cases...

[22:30:02.0000] <Hixie>
i wonder what rob is basing that on

[22:31:00.0000] <G0k>
good god is this still being argued?

[22:32:00.0000] <G0k>
is there seriously a legitimate use for image maps at all nowadays?

[22:34:00.0000] <Hixie>
i dunno i'm not entering this discussion any more

[22:34:01.0000] <G0k>
deprecate with extreme prejudice!

[22:35:00.0000] <Lachy>
Hixie, Rob is looking for usage that matches the way he imagines <input usemap> should work, not the way it has ever been defined or implemented

[22:36:00.0000] <Lachy>
and thus, he rejects any observation of <input usemap> on the basis that what he's looking for automatically can't exist

[22:41:00.0000] <Hixie>
anyone tried explaining the whole thing about how scientific research and such is all about looking at what actually is there, and not what you're expecting?

[22:43:00.0000] <Lachy>
I think he misses the point of the whole observation stage.  I think he wants to start directly with some hypothesis developed without any observation at all and then test it

[22:44:00.0000] <Lachy>
I don't think he understands that a hypothesis should be developed to try and explain your observations, when then gets tested and refined

[22:47:00.0000] <Lachy>
maybe someone needs to explain that the statistics he's objecting to so much are used as a rough guide only. The more important information is obtained from looking at how and why something gets used, which is what our decisions are based on

[22:47:01.0000] <Hixie>
/me shrugs

[22:49:00.0000] <G0k>
maybe it should be decided by a duel.

[22:50:00.0000] <Hixie>
on a more positive note, i do like how much the various studies correlate

[22:51:00.0000] <G0k>
anyone know who operates http://validator.nu/ ?

[22:55:00.0000] <Hixie>
hsivonen

[22:55:01.0000] <karlUshi>
/me would prefer that people do not use "scientific research" to justify, specifically when things are not really done that way. :) Or at least that there are many ways of doing scientific research.

[22:56:00.0000] <karlUshi>
/me has the feeling that people are mixing experimental research with scientific research. :)

[22:58:00.0000] <Lachy>
karlUshi, observation and experementation is part of the whole scientific process, so I don't understand what you mean

[22:59:00.0000] <karlUshi>
Lachy: My whole academic background is… scientific research :)))

[22:59:01.0000] <karlUshi>
Astrophysics exactly

[22:59:02.0000] <karlUshi>
there are many ways of doing sciences, not only one. 

[23:00:00.0000] <karlUshi>
You can start from observations. You can also start from a model and looking for evidences by observing.

[23:00:01.0000] <karlUshi>
Observations don't necessary come first in the scientific method.

[23:12:00.0000] <G0k>
yeah but...once you fail to find evidence, that's not evidence of a conspiracy to undermine your model.

[23:39:00.0000] <Hixie>
i hope the hixie.ch dns gets fixed soon so i don't lose too much e-mail

[23:42:00.0000] <othermaciej>
is Rob just engaging in a brilliantly subtle attempt to drain all productivity?

[23:42:01.0000] <othermaciej>
I'm starting to be amazed at his ability to disagree with even the most basic things

[23:48:00.0000] <G0k>
is this on public-html?

[23:52:00.0000] <othermaciej>
yes

[23:53:00.0000] <othermaciej>
public-html is now mostly useful content plus him plus replies to him

[23:53:01.0000] <G0k>
who is he?

[23:53:02.0000] <othermaciej>
Robert Burns?

[23:54:00.0000] <G0k>
yeah what's he do?

[23:54:01.0000] <othermaciej>
it's hard to explain without experiencing it

[23:54:02.0000] <G0k>
i mean does he have an occupation?

[23:54:03.0000] <othermaciej>
oh

[23:55:00.0000] <othermaciej>
I believe he is a grad student

[23:55:01.0000] <karlUshi>
http://www.robburns.com/

[23:56:00.0000] <karlUshi>
I guess he will be able to come at the Tech Plenary being in Mass.

[23:56:01.0000] <G0k>
does he design pages or something?

[23:57:00.0000] <G0k>
eeeks econ say no more

[23:58:00.0000] <othermaciej>
marxist econ

[23:58:01.0000] <karlUshi>
G0k: this is an innapropriate comment, except if you let me say "Computer Programmer… oh these binary people! say no more.". :/

[23:59:00.0000] <Lachy>
what is Marxian philosophy?

[23:59:01.0000] <G0k>
i make try to make as few appropriate comments as possible

[23:59:02.0000] <G0k>
it is my intention to maximize my offendees

[23:59:03.0000] <othermaciej>
presumably the philosophy of Karl Marx, his students, and intellectual followers

[23:59:04.0000] <G0k>
so yeah bring it on

[00:00:00.0000] <othermaciej>
I don't have a problem with economists as a class

[00:00:01.0000] <karlUshi>
othermaciej: me neither. 

[00:00:02.0000] <karlUshi>
and computer programmer as well.

[00:00:03.0000] <othermaciej>
most of them are good at thinking in terms of incentives, spontaneous order, and unintended consequences

[00:00:04.0000] <karlUshi>
There are people of different nature and different style, specifically online/offline.

[00:00:05.0000] <othermaciej>
all of which are are useful concepts to apply to global technical standards

[00:01:00.0000] <karlUshi>
I know a few persons on this channel who behave differently online and offline.

[00:01:01.0000] <G0k>
i have several economist friends, they're just all very likely to get into an argument with me

[00:02:00.0000] <karlUshi>
maybe you have to ask questions about yourself then G0k ;)

[00:02:01.0000] <G0k>
yeah i don't deny that i'm likely to be wrong usually

[00:54:00.0000] <Lachy>
Hixie, would it be possible for the spec to provide some sort of heuristic guidelines for programmatically determining when should be provided?  e.g. if you have <h1><img src=""></h1>, it should be clear for a validator that the lack of alt text is an error.

[00:55:00.0000] <othermaciej>
you could require alt in any context that would require text if the embedded content wasn't there

[00:55:01.0000] <Lachy>
could it be defined that alt="" can only ever be omitted when contained within a <figure>, since that would represent it as a key part of the content

[00:55:02.0000] <othermaciej>
or something

[00:56:00.0000] <Lachy>
currently, all the examples in that section where alt can be omitted are contained within figure too

[00:58:00.0000] <othermaciej>
allowing missing alt only in <figure> would address the photo gallery use case but not the "pasting images into an end-user rich text editor" use case

[00:59:00.0000] <Lachy>
hmm, yeah

[01:56:00.0000] <jgraham>
Lachy: I would be _very_ wary of trying to subset where alt="" is allowed.

[01:56:01.0000] <jgraham>
Oh wait a sec, I misread that

[01:56:02.0000] <jgraham>
Ignore me

[01:56:03.0000] <Lachy>
right

[02:13:00.0000] <virtuelv>
I really wish people could refrain from bad ad hominems on random blogs

[02:13:01.0000] <Lachy>
virtuelv, ?

[02:13:02.0000] <Lachy>
pointer

[02:18:00.0000] <virtuelv>
http://www.456bereastreet.com/archive/200708/the_html_5_circus_why_i_left_and_rejoined_the_w3c_html_working_group/#comment3

[02:21:00.0000] <krijnh>
Yeah, thanks for DOSing my server Roger ;]

[02:24:00.0000] <krijnh>
Ah, the comment

[02:25:00.0000] <othermaciej>
looks like Google has secretly invented the fountain of youth

[02:27:00.0000] <virtuelv>
krijnh: yes, I was specifically refering to the comment itself

[02:28:00.0000] <virtuelv>
While I may not agree with all of what Roger says, he is at least not resorting to fallacies to argue his position

[02:30:00.0000] <othermaciej>
Roger's actual post seems reasonable, though looking back at his posts to the group and the replies to them, I'm not sure why he felt attacked

[02:30:01.0000] <Lachy>
"I am far from the only person to think that the current process of editing the specification for the next version of HTML is… less than ideal." -  we really need to address this issue somehow.

[02:30:02.0000] <Lachy>
although the consensus based approach some people are advocating won't work, I think we need better PR to improve the current process

[02:31:00.0000] <krijnh>
/me agrees

[02:31:01.0000] <krijnh>
Although I don't know how to fix it

[02:31:02.0000] <hsivonen>
the IETF has *rough* consensus and *running* code

[02:32:00.0000] <hsivonen>
much better than consensus on theory

[02:32:01.0000] <krijnh>
Perhaps it's the whatwg "vs." w3c attitude

[02:32:02.0000] <othermaciej>
I like the idea of the IETF process

[02:32:03.0000] <othermaciej>
though I have no real experience with the practice

[02:32:04.0000] <Lachy>
Roger just suggested to me that, whenever Hixie sends one of his summary emails responding to an issue on whatwg, we should notify public-html of it too

[02:33:00.0000] <krijnh>
whatwg is easier to track though

[02:33:01.0000] <hsivonen>
Lachy: good idea

[02:34:00.0000] <krijnh>
Why isn't that done already?

[02:34:01.0000] <krijnh>
Seems logical to do so

[02:34:02.0000] <Lachy>
getting a schedule from Hixie too would be good, and searching the whatwg archives and providing pointers to previous discussions beforehand may also help

[02:36:00.0000] <othermaciej>
there's lots of organizational and informational things that could be done, but ideally by someone other than Hixie since his plate is pretty full

[02:36:01.0000] <othermaciej>
I feel like the sense that reported issues are tracked is not solid enough yet, which may make people feel ignored

[02:36:02.0000] <hsivonen>
Lachy: last time when Hixie had an open-secret schedule, it didn't hold. I think the best we could get is a topic list without calendar time

[02:36:03.0000] <Lachy>
yeah, I'm sure others will volunteer to find all that info.  We just need to know what the upcoming topics are

[02:36:04.0000] <othermaciej>
a short list of the next few topics would be helpful

[02:37:00.0000] <othermaciej>
and could guide review as well

[02:37:01.0000] <othermaciej>
(more useful to review soon-to-be-revisited areas)

[03:10:00.0000] <virtuelv>
I'd like to have something like trac for open issues

[03:20:00.0000] <hsivonen>
hrm. looks like the Dreamhost DNS is down

[03:21:00.0000] <virtuelv>
hsivonen: appears that way, all my sites are down as well

[03:46:00.0000] <Lachy>
is the only purpose behind the suggested noalt="" attribute (or equivlent) just a hack to work around the validation issue?

[03:48:00.0000] <hsivonen>
Lachy: the purpose is to make hand coders perform an affirmative act of typing instead of just leaving something out

[03:49:00.0000] <hsivonen>
(fwiw, I don't like the noalt idea. It won't help with non-hand-coded source and it will majorly annoy hand coders)

[03:49:01.0000] <Lachy>
hmm. Ok.  But I think it's perfectly fine for authoring tools and conformance checkers to provide warnings when an alt attribute has been omitted, if the author wants such information

[03:50:00.0000] <Lachy>
indeed, and it doesn't acctually address the issue of getting people to provide it, or just abusing it to pass validation

[03:50:01.0000] <hsivonen>
yeah, warnings requested by the author are very different from baking something into the conformance definition

[03:53:00.0000] <Lachy>
you summed it up nicely here http://www.mail-archive.com/whatwg⊙lwo/msg06146.html

[03:54:00.0000] <hsivonen>
btw, if you feel there are particular issues that I should develop optional warnings for, please start a page on the WHATWG wiki

[04:02:00.0000] <Lachy>
I think there is a page on the wiki for that stuff somewhere, isn't there?

[04:02:01.0000] <Lachy>
I'll have to check when the server is back up

[04:55:00.0000] <Whiskey_M>
'lo

[05:52:00.0000] <gsnedders>
is it just me or is whatwg.org down?

[06:14:00.0000] <mgdm>
gsnedders: I can't resolve it

[06:14:01.0000] <mgdm>
in DNS, I mean

[06:14:02.0000] <Lachy>
all of Hixie's sites are down.  I think it's a problem with dreamhost

[06:14:03.0000] <mgdm>
Ah, I tried to get to simplebits.com earlier and it was broken too

[06:20:00.0000] <Philip`>
dreamhoststatus.com appears to be broken too, which doesn't help immensely

[07:28:00.0000] <gsnedders>
what's the point of the mask in "Content-Type sniffing: unknown type"?

[07:29:00.0000] <gsnedders>
ah. to make them case insensitive, or not.

[09:28:00.0000] <Lachy>
would anyone like to review this blog post for the whatwg blog (which I'll publish when the site is back up) http://lachy.id.au/temp/alt

[09:40:00.0000] <zcorpan_>
Lachy: looks good to me... though you may want to include more examples of where it is ok to omit alt

[09:41:00.0000] <hsivonen>
Lachy: I'd do s/the reality of the situation/the situation/ to avoid the impression that others are missing the reality

[09:41:01.0000] <Lachy>
I'm going to be linking to the spec, should I just borrow some of those?

[09:41:02.0000] <Lachy>
I like talking about reality, but ok

[09:42:00.0000] <Lachy>
I suppose it hasn't gone down well with some people

[09:43:00.0000] <jonbarnett>
my lynx doesn't behave as described here

[09:44:00.0000] <Lachy>
it doesn't?  There's emails in the whatwg archive that say it does

[09:45:00.0000] <zcorpan_>
jonbarnett: you have a test url?

[09:45:01.0000] <jonbarnett>
for me, lynx has options to ignore images, treat images as labels, or treat them as links.  When treated as labels, the filename is used when alt is omitted, and nothing is used when alt is blank.  When treated as links, the filename is used both when alt is omitted and when alt is blank.  It's not that big of a difference: lynx is still making a distinction, and your version may vary

[09:47:00.0000] <Lachy>
I'm not sure what version was tested, it wasn't mentioned

[09:47:01.0000] <jonbarnett>
(I may have been the one mentioning it and not having tested thoroughly *blush*)

[10:01:00.0000] <Philip`>
/me wonders if it could be required for conformant <table>s to contain at least one <th>

[10:02:00.0000] <Philip`>
since I can't think of any obvious cases where you'd have a table with no headings, unless it's a layout table

[10:02:01.0000] <zcorpan_>
Philip`: i've used tables without headers that aren't layout tables

[10:03:00.0000] <zcorpan_>
genealogical table

[10:08:00.0000] <Lachy>
Philip`, the question you need to ask is, in the event that there are use cases for tables without headers, what harm is caused by the lack of th?

[10:09:00.0000] <hsivonen>
Philip`: four-cell tables in MBA material

[10:09:01.0000] <hsivonen>
Philip`: unless you consider all four to me <th>s

[10:09:02.0000] <hsivonen>
s/to me/to be/

[10:18:00.0000] <Philip`>
Lachy: If there are uses cases for headerless data tables, then we can't require headers; but if there aren't, then conformance checkers could know that any headerless table is a non-data table and is therefore non-conforming

[10:18:01.0000] <Philip`>
(e.g. http://www.noraradcliffe.co.uk/ is (very nearly) conforming XHTML1, designed by people who "exceed accessibility requirements", but uses tables for layout)

[10:18:02.0000] <hsivonen>
Philip`: which would only lead to people including bogus or barely-semantic <th>s in their tables

[10:18:03.0000] <Lachy>
maybe, but it's very difficult to prove that there aren't any

[10:19:00.0000] <Philip`>
(Er, I mean it's (very nearly) valid XHTML1)

[10:23:00.0000] <Philip`>
http://www.w3.org/TR/xhtml-role/ - "A conforming user agent MUST support all of the features required in this specification." - hmm, I can't actually find any required UA features in the specification

[10:26:00.0000] <zcorpan_>
Philip`: indeed

[10:28:00.0000] <zcorpan_>
e.g. http://www.pointerklubben.se/stamtavla.asp?Id=S27834/2001 is a table with no headers

[10:42:00.0000] <hsivonen>
zcorpan_: interesting dog names

[10:58:00.0000] <doublec>
is www.whatwg.org down?

[10:59:00.0000] <Philip`>
Yes

[11:00:00.0000] <Philip`>
(Dreamhost's problem, I think)

[11:02:00.0000] <doublec>
thanks 

[11:23:00.0000] <h3h>
hooray Dreamhost!

[11:23:01.0000] <h3h>
/me rolls eyes

[11:32:00.0000] <jonbarnett>
... and hixie's site is down, too?

[11:34:00.0000] <Philip`>
Yes, for the same reason

[11:36:00.0000] <Hixie>
this is the worst dreamhost outage yet

[11:36:01.0000] <Hixie>
if it wasn't for the fact that i pay $0, i'd be outraged

[11:36:02.0000] <Hixie>
yesterday i was worried that i was losing whatwg e-mail

[11:37:00.0000] <Hixie>
but now that whatwg isn't resolving either, i'm less concerned :-P

[11:37:01.0000] <Philip`>
At least http://www.dreamhost.com/ is up and says "Unfortunately we are experiencing a return of similar network problems that we experienced yesterday. Please keep your eye on this post as we will update it as soon as we have mor..." then links to the dead dreamhoststatus.com

[11:37:02.0000] <Philip`>
so presumably they have actually noticed there's a problem

[11:37:03.0000] <jonbarnett>
my personal space is on DH, and it's up (I guess I'm just tauting)

[11:38:00.0000] <met_>
http://ajaxian.com/archives/web-forms-20-cross-browser-implementation

[11:41:00.0000] <zcorpan_>
hmm, some of the tests don't have any pass condition

[11:42:00.0000] <zcorpan_>
and all have too much junk around the actual test... :)

[11:47:00.0000] <Hixie>
they do know there's a problem

[11:47:01.0000] <Hixie>
they've been working overnight trying to fix it

[11:47:02.0000] <Hixie>
they update their internal status message ever few minutes

[11:48:00.0000] <Hixie>
something to do with lost dns records, they're rebuilding their dns config from scratch using a slow-working script

[11:48:01.0000] <Hixie>
the discussion in #html-wg was interesting

[11:49:00.0000] <Hixie>
can someone summarise in an unbiased way what robburns' concerns were?

[11:51:00.0000] <Hixie>
Lachy: btw i apparently have ~3800 or so (+/- 200, i'm going from memory) e-mails in my queue right now

[11:52:00.0000] <Lachy>
oh, nice! It's reduced :-)

[11:52:01.0000] <zcorpan_>
aiui, he wanted the html wg to discuss issues raised outside the group before the spec is edited, instead of the editors editing based on external feedback and then letting the wg discuss further... but i might have missed something

[11:52:02.0000] <jonbarnett>
Hixie: i haven't been on #html-wg.  is it about <img> or your editing process?

[11:53:00.0000] <jonbarnett>
nm

[11:55:00.0000] <hsivonen>
Hixie: my understanding was that Rob was unhappy that 1) you are focusing on stuff other than the discussions of public-html and 2) that your recent edits (in his opinion) run counter the thrust of discussions on public-html. (this may well be a biasod account)

[11:56:00.0000] <Hixie>
ok

[11:57:00.0000] <Hixie>
2) may well be true, i've seen no real consensus on public-html so it's hard to know what the thrust of discussion is

[11:57:01.0000] <Hixie>
i guess i could start picking htmlwg issues to deal with if it makes them happier

[11:57:02.0000] <Lachy>
Hixie, I believe Rob would agree with these suggestions http://krijnhoetmer.nl/irc-logs/whatwg/20070817#l-320aren't 

[11:57:03.0000] <Hixie>
doesn't really make much difference what order i deal with issues in

[11:57:04.0000] <hsivonen>
I've been concerned about the formation of a false consensus on public-html

[11:57:05.0000] <Lachy>
http://krijnhoetmer.nl/irc-logs/whatwg/20070817#l-320

[11:58:00.0000] <hsivonen>
Rob's view suggests that my concern is valid

[11:58:01.0000] <hsivonen>
Hixie: well, it does make a PR difference

[11:58:02.0000] <Hixie>
i can certainly bcc public-html on any mail to whatwg

[11:58:03.0000] <hsivonen>
Hixie: that would probably help

[11:59:00.0000] <Hixie>
the reason i haven't is that it would cause a flood of e-mail to whatwg

[11:59:01.0000] <Hixie>
since every e-mail i send to whatwg is just replied by people with actual experience, but e-mail sent to public-html immediately devolves into a flamewar

[12:00:00.0000] <hsivonen>
but it's really tiresome to avoid the formation of false consensus by keeping parroting old points

[12:00:01.0000] <Hixie>
don't worry about stopping false consensus

[12:00:02.0000] <hsivonen>
Hixie: true

[12:00:03.0000] <Lachy>
I think the alt attribute thread has actually become somewhat productive now

[12:01:00.0000] <Hixie>
just make sure that the wiki page for the issue clearly states all points

[12:01:01.0000] <Lachy>
although, started out as another near-flameware

[12:01:02.0000] <Hixie>
Lachy: really? you mean i should make more changes beyond what the spec says?

[12:01:03.0000] <Lachy>
Steven Faulkner wrote a constructive suggestion. I'm not sure I agree with it, but it was mostly well written

[12:03:00.0000] <Hixie>
you mean http://lists.w3.org/Archives/Public/public-html/2007Aug/0660.html ?

[12:03:01.0000] <Hixie>
i don't see how that could work

[12:03:02.0000] <Lachy>
nor do I.  I was just pointing out that it was well intended, not a flame

[12:04:00.0000] <Hixie>
(btw, re your idea, <h1></h1> isn't non-conforming -- in fact we're dropping the concept of significant inline content)

[12:04:01.0000] <Hixie>
sure

[12:04:02.0000] <Lachy>
hmm. that's what I thought you'd say

[12:05:00.0000] <Hixie>
it's a good idea, though, i'd do it if we weren't doing the other thing

[12:05:01.0000] <Hixie>
is there a wiki page with a list of problems that have been fully described and are stable?

[12:06:00.0000] <Lachy>
I suppose if conformance checkers implement a feature like that, you could spec it.

[12:07:00.0000] <Lachy>
in the HTMLWG wiki?  I don't think anything is particularly stable there

[12:08:00.0000] <Hixie>
maybe i should send mail saying that i'm still lwaiting for a stable wiki page

[12:09:00.0000] <Lachy>
the table headers issues is probably the most stable, and addressing that sooner rather than later would probably really help calm people down

[12:11:00.0000] <Lachy>
good night everyone.

[12:12:00.0000] <Hixie>
nn

[12:13:00.0000] <Hixie>
hmm

[12:14:00.0000] <Hixie>
i'm gonna go to work and consider how else to help keep public-html in the loop

[12:15:00.0000] <Hixie>
i don't think forwarding all whatwg mail to public-html (or even my own e-mails to public-html) would be productive

[12:15:01.0000] <Hixie>
but maybe we should try that anyway

[12:16:00.0000] <Hixie>
hmm

[12:16:01.0000] <Hixie>
afk

[12:17:00.0000] <takkaria>
I think that just having a public list with the next few things you're thinking of working on, with some semi-accurate dates, would go a long way

[13:08:00.0000] <Hixie>
takkaria: yeah... the problem is right now what i'm working on is "pick random piece of feedback, reply to it"

[13:09:00.0000] <takkaria>
may I suggest such a list might also be useful to you, then? :)

[13:09:01.0000] <Hixie>
woot, my site is back up :-D

[13:09:02.0000] <Philip`>
That sounds like it could be extended to "pick several random pieces of feedback, put them in a list, and reply to them in that order"

[13:09:03.0000] <Hixie>
takkaria: i dunno, i find i work most productively on the opposite of what i'm supposed to work on :-)

[13:10:00.0000] <takkaria>
ah, so you're not superhuman either?  shame

[13:10:01.0000] <Hixie>
takkaria: literally, to the point where i've sometimes made lists of what i Must Do Now that are actually the opposite of what i really have to do to trick myself into doing it :-P

[13:10:02.0000] <takkaria>
heh

[13:11:00.0000] <takkaria>
I can't imagine a list of things you're not going to work on being very useful to the WG, since that'd be quite a long list

[13:11:01.0000] <gsnedders>
/me can think of one think to do:

[13:11:02.0000] <gsnedders>
/me faceplams

[13:11:03.0000] <gsnedders>
*facepalms

[13:11:04.0000] <gsnedders>
Hixie: the worst part is I'm not actually that surprised at you doing that.

[13:12:00.0000] <Hixie>
heh

[13:12:01.0000] <Hixie>
so, anyone want to take bets on what lists i got unsubscribed from due to my mail bouncing?

[13:13:00.0000] <gsnedders>
I'm tempted to guess public-xhtml

[13:14:00.0000] <Philip`>
Mailman seems to default to needing several weeks of bouncing before unsubscribing someone

[13:14:01.0000] <Hixie>
cool

[13:15:00.0000] <Philip`>
(but I could be wrong, or people could have non-default settings)

[14:19:00.0000] <kingryan>
does anyone know if anyone has implemented "4.7.4. Content-Type sniffing: feed or HTML"

[14:19:01.0000] <kingryan>
?

[14:20:00.0000] <kingryan>
sorry, my client is acting up

[14:20:01.0000] <gsnedders>
kingryan: yes. me.

[14:20:02.0000] <kingryan>
gsnedders: do you have a test suite? ;)

[14:20:03.0000] <gsnedders>
kingryan: no

[14:20:04.0000] <gsnedders>
kingryan: the implementation is literally just finished :)

[14:20:05.0000] <kingryan>
btw, url: http://www.whatwg.org/specs/web-apps/current-work/#content-type3

[14:21:00.0000] <kingryan>
gsnedders: is your code available?

[14:21:01.0000] <kingryan>
open source?

[14:21:02.0000] <zcorpan_>
/me could perhaps write some tests on content sniffing

[14:21:03.0000] <gsnedders>
zcorpan_: I'll write some tonight

[14:21:04.0000] <kingryan>
I'll contribute some tests too

[14:21:05.0000] <zcorpan_>
gsnedders: ok, great :)

[14:22:00.0000] <gsnedders>
kingryan: it will be, soon

[14:22:01.0000] <kingryan>
I have an interpretation question though

[14:22:02.0000] <gsnedders>
BTW, for step 10: http://lists.w3.org/Archives/Public/public-html/2007Aug/0671.html

[14:22:03.0000] <kingryan>
in step 6 of that algorithm

[14:22:04.0000] <gsnedders>
the double otherwise?

[14:22:05.0000] <kingryan>
both substeps 3 and 4 start with "otherwise"

[14:22:06.0000] <kingryan>
:) yes

[14:22:07.0000] <gsnedders>
a mistake, without question

[14:22:08.0000] <gsnedders>
for the algorithm to make sense they need to be a single "otherwise"

[14:22:09.0000] <kingryan>
did you get a clarification from Hixie or whoever else contributed that?

[14:23:00.0000] <zcorpan_>
how did you implement it? if ... else ... else ? :)

[14:23:01.0000] <gsnedders>
not yet, but there's no other way the algorithm would even work

[14:23:02.0000] <gsnedders>
(<http://lists.w3.org/Archives/Public/public-html/2007Aug/0672.html>)

[14:23:03.0000] <kingryan>
ok, I'll try that approach

[14:24:00.0000] <gsnedders>
how should test cases be done?

[14:24:01.0000] <gsnedders>
as the HTTP headers are relevant…

[14:24:02.0000] <gsnedders>
just entire HTTP responses? JSON objects of header name => value?

[14:25:00.0000] <zcorpan_>
hmm

[14:25:01.0000] <kingryan>
I know 4.7.4 isn't self contained, but I don't see any references in that sectoin to http

[14:25:02.0000] <gsnedders>
kingryan: but section 4.7 as a whole

[14:25:03.0000] <zcorpan_>
gsnedders: just using .htaccess should work, no?

[14:25:04.0000] <kingryan>
right. I'm only focused on 4.7.4 right now

[14:26:00.0000] <kingryan>
but for testing the section as a whole...

[14:26:01.0000] <gsnedders>
zcorpan_: I currently have all my tests in SVN though, and not served through Apache

[14:26:02.0000] <kingryan>
what I usually do is have pairs of files "foo.html" and "foo.json"

[14:27:00.0000] <kingryan>
s/html/http/

[14:27:01.0000] <kingryan>
where "foo.http" is a full http response (headers and all)

[14:27:02.0000] <kingryan>
and "foo.json" includes whatever data we hope to extract

[14:27:03.0000] <Philip`>
/me wonders if that content-type sniffing algorithm can be correctly implemented as a regular expression

[14:27:04.0000] <kingryan>
Philip`: I think most state machines can be mapped to regexen

[14:28:00.0000] <gsnedders>
Philip`: depends on the regular expression engine :P

[14:29:00.0000] <Philip`>
I guess it'd be somewhat like ...

[14:29:01.0000] <Philip`>
my $ct = 'text/html'; $ct = { rss => 'application/rss+xml', feed => 'application/atom+xml' }->{$1} if substr($data, 0, 512) =~ /^(?:\s+(?:<!--.*?-->|!.*?>|\?.*?\?>))+?(rss|feed)/;

[14:29:02.0000] <gsnedders>
kingryan: so .json would just have the MIME type, in this case?

[14:29:03.0000] <Philip`>
at least if you ignore the rdf:RDF bit

[14:29:04.0000] <kingryan>
gsnedders: yes

[14:29:05.0000] <Philip`>
but that's quite utterly untested

[14:29:06.0000] <gsnedders>
Philip`: see the former of the two emails I linked to about rdf:RDF

[14:30:00.0000] <gsnedders>
Saf just treats <rdf:RDF as application/rss+xml, and I'm unaware of anything that breaks due to it

[14:31:00.0000] <Philip`>
I guess /^(?:\s+(?:<!--.*?-->|!.*?>|\?.*?\?>))+?<(rss|feed|rdf:RDF)/ might do that, then

[14:32:00.0000] <Philip`>
Urgh, but I'm still missing some <s

[14:32:01.0000] <Philip`>
/^(?:\s+(?:<!--.*?-->|<!.*?>|<\?.*?\?>))+?<(rss|feed|rdf:RDF)/

[14:32:02.0000] <Philip`>
/^(?:\s+(?:<!--.*?-->|<!.*?>|<\?.*?\?>))+?<(rss|feed|rdf:RDF)/

[14:33:00.0000] <Philip`>
/me finds that much easier to read than what's in the spec

[14:33:01.0000] <Philip`>
though it's still wrong

[14:33:02.0000] <Philip`>
/^(?:\s*(?:<!--.*?-->|<!.*?>|<\?.*?\?>))*?<(rss|feed|rdf:RDF)/

[14:34:00.0000] <Philip`>
/^(?:\s*(?:<!--.*?-->|<!.*?>|<\?.*?\?>))*?<(rss|feed|rdf:RDF|)/

[14:34:01.0000] <Philip`>
/me will get there eventually

[14:36:00.0000] <Philip`>
Hmm, still wrong

[14:36:01.0000] <Philip`>
/^(?:\s*(?:<!--.*?-->|<!.*?>|<\?.*?\?>))*?<(rss|feed|rdf:RDF|(?!!--|!|?))/

[14:37:00.0000] <Philip`>
Okay, maybe it's getting less easy to read

[14:40:00.0000] <gsnedders>
how about /^<\?xml[\x20\x9\xD\xA]+version([\x20\x9\xD\xA]+)?=([\x20\x9\xD\xA]+)?(\"1.0\"|'1.0'|\"1.1\"|'1.1')[\x20\x9\xD\xA]+encoding([\x20\x9\xD\xA]+)?=([\x20\x9\xD\xA]+)?(\"[A-Za-z][A-Za-z0-9._\-]*\"|'[A-Za-z][A-Za-z0-9._\-]*')([\x20\x9\xD\xA]+standalone([\x20\x9\xD\xA]+)?=([\x20\x9\xD\xA]+)?(\"(yes|no)\"|'(yes|no)'))?([\x20\x9\xD\xA]+)?\?>/

[14:40:01.0000] <gsnedders>
:P

[14:41:00.0000] <gsnedders>
if that's easy to read…

[14:43:00.0000] <Philip`>
It'd be a bit nicer if you split the common whitespace bits into a variable and did /^<\?xml $ws+ version $ws* = .../x or something :-)

[14:44:00.0000] <gsnedders>
it's a good question why I put it all together before saving it. I wrote it with variables like that.

[14:44:01.0000] <Philip`>
I wonder if I can compile the tokeniser algorithm into a Perl regexp...

[14:44:02.0000] <gsnedders>
Philip`: try! I doubt it, but I want to see the result if you can

[14:44:03.0000] <gsnedders>
kingryan: actually, just storing whole HTTP responses gets annoying when you have binary responses

[14:46:00.0000] <Philip`>
I don't remember seeing any bits of the algorithm that would cause annoying pumping lemma problems when trying to make a (theoretical) regular expression of it, and in any case nobody cares about that since real regexps are not regular...

[14:47:00.0000] <kingryan>
gsnedders: why is that annoying?

[14:47:01.0000] <takkaria>
being able to define your own character classes could increase readability a lot

[14:47:02.0000] <Philip`>
(I don't think a great number of theoretical models of regular expressions even attempt to cope with the ability to include arbitrary Perl code in the middle of one)

[14:48:00.0000] <gsnedders>
kingryan: text editors tend to not like arbitrary binary data

[14:48:01.0000] <kingryan>
right

[14:49:00.0000] <gsnedders>
kingryan: you want to do all of feed/html?

[14:50:00.0000] <kingryan>
gsnedders: not sure what you're asking

[14:50:01.0000] <gsnedders>
kingryan: of the test cases

[14:50:02.0000] <gsnedders>
kingryan: if you do the feed/html ones, I'll just do the rest

[14:50:03.0000] <kingryan>
cool

[14:51:00.0000] <kingryan>
yeah, I'll do as many as it takes for my software to work right :)

[14:51:01.0000] <gsnedders>
the feed/html ones are actually the ones I need, too :P

[14:52:00.0000] <kingryan>
so, how do we want to organize these?

[14:52:01.0000] <gsnedders>
that's what I'm wondering.

[14:52:02.0000] <kingryan>
the feed/html ones aren't dependent on HTTP

[14:52:03.0000] <kingryan>
is "content-type" the only HTTP header we need?

[14:53:00.0000] <gsnedders>
there's one check that content-encoding isn't set

[14:53:01.0000] <gsnedders>
but that only applies for certain content-type headers anyway

[14:53:02.0000] <gsnedders>
(step 1, 4.7)

[14:54:00.0000] <kingryan>
in other words, do we need arbitrary http headers?

[14:54:01.0000] <kingryan>
if not, we can just use JSON

[14:55:00.0000] <takkaria>
surely you could use JSON anyway?

[14:55:01.0000] <kingryan>
takkaria: the question was whether we need full http headers

[14:55:02.0000] <kingryan>
which I doubt we do

[14:55:03.0000] <gsnedders>
kingryan: we could just have tests for sections 4.7.[1-4]

[14:55:04.0000] <kingryan>
gsnedders: it'd be useful for me to have tests for the subsections, but we should probably also do overall tests

[14:55:05.0000] <gsnedders>
kingryan: actually no

[14:56:00.0000] <kingryan>
no what?

[14:56:01.0000] <gsnedders>
kingryan: the image section relies on the outer algorithm if it doesn't find anything

[14:56:02.0000] <kingryan>
ah, ok

[14:56:03.0000] <kingryan>
nevermind then

[14:56:04.0000] <gsnedders>
unknown jumps to text or binary

[14:57:00.0000] <kingryan>
we can still separate them logically so that if you have only implemented, for example "html vs. feed", you can get that section of tests to pass

[14:57:01.0000] <gsnedders>
yeah

[14:57:02.0000] <kingryan>
how about we structure the tests like so:

[14:57:03.0000] <kingryan>
[{"input": "512 byte string", "result": "mime type"},...]

[14:58:00.0000] <gsnedders>
always 512 bytes?

[14:58:01.0000] <kingryan>
isn't that the limit?

[14:58:02.0000] <kingryan>
unless I'm mistaken

[14:58:03.0000] <gsnedders>
"512 or more"

[14:58:04.0000] <gsnedders>
it's the min

[14:59:00.0000] <kingryan>
it seems to me like there non-determinism there

[14:59:01.0000] <gsnedders>
we better keep them to 512 bytes, so the most pessimistic implementations pass

[15:00:00.0000] <kingryan>
that seems reasonable to me

[15:00:01.0000] <gsnedders>
what do you want to do for the "Big Issue"?

[15:00:02.0000] <kingryan>
nothing for now?

[15:00:03.0000] <kingryan>
we have plenty of non-"big issue" things to implement

[15:01:00.0000] <gsnedders>
that's true.

[15:02:00.0000] <gsnedders>
kingryan: why did you use [{…}] though?

[15:02:01.0000] <gsnedders>
why the outer array?

[15:02:02.0000] <gsnedders>
you mean to have them all in one file?

[15:02:03.0000] <kingryan>
I was just saying that we'd have a list of them in json

[15:03:00.0000] <kingryan>
yes, I mean to have more than one per file

[15:03:01.0000] <gsnedders>
right.

[15:03:02.0000] <kingryan>
and we should do a file for each sub-section

[15:04:00.0000] <kingryan>
so binary.json, unknown.json, image.json, feed.json

[15:06:00.0000] <gsnedders>
JSON has no sort of hex encoding, only unicode, doesn't it?

[15:06:01.0000] <kingryan>
I think so

[15:06:02.0000] <gsnedders>
doesn't that make it impossible to represent some of the bytes we need?

[15:07:00.0000] <kingryan>
like what?

[15:08:00.0000] <kingryan>
/me has only really looked at the feeds sectoin

[15:08:01.0000] <gsnedders>
/me realises.

[15:08:02.0000] <gsnedders>
I can just do \u00FE\u00FF

[15:08:03.0000] <gsnedders>
allowing a UTF-16BE BOM without using the unicode code point for it.

[15:08:04.0000] <kingryan>
gsnedders: we could also do arrays of numbers

[15:08:05.0000] <gsnedders>
kingryan: no, strings are easier

[15:08:06.0000] <kingryan>
I agree

[15:09:00.0000] <gsnedders>
kingryan: do you have commit access to put them in html5lib?

[15:09:01.0000] <kingryan>
yes

[15:09:02.0000] <gsnedders>
ah. I don't.

[15:10:00.0000] <kingryan>
I can commit tests if you send them to me

[15:10:01.0000] <gsnedders>
/me instinctively types \xFF and not \u00FF

[15:11:00.0000] <gsnedders>
kingryan: @technorati.com one?

[15:11:01.0000] <kingryan>
either that or @theryanking.com

[15:11:02.0000] <kingryan>
doesn't matter

[15:11:03.0000] <gsnedders>
I just looked at what I had on mailing lists from you :)

[15:12:00.0000] <gsnedders>
jgraham: around?

[15:12:01.0000] <jgraham>
Yep

[15:13:00.0000] <gsnedders>
jgraham: possible to add me to html5lib so I can add tests?

[15:14:00.0000] <jgraham>
Sure. foolistbar? or some other address?

[15:14:01.0000] <gsnedders>
geoffers

[15:14:02.0000] <jgraham>
/me needs a better gmail address

[15:15:00.0000] <gsnedders>
foolistbar sometimes seems so appropriate

[15:15:01.0000] <jgraham>
OK it should be done

[15:15:02.0000] <Philip`>
/me needs a better one too

[15:16:00.0000] <gsnedders>
jgraham: thanks

[15:16:01.0000] <Philip`>
If someone could kick off all the early users who got the good usernames, that'd be very helpful

[15:16:02.0000] <gsnedders>
like me?

[15:16:03.0000] <gsnedders>
/me ducks

[15:16:04.0000] <gsnedders>
I got foolistbar quite late on, actually

[15:16:05.0000] <gsnedders>
they were already using googlemail.com by then

[15:16:06.0000] <jgraham>
gsnedders: You don't have anything like as common a name as Philip` or I though

[15:16:07.0000] <gsnedders>
jgraham: that's true

[15:18:00.0000] <jgraham>
/me decides jgrham.html might be good for html related bits

[15:18:01.0000] <jgraham>
er jgraham that is

[15:19:00.0000] <jgraham>
or if not good at least better

[15:23:00.0000] <gsnedders>
god do I hate typing JSON

[15:23:01.0000] <gsnedders>
practice makes perfect, though

[15:25:00.0000] <gsnedders>
(all of the tests I've so far written read "Hello, World!")

[15:25:01.0000] <takkaria>
I like using Jabberwocky when string data is required for tests

[15:26:00.0000] <jgraham>
Has anyone implemented the table model stuff

[15:26:01.0000] <jgraham>
?

[15:27:00.0000] <jgraham>
/me can't see where the algorithm is supposed to move on to the next row

[15:30:00.0000] <gsnedders>
http://geoffers.no-ip.com/svn/php-html-5-direct/tests/type-sniffing/textOrBinary — that's horrid to work on :\

[15:31:00.0000] <kingryan>
gsnedders: that does look horrid

[15:32:00.0000] <kingryan>
working with binary data in a text editor is never fun though

[15:32:01.0000] <gsnedders>
those aren't binary.

[15:32:02.0000] <gsnedders>
they're specific unicode encodings.

[15:32:03.0000] <kingryan>
ah, ok

[15:32:04.0000] <gsnedders>
BOM + "Hello, World!"

[15:33:00.0000] <zcorpan_>
don't \u... represent characters, not bytes?

[15:33:01.0000] <gsnedders>
zcorpan_: yes, but there's no other way to do it :(

[15:33:02.0000] <zcorpan_>
ok

[15:33:03.0000] <kingryan>
maybe JSON isn't the best approach then?

[15:33:04.0000] <gsnedders>
everything is going to be horrid

[15:34:00.0000] <zcorpan_>
[0xFE,0xFF,0x00, ...

[15:34:01.0000] <gsnedders>
that's the alternative in JSON

[15:34:02.0000] <kingryan>
gsnedders: we *could* do integer arrays

[15:34:03.0000] <gsnedders>
yeah

[15:34:04.0000] <gsnedders>
probably better, in actuality

[15:34:05.0000] <kingryan>
and that'd be closer to how the spec writes it

[15:35:00.0000] <kingryan>
how about we say  the tests runners should be able to take either a string or an array of integers representing bytes?

[15:35:01.0000] <gsnedders>
actually, no hex forms in JSON.

[15:35:02.0000] <gsnedders>
ergh.

[15:35:03.0000] <kingryan>
ick

[15:36:00.0000] <gsnedders>
if we're doing JSON, the way it is is the only hex way.

[15:36:01.0000] <gsnedders>
yuk.

[15:37:00.0000] <Philip`>
Maybe you could have an array of string+ints, so you can write [0xFE,0xFF,"Hello",0x20,"world"]

[15:37:01.0000] <gsnedders>
Philip`: no. hex. numbers.

[15:37:02.0000] <Philip`>
Oops - [254,255,"Hello",32,"world"]

[15:38:00.0000] <gsnedders>
that means converting everything from the spec

[15:38:01.0000] <gsnedders>
and I can't convert hex to dec in my head

[15:38:02.0000] <gsnedders>
slowwww :(

[15:38:03.0000] <Philip`>
JSON5!

[15:38:04.0000] <gsnedders>
:)

[15:38:05.0000] <Philip`>
Or just say it's JS

[15:38:06.0000] <gsnedders>
yes, but half the point of JSON is it is supported in other languages

[15:39:00.0000] <Philip`>
Or use strings like "FE FF 00 48" and get people to translate it after the JSON parsing

[15:39:01.0000] <gsnedders>
XML + numeric entities suddenly looks tempting

[15:46:00.0000] <kingryan>
gsnedders: maybe a microformat? ;)

[15:46:01.0000] <gsnedders>
kingryan: what an out of the world suggestion from you :)

[15:47:00.0000] <kingryan>
:)

[15:47:01.0000] <gsnedders>
kingryan: <var> could be used for input and result… but their content?

[15:48:00.0000] <gsnedders>
<samp> for result?

[15:48:01.0000] <kingryan>
sure

[15:48:02.0000] <gsnedders>
and for the input?

[15:49:00.0000] <kingryan>
<code> ?

[15:50:00.0000] <gsnedders>
I'm slightly dubious about that, but I can't think of anything better

[15:50:01.0000] <kingryan>
<pre> ?

[15:50:02.0000] <gsnedders>
probably better we use XHTML, so there is no need for a full blown HTML parser

[15:51:00.0000] <gsnedders>
<pre> suits us insofar as that the entities (as they'll have to be) are the preformated input

[15:52:00.0000] <gsnedders>
actually, we can't have &#00;

[15:52:01.0000] <kingryan>
/me likes the idea of requiring an HTML parser for the HTML test suite

[15:53:00.0000] <kingryan>
ok, so are we back to JSON, then?

[15:53:01.0000] <gsnedders>
what happens to &#00 in HTML5? Hixie?

[15:53:02.0000] <gsnedders>
*&#00;

[15:54:00.0000] <kingryan>
I don't see any references to that in html5

[15:54:01.0000] <Philip`>
"if the number is zero, if the number is higher than 0x10FFFF, or if it's one of the surrogate characters (characters in the range 0xD800 to 0xDFFF), then this is a parse error; return a character token for the U+FFFD REPLACEMENT CHARACTER character instead"

[15:55:00.0000] <Philip`>
( http://www.whatwg.org/specs/web-apps/current-work/multipage/section-tokenisation.html#tokenising )

[15:55:01.0000] <kingryan>
ah

[15:55:02.0000] <gsnedders>
I knew if it was unescaped it got replaced

[15:55:03.0000] <gsnedders>
/me sighs

[15:56:00.0000] <gsnedders>
damn. null. bytes.

[15:56:01.0000] <Philip`>
You could store the whole document with ROT-1, so the null bytes map onto &#x01;

[15:57:00.0000] <takkaria>
/me chuckles

[15:57:01.0000] <kingryan>
I think we're back to JSON

[15:57:02.0000] <gsnedders>
ROT-1 sounds like a nice idea

[15:57:03.0000] <gsnedders>
as long as we never need 01

[15:58:00.0000] <kingryan>
you mean FE?

[15:58:01.0000] <gsnedders>
it means we can't test one byte in a range of bytes

[15:58:02.0000] <gsnedders>
kingryan: no, with rot-1

[15:58:03.0000] <gsnedders>
kingryan: map &#00; to &#01;

[15:58:04.0000] <kingryan>
that mens FE maps to 00, right?

[15:58:05.0000] <gsnedders>
kingryan: ah. true.

[15:59:00.0000] <gsnedders>
kingryan: FF, surely?

[15:59:01.0000] <Philip`>
You've got the whole of Unicode to work in - use &#x100;

[15:59:02.0000] <Philip`>
(for &#xFF + 1)

[15:59:03.0000] <kingryan>
true

[15:59:04.0000] <gsnedders>
or just replace 00 with 01

[15:59:05.0000] <gsnedders>
and keep the rest as is

[15:59:06.0000] <kingryan>
as long as we don't need FFFE :)

[15:59:07.0000] <gsnedders>
FFFE is UTF-16 BOM

[16:00:00.0000] <kingryan>
I know ;)

[16:00:01.0000] <gsnedders>
I expected you did

[16:00:02.0000] <gsnedders>
replace 0 with 1 seems the simple solution

[16:02:00.0000] <kingryan>
are we still talking about doing this in html?

[16:03:00.0000] <gsnedders>
yes.

[16:08:00.0000] <gsnedders>
RFC: http://geoffers.no-ip.com/svn/php-html-5-direct/tests/type-sniffing/textOrBinary

[16:09:00.0000] <gsnedders>
wow. we've taken 75 minutes to get this far.

[16:09:01.0000] <tantek>
/me volunteers gsnedders to go to a W3C f2f meeting. 

[16:10:00.0000] <gsnedders>
tantek: if you can get me out of school and get me paid for, sure :P

[16:10:01.0000] <gsnedders>
/me wonders what is up with tantek's connection

[16:11:00.0000] <gsnedders>
tantek: if you can get me out of school and get me paid for, sure :P

[16:11:01.0000] <gsnedders>
tantek: what's up with your connection, BTW?

[16:12:00.0000] <kingryan>
I think tantek's at a cafe

[16:12:01.0000] <gsnedders>
and bad wi-fi?

[16:14:00.0000] <gsnedders>
kingryan: should we go with HTML then?

[16:14:01.0000] <gsnedders>
kingryan: just have it marked up like that?

[16:15:00.0000] <gsnedders>
tantek: why me?

[16:18:00.0000] <tantek_>
gsnedders, you mentioned something about taking so many minutes to get so far. indicating that you've clearly learned some patience. a very useful quality.

[16:19:00.0000] <gsnedders>
tantek_: not patience, just persistence

[16:19:01.0000] <tantek_>
another useful quality.

[16:20:00.0000] <jgraham>
Since test case formats are tonight's hot topic, would anyone like to suggest a format for table model tests? They need to take HTML as an input and check that the cells appear in the right places in a 2D grid

[16:20:01.0000] <jgraham>
JSON would work but it would be really verbose

[16:21:00.0000] <jgraham>
I guess some custom plaintext format might be best

[16:22:00.0000] <jgraham>
Oh and each slot can have >1 or 0 cells

[16:22:01.0000] <jgraham>
and each cell has extra metadata which needs to be checked :(

[16:24:00.0000] <Philip`>
You could represent the 2D grid with an HTML table

[16:26:00.0000] <kingryan>
gsnedders: as cool as using HTML would be, many of the tests are already in JSON and adding more dependencies, even if self-dependencies adds to the cost of collaboration

[16:26:01.0000] <kingryan>
jgraham: I'd use a multidimensional JSON array

[16:27:00.0000] <kingryan>
[[cell, cell], [cell, [double, cell]]]

[16:27:01.0000] <gsnedders>
kingryan: we'd need to add something like "Every unicode character must be interpreted as a byte with the value of the unicode code point the character represents"

[16:27:02.0000] <kingryan>
yes

[16:28:00.0000] <jgraham>
kingryan: That's sort of thing is basically what I was thinking too. It's just slightly irritating that json has such a rigid syntax. However it's less verbose than HTML

[16:29:00.0000] <Philip`>
/me likes YAML

[16:29:01.0000] <kingryan>
right, but json is always utf-8, so the code-point=>character=>byte relationship is fixed

[16:29:02.0000] <kingryan>
Philip`: json is yaml :)

[16:30:00.0000] <kingryan>
http://redhanded.hobix.com/inspect/yamlIsJson.html

[16:30:01.0000] <kingryan>
/me uses a yaml parser for json sometimes

[16:30:02.0000] <Philip`>
JSON is only a subset of YAML, so YAML necessarily has less of a rigid syntax :-)

[16:31:00.0000] <kingryan>
true

[16:31:01.0000] <kingryan>
and it does allow hex numbers

[16:32:00.0000] <Philip`>
YAML seems hugely more complex to parse than JSON, but I don't care because I just say "use YAML::Syck" and it works magically

[16:32:01.0000] <kingryan>
indeed

[16:32:02.0000] <kingryan>
I'd rather not add another dependency to html5lib, though

[16:32:03.0000] <kingryan>
(though that's a no-op for ruby, since yaml is in the stdlib)

[16:32:04.0000] <gsnedders>
kingryan: "JSON text SHALL be encoded in Unicode.  The default encoding is UTF-8. "

[16:33:00.0000] <kingryan>
/me stands corrected

[16:33:01.0000] <gsnedders>
/me has RFC4627 open due to his sheer lack of knowledge of JSON and remembers noticing that earlier

[16:41:00.0000] <gsnedders>
kingryan: you go and make up some sort of test case format

[16:41:01.0000] <gsnedders>
time for me to sleep

[16:41:02.0000] <kingryan>
k


2007-08-18
[17:19:00.0000] <Philip`>
/me sees that HTML 3.2 said <input type=image> should always send x and y coordinates, then HTML 4 instead said 'ismap' was required for that (though the DTD didn't allow it on <input> until HTML 4.01)

[17:21:00.0000] <jacobolus>
hey.  just curious: are there any recent summaries of HTML5 progress (either through W3C or WHATWG)?

[17:22:00.0000] <zcorpan_>
jacobolus: http://html5.org/tools/web-apps-tracker ?

[17:23:00.0000] <Philip`>
How recent, and how summarised?

[17:23:01.0000] <jacobolus>
Philip`: sometime in the last few months.  paragraphs not commit logs. :)

[17:24:00.0000] <jacobolus>
Philip`: big ideas, not minor fixes, etc.

[17:24:01.0000] <zcorpan_>
don't think so

[17:24:02.0000] <jacobolus>
okay, no worries. :)

[17:24:03.0000] <Philip`>
Hmm, I can't even remember any big changes recently

[17:24:04.0000] <Philip`>
which is probably caused by my memory rather than by a lack of big changes

[17:25:00.0000] <zcorpan_>
http://lists.whatwg.org/pipermail/help-whatwg.org/2007-August/000070.html

[17:25:01.0000] <zcorpan_>
<img alt> has recently been revamped

[17:25:02.0000] <jacobolus>
also, is there any rough idea from browser vendors when they plan to start abiding by html 5's recommendations?

[17:26:00.0000] <jacobolus>
even for existing elements (i.e. not counting all the fun new stuff)

[17:26:01.0000] <othermaciej>
it will be gradual

[17:26:02.0000] <othermaciej>
there aren't a lot of changes to existing elements though

[17:26:03.0000] <othermaciej>
(other than global stuff that applies to all elements)

[17:27:00.0000] <jacobolus>
i thought html5 was trying to get the browsers to act the same way in several places that they currently interpret the html spec differently

[17:28:00.0000] <jacobolus>
or are those mostly global things?

[17:29:00.0000] <othermaciej>
well, parsing is one big area where that is the case

[17:29:01.0000] <othermaciej>
but, for instance, there's nothing about the <b> element that would change in browsers

[17:29:02.0000] <othermaciej>
or <img>

[17:29:03.0000] <Philip`>
<img style="display:none"> should change in some browser

[17:30:00.0000] <othermaciej>
what is the change?

[17:30:01.0000] <zcorpan_>
othermaciej: <img> has some dom interfaces that might need changing (like .complete)

[17:30:02.0000] <zcorpan_>
er, members of the HTMLImageElement interface

[17:30:03.0000] <Philip`>
Opera doesn't download the image, which I've found annoyingly inconsistent with other browsers

[17:30:04.0000] <othermaciej>
yeah, all other browsers will load images that are unrendered or not in the DOM

[17:30:05.0000] <jacobolus>
othermaciej, others: okay.  thanks :)

[17:32:00.0000] <jacobolus>
othermaciej: know if there's been any decision about <a>'s with display:block being allowed/allowing block elements inside or not?

[17:32:01.0000] <zcorpan_>
Philip`: iirc we do because it saves bandwidth... which is important on mobiles

[17:33:00.0000] <zcorpan_>
jacobolus: 'display' doesn't in any way change nesting rules

[17:33:01.0000] <othermaciej>
jacobolus: putting display: block on an <a> element is a consideration that is out of scope for the HTML spec

[17:33:02.0000] <othermaciej>
it's purely at the CSS level

[17:33:03.0000] <othermaciej>
and it is allowed

[17:33:04.0000] <jacobolus>
ah, okay

[17:33:05.0000] <othermaciej>
nesting block elements in <a> remains non-conforming I believe

[17:34:00.0000] <jacobolus>
othermaciej: I believe putting block elements in an <a> w/ display:block works just fine in all browsers, though :)

[17:34:01.0000] <othermaciej>
though in practice it is handled (not sure if the spec has full detail on how to handle it yet)

[17:34:02.0000] <zcorpan_>
though it has been suggested that it should be allowed to nest blocks in lins

[17:34:03.0000] <zcorpan_>
links

[17:34:04.0000] <othermaciej>
jacobolus: putting block elements in <a> without display:block also works

[17:34:05.0000] <Philip`>
zcorpan_: hendry suggested the same reason

[17:34:06.0000] <zcorpan_>
Philip`: what reason?

[17:35:00.0000] <jacobolus>
othermaciej: hmm, it caused some pretty nasty rendering issues in firefox though, when I tested it a few months ago (don't remember what version)

[17:35:01.0000] <Philip`>
The saving-bandwidth-on-mobiles reason

[17:35:02.0000] <zcorpan_>
Philip`: ah. yep

[17:35:03.0000] <Philip`>
It's just annoying when I'm writing <canvas> tests and using drawImage with <img>s that shouldn't be visible, though maybe that's not an especially common use case :-)

[17:36:00.0000] <Philip`>
(Now I tend to use visibility:hidden and stick them at the end of the document where nobody will notice them)

[17:39:00.0000] <othermaciej>
Philip`: new Image() creates an image that is not in the document but which generally loads and fires a load event

[17:39:01.0000] <othermaciej>
Philip`: perhaps that works in Opera even if <img style="display: none"> doesn't

[17:40:00.0000] <Philip`>
That does work in Opera, but I vaguely remember there being some reason why I didn't want to do that

[17:41:00.0000] <Philip`>
but I have no idea what the actual reason was

[17:44:00.0000] <Philip`>
Hmm, maybe it was just because I was lazy and preferred to write one line of HTML instead of two lines of JS

[19:05:00.0000] <Lachy>
good morning

[19:30:00.0000] <Lachy>
I don't understand what Sam Ruby is ranting about.  What's wrong with the way Firefox handles his test case? http://feedvalidator.org/testcases/atom/1.1/brief-noerror.xml

[19:34:00.0000] <Philip`>
He "wanted to view the data in an XML parse tree" but Firefox thinks it's a feed

[19:46:00.0000] <Lachy>
so what's wrong with that?  it is a feed.  He wouldn't expect an XHTML or SVG file delivered as application/xml to be rendered as a tree would he?

[19:50:00.0000] <Philip`>
If you do 'data:application/xml,the stuff from that file' then it does display the tree instead of thinking it's a feed

[19:58:00.0000] <Philip`>
(Not that that's a particularly relevant observation...)

[20:37:00.0000] <Philip`>
/me wonders how any UA could ever distinguish between missing-alt and alt="", given that existing content presumably(?) mixes the two forms freely and future content will keep doing it wrong

[20:38:00.0000] <othermaciej>
Philip`: at least some UAs already treat the two cases differently

[20:43:00.0000] <Philip`>
Hmm, Lynx is the only one I have that seems to care

[20:44:00.0000] <Philip`>
(Links doesn't - it always ignores the image, unless it's in a link and then it says [IMG] instead)

[04:40:00.0000] <jgraham>
Hixie: yt?

[05:40:00.0000] <Whiskey_M>
'lo

[05:42:00.0000] <Philip`>
Greetings

[06:24:00.0000] <Philip`>
Lynx supports a surprising number of HTML 3.0 features, which no other browser does that I'm aware of

[06:24:01.0000] <Philip`>
(<tab>, <note>, <fn>, etc)

[07:07:00.0000] <Lachy>
what does <note> do that's special?

[07:10:00.0000] <Philip`>
"Lynx implements the HTML 3.0 NOTE element (Admonishment) as a labeled block, i.e., as if a new paragraph were indicated whether or not paragraphing markup is included in its content, with greater than normal left and right margins, and with the type of note indicated by an emphasized label based on the value of its CLASS or ROLE attribute. If no CLASS or ROLE attribute is included, the default label NOTE: will be used. Lynx recognizes the values cauti

[07:10:01.0000] <Philip`>
... Lynx recognizes the values caution and warning, for which, respectively, the labels CAUTION: or WARNING: will be used."

[07:11:00.0000] <Philip`>
http://www.w3.org/MarkUp/html3/notes.html

[07:12:00.0000] <Philip`>
Hmm, it even lets you specify an image to accompany the note and an MD5 checksum of the image to ensure it hasn't been tampered with

[07:13:00.0000] <Philip`>
Sounds like it'd be vaguely relevant to the various warning/note/etc boxes Wikipedia puts all over the place, like in the top of http://en.wikipedia.org/wiki/HTML5

[07:18:00.0000] <Lachy>
that might be an interesting element to investigate.  Didn't we have class="note" predefined at one stage? Maybe that could be a non-controversial replacement

[07:25:00.0000] <Philip`>
What would the value be, compared to writing plain text like "Warning: Leopards spotted"?

[07:32:00.0000] <Lachy>
don't know.  That would be something that needed investigation. The best I can think of at this stage would be as a styling hook, but that's hardly good enough justification

[07:38:00.0000] <Philip`>
Shouldn't you think of problems first and then find solutions, rather starting with the <note> solution and then thinking of problems for it to solve? :-p

[07:39:00.0000] <Philip`>
*rather than

[08:09:00.0000] <tantek>
Lachy re: class="note".  We have defined that in hCard http://microformats.org/wiki/hcard per the NOTE property from RFC2426 section 3.6.2 NOTE Type Definition

[08:09:01.0000] <Lachy>
Philip`, yes, that's right

[08:10:00.0000] <Lachy>
back later, bye

[09:09:00.0000] <Philip`>
It'd be nice if HTML5 <xmp> parsing was more like IE, and more like <pre>/<textarea> in all browsers, by ignoring a leading newline

[09:20:00.0000] <zcorpan_>
Philip`: send it to the list? :) it would also affect getting innerHTML

[09:20:01.0000] <zcorpan_>
what about <listing>? are there others?

[09:23:00.0000] <Philip`>
I bet someone would just tell me that I shouldn't be writing new content with <xmp> :-(

[09:24:00.0000] <Philip`>
but there's no way I'm going to bother with writing &lt;s everywhere

[09:25:00.0000] <zcorpan_>
that's not relevant to the proposal though

[09:26:00.0000] <zcorpan_>
ie ignores the leading newline with listing and plaintext too

[09:28:00.0000] <Philip`>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%0D%0A%3Cstyle%3Ebody%20*%20%7B%20border%3A%201px%20solid%20%7D%3C/style%3E%0D%0A%3Cbody%3E%0D%0A%0D%0A%3Cdiv%20style%3D%22white-space%3A%20pre%22%3E%0D%0A%0D%0Adiv%0D%0A%0D%0A%3C/div%3E%0D%0A%0D%0A%3Cpre%3E%0D%0A%0D%0Apre%0D%0A%0D%0A%3C/pre%3E%0D%0A%0D%0A%3Ctextarea%20rows%3D5%3E%0D%0A%0D%0Atextarea%0D%0A%0D%0A%3C/textarea%3E%0D%0A%0D%0A%3Cxmp%3E%0D%0A%0D%0Axmp%0D%0A%0D%0A%3

[09:28:01.0000] <Philip`>
Urgh, too long

[09:29:00.0000] <Philip`>
Anyway, IE seems to not render the leading newline with div either

[09:29:01.0000] <Philip`>
nor include it in the innerHTML output

[09:35:00.0000] <zcorpan_>
but it's probably not needed for <div>

[09:36:00.0000] <Philip`>
Oh, how nice - IE's innerHTML changes depending on whether you've got CSS white-space:pre or not

[09:36:01.0000] <zcorpan_>
why am i not surprised

[10:08:00.0000] <zcorpan_>
hmm, where's the zombie dom viewer?

[10:10:00.0000] <Philip`>
http://canvex.lazyilluminati.com/misc/dom-viewer/

[10:11:00.0000] <Philip`>
Useful for http://canvex.lazyilluminati.com/misc/dom-viewer/?%3C%21DOCTYPE%20html%3E%0D%0A%3Cplaintext%3Efoo in Opera

[10:11:01.0000] <zcorpan_>
yeah

[10:13:00.0000] <zcorpan_>
does w() work with the dead dom viewer?

[10:14:00.0000] <Philip`>
No

[10:15:00.0000] <zcorpan_>
we ignore the leading newline with <listing> but not <xmp> or <plaintext>

[10:15:01.0000] <Philip`>
since I can't think of any way to inject the function into the page

[10:15:02.0000] <zcorpan_>
ok

[10:15:03.0000] <Philip`>
Everyone seems to do <listing> the same

[10:15:04.0000] <Philip`>
(i.e. not the same as HTML5)

[10:16:00.0000] <zcorpan_>
firefox crashes for me when i try <listing>x in the live dom viewer

[10:19:00.0000] <zcorpan_>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%3Cscript%3Ewindow.onload%3Dfunction(){w(document.body.firstChild.firstChild.data.length)}%3C/script%3E%3Cbody%3E%3Clisting%3E

[10:19:01.0000] <zcorpan_>
then type an x

[10:19:02.0000] <Philip`>
IE really is very very crazy

[10:20:00.0000] <Philip`>
Non-deterministically so, as far as I can see...

[10:23:00.0000] <Philip`>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%0D%0A%3Cstyle%3Ebody%20*%20%7B%20border%3A%201px%20solid%20%7D%3C/style%3E%0D%0A%3Cbody%20onload%3D%22with%28document.body%29%7Bfor%28var%20i%3D0%3Bi%3CchildNodes.length%3B++i%29w%28%27%3E%3E%3E%27+childNodes%5Bi%5D.firstChild.nodeValue+%27%3C%3C%3C%27%29%7D%22%3E%3Cdiv%20style%3D%22white-space%3A%20pre%22%3E%0D%0A%0D%0Adiv%20white-space%3Apre%0D%0A%0D%0A%3C/div%3E

[10:23:01.0000] <Philip`>
If I add a character onto the end, then an extra newline appears in the top of the rendered div

[10:23:02.0000] <Philip`>
but when I use my copy of the DOM viewer, the newline randomly appears and disappears when I add or remove characters on the end

[10:23:03.0000] <Philip`>
(in IE7)

[10:24:00.0000] <Philip`>
I don't get the <listing> crash in FF2

[10:24:01.0000] <zcorpan_>
firefox 3

[10:25:00.0000] <zcorpan_>
ok, so the leading newline is only needed for <listing>. not <xmp> or <plaintext>

[10:27:00.0000] <Philip`>
/me will post to the list

[10:28:00.0000] <Philip`>
(hopefully without too many errors)

[10:29:00.0000] <Philip`>
I'd just be happy if html5lib stripped the newline from <xmp>, regardless of what browsers do, so I wouldn't have to write lines of code to manipulate the DOM and fix up the text while transforming it into <pre><code> :-)

[10:30:00.0000] <Philip`>
(though maybe that's a selfish reason, and it's only about two lines of code anyway)

[10:32:00.0000] <Philip`>
Argh

[10:32:01.0000] <Philip`>
/me replies to himself

[10:37:00.0000] <Philip`>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%0A%3Cbody%20onload%3D%22w%28%27ok%27%29%7D%22%3E - why does Firefox accept that?

[10:54:00.0000] <Philip`>
Oh, it's fixed in FF3

[10:54:01.0000] <Philip`>
and <listing> crashes for me in FF3

[11:52:00.0000] <zcorpan_>
Philip`: will Hixie find that message again when it doesn't contain "detailed review" in the subject...? :)

[12:15:00.0000] <Philip`>
It's not part of a detailed review at all - it's just a random comment about something that I only noticed because I'm being silly and using <xmp> - so I wouldn't want to be misleading in the subject :-)

[12:21:00.0000] <zcorpan_>
/me uses "detailed review" in his public-html messages that he wants the editors to find, even though they're random comments

[12:21:01.0000] <zcorpan_>
<listing> isn't silly though as there's interop already

[12:22:00.0000] <Philip`>
I've added it to the SpecReviews wiki page, since hopefully things there won't get too lost

[12:25:00.0000] <zcorpan_>
that should work too, indeed

[12:27:00.0000] <Philip`>
Perhaps I could also set up a blog anywhere on the internet and post it there, and I expect Hixie would find it through his magic finding-anything-anybody-says-about-HTML5-on-blogs powers

[12:30:00.0000] <Dashiva>
He probably uses (shock) google :)

[12:31:00.0000] <Philip`>
Nah, I bet he's a closet Live Search user

[12:31:01.0000] <Philip`>
How can you not like a search engine whose front page says "Get Pimped!"?

[12:36:00.0000] <hsivonen>
Philip`: has the live.com front page actually said that?

[12:37:00.0000] <Philip`>
It said it when I looked at it just now

[12:37:01.0000] <Philip`>
"Get Pimped! FREE emoticon packs for Messenger" linking to http://www.pimpmylive.co.uk/

[12:37:02.0000] <hsivonen>
ooh. there's a changing ad on reload

[12:39:00.0000] <hsivonen>
well, MTV still has enough impact to make the word "pimp" suitable for the front page of a corporate property

[12:41:00.0000] <hsivonen>
btw, does live.com have live search feeds for a term like Google and Technorati?

[12:43:00.0000] <hsivonen>
hah. live.com uses bitmaps for non-Latin-script language names.

[12:45:00.0000] <hsivonen>
can't Microsoft just install the full load of fonts and locale-specific algorithms with the default install and get rid of the IE language pack nag dialogs?

[12:46:00.0000] <Philip`>
http://blogs.msdn.com/michkap/archive/2006/10/10/813347.aspx - sounds like in Vista they decided to just install everything

[12:46:01.0000] <hsivonen>
cool

[12:47:00.0000] <Philip`>
and then people complain about bloat :-)

[12:47:01.0000] <hsivonen>
disk is cheap. configuration-specific quality assurance isn't


2007-08-19
[17:45:00.0000] <Philip`>
/me wishes Gmail didn't destroy his careful formatting by using a variable-width font

[17:55:00.0000] <colione>
/me thinks the gmail devil is laughing

[07:59:00.0000] <hendry>
i don't quite understand the difference beween <object> and <embed>

[08:01:00.0000] <hendry>
perhaps its a legacy thing

[08:01:01.0000] <hendry>
though i've seen <embed> as a child of <object>

[08:05:00.0000] <gsnedders>
hendry: <object> has fallback content, <embed> does not. that's the main difference

[08:30:00.0000] <hendry>
gsnedders: thanks

[08:38:00.0000] <hsivonen>
hendry: seen this? http://www.faculty.idc.ac.il/arik/IMRet-All.mov

[08:38:01.0000] <hsivonen>
(content adaptation)

[08:38:02.0000] <jgraham>
(browser crash)

[08:39:00.0000] <jgraham>
It seems to be working this time :)

[08:42:00.0000] <hendry>
hsivonen: woo, thanks for the link

[08:52:00.0000] <hendry>
hsivonen: that's amazing. though I guess it's going to require a 'bit of work', hence server side. Also that stuff with faces....

[08:57:00.0000] <hendry>
hsivonen: i remember you doing some jp2 tests. any future thoughts? I guess we will be playing around with jpg/png for sometime to come...

[08:59:00.0000] <hsivonen>
hendry: PNG has much better characteristics when it comes to downloading less data in order to render a smaller image

[08:59:01.0000] <hsivonen>
(assuming Adam interlacing)

[09:02:00.0000] <hsivonen>
hendry: the problem is that the current crop of browsers downloads full PNGs, and one would want the current browsers to load quarter data and new browsers versions for HD desktop displays to load full images

[09:02:01.0000] <hendry>
that would make sense

[09:02:02.0000] <hendry>
hsivonen: do you have an opinion on Opera's img { display: none;} ?

[09:03:00.0000] <hsivonen>
hendry: I don't.

[09:03:01.0000] <hsivonen>
hendry: is it that Opera doesn't load the HTTP resource but other browsers do?

[09:03:02.0000] <hendry>
hsivonen: yeah

[09:03:03.0000] <hsivonen>
seems reasonable

[09:04:00.0000] <hsivonen>
the canonical way to preload images from a script is new Image

[09:04:01.0000] <hendry>
in html5 it's speced that img must download

[09:05:00.0000] <hsivonen>
hendry: I don't understand the problem space well enough to have an educated opinion right now

[09:06:00.0000] <hendry>
hsivonen: how do you come across this 'Adam interlacing' stuff? do you have tests?

[09:06:01.0000] <hsivonen>
hendry: in the PNG spec :-)

[09:07:00.0000] <hsivonen>
hendry: http://www.schaik.com/pngsuite/pngsuite_int_png.html

[09:08:00.0000] <hendry>
is there some reason why "the current crop of browsers" don't implement interlaced aware downloads?

[09:09:00.0000] <hendry>
because I am always thinking that if images were somehow well dealt with for mobile env., we'd see a massive break through

[09:10:00.0000] <hsivonen>
hendry: I'm not sure. but I'd be surprised if current browsers stopped PNG downloads after a quarter of data when the display size to device pixels only needs a quarter of the samples

[09:10:01.0000] <hsivonen>
hendry: I haven't actually tested

[09:11:00.0000] <hsivonen>
hendry: I'd be happy to have guessed wrong :-)

[09:11:01.0000] <hsivonen>
gotta go

[09:14:00.0000] <hendry>
hsivonen: cheers

[09:37:00.0000] <Philip`>
http://canvex.lazyilluminati.com/misc/dom-viewer/?%3C%21DOCTYPE%20html%3E%0D%0A%3Ch1%20id%3D%22logo%22%3E%3Cimg%20alt%3D%22WordPress%22%20src%3D%22wp-admin/images/wordpress-logo.png/%3E%3C/h1%3E%0D%0A%3Cp%3E%0D%0A%3Ch1%3EError%20establishing%20a%20database%20connection%3C/h1%3E

[09:38:00.0000] <Philip`>
http://james.html5.org/cgi-bin/parsetree/parsetree.py?source=%3C%21DOCTYPE+html%3E%0D%0A%3Ch1+id%3D%22logo%22%3E%3Cimg+alt%3D%22WordPress%22+src%3D%22wp-admin%2Fimages%2Fwordpress-logo.png%2F%3E%3C%2Fh1%3E%0D%0A%3Cp%3E%0D%0A%3Ch1%3EError+establishing+a+database+connection%3C%2Fh1%3E

[09:38:01.0000] <Philip`>
Is that a bug in HTML5, since it loses information that IE/Firefox/Opera all preserve?

[09:38:02.0000] <Philip`>
(on real-world content, too)

[09:45:00.0000] <Philip`>
Hmm, Safari does about the same as HTML5 (except it doesn't even create the img element)

[09:50:00.0000] <Philip`>
http://hasteworld.co.uk/ has exactly the same error message too - the whole page is blank in Safari (and HTML5)

[12:45:00.0000] <Hixie>
Lachy: someone sent me mail saying a blog post was pending

[13:57:00.0000] <Philip`>
Hixie: I made some changes to the multipage-generator so it splits into more sensible-sized sections and tries to redirect users when things have moved - would it be possible/convenient some time to set that up on whatwg.org?

[13:57:01.0000] <Hixie>
iirc i had to make a bunch of changes to get it to work in my current setup

[13:57:02.0000] <Hixie>
i'll have to look into it

[13:57:03.0000] <Hixie>
ping me during the week some time?

[13:58:00.0000] <Philip`>
Okay, sure

[13:58:01.0000] <Hixie>
the current setup on whatwg.org is complicated -- the files get sent off to a server in the uk for splitting, and that's where your script actually runs

[13:58:02.0000] <Hixie>
then it gets sent back to whatwg.org as a tarball and extracted

[14:15:00.0000] <Hixie>
http://www.whatwg.org/issues/

[14:28:00.0000] <Philip`>
Hixie: When registering, I got the "Your code is: ..." email twice

[14:29:00.0000] <Philip`>
(I suppose I might have accidentally double-clicked, though)

[14:29:01.0000] <Philip`>
(but I don't think I did)

[14:30:00.0000] <Philip`>
After seemingly having logged in with that code, it still says "You must be logged in to vote."

[14:31:00.0000] <Philip`>
If I log in with an invalid address/code, it doesn't actually do anything different - it just reloads the page with ?email=x%40y.com&code=z and nothing much happens

[14:32:00.0000] <Philip`>
In FF3, the whole page just says "Scripting error: Script error. in javascript:%20eval(__firebugTemp__); line 0" (plus a Reload button)

[14:33:00.0000] <Philip`>
(perhaps partly because I've got a half-broken Firebug installed)

[14:33:01.0000] <Philip`>
Oh, now it says "Scripting error: document.getElementById("no-js-warning") has no properties in http://www.whatwg.org/issues/ line 157"

[14:35:00.0000] <Philip`>
IE6 (in Wine) has nothing other than a single empty button on the page

[14:36:00.0000] <Philip`>
(My earlier problems were in Opera)

[14:37:00.0000] <Philip`>
Aha, Konqueror can log in

[14:37:01.0000] <Philip`>
but nothing happens when clicking a "vote" button

[14:37:02.0000] <Philip`>
(Browser interoperability is fun!)

[14:39:00.0000] <Philip`>
Nice idea, though :-)

[14:41:00.0000] <Philip`>
I think it would be nice if you could vote on a whole category, rather than having to choose specific emails, since e.g. someone implementing <video> would probably want all of the graphics-video issues to be addressed

[14:46:00.0000] <Hixie>
Philip`: what browser?

[14:46:01.0000] <Hixie>
i only tested firefox3 and safari3

[14:47:00.0000] <Philip`>
I was using Opera 9.2 first

[14:48:00.0000] <Hixie>
yeah opera has some weird bugs

[14:48:01.0000] <Hixie>
i didn't work around any bugs in this

[14:48:02.0000] <Hixie>
you'll need a 100% compliant browser :-)

[14:49:00.0000] <Hixie>
(for the things i use)

[14:49:01.0000] <Hixie>
interesting idea regarding voting for a category

[14:49:02.0000] <Hixie>
though i don't really want to allow that since the categories can change on a whim

[14:51:00.0000] <Philip`>
Aha, it seems to work in FF3 if I remove Firebug

[14:52:00.0000] <Philip`>
It also looks like the vote I made in Konqueror did actually get counted, though it didn't give me any feedback in the browser (whereas it does give feedback in FF3)

[14:54:00.0000] <Hixie>
cool

[14:54:01.0000] <zcorpan_>
/me will hunt down the bugs in opera tomorrow

[14:55:00.0000] <Philip`>
Is it intentional that it doesn't say who each email is from?

[14:55:01.0000] <Hixie>
yes

[15:04:00.0000] <mpt>
So much for HTML 5 being backward-compatible :-P

[15:04:01.0000] <jgraham>
Hixie: Am I blind or is there a step missing from the table model construction algorithm? Somewhere around step 18 I expect current element to be set to the next child of the table element but I can't see where that happens

[15:04:02.0000] <jgraham>
/me can send email if you don't want to look now

[15:29:00.0000] <zcorpan_>
/me voted for "Re: [whatwg] getElementsByClassName"

[15:30:00.0000] <zcorpan_>
Hixie: when pressing the Vote button, shouldn't there be some feedback that the vote was successful or something?

[15:31:00.0000] <Philip`>
zcorpan_: There was feedback when I did it in FF3

[15:31:01.0000] <zcorpan_>
Philip`: oh. i didn't get any feedback. Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a8pre) Gecko/2007081505 Minefield/3.0a8pre

[15:33:00.0000] <Philip`>
It changes into a "Remote Vote" button and lists me as having voted for it, as soon as I click

[15:33:01.0000] <Philip`>
in Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9a8pre) Gecko/2007081904 Minefield/3.0a8pre

[15:38:00.0000] <zcorpan_>
oh yep, works in today's build

[15:54:00.0000] <zcorpan_>
http://www.youtube.com/watch?v=Har-PRP4X9U svg + <video>

[15:56:00.0000] <hasather_>
zcorpan_: cool

[15:56:01.0000] <zcorpan_>
http://www.bluishcoder.co.nz/2007/08/svg-video-demo.html is the post about it


2007-08-20
[21:23:00.0000] <Hixie>
jgraham: always send mail :-)

[21:24:00.0000] <Hixie>
mpt: i don't think the problems are with html5 implementations, i think they're bugs with the implementations of other things (like xmlhttprequest, html4 forms, etc)

[21:49:00.0000] <mpt>
Hixie, perhaps, but it gives a bad impression nonetheless

[21:57:00.0000] <Hixie>
mpt: feel free to work out what workarounds i should use

[22:00:00.0000] <mpt>
No, sorry, I'm working

[22:10:00.0000] <Hixie>
mpt: ditto :-)

[23:32:00.0000] <Molly>
Hello

[00:21:00.0000] <karlcow>
/me wonders if Molly has a limit of 50 comments by post. I had replied and it has not been published.

[00:23:00.0000] <zcorpan_>
karlcow: she doesn't

[00:33:00.0000] <zcorpan_>
does anyone know already why http://www.whatwg.org/issues/ doesn't work in opera? :)

[00:36:00.0000] <zcorpan_>
status is undefined

[00:36:01.0000] <zcorpan_>
hmm

[00:36:02.0000] <othermaciej>
what aspect of it does not work?

[00:38:00.0000] <zcorpan_>
regestring doesn't work. because it fails to update the status

[00:38:01.0000] <zcorpan_>
which is because of a parsing bug, it seems. <p><form><p><span>

[00:38:02.0000] <zcorpan_>
vs <p></p><p><span>

[00:38:03.0000] <zcorpan_>
er

[00:38:04.0000] <zcorpan_>
<p></p><form><p><span>

[00:49:00.0000] <Whiskey_M>
'lo

[00:50:00.0000] <zcorpan_>
hey Whiskey_M

[02:43:00.0000] <Hixie>
ok i worked around tehe parsing bug in opera

[02:43:01.0000] <Hixie>
(by putting in the missing </p>s)

[02:43:02.0000] <Hixie>
and made it detect the IEissues so it wouldn't fail later

[02:44:00.0000] <Hixie>
Lachy: btw dunno if you saw but someone else had a blog post ready

[02:44:01.0000] <Hixie>
(or so they tell me)

[02:44:02.0000] <Lachy>
yeah, I saw your message earlier.  I just had to go get myself a new keyboard before I could do anything about it

[02:45:00.0000] <Hixie>
heh

[02:45:01.0000] <Hixie>
what did you do to your old one?

[02:45:02.0000] <Hixie>
also in case anyone missed my earlier comment, i'm experimenting with http://www.whatwg.org/issues/

[02:45:03.0000] <Hixie>
i'll probably send a message about it to public-html tomorrow

[02:45:04.0000] <Hixie>
assuming it works well

[02:46:00.0000] <Lachy>
it was raining yesterday and it got wet when I pulled it out of my car, and even though I tried it, it appears that some of the wires inside had corroded and finally gave up

[02:46:01.0000] <Lachy>
s/tried/dried/

[02:47:00.0000] <Hixie>
ah

[02:47:01.0000] <Hixie>
that sucks

[02:49:00.0000] <Lachy>
is that the list of issues created from your existing email archive folders?

[02:50:00.0000] <annevk>
hmm, 1035 new e-mails

[02:51:00.0000] <Lachy>
Hixie, who was it that emailed you about the blog entry?

[02:51:01.0000] <hsivonen>
annevk: welcome back

[02:51:02.0000] <hsivonen>
now it is my turn to go away: I'll be away from tomorrow until Sept 2

[02:51:03.0000] <annevk>
hi all, thanks

[02:52:00.0000] <annevk>
oh, I will be away again starting tomorrow too...

[02:52:01.0000] <Lachy>
hi annevk 

[02:52:02.0000] <Lachy>
where are you going?

[02:52:03.0000] <annevk>
/me will be back for "real" September 5

[02:52:04.0000] <annevk>
some Dutch island (Schiermonnikoog) and after that hiking around the Mont Blanc (France, Switerzerland, Italy)

[02:53:00.0000] <hsivonen>
Lachy: Romania via Hungary

[02:53:01.0000] <Lachy>
hsivonen, for holiday or work?

[02:53:02.0000] <Lachy>
or other?

[02:53:03.0000] <hsivonen>
Lachy: holiday

[02:54:00.0000] <annevk>
have fun hsivonen!

[02:54:01.0000] <hsivonen>
annevk: thanks. you too

[02:56:00.0000] <Lachy>
I'm trying to figure out if this is a translation of a previous article or an original http://blog.whatwg.org/wp-admin/post.php?action=edit&post=48

[02:56:01.0000] <Lachy>
the title of it is "Présentation : comment HTML 5 peut il être utilisé aujourd’hui ?"

[02:56:02.0000] <hsivonen>
Presentation: how can HTML 5 be used today?

[02:57:00.0000] <Lachy>
ah, right! it's a translation of http://blog.whatwg.org/html5-geekmeet

[02:57:01.0000] <annevk>
hmm, why did people oppose to dropping <input usemap>?

[02:58:00.0000] <Lachy>
because it has some purported accessibility benefits

[02:58:01.0000] <Lachy>
despite the fact that HTML4 didn't define how it worked and no implementation of it actually provded any accessibility benefits

[02:59:00.0000] <Lachy>
eventually, the discussion essentially turned into a proposal for a new feature to provide those accessibility benefits, but no-one presented any use cases for using an image map for form submission

[02:59:01.0000] <annevk>
right...

[03:08:00.0000] <Lachy>
aargh! Who does Rob think he is, being able to tell people when they should or should not respond? http://lists.w3.org/Archives/Public/public-html/2007Aug/0716.html

[03:12:00.0000] <Hixie>
Lachy: it was someone posting about a web forms 2 impl

[03:13:00.0000] <Hixie>
Lachy: and i presume rob burns would be the most knowledgable about when to reply or not reply, since he replies to pretty much every thread

[03:14:00.0000] <Hixie>
and yes, the list mentioned above (whatwg.org/issues, for anne's benefit) is derived from my imap folders

[03:15:00.0000] <othermaciej>
Ironically, Rob seems unaware of his eponymous rules of order

[03:16:00.0000] <Lachy>
Hixie, I didn't see anything in the various WF3 lists about <input placeholder>

[03:16:01.0000] <Hixie>
Lachy: it's there somewhere

[03:16:02.0000] <Hixie>
might be in WF2

[03:18:00.0000] <Lachy>
maybe none of the placeholder discussions have "placeholder" in the subject

[03:18:01.0000] <Lachy>
oh well, doesn't matter

[03:19:00.0000] <Hixie>
possibly

[03:36:00.0000] <zcorpan_>
<input disabled tabindex=0> -- focusable or not?

[03:38:00.0000] <Hixie>
zcorpan_: nothing disabled is focusable

[03:38:01.0000] <rburns>
othermaciej: hey. I'm not unaware of my pseudo namesake, I spelled it Robert's instead of Roberts' a subtle pun (I thought that would be clear to those in the know)

[03:39:00.0000] <rburns>
Anne: the question of the elimination of <input usemap> was concern that insufficient research had been done to justify eliminating it.

[03:41:00.0000] <othermaciej>
rburns: telling people they shouldn't have replied to you is not appropriate behavior on a mailing list; nor is it proper meeting procedure either

[03:41:01.0000] <othermaciej>
if you want to make a formal motion then the usual procedure would be to wait for the chair to recognize you

[03:41:02.0000] <rburns>
well I would say replying to a motion without a second is improper meeting procedure (especially when that norm is raised in the very email being replied to)

[03:41:03.0000] <othermaciej>
and it is the chair's job to make sure rules of order are followed, not any random person who says something

[03:42:00.0000] <othermaciej>
if you think the chair is not doing his job, you raise a point of order with the chair

[03:42:01.0000] <rburns>
the chair could rule my motion out of order and stop a seconding from happening. That would be the email analogous thing to happen

[03:43:00.0000] <rburns>
I think the chair is doing fine.

[03:43:01.0000] <othermaciej>
if you want to apply rules of procedure to a mailing list, even notwithstanding that this is a really bad idea and was not agreed to by any one, you can't pick and choose just the ones you like

[03:43:02.0000] <rburns>
othermaciej: You seem to be the one complaining that he didn't rule my motion out of order

[03:43:03.0000] <othermaciej>
no, I'm not complaining

[03:43:04.0000] <othermaciej>
I think you can phrase your email however you like

[03:44:00.0000] <othermaciej>
but the fact that you called it a motion doesn't give you the authority to tell people not to reply

[03:44:01.0000] <rburns>
othermaciej: I'mnot picking an choosing. I've read complaints about how the email list is unwieldy. I made a motion to help with that. That motion wasn't seconded. I don't see how that's out of order.

[03:44:02.0000] <othermaciej>
I'm not complaining about your motion email

[03:44:03.0000] <othermaciej>
I just think you were out of line in telling people they should not have replied to it

[03:44:04.0000] <rburns>
Well why would someone reply to it if they don't like the volume of email on the list and they have no intention of seconding it. Just let it die then.

[03:45:00.0000] <rburns>
othermaciej : Well I disagree

[03:45:01.0000] <Lachy>
rburns, I think the list needs much less bureaucracy

[03:45:02.0000] <othermaciej>
it's not the first time you've told people they shouldn't have replied to one of your emails, either

[03:45:03.0000] <rburns>
othermaciej: I think you're being over -sensitive here and missing the irony of the situtation,

[03:46:00.0000] <rburns>
othermaciej: It's not?

[03:46:01.0000] <othermaciej>
no, it's not

[03:47:00.0000] <rburns>
Lachy: OK, that's fine. It seems a bit strange to complain about the list volume, but then not want to do anything about it. My proposal would have helped organize emails into meaningful categories.

[03:47:01.0000] <rburns>
othermaciej: It's not?

[03:47:02.0000] <hsivonen>
rburns: taxonomies don't lower the volume

[03:48:00.0000] <hsivonen>
rburns: not suggesting taxonomies doesn't add to the volume

[03:48:01.0000] <Hixie>
the best solution to the list volume issue (if one believes it to be an issue, which i don't) is to reduce the number of e-mails sent to hte list, e.g. by not posting more than 2 or 3 a day

[03:48:02.0000] <rburns>
hsivonen: no, taxonomies do not lower the volume but they help manage the volume

[03:49:00.0000] <othermaciej>
rburns: people substantively disagree with you that your specific proposal is a good way to address the issue of the list volume; that's not the same as not wanting to do anything about it

[03:49:01.0000] <Hixie>
sending an e-mail discussing the process for categorising e-mail isn't a solution to the e-mail volume)
(

[03:50:00.0000] <rburns>
Othermaciej: that's fine to disagree. Though why continue a motion that hasn't even been seconded. It goes against the very concept of wanting to control list volume.

[03:50:01.0000] <othermaciej>
but in any case, telling them they shouldn't express their disagreement or alternate proposals because it would violate a rule that you proposed in that very email is out of line

[03:51:00.0000] <rburns>
Hixie: certainly it is. If the volume is high, and it's easy to separate announcements, from reviews, from motions, from general discussion (the bulk of the volume), then it's much easier to handle that volume.

[03:51:01.0000] <othermaciej>
it's like if I sent an email that says "I am now king of the HTML Working Group. I command that no one reply to this message, because I am king."

[03:51:02.0000] <othermaciej>
I hope if I did that, people would rightly see it as invalid and tell me so

[03:51:03.0000] <othermaciej>
and perhaps cart me off to the funny farm

[03:52:00.0000] <othermaciej>
but in any case I don't think the list has really been used for motions

[03:52:01.0000] <rburns>
Hixie: so that to me is a solution to the volume. It's like the volume of web content is a problem to deal with. But some folks thought of creating search engines to deal with that volume. The search engine is a solution to the volume of websites.  No need to starting DO attacks on servers instead.

[03:52:02.0000] <rburns>
Othermaciej: but you do see the irony don't you? :-)

[03:53:00.0000] <Lachy>
rburns, conventions evolve over time, they're not dictated by authorities. Trying to force people to adopt a specific convention is not the right approach

[03:53:01.0000] <Hixie>
rburns: the problem people have with high volume is that they can't read all the e-mail. categorising it doesn't help with that.

[03:53:02.0000] <Hixie>
rb we already have threading mail clients that do a quite good job of categorising e-mails on a per-topic basis.

[03:53:03.0000] <Hixie>
[24~rb the volume of websites is di

[03:53:04.0000] <Hixie>
fferent pbecause you're not expected to read all websites

[03:53:05.0000] <rburns>
othermarcij: if you sent me that email I would just laugh. I thought my reply would cause the same reaction.

[03:53:06.0000] <othermaciej>
rburns: I do see the irony of you sending email to suggest ways to reduce email volume when you are one of the top posters

[03:53:07.0000] <othermaciej>
rburns: and I see the further irony of you replying to people who reply to your message, to tell them their replies needlessly add to list volume

[03:54:00.0000] <rburns>
othermaciej: I see that irony too. So what?

[03:54:01.0000] <rburns>
othermaciej: does that make it impossible to see the other irony I pointed out?

[03:54:02.0000] <Hixie>
wow, my client is chopping up my text real bad. silly lag issues.

[03:55:00.0000] <Hixie>
othermaciej: he's not "one of" the top posted, jhe has the honour of being the top poster, by a factor of 2 more volume than the next person.

[03:55:01.0000] <hsivonen>
rburns: well, there's a way to lower volume that is in your disposal and that doesn't even require the cooperation of the group

[03:55:02.0000] <rburns>
Hixie: categorizing can help nonetheless. It can help decide how quickly to skim a message or even whether to skip it that isn't just based on the name it came from.

[03:55:03.0000] <Hixie>
i should point out that i have nothing against rob posting a lot

[03:56:00.0000] <Hixie>
i don't think we want working group members to be skipping any e-mail

[03:56:01.0000] <rburns>
hsivonen: except I'm not really interested in that way of lowering volume.

[03:56:02.0000] <Hixie>
that would not seem to be hte best way for working group members to stay up to speed on issues and f

[03:56:03.0000] <Hixie>
or us to get consensus on the spec

[03:57:00.0000] <rburns>
Hixie: I don't feel like we're getting consensus on the spec regardless of volume.

[03:57:01.0000] <Lachy>
I don't mind anyone posting a lot if their emails are actually productive

[03:57:02.0000] <rburns>
Lachy: that's why I suggested you shouldn't have replied.

[03:58:00.0000] <Lachy>
rburns, my email was a productive response to your suggestion pointing out existing conventions.  Yours telling me not to respond was not

[03:58:01.0000] <Hixie>
rburns: oh i agree, we're years from the spec being near ready enough to have any sort of significant consensus

[03:58:02.0000] <Hixie>
html5 is a multi-year effort

[03:58:03.0000] <jgraham>
Seriously, the best way of lowering volume is for people to think more about what they are replying to before they post. Lots of discussion has been cyclical patterns of messages with tiny tiny increments of understanding per message.

[03:58:04.0000] <Hixie>
we've been doing it since 2004, i don't expect it to be done before 2014 (the spec, that is, before even considering test suites)

[03:59:00.0000] <rburns>
Lachy: if the message I posted said it needed a second or no further discussion was necessary,, how could a response to it that didn't second the message possibly be productive. What end could it serve,. for the WG that is.

[03:59:01.0000] <Hixie>
(which iwll probably take at least 5 more years beyond that)

[03:59:02.0000] <rburns>
jgraham: I agree.

[04:02:00.0000] <rburns>
jgraham: thtat's also why I felt tagging subject lines would help. Similar to distilling things to the wiki, general discussion could take place that might just be about building understanding among the participants in those discussions. Those would be the message everyone could safely ignore if they were not interested in the subject 

[04:02:01.0000] <rburns>
jgraham: then as members reached some level of understanding of the topic, reviews, proposals or motions could be made. These would be the high-grade ore on the list. 

[04:03:00.0000] <jgraham>
rburns: But saying "don't read stuff you don't think is interesting" isn't a good solution. People should be able to follow the discussions

[04:04:00.0000] <jgraham>
However that requires participants in the discussions to think hard about the other point of view and why it makes sense to the other person before posting

[04:04:01.0000] <jgraham>
and to prefer one well-researched post over many hastily written ones

[04:04:02.0000] <othermaciej>
One possible thing to think about is, "how far removed is the point I am about to make from the deliverables of this group"

[04:05:00.0000] <jgraham>
and to try, as far as possible, to express themselves concisely

[04:05:01.0000] <jgraham>
/me is especially bad the concise thing

[04:05:02.0000] <othermaciej>
I have at times been guilty of replying to emails only to make a meta-meta-meta-point that no longer has much concrete relevance to the spec, our other planned publications, the test suite, or any other deliverables

[04:06:00.0000] <othermaciej>
but I am certainly not the only one who needs to count the metas before hitting send

[04:06:01.0000] <rburns>
jgraham: I think conciseness has been a problem leading to more volume. I've seen half sentence replies that trigger dozens of misconstrued responses.

[04:08:00.0000] <jgraham>
Of course there is balance. But nothing puts me off a message like seeing it is tens of kb of text, especially if the discussion isn't producing much of value

[04:08:01.0000] <rburns>
jgraham: however, by categorizing the messages, I'm not saying others should be allowed to join in and read the message or reply to them. I'm saying those without tags could be more safely ignored than the others.  And you can't tell me messages aren't being ignored anyway, because I see the responses to threads I'm not even necessarily involved in and I can tell the responder hasn't even read the thread. Categorizing would make the p

[04:08:02.0000] <hsivonen>
rburns: "Categorizing would make  the p

[04:08:03.0000] <hsivonen>
"

[04:08:04.0000] <hsivonen>
rburns: you got cut off again. (ironically :-)

[04:08:05.0000] <rburns>
othermaciej: I really don't see a lot of problems with out-of-scope emails. The scope HTML5 has set for itself is pretty far-reaching.

[04:10:00.0000] <rburns>
" Categorizing would make the process of filtering easier."

[04:11:00.0000] <rburns>
othermaciej: the most seirous volume problems I've seen is members going round and round not understanding what other members are saying. and each time incrementing the number of emails.

[04:12:00.0000] <othermaciej>
I agree that arguing in circles is a problem

[04:12:01.0000] <rburns>
othermaciej: I've intervened from time to time and illuminated the situation and the thread ends, because the participants see what the disagreement was about.

[04:12:02.0000] <othermaciej>
but I don't think labelling such discussions with a one-letter code will in any way help

[04:12:03.0000] <rburns>
othermaciej: that's fine. I can see the drawbacks to that too.

[04:13:00.0000] <othermaciej>
two things I'm going to try to help with that are: (a) read the whole thread before replying; and (b) if I want to reply to many messages, try to consolidate my reply

[04:13:01.0000] <othermaciej>
those are in general good practices to avoid adding too much noise to a discussion

[04:13:02.0000] <rburns>
othermaciej:  I'm not bothered if no one agrees with the proposal. I was simply taking feedback and remarks I'd seen and trying to formulate a solution

[04:14:00.0000] <rburns>
othermaciej: that sounds like good advice. I try to do the same (particularly the combining part).

[04:27:00.0000] <rburns>
annevk: all but 5 character references have to be handled by the next level above XML. XML is unaware of any other entity references.

[04:27:01.0000] <Hixie>
http://whatwg.org/issues/ now has a companion http://whatwg.org/issues/top

[04:27:02.0000] <Hixie>
i'll mail the list about them tomorrow

[04:28:00.0000] <rburns>
s/character references/entity references

[04:28:01.0000] <Hixie>
now i must sleep

[04:28:02.0000] <hsivonen>
rburns: not correct. entity references are expanded on the XML except when a non-validating XML processor has skipped an external entity

[04:29:00.0000] <hsivonen>
s/on the XML/on the XML layer/

[04:29:01.0000] <rburns>
hsivonen: Wouldn't these be non-validating XML processors?

[04:30:00.0000] <hsivonen>
rburns: what's the HTML 5 spec to say what mode to run the XML processor in?

[04:31:00.0000] <hsivonen>
rburns: though in practice, in sane browser implementations, the XML processor is non-validating

[04:31:01.0000] <rburns>
hsivonen: I guess I'm saying that IF the XHTML5 processor is acting as a non-validating XML processor then...

[04:31:02.0000] <hsivonen>
rburns: there's also the other condition: the non-validating processor has to skip an external entity for the loop hole to kick in

[04:32:00.0000] <hsivonen>
rburns: generally, XML processors cannot be configured to skip external entities conditionally

[04:32:01.0000] <rburns>
hsivonen: it's not a  big deal, but I feel like when I do encounter draconian error-handling its often  because of an over-zealoous exclusion of entity references

[04:32:02.0000] <hsivonen>
(implementation limitation)

[04:33:00.0000] <rburns>
hsivonen: Also  I wasn't suggesting no error, I was just suggestion a non-fatal-error

[04:33:01.0000] <hsivonen>
rburns: like I explained before, the loophole does not kick in in expat in Gecko, because as far as expat is concerned, it has not skipped anything

[04:33:02.0000] <rburns>
hsivonen: like a big black cockroach or something where the entity should be :-)

[04:34:00.0000] <hsivonen>
rburns: what you are asking is rigging both browsers and content is a very specific way in order to exploit a loophole in the XML spec

[04:34:01.0000] <rburns>
hsivonen: expat?

[04:34:02.0000] <hsivonen>
rburns: the XML parser

[04:34:03.0000] <rburns>
hsivonen: oh

[04:35:00.0000] <rburns>
hsivonen: OK. thanks for the info (I didn't know they named their xml parser) :-).

[04:35:01.0000] <hsivonen>
rburns: now, the question is, why do you want to use XML, if you don't like the way XML is to the point of wanting to standardize a particular loop hole exploitation?

[04:36:00.0000] <hsivonen>
and the way XML is is that the speccers of XML gave us 5 reliable entities and UTF-8

[04:36:01.0000] <rburns>
hsivonen: I just think the draconian part is taken too far there with the entity references. Most of the other fatal-errors make sense from the XML perspective. They have wider implications for the whole document. These unknown character references don't seem to so much.

[04:37:00.0000] <takkaria>
but you should be using Unicode anyway

[04:37:01.0000] <rburns>
hsivonen: perhaps most of the problem is being fixed anyway. It wasn't long ago when the major browsers would choke even on known HTML entity references when processing XHTML

[04:37:02.0000] <hsivonen>
rburns: I think that the situation sucks too, but the thing is that the value of XML is in XML parsers (however configured!) being ubiquitous. for this reason, XML has to be what it is and we don't guess to second guess

[04:38:00.0000] <rburns>
hsivonen: Now they seem to be repairing those bugs for the most part. My feeling was just why choke on any well-formed though invalid entity reference.

[04:39:00.0000] <rburns>
hsivonen: well we do second-guess a lot of other standards tat I think have wider problems than well-formed though invalid entity references in XML (authoritative content-type headers for example).

[04:39:01.0000] <hsivonen>
rburns: they weren't bugs to begin with and the issue has not been "fixed". there's just an ugly workaround for a handful of common cases

[04:40:00.0000] <hsivonen>
rburns: yes we do, but second-guessing XML has been implicitly out-of-bounds

[04:40:01.0000] <rburns>
hsivonen: I'm not sure what you  mean. The XHTML 1.0 recommendation includes a complete list of HTML entity references.

[04:40:02.0000] <hsivonen>
XML 1.0 is kinda sacred

[04:40:03.0000] <rburns>
hsivonen: Yeah I understand. It's just HOW sacred.

[04:41:00.0000] <hsivonen>
rburns: well, second-guessing it defeats the point of using XML

[04:41:01.0000] <hsivonen>
rburns: since the rationale for using XML is to leverage the network effects of the installed base of XML tooling

[04:42:00.0000] <rburns>
hsivonen: I don't think so. Not in these few isolated areas. As long as their are not wider-reaching implication (which it is sometimes hard to determine).

[04:43:00.0000] <hsivonen>
rburns: what was "I don't think so" in reference to? to the network effects?

[04:44:00.0000] <hsivonen>
rburns: don't you consider micromanaging the parser configuration of every XHTML5 consuming XML parser far reaching?

[04:44:01.0000] <rburns>
hsivonen:  no that was in reference to defeating the purpose (which I guess than you added to it about the xml tooling)

[04:44:02.0000] <rburns>
hsivonen: but to implement any namespace in XML without retrieving schema, ,an application has to do something else with the XML just to be able to handle entity references. So it can't just be using the purse XML stuff.

[04:44:03.0000] <hsivonen>
rburns: what's your purpose for using XML?

[04:45:00.0000] <hsivonen>
purse?

[04:45:01.0000] <othermaciej>
I assume he meant "pure"

[04:45:02.0000] <hsivonen>
oh

[04:46:00.0000] <hsivonen>
rburns: the XML processor is a sacred black box. you get to write app code outside it but you don't get to hack its internals

[04:46:01.0000] <rburns>
hsivonen: To me, extensibility.  common parsing of elements. separating well-formedness from validity. The fatal error handling has some benefits, I'm just not quite sure how they relate to entity references and these more isolated parts of the parsing behavior.

[04:46:02.0000] <rburns>
othermaciej: yes "pure"

[04:47:00.0000] <hsivonen>
rburns: they do relate and we don't get to pick and choose which parts of XML we like

[04:47:01.0000] <rburns>
hsivonen: Like I said before. I think it should be an error. It could even be an obvious ugly error on an XHTML page. I'm just not quite sure what making it a fatal error does for anyone. (in the entity reference case)

[04:47:02.0000] <othermaciej>
as I understand it, the main reason to have an XML serialization is to interoperate with all of the deployed XML stuff that's out there

[04:47:03.0000] <hsivonen>
rburns: cause if you hack the internals of the XML processor, it ain't an XML processor anymore

[04:48:00.0000] <othermaciej>
having an almost-XML serialization would not really have the same benefits

[04:48:01.0000] <hsivonen>
othermaciej: exactly

[04:48:02.0000] <hsivonen>
othermaciej: hence "network effects" above

[04:48:03.0000] <rburns>
hsivonen: it would to me. Even more benefits I think.

[04:49:00.0000] <othermaciej>
at which point, you may as well have only the HTML serialization, since it already works in clients and with well-defined parsing rules is feasible to process with tree-based tools (instead of just with text processing tools)

[04:50:00.0000] <othermaciej>
XML does also have the benefit (some would say) of namespaces

[04:50:01.0000] <rburns>
othermaciej: I don't mean an almost-xml serialization. I mean an all xml serialization that is processed without fatal errors for entity references. In other words the HTML XML application would define an anyname entity reference mapped to the Unicode replacement character. 

[04:50:02.0000] <othermaciej>
but grafting namespaces onto HTML is a better way to get namespaces + looser parsing than loosening XML parsing rules

[04:51:00.0000] <rburns>
tothermaciej: but were already going to have XML and namespaces, whatever we do with namespaces for the text/html serialization.

[04:51:01.0000] <rburns>
othermaciej: adding namespaces to the text/html serialization sounds scary to me.

[04:51:02.0000] <rburns>
othermaciej: though I think it would be great if we could pull it off.

[04:53:00.0000] <rburns>
I know other schemas have used an anyAttributeName in their schemas. This doesn't seem that different. It's just an NCName for any entity reference rather than any attribute name.

[04:54:00.0000] <rburns>
s/NCName/NCNames (as in the infinity of them)

[06:36:00.0000] <Lachy>
anyone who speaks French in here? What would be an appropriate post slug for the article "Présentation: comment HTML 5 peut-il être utilisé aujourd'hui ?"

[06:37:00.0000] <Lachy>
(the post slug is the path in the URL on the blog, e.g. in http://blog.whatwg.org/html-contre-xhtml the post slug is "html-contre-xhtml")

[06:37:01.0000] <Lachy>
hsivonen?

[06:38:00.0000] <hsivonen>
html5-aujourd-hui perhaps

[06:38:01.0000] <Lachy>
thanks, that'll do

[06:38:02.0000] <Lachy>
http://blog.whatwg.org/html5-aujourd-hui

[08:40:00.0000] <Lachy>
/me hopes his latest post to public-html raises the bar for quality on public-html, and that others follow the example for their own proposals

[08:40:01.0000] <gsnedders>
Lachy: quality? who needs no stinkin' quality!?

[08:41:00.0000] <Lachy>
gsnedders, well, we could continue with the current state of unproductive discussion if you like?

[08:41:01.0000] <gsnedders>
Lachy: I doubt whether quality posts will help

[08:41:02.0000] <Lachy>
I have my doubts too, but it never hurts to have a little hope

[08:41:03.0000] <Whiskey_M>
ahhh, that is what is meant by placeholder ;-)

[12:49:00.0000] <aroben>
Does anyone know of a way to determine whether a document was interpreted as HTML or XHTML from within the document itself?

[12:51:00.0000] <aroben>
Rather, a way to do so in an HTML5-compliant user agent

[12:52:00.0000] <annevk>
isHtml = document.documentElement.tagName == "HTML" ? true : false

[12:53:00.0000] <aroben>
annevk: thanks!

[12:53:01.0000] <annevk>
yw

[12:55:00.0000] <Philip`>
<html xmlns="http://www.w3.org/1999/xhtml"><body onload="alert('This is '+(document.getElementById('this-is-html')?'':'X')+'HTML')"><script>document.write('\x3cp id=this-is-html>\x3c/p>')</script></body></html>

[13:06:00.0000] <Hixie>
Lachy: i specifically didn't add a vote button to the "top" page so that we wouldn't have issues get skewed votes

[13:06:01.0000] <Lachy>
ok

[13:08:00.0000] <Hixie>
and yeah, i don't do any encoding converstion

[13:09:00.0000] <Hixie>
if you know of a perl library that would do that for me let me know, 'll plug it in

[13:09:01.0000] <Hixie>
http://www.cssglobe.com/post.asp?id=863: "A cross-browser implementation that has been adopted by the W3C HTML Working Group of the WHATWG Web Forms 2.0 specification."

[13:12:00.0000] <Lachy>
aroben, the more accurate test would be

[13:12:01.0000] <Lachy>
 var isHTML = (function() {

[13:12:02.0000] <Lachy>
  var htmlElement = document.getElementsByTagName("html").item(0);

[13:12:03.0000] <Lachy>
  if (htmlElement) {

[13:12:04.0000] <Lachy>
   return (htmlElement.tagName == "HTML");

[13:12:05.0000] <Lachy>
  } else {

[13:12:06.0000] <Lachy>
   return false;

[13:12:07.0000] <Lachy>
  }

[13:12:08.0000] <Lachy>
 })();

[13:12:09.0000] <Lachy>
That handles obscure XML cases like <HTML xmlns="http://example.org/NotHTML">

[13:14:00.0000] <Lachy>
Hixie, there's probably email clients that are written in perl that would use libraries like that, though I don't know of any.

[13:14:01.0000] <annevk>
or simply

[13:14:02.0000] <annevk>
isHtml = document.documentElement.tagName == "HTML" && document.documentElement.namespaceURI = "http://www.w3.org/1999/xhtml" ? 1 : 0

[13:15:00.0000] <Lachy>
that will work for HTML5, but not current implementations, especially IE

[13:15:01.0000] <Hixie>
isHTML = document.createElement('html').tagName == 'HTML';

[13:16:00.0000] <Lachy>
Hixie wins! :-)

[13:17:00.0000] <annevk>
"Rather, a way to do so in an HTML5-compliant user agent"

[13:17:01.0000] <annevk>
fwiw

[13:17:02.0000] <Philip`>
isHTML = true; and then you'll be right 99.99% of the time

[13:19:00.0000] <annevk>
you can now DOS-attack the (X)HTML5 tracker easier if you know the source code

[13:21:00.0000] <annevk>
this is considered to be a feature, until it's abused so heavily it will be removed :)

[13:21:01.0000] <Lachy>
annevk, can I have a copy of the source code?

[13:23:00.0000] <annevk>
sure: http://html5.googlecode.com/svn/trunk/web-apps-tracker/web-apps-tracker

[13:24:00.0000] <annevk>
the bit with form["limit"] is relevant

[13:24:01.0000] <Lachy>
annevk, there's a bug in Firefox.  because you have <input ... value= name=from>, it displays name=from in the text field

[13:25:00.0000] <annevk>
hmm

[13:25:01.0000] <annevk>
that's silly

[13:25:02.0000] <Lachy>
you need to add quotes to value="" when there's no value, or omit the attribute

[13:25:03.0000] <annevk>
that seems rather obvious, yes

[13:25:04.0000] <Philip`>
Why parse the SVN log text rather than using "svn log --xml"?

[13:26:00.0000] <annevk>
not sure if dreamhost supports --xml, it might though

[13:27:00.0000] <annevk>
Lachy, that parsing behavior seems consistent with IE7

[13:27:01.0000] <annevk>
Lachy, it seems that after the equal sign whitespace is ignored

[13:27:02.0000] <Lachy>
yes. I believe it's also what HTML5 has specced

[13:28:00.0000] <annevk>
and what every browser implements

[13:28:01.0000] <annevk>
fixed...

[13:29:00.0000] <annevk>
Philip`, you're welcome to provide enhancements

[14:46:00.0000] <gsnedders>
kingryan: the htmlOrFeed.json isn't being parsed by PHP. Can't work out why, though :\

[14:46:01.0000] <kingryan>
which json parser are you using?

[14:46:02.0000] <gsnedders>
PHP 5.2's own

[14:46:03.0000] <kingryan>
hmm

[14:50:00.0000] <gsnedders>
kingryan: line 22

[14:50:01.0000] <kingryan>
what about it?

[14:51:00.0000] <gsnedders>
kingryan: it seems to be one of several lines causing it to fail

[14:51:01.0000] <gsnedders>
kingryan: it's the \digits that is causing it

[14:52:00.0000] <kingryan>
oh, whoops

[14:52:01.0000] <kingryan>
yeah, that's not valid JSON

[14:52:02.0000] <gsnedders>
PHP's parsing is very draconian

[14:53:00.0000] <gsnedders>
kingryan: you want me to fix that in SVN?

[14:53:01.0000] <kingryan>
I'll fix it now

[14:54:00.0000] <kingryan>
I see only 2 occurences of that

[14:54:01.0000] <kingryan>
are you getting more than 2 errors?

[14:54:02.0000] <gsnedders>
kingryan: lines 33 and lines 22

[14:54:03.0000] <kingryan>
got those

[14:55:00.0000] <gsnedders>
just those two lines

[14:56:00.0000] <kingryan>
alright, fix checked in

[14:57:00.0000] <gsnedders>
right. my paranoia about now properly using my untested PHP implementation was right

[15:04:00.0000] <gsnedders>
http://geoffers.no-ip.com/hg/foleeo?f=89f835292a7b;file=sniffed.php;style=gitweb is my implementation, BTW. still fails various tests, though

[15:07:00.0000] <gsnedders>
well, that revision's child passes all of the tests, now

[15:09:00.0000] <kingryan>
gsnedders: cool

[15:09:01.0000] <kingryan>
you have tests for binary?

[15:39:00.0000] <gsnedders>
kingryan: no, I mean those tests

[15:39:01.0000] <kingryan>
ah, gotcha

[15:39:02.0000] <gsnedders>
kingryan: if I had tests, they'd be in SVN

[15:40:00.0000] <gsnedders>
/me sighs. school tomorrow.


2007-08-21
[18:05:00.0000] <kingryan>
it seems that the sniffing section doesn't mention BOMs

[18:05:01.0000] <kingryan>
I wonder if that's on purpose

[18:07:00.0000] <kingryan>
Hixie ?

[18:40:00.0000] <Hixie>
kingryan: here now

[18:40:01.0000] <kingryan>
I was wondering about BOMs in the html vs feed sniffing

[18:41:00.0000] <kingryan>
the other sniffing subsections mentoin BOMs, but not that one. is it for a reason?

[18:41:01.0000] <Hixie>
/me looks

[18:42:00.0000] <kingryan>
I also sent email about this a minute ago

[18:42:01.0000] <kingryan>
http://www.whatwg.org/specs/web-apps/current-work/#content-type3 is the relevant section

[18:44:00.0000] <Hixie>
oh you mean why doesn't it skip over a UTF-8 BOM if it sees one at the start?

[18:45:00.0000] <Hixie>
i suppose i could add a step between steps 4 and 5 that skips a BOM if there is one

[18:45:01.0000] <kingryan>
yes, that's what I mean

[18:46:00.0000] <kingryan>
it'd be useful to add a step there. I implemented that section very literally, then it failed on test cases that had BOMs

[18:47:00.0000] <kingryan>
I was able to figure out the problem quickly, but it'd be nice if the spec was explicit about that

[18:50:00.0000] <Hixie>
kingryan: k

[18:51:00.0000] <kingryan>
thanks

[18:51:01.0000] <Hixie>
kingryan: i've saved your e-mail to my parsing/sniffing pile

[18:51:02.0000] <Hixie>
hopefully it was clear enough that i'll understand what it means when i look at it :-)

[18:51:03.0000] <kingryan>
I hope so too

[23:17:00.0000] <Lachy>
Is there anything Rob won't object to?https://bugzilla.mozilla.org/show_bug.cgi?id=189643#c7

[23:17:01.0000] <othermaciej>
impressive

[23:19:00.0000] <karlUshi>
Lachy: given that people banned +1 on the list :) I guess yes when people expressed themselves is usually to disagree. Then after it's always difficult to consider how much energy and bandwidth do we have when we object.

[23:19:01.0000] <karlUshi>
Sometimes I don't fight for some issues because I just do not have time for it

[23:19:02.0000] <karlUshi>
and I expect someone else will do it later on.

[23:20:00.0000] <Hixie>
/me closes the aforementioned bug

[23:21:00.0000] <Hixie>
and as qa contact, it's my responsibility to do so!

[23:21:01.0000] <othermaciej>
conflict of interest!

[23:21:02.0000] <karlUshi>
heheh

[23:22:00.0000] <Lachy>
I also filed https://bugzilla.mozilla.org/show_bug.cgi?id=392994 (Rob hasn't objected there yet)

[23:23:00.0000] <Hixie>
he probably didn't know about it before you posted it on irc just now

[23:24:00.0000] <Lachy>
yeah, I know, but I figured there would be others in here who would be interested in it despite that

[23:25:00.0000] <Hixie>
lol, he just commented

[23:26:00.0000] <molly>
:: shakes off whatever logic might have been stuck to her ::

[23:26:01.0000] <molly>
Hi everyone!

[23:26:02.0000] <Hixie>
hey molly

[23:26:03.0000] <molly>
Ian, how are you?

[23:26:04.0000] <molly>
nice to see you

[23:28:00.0000] <molly>
who else is around?

[23:28:01.0000] <molly>
I never know, I'm so time-zone impaired

[23:29:00.0000] <molly>
Lachy?

[23:29:01.0000] <karlUshi>
hi molly

[23:29:02.0000] <molly>
Hiya

[23:29:03.0000] <Lachy>
molly?

[23:29:04.0000] <Hixie>
it's pretty much active 24h/day around here

[23:29:05.0000] <molly>
Hi Lachy!

[23:29:06.0000] <molly>
just have to poke 'em ;-)

[23:29:07.0000] <karlUshi>
2:37pm here

[23:29:08.0000] <molly>
10:38 Monday night here

[23:29:09.0000] <molly>
i have ice cream! Anyone want ice cream?

[23:30:00.0000] <othermaciej>
hello Molly

[23:30:01.0000] <karlUshi>
with the 35deg C out there that would be welcome

[23:30:02.0000] <molly>
hey Maciej (I just cannot call you othermaciej, you are my only maciej)

[23:30:03.0000] <karlUshi>
ooops 32ºC only today http://weather.yahoo.co.jp/weather/jp/14/4610/14205/2520816.html

[23:30:04.0000] <molly>
hey, I'll up you on that karlUshi, it's hotter here

[23:30:05.0000] <Lachy>
/me wants ice cream

[23:31:00.0000] <karlUshi>
maybe at the TPAC molly ;)

[23:31:01.0000] <karlUshi>
in the so hoooot Boston

[23:31:02.0000] <molly>
I have three flavors: Triple Chocolate, Butter Pecan, and Mint Chocolate Chip

[23:31:03.0000] <molly>
Hey KarlUshi, i'll be there

[23:31:04.0000] <molly>
awesome!

[23:32:00.0000] <molly>
I like getting to talk to folks F2f and meet up

[23:32:01.0000] <karlUshi>
good luck molly, you know there are these folks who only talk through IRC :) bring your computer. F2F by IRC :p (half joking)

[23:33:00.0000] <karlUshi>
Though last time at Xtech in Paris, it was quite lively

[23:33:01.0000] <molly>
karlUshi, I'm a girl. We're known for our multitasking skills!

[23:33:02.0000] <Hixie>
aw man, rob burns is starting to send his e-mails twice under different subject lines

[23:33:03.0000] <molly>
I liked it. It was a beginning.

[23:33:04.0000] <Hixie>
/me isn't going to be able to keep up with the e-mail flood if people start sending dupes

[23:34:00.0000] <molly>
@hixie I don't know how you do now

[23:34:01.0000] <molly>
the single biggest barrier to entry

[23:34:02.0000] <molly>
too much stuff to read.

[23:35:00.0000] <hober>
I've managed to stay on top of the list, but that doesn't leave much time to actually write posts to it. :)

[23:35:01.0000] <Hixie>
molly: i read pretty much all the mail on the list at the moment

[23:36:00.0000] <Lachy>
/me updates Selectors-API based based on Working Group vote and hopes the naming issue is resolved for good! http://dev.w3.org/cvsweb/~checkout~/2006/webapi/selectors-api/Overview.html?content-type=text/html;%20charset=utf-8

[23:36:01.0000] <molly>
Hixie, that's so cool. You know I'm in awe.

[23:36:02.0000] <molly>
Truly, truly.

[23:36:03.0000] <karlUshi>
Too much too read? :)))) un chef d'oeuvre - http://en.wikipedia.org/wiki/La_Com%A9die_humaine

[23:36:04.0000] <molly>
needs to be even more simplistic

[23:36:05.0000] <molly>
sad but true

[23:36:06.0000] <molly>
I'm serious

[23:37:00.0000] <molly>
that's why the red-mark idea is so brilliant

[23:37:01.0000] <molly>
what do you think Hixie?

[23:37:02.0000] <molly>
I thought that was awesome, and I want to work with that

[23:38:00.0000] <Hixie>
which idea? marking the big known issues in the draft?

[23:38:01.0000] <othermaciej>
Lachy: I just thought of a great new name!

[23:38:02.0000] <othermaciej>
/me ducks

[23:38:03.0000] <molly>
Hixie: yes, please

[23:39:00.0000] <molly>
othermaciej: for what

[23:39:01.0000] <molly>
and you best duck for Lachy, he's really tall. I've met him!

[23:39:02.0000] <Lachy>
othermaciej, is it goAndGetMeStuff() and goAndGetMeAllStuff()?

[23:39:03.0000] <molly>
;-)

[23:39:04.0000] <karlUshi>
for the selectors API doc

[23:39:05.0000] <othermaciej>
Molly: the Selectors API naming debate has been a watershed for bike sheds

[23:39:06.0000] <karlUshi>
sudoGoAndGetMeStuff() ?

[23:39:07.0000] <othermaciej>
Lachy: wow, those are even better

[23:40:00.0000] <Hixie>
molly: that's already done

[23:41:00.0000] <molly>
So Hixie, All - if we take those one by one, can we do a good, easy explanation for each "hot topic"?

[23:41:01.0000] <molly>
also, I need the best description of WHY HTML 5 is necessary. I have a great quote from Lachy, and another from Dean Edwards

[23:41:02.0000] <Lachy>
what quote from me?

[23:41:03.0000] <molly>
but, not completely convincing to the complacent world at large

[23:42:00.0000] <molly>
damn, he would ask, hang on ;-)

[23:42:01.0000] <hober>
I liked Hixie's explanation along the lines of "very long-term survival of our cultural output"

[23:42:02.0000] <othermaciej>
if we've upgraded the web to 2.0, we should really bump the version number of the standards too

[23:42:03.0000] <molly>
hober, I love that too, but it isn't explicit. It's the way I might describe something and people call me weird when I do it

[23:43:00.0000] <Hixie>
molly: http://blog.whatwg.org/faq covers most of those issues, you might want to crib from there

[23:43:01.0000] <molly>
they just call Hixie really smart when he does it :P

[23:43:02.0000] <Hixie>
people generally call me an asshole, actually :-P

[23:43:03.0000] <molly>
that's because they haven't met you

[23:43:04.0000] <othermaciej>
a little from column A, a little from column B

[23:44:00.0000] <molly>
lol!

[23:44:01.0000] <Hixie>
one big reason for html5 is that the languages that the web are built are are showing their age, and are facing competition from proprietary technologies like flash and silverlight

[23:45:00.0000] <molly>
I think we need to distill it in real ways like "this is why you would use this element and how"

[23:45:01.0000] <molly>
and also tell that to browser implementors as well as tools folks

[23:45:02.0000] <Hixie>
we need to bring the web's open, standard, vendor-neutral technologies up to the state of the art if we are to keep the web's openness safe from vendor lockin

[23:45:03.0000] <molly>
right now, I think people are just sort of like, okay, let's implement css 2.1

[23:45:04.0000] <molly>
:: bangs head on desk ::

[23:45:05.0000] <molly>
I mean, look at iPhone

[23:45:06.0000] <othermaciej>
what about it?

[23:46:00.0000] <molly>
beautiful device

[23:46:01.0000] <othermaciej>
I think it has a pretty decent implementation of css 2.1

[23:46:02.0000] <molly>
how about just making sure it also implements handheld css

[23:46:03.0000] <karlUshi>
http://blog.wired.com/monkeybites/2007/08/the-iphone-is-i.html

[23:46:04.0000] <karlUshi>
"The IPhone Is Internet Explorer 4 All Over Again"

[23:46:05.0000] <othermaciej>
we don't claim to be the handheld media type because in the cases where it makes a difference, that leads to a worse experience for the user

[23:47:00.0000] <karlUshi>
http://www.w3.org/QA/2007/08/iphone_developer_guidelines_pr.html

[23:47:01.0000] <molly>
i actually would challenge that. I just saw an "iPhone enhanced" experience which meant really using webcore

[23:47:02.0000] <karlUshi>
"iPhone Developer Guidelines Promote One Web, Open Standards"

[23:48:00.0000] <karlUshi>
molly: this is the point of the first article

[23:48:01.0000] <molly>
i'm so cool with awesome stuff. I mean, blow my mind. Always

[23:48:02.0000] <molly>
BUT

[23:48:03.0000] <karlUshi>
to have this iPhone enhanced

[23:48:04.0000] <molly>
do it with the baseline there

[23:48:05.0000] <molly>
My brother Linus always told me I was going to end up the Ralph Nader of the Web

[23:48:06.0000] <othermaciej>
we encourage standards-based design first and foremost

[23:48:07.0000] <molly>
that was a decade ago

[23:48:08.0000] <molly>
sigh.

[23:49:00.0000] <molly>
wait, you guys don't get it

[23:49:01.0000] <molly>
I have nothing against WHAT WG

[23:49:02.0000] <molly>
never have!

[23:49:03.0000] <karlUshi>
"The first design rules for web applications on iPhone are to stick with web standards and follow established web design practices." -- http://developer.apple.com/iphone/designingcontent.html

[23:49:04.0000] <molly>
I love the entire idea and exercise

[23:49:05.0000] <othermaciej>
iphone-optimized content generally works in other browsers

[23:49:06.0000] <molly>
I'm just not fucking certain the W3C actually  HELPS YOU

[23:49:07.0000] <othermaciej>
there are sadly some sites that lock non-iphone browsers out of the iphone version

[23:49:08.0000] <molly>
or vice versa

[23:49:09.0000] <karlUshi>
ohoh

[23:49:10.0000] <karlUshi>
:)))

[23:50:00.0000] <molly>
what, patents?

[23:50:01.0000] <karlUshi>
schizophrenia ahead

[23:50:02.0000] <molly>
why do you say that KarlUshi?

[23:51:00.0000] <karlUshi>
because many people here are part of the two :) 

[23:51:01.0000] <molly>
so what

[23:51:02.0000] <molly>
one could say I am, sorta

[23:51:03.0000] <karlUshi>
[14:58]  <molly> I'm just not fucking certain the W3C actually  HELPS YOU

[23:51:04.0000] <karlUshi>
:)

[23:51:05.0000] <molly>
KarlUshi, that's fair

[23:51:06.0000] <molly>
I can't argue that at all, and I won't

[23:52:00.0000] <molly>
I wish it could though.

[23:52:01.0000] <molly>
I always, perhaps innocently

[23:52:02.0000] <molly>
looked at the W3C as the wisdom

[23:52:03.0000] <karlUshi>
for me it's like your left brain saying "be careful, your right brain will not help you" 

[23:52:04.0000] <karlUshi>
eeeek

[23:52:05.0000] <molly>
my brain isn't that clearly split. It's kinda mixed ;-)

[23:52:06.0000] <karlUshi>
wisdom?

[23:53:00.0000] <molly>
I've been in IT since 1988 and the web since 1992.5.  Weird

[23:53:01.0000] <karlUshi>
why people create always religion of organizations :) it is a lot more basic than that. 

[23:53:02.0000] <molly>
I like to think maybe my longevity

[23:53:03.0000] <molly>
has given me *some*

[23:53:04.0000] <molly>
maybe not a lot

[23:53:05.0000] <molly>
but *some* wisdom? Anyone?

[23:53:06.0000] <othermaciej>
given that the w3c wanted to have an effort to make a new official version of HTML, it seems to make sense to work together

[23:53:07.0000] <molly>
Help me out here.

[23:53:08.0000] <othermaciej>
benefits of working with w3c:

[23:53:09.0000] <karlUshi>
othermaciej: yes. That's the point. 

[23:53:10.0000] <othermaciej>
- clear patent policy

[23:54:00.0000] <othermaciej>
- brand value of having the premier web standards body support the spec

[23:54:01.0000] <othermaciej>
- potential for companies like Microsoft to participate

[23:54:02.0000] <othermaciej>
- avoidance of needlessly fragmenting the marketplace

[23:55:00.0000] <karlUshi>
(wider reviews by wider audience)

[23:55:01.0000] <molly>
I think we have our main headers for our first WHY HTML5 article for the public ;-)

[23:55:02.0000] <molly>
seriously, that's awesome

[23:55:03.0000] <molly>
broadly

[23:56:00.0000] <molly>
then there are specifics

[23:56:01.0000] <molly>
the accessibility issue, Ruby? That's cool but how many people have really adopted XHTML 1.1 to support RUBY? c'mon.

[23:56:02.0000] <Hixie>
the only two reasons we considered going to the w3c was to get the pateny policy easily and to get microsoft on board (which was actually the same thing, they wanted whatwg to have a patent policy)

[23:56:03.0000] <Hixie>
the w3c brand actually didn't matter at all

[23:57:00.0000] <molly>
I know. I speak to Chris a lot due to my work at MS (which I have to say is surprisingly awesome) 

[23:57:01.0000] <molly>
but he is clearly upset about IP 

[23:57:02.0000] <molly>
I have opposed him as WG co-lead anyway

[23:57:03.0000] <molly>
long ago and far away

[23:57:04.0000] <karlUshi>
/me is smiling at illusions

[23:57:05.0000] <molly>
he's my friend, and my colleague, and certainly smart enough and more knowledgeable than most of us here

[23:58:00.0000] <molly>
but WTF?

[23:58:01.0000] <molly>
am I nuts (okay, yes, I am but.)

[23:59:00.0000] <karlUshi>
you are just molly (c)

[23:59:01.0000] <othermaciej>
in its current incomplete state I would summarize the benefits as:

[23:59:02.0000] <othermaciej>
- better cross-browser compatibility (since things are more explicitly defined)

[23:59:03.0000] <othermaciej>
- better support for the semantics of modern web documents like weblogs and wikis

[00:00:00.0000] <othermaciej>
- better support for rich web applications

[00:00:01.0000] <othermaciej>
- better built-in support for non-text media

[00:01:00.0000] <othermaciej>
of course that kind of shows my personal biases of what I care about

[00:01:01.0000] <othermaciej>
others with different priorities might give a different summary

[00:01:02.0000] <molly>
anyone ever have a great experience where everything slows down and there's a great soundtrack? And everything you ever dreamed of, well, it's right there

[00:02:00.0000] <othermaciej>
that seems inappropriate to discuss on a family channel

[00:02:01.0000] <molly>
tangent:

[00:02:02.0000] <molly>
I'm collecting stats

[00:02:03.0000] <molly>
the W3C staff notes 52 males to 15 females

[00:02:04.0000] <molly>
WHAT WG?

[00:02:05.0000] <molly>
any stats or assumed stats?

[00:03:00.0000] <othermaciej>
it doesn't have any staff

[00:03:01.0000] <Hixie>
there's no WHATWG staff

[00:03:02.0000] <molly>
no the W3C

[00:03:03.0000] <Hixie>
so we have a perfect balance of male and female

[00:03:04.0000] <othermaciej>
so 0 to 0

[00:03:05.0000] <molly>
who they pay. I also want to collect info as to WGs and alternative groups such as WHAT WG and microformats.org

[00:03:06.0000] <molly>
etc.

[00:04:00.0000] <molly>
Hixie, what a spin, wheeeee......

[00:04:01.0000] <molly>
I love that!

[00:04:02.0000] <molly>
How'd he do that guys?

[00:05:00.0000] <molly>
honestly, how many girls stop by here

[00:05:01.0000] <molly>
day to day

[00:05:02.0000] <Hixie>
no idea

[00:05:03.0000] <molly>
include me, even if I'm no longer a girl

[00:05:04.0000] <molly>
okay, who PARTICIPATE

[00:05:05.0000] <Hixie>
/me doesn't actually know the genders of most of the people here to be honest

[00:05:06.0000] <molly>
Hixie, that's okay. I think you scare most people anyway.

[00:06:00.0000] <karlUshi>
molly: we don't ask for genders when people participate in the groups

[00:06:01.0000] <othermaciej>
there's really no way to estimate, except to look at the mailing list and guess based on name

[00:06:02.0000] <Hixie>
i actively try to not know the affiliations and personal backgrounds of most contributors, as i don't think it affects the value of their input

[00:06:03.0000] <molly>
okay, just doing my stats homework

[00:06:04.0000] <molly>
I will say just that

[00:06:05.0000] <karlUshi>
as the staff contact of the HTML WG and approving all applications, I have not counted but that would mean I know all type of names

[00:06:06.0000] <molly>
I wasn't trying to be gender aggressive, I really prefer men anyway

[00:06:07.0000] <karlUshi>
the most common example is

[00:06:08.0000] <molly>
it's just that I was asked to give some stats

[00:06:09.0000] <karlUshi>
anne

[00:06:10.0000] <karlUshi>
girl or boy ;)

[00:07:00.0000] <karlUshi>
but molly you can try to guess here

[00:07:01.0000] <karlUshi>
v

[00:07:02.0000] <karlUshi>
http://www.w3.org/2000/09/dbwg/details?group=40318&public=1

[00:07:03.0000] <molly>
oh well, I met Anne when he still has spots. They cleared up last time I saw him though. He's clearly male. And a beautiful brilliant boy at that

[00:07:04.0000] <molly>
don't mistake me for someone who just shows up to cause trouble

[00:08:00.0000] <molly>
I really only show up

[00:08:01.0000] <molly>
to cause trouble ;-)

[00:08:02.0000] <molly>
But I really do show up ;-)

[00:10:00.0000] <molly>
I mean, clearly

[00:10:01.0000] <karlUshi>
molly: and if you want a distribution by country, which is far to be complete

[00:10:02.0000] <karlUshi>
http://esw.w3.org/topic/PeopleLocation

[00:10:03.0000] <Lachy>
molly, I don't understand why it matters what the gender stats are

[00:10:04.0000] <molly>
I'm sure you don't

[00:11:00.0000] <molly>
sorry, I didn't start this

[00:11:01.0000] <karlUshi>
:) 

[00:11:02.0000] <molly>
I was asked

[00:11:03.0000] <molly>
but honestly guys

[00:11:04.0000] <molly>
think about it

[00:11:05.0000] <molly>
you can say "no gender bias" all you want

[00:11:06.0000] <molly>
give me 10 women active

[00:11:07.0000] <Hixie>
there's clearly a gender bias

[00:11:08.0000] <molly>
and I'll shut the fuck up

[00:11:09.0000] <Hixie>
it's across the entire tech industry

[00:11:10.0000] <molly>
actually, I'll STFU anyway

[00:11:11.0000] <molly>
yes, that's my point. I'm NOT drawing you guys out.

[00:12:00.0000] <molly>
:X

[00:12:01.0000] <Hixie>
/me participates in FIRST high school programmes to encourage gender-balance tech participation

[00:12:02.0000] <molly>
please don't get defensive

[00:12:03.0000] <Hixie>
(amongst other things)

[00:12:04.0000] <Hixie>
just sayin' :-)

[00:12:05.0000] <Hixie>
i encourage everyone to do that kind of stuff

[00:12:06.0000] <Hixie>
it's hella fun

[00:12:07.0000] <molly>
so my job sometimes means saying "look at this, it's Fugly"

[00:12:08.0000] <karlUshi>
molly: I think that some issues are difficult to understand when people do not live them. Gender, racial, Culture, etc

[00:12:09.0000] <molly>
right? Yes.

[00:12:10.0000] <molly>
Me too.

[00:12:11.0000] <othermaciej>
given the gender composition of the tech industry in general, of standards groups in particular, there's probably significantly more men actively involved than women

[00:12:12.0000] <molly>
I agree Karl

[00:13:00.0000] <molly>
I think I *might* know something about this

[00:13:01.0000] <molly>
maybe not that much

[00:13:02.0000] <molly>
but a little?

[00:13:03.0000] <othermaciej>
there have been at least some number of people with female-sounding names that I don't spefically know to be male who have posted on public-html

[00:13:04.0000] <karlUshi>
yep othermaciej 

[00:13:05.0000] <molly>
it's not about HTML5

[00:14:00.0000] <molly>
as Hixie mentioned, it's industry wide, world wide in fact

[00:14:01.0000] <molly>
but don't you think having some girls around is good?

[00:14:02.0000] <molly>
I mean really!

[00:15:00.0000] <karlUshi>
/me agrees

[00:15:01.0000] <molly>
we smell good

[00:16:00.0000] <molly>
and bring food.

[00:16:01.0000] <othermaciej>
Molly: we're just doing our best to give you the info you asked for -- there aren't good stats on this, so all we can do is extrapolate

[00:16:02.0000] <karlUshi>
molly: not all girls smell good, but the same goes with guys ;) 

[00:17:00.0000] <molly>
I know, but I also felt some defensiveness. I think mostly you're a lot of guys who are really smart and don't have any idea why more women are involved with HTML5 and tech in general

[00:17:01.0000] <molly>
but would welcome them

[00:17:02.0000] <molly>
particularly if they smelled good and brought food.

[00:17:03.0000] <molly>
LOL Karl

[00:17:04.0000] <molly>
Don't want to ask who YOU've been hanging with!

[00:17:05.0000] <karlUshi>
:D

[00:17:06.0000] <othermaciej>
I don't tend to have that particular thought

[00:17:07.0000] <othermaciej>
I try to treat people as people

[00:18:00.0000] <karlUshi>
I'm a cow

[00:18:01.0000] <Hixie>
yeah like i said, i don't care what gender contributors have

[00:18:02.0000] <molly>
I'm proud of all of you for that

[00:18:03.0000] <molly>
maybe I'm just a different generation

[00:18:04.0000] <molly>
that still feels it needs to provide good pathways

[00:18:05.0000] <molly>
etc. etc.

[00:18:06.0000] <othermaciej>
fwiw the engineering team that reports to me has an unusually high proportion of women for a software engineering team, and I have encouraged all of them to participate more in web standards, but I am not going to force anybody to post

[00:19:00.0000] <molly>
i think the workforce is changing

[00:19:01.0000] <othermaciej>
(40%)

[00:19:02.0000] <molly>
it's just at the upper layers - speakers, spec writers, etc.

[00:19:03.0000] <Lachy>
I just did a quick skim of the HTMLWG members and counted about 20 names that could possibly be female, and I'm probably wrong about 15 of them :-)\

[00:19:04.0000] <molly>
hahaha

[00:19:05.0000] <molly>
I love you Lachlan 

[00:19:06.0000] <Hixie>
yeah google as a company has a much higher than average female:male engineer ratio

[00:19:07.0000] <molly>
that's great.

[00:19:08.0000] <Hixie>
and does a lot to encourage that as an industry trend

[00:19:09.0000] <Hixie>
(like sponsor FIRST)

[00:20:00.0000] <karlUshi>
interesting

[00:20:01.0000] <Lachy>
there were about 200 names that were complete foreign and ambiguous though, so my count isn't at all accurate

[00:20:02.0000] <Hixie>
the problem mostly stems from the education sector, though, i think

[00:20:03.0000] <othermaciej>
I'm lucky enough that if I told my team that what I like best about them is that they bring food and smell good, they would probably laugh at me rather thangetting angry

[00:20:04.0000] <karlUshi>
we often talk about female:male engineer ratio, but never about female:male comm ratio, or female:male admin ratio

[00:20:05.0000] <molly>
this is why I'm doing the train the trainer program

[00:20:06.0000] <molly>
it's not all about women either

[00:21:00.0000] <molly>
I mean how many black americans? Hispanics? Etc.

[00:21:01.0000] <karlUshi>
how many cows? !!!!

[00:21:02.0000] <molly>
the IT world is still one based on class

[00:21:03.0000] <molly>
Karl, you still can't let go of the cow?

[00:21:04.0000] <molly>
I mean, isn't this a zen thing

[00:21:05.0000] <karlUshi>
;) that's my blood

[00:21:06.0000] <molly>
"I let go of the cow"

[00:22:00.0000] <othermaciej>
I think large parts of the IT business world are pretty egalitarian in attitude -- most engineering types don't care too much who's doing the work if they can code

[00:23:00.0000] <othermaciej>
although there are clearly pockets of bad attitude and inappropriate behavior

[00:23:01.0000] <karlUshi>
http://www.normandieweb.org/nature/vache/images/JMLvache01.jpg

[00:23:02.0000] <othermaciej>
I would say the biggest issue is the education pipeline

[00:23:03.0000] <molly>
I am totally the wrong person to talk about women in the IT world. My best mentors have been men, men have helped me all my life. I love men. I trust them more than women.

[00:24:00.0000] <othermaciej>
engineering is also an occupation that's unusually easy to enter as a self-taught practitioner; but very few women choose that path

[00:24:01.0000] <karlUshi>
molly: interesting I had almost the same exact opposite thing

[00:24:02.0000] <molly>
I guess we can call it meritocracy

[00:24:03.0000] <karlUshi>
prior to W3C, all my boss have been... women, even in Astrophysics Research

[00:24:04.0000] <molly>
if you earn it

[00:24:05.0000] <molly>
you get the respect. That's been my sense

[00:24:06.0000] <molly>
I've only been insulted by trolls. True colleagues have disagreed, but always respected

[00:25:00.0000] <molly>
that makes me feel very good as a person

[00:25:01.0000] <molly>
I always have men as bosses

[00:25:02.0000] <molly>
except my mother

[00:25:03.0000] <molly>
lol

[00:25:04.0000] <molly>
gay men a lot, too

[00:25:05.0000] <molly>
wonder why

[00:25:06.0000] <molly>
not that it matters

[00:26:00.0000] <molly>
weren't we talking about gender indifference?

[00:26:01.0000] <molly>
I wish it were that way, young men

[00:26:02.0000] <molly>
but it's not, not quite

[00:26:03.0000] <molly>
but you guys are paving the cowpaths

[00:26:04.0000] <molly>
:: cough cough :

[00:30:00.0000] <Hixie>
ok so

[00:30:01.0000] <Hixie>
i have a few minutes to work on whatwg stuff before bed time

[00:31:00.0000] <molly>
Hixie: Peter from WSG, with 4k members, wants to help organize a world-wide F2F

[00:31:01.0000] <molly>
I'd like to work on that 

[00:31:02.0000] <Hixie>
isn't that what we're doing in november

[00:31:03.0000] <molly>
well, there's the Technical Plenary and we talked about that

[00:31:04.0000] <molly>
but nothing was confirmed that I know of

[00:31:05.0000] <Lachy>
how would a world wide F2F work?

[00:32:00.0000] <molly>
to do some worldwide F2F

[00:32:01.0000] <molly>
Lachy, all ideas welcome

[00:32:02.0000] <molly>
we need several core cities

[00:32:03.0000] <Lachy>
meet in various locations around the world and hook up via satellite and IRC

[00:32:04.0000] <molly>
where people can gather

[00:32:05.0000] <molly>
yep

[00:32:06.0000] <Hixie>
just for socialising? or?

[00:32:07.0000] <molly>
we have to figure out the time/space continuum but I'm sure we can handle that

[00:33:00.0000] <molly>
Hixie, you mentioned that it's suffering to work on specs F2F

[00:33:01.0000] <molly>
and I agreed

[00:33:02.0000] <Hixie>
/me doesn't think this is something he can really do in the next hour or so :-)

[00:33:03.0000] <karlUshi>
world-wide F2F on How To Make Love. Great!

[00:33:04.0000] <molly>
so I think it shouldn't be actual spec work  but discussion around key points

[00:33:05.0000] <Hixie>
molly: well, it's hard to ahve meetings of any productive kind that are more than about 3 people

[00:33:06.0000] <Hixie>
molly: but social gathering do help oil the wheels, as it were

[00:33:07.0000] <Hixie>
i don't know what i can do about it myself though

[00:33:08.0000] <molly>
Yes, Ian, I know, as I've witnessed 

[00:34:00.0000] <molly>
with you there ;-)

[00:34:01.0000] <molly>
so my point is the F2F brings the good connections

[00:34:02.0000] <molly>
and we should discuss key points

[00:34:03.0000] <molly>
but we should bring those back to discussion

[00:34:04.0000] <molly>
not make it an actual spec meeting

[00:34:05.0000] <Lachy>
hmm. a bar camp style meeting would probably work

[00:35:00.0000] <molly>
I think so

[00:35:01.0000] <molly>
but more focused

[00:35:02.0000] <molly>
anyway, that's my thought

[00:36:00.0000] <Hixie>
well i suggest we try that and see what the one-location meeting in november works out like first, but yes

[00:36:01.0000] <Hixie>
it'd be great if we got an agenda for that meeting soon though

[00:36:02.0000] <molly>
well, we did a one location gig in Paris

[00:36:03.0000] <molly>
you weren't there, but we had quite a turnout with Anne and Henri, etc

[00:37:00.0000] <Lachy>
my major concern with with idea is the problem of getting 400+ people meeting up simultaneously around the world isn't easy

[00:37:01.0000] <molly>
well, the agenda has to come out of hot topics, which I need some help on

[00:37:02.0000] <molly>
I can write the stuff

[00:37:03.0000] <molly>
I need someone to help me analyze it

[00:37:04.0000] <molly>
Lachy, we aim for the most

[00:37:05.0000] <molly>
and people who are truly motivated

[00:37:06.0000] <othermaciej>
I think doing it barcamp/unconference style reduces the need for a formal agenda

[00:37:07.0000] <molly>
well, those are your winners

[00:38:00.0000] <othermaciej>
though suggestions of possible topics can be helpful

[00:38:01.0000] <molly>
how would you see it maciej? Because that format is very creative and popular

[00:38:02.0000] <molly>
I think we can come up with a few hot topics here :D

[00:38:03.0000] <molly>
I'm not so worried about that lol

[00:38:04.0000] <Lachy>
/me is giong to a BarCamp next weekend

[00:39:00.0000] <Hixie>
by "agenda" i meant a decision from DanC about whether november's meeting would be an unconference or not

[00:39:01.0000] <hober>
I think the hot topics would naturally bubble to the forefront in an unconference-style gathering

[00:39:02.0000] <molly>
Hixie, wow. I didn't think of it like that

[00:39:03.0000] <molly>
I thought they'd do that W3C style. Do you want to advocate for a barcamp style?

[00:39:04.0000] <molly>
I'd work with you

[00:39:05.0000] <molly>
I mean, 400 some odd members

[00:40:00.0000] <molly>
NOT your average Working Group

[00:40:01.0000] <Hixie>
i asked for us to do that in response to the last meeting we had (last week)

[00:40:02.0000] <molly>
hey that's awesome

[00:40:03.0000] <Hixie>
danc said he's noodle on it or something

[00:40:04.0000] <molly>
How did it go over?

[00:40:05.0000] <molly>
ok

[00:40:06.0000] <molly>
what about Chris?

[00:40:07.0000] <molly>
others?

[00:41:00.0000] <Hixie>
i have not heard anything from chris

[00:41:01.0000] <Hixie>
response from others has been mostly positive from what i hear but Dan's the one who has to decide

[00:41:02.0000] <molly>
wow, okay

[00:41:03.0000] <molly>
thanks, that was good to know

[00:41:04.0000] <karlUshi>
:)

[00:41:05.0000] <molly>
hey, I'm going to bed you guys. Thanks for a great chat as always. More discussion tomorrow I hope.

[00:42:00.0000] <molly>
ttys

[00:44:00.0000] <Hixie>
you know given how many complaints there were about lack of transparency about the issues i had, there has been remarkably little response to the e-mail i sent

[00:44:01.0000] <Hixie>
about the issues list

[00:46:00.0000] <Lachy>
the complaints about the lack of transparency always confused me, since the whatwg and htmlwg have been the most open and transparent groups compared with many others

[00:47:00.0000] <Hixie>
Lachy: yeah, well, the issues list was closed, which was a valid complaint. but it is no longer!

[00:49:00.0000] <hober>
The repeated transparency calls have really given me a strong impression of bureaucray trolling (http://tantek.pbwiki.com/TrollTaxonomy#Bureaucraytroll)

[00:50:00.0000] <Lachy>
well, the text box placeholder email I sent seemed to get a fairly positive response so far. maybe it would help to send a summary email like that for other issues shortly before you start addressing them

[00:51:00.0000] <Hixie>
Lachy: that wouldn't really fit my working model

[00:51:01.0000] <Hixie>
which is to select an e-mail to reply to, and to fix the spec as i'm replying to it

[00:51:02.0000] <Lachy>
hmm. yeah

[00:51:03.0000] <Lachy>
I like your model

[00:52:00.0000] <Lachy>
I'm just trying to think of a way to address people's concerns, even though I actually disagree with them

[00:54:00.0000] <Hixie>
so... offline web apps, huh

[01:02:00.0000] <Hixie>
 * User goes to a page, then goes offline and uses it, then goes back

[01:02:01.0000] <Hixie>
   online and uses it. The page and its subresources are always at

[01:02:02.0000] <Hixie>
   their most up-to-date. Interactions with the page while offline are

[01:02:03.0000] <Hixie>
   synced to the server when going online.

[01:04:00.0000] <Lachy>
which folder in your issues list contains the relevenat discussions?

[01:04:01.0000] <Hixie>
filesystem

[01:27:00.0000] <Hixie>
hey, i complained, and now people are posting about it

[01:27:01.0000] <Hixie>
go figure

[01:37:00.0000] <Lachy>
which DOM3 core features are required that IE7 doesn't support?

[01:37:01.0000] <Lachy>
for the issues page

[01:40:00.0000] <Lachy>
Hixie, couldn't you use innerHTML instead of textContent?

[01:42:00.0000] <Hixie>
not unless i wanted to get pwned

[01:42:01.0000] <Lachy>
?

[01:42:02.0000] <Hixie>
that would open me to XSS bugs

[01:42:03.0000] <Hixie>
i could use innerText, maybe, but that isn't compliant

[01:43:00.0000] <Hixie>
and frankly, i don't care if it doesn't work in IE7

[01:43:01.0000] <Hixie>
that's IE7's prblem

[01:43:02.0000] <Lachy>
I didn't know there were security issues with innerHTML

[01:43:03.0000] <Hixie>
innerHTML parses HTML

[01:43:04.0000] <Lachy>
oh, so it would parse HTML in the emails

[01:43:05.0000] <Hixie>
yup

[01:44:00.0000] <Hixie>
for instance

[01:44:01.0000] <Lachy>
then use .innerHTML = escape(email); where escape is a function that converts < to &lt;, & and &amp;, etc.

[01:44:02.0000] <annevk>
.firstChild.data ?

[01:44:03.0000] <Lachy>
that would work too

[01:45:00.0000] <Hixie>
.firstChild.data wouldn't blow away the second text node, other elements, etc

[01:45:01.0000] <Hixie>
what's wrong with textContent?

[01:45:02.0000] <Hixie>
it's standards compliant and exactly what i want

[01:45:03.0000] <Lachy>
just that IE users are complaining

[01:45:04.0000] <Hixie>
that's their problem

[01:45:05.0000] <Hixie>
they need a better browser

[01:45:06.0000] <annevk>
x.innerHTML = "foo"; .x.firstChild.data = bar

[01:46:00.0000] <annevk>
not very nice though

[01:46:01.0000] <Hixie>
i don't want to instantiate an HTML parser just to remove nodes, that's a performance nightmare

[01:46:02.0000] <Hixie>
this is slow enough as it is

[01:59:00.0000] <annevk>
/me wonders how closely RB read the XML specification...

[02:10:00.0000] <annevk>
/me wonders why hsivonen didn't opt for the forms TF

[02:26:00.0000] <annevk>
the bugzilla bugs mentioned earlier are funny

[02:27:00.0000] <Hixie>
btw if anyone has an urge to write their own frontend to the WHATWG issues thingy, the API is described here: http://www.whatwg.org/issues/API

[02:30:00.0000] <Hixie>
annevk: feel free to take over on those bugs, i'm leaving it up to the gecko devs at this point

[02:30:01.0000] <Hixie>
no point arguing back and forth

[02:31:00.0000] <annevk>
I'm tempted to mark the one he filed INVALID, but since he's reporter he can reopen I believe so it's kind of pointless to do so probably

[02:31:01.0000] <annevk>
Especially since he his this weird untrackable train of thought on how things are done in the real world and specified in HTML4...

[02:34:00.0000] <Hixie>
as i understand it basically what he proposes is that usemap="" on <input type=image> only be used for giving tooltips for the <input>, but i don't understand why that would be useful for accessibility, and i don't see any evidence that authors are looking for that

[02:34:01.0000] <Hixie>
(at least not in numbers that would support this feature in view of hte Baby Steps principle)

[02:35:00.0000] <annevk>
I suppose they would also become focusable regions

[02:35:01.0000] <Hixie>
no because he's saying that there'd be no href="" on them

[02:35:02.0000] <Hixie>
so they'd all be image-map-dead areas

[02:35:03.0000] <Hixie>
(he's arguing that the clicks on those areas should go through to the <input> element)

[02:35:04.0000] <Hixie>
(as opposed to being caught by the <area> as you might otherwise expect in dead <area>s)

[02:36:00.0000] <annevk>
in that case it seems pretty pointless

[02:37:00.0000] <annevk>
actually, he mentioned yesterday that he would want the <area id> to be submitted as well or something

[02:40:00.0000] <zcorpan_>
http://blog.whatwg.org/issues

[02:41:00.0000] <annevk>
maybe update it to reference the API in case people want to hack their own tracker that works in IE?

[02:50:00.0000] <Hixie>
thanks for fixing my typo :-)

[02:50:01.0000] <Hixie>
right, bed time

[02:50:02.0000] <Hixie>
nn

[02:51:00.0000] <annevk>
g'night

[02:54:00.0000] <Hixie>
(he reopened the bug for the 2nd time btw)

[02:54:01.0000] <annevk>
right

[02:55:00.0000] <annevk>
/me gives up

[02:58:00.0000] <mgdm>
This is great... by tracking the development of new Web stuff, I can learn about things I never new existed... http://www.w3.org/TR/ruby/

[03:00:00.0000] <zcorpan_>
Lachy: s#XHTML 1.0#XHTML 1.0 as text/html# in the proposed design principles example

[03:00:01.0000] <Lachy>
zcorpan_, yes

[03:01:00.0000] <annevk>
you can use s# too? interesting

[03:01:01.0000] <Lachy>
I assumed he meant s/.../.../ though I'm not sure how one would make such a typo

[03:02:00.0000] <annevk>
no, he used a / in the replacement, so using # makes some sense if it's allowed

[03:02:01.0000] <zcorpan_>
in perl, you can use almost any punctation character

[03:02:02.0000] <annevk>
ah yeah, I remember something like that from my Perl book

[03:03:00.0000] <zcorpan_>
s!...!...! works too for instance

[03:03:01.0000] <annevk>
/me made a few hello world scripts once

[03:03:02.0000] <annevk>
/me then learned Python

[03:05:00.0000] <Lachy>
hmm. I don't think it's worth responding to Rob in bugzilla anymore.  There doesn't seem to be anything that can be said, which would help.

[03:08:00.0000] <Lachy>
wow, this is contradictory: "Safari and IE both support submission of forms with <input type=image usemap=#IDREF >, where IDREF is the name of a MAP element. This is exactly the behavior I'm proposing for Gecko."

[03:08:01.0000] <Lachy>
So he wants to implement <area> the same way IE and Safari do (which is not at all), but at the same time object to removing support for it.

[03:11:00.0000] <annevk>
IE is case-insensitive for <label for> too

[03:11:01.0000] <annevk>
probably Unicode case-insensitive

[03:11:02.0000] <annevk>
I suspect it might do that for all ID values

[03:12:00.0000] <Lachy>
I don't think IE implements case sensitivity correctly according to Unicode

[03:12:01.0000] <annevk>
"In keeping with the spirit of the the recommendation" lol

[03:13:00.0000] <annevk>
Lachy, that would make sense; it's hardly conforming to anything :)

[03:13:01.0000] <annevk>
(much like any other browser for that matter, in case someone takes offense at the above)

[03:13:02.0000] <Lachy>
I find it amusing how he asks us to cite the relevant part of the specifcation to support our argument, but he's just making up the implementation requirements as he goes

[03:14:00.0000] <Lachy>
I did some testing for Unicode case folding recently

[03:16:00.0000] <Lachy>
/me is looking for the test cases he made

[03:24:00.0000] <Lachy>
found them. http://lachy.id.au/dev/unicode/test/case-folding (note: may take a while to load)

[03:35:00.0000] <zcorpan_>
hmmm. we're not happy with implementing all supported interfaces on Document objects 

[03:36:00.0000] <zcorpan_>
we'd rather move members that are useful for all types of documents to the Document interface than to implement HTMLDocument on all documents

[03:36:01.0000] <zcorpan_>
like cookie and getElementsByClassName

[03:38:00.0000] <zcorpan_>
perhaps also move getElementsByClassName from HTMLElement to Element

[03:42:00.0000] <annevk>
that certainly makes sense

[03:42:01.0000] <annevk>
I still think it makes sense for all documents to implement HTMLDocument for the compound document scenario

[03:43:00.0000] <zcorpan_>
why?

[03:44:00.0000] <annevk>
let me put it differently, how to decide which interfaces to implement?

[03:45:00.0000] <zcorpan_>
yeah, that's also an issue. deciding on the root element's namespace makes sense to me

[03:45:01.0000] <annevk>
that's to late

[03:45:02.0000] <zcorpan_>
it's what opera does, no? :)

[03:45:03.0000] <annevk>
I don't think so

[03:46:00.0000] <annevk>
I think we do it based on MIME type which is clearly not desirable

[03:46:01.0000] <zcorpan_>
no, that's what firefox does

[03:46:02.0000] <zcorpan_>
we decide on root namespace

[03:46:03.0000] <annevk>
sure?

[03:46:04.0000] <zcorpan_>
last time i checked

[03:47:00.0000] <annevk>
how does that work with createDocument() and such?

[03:47:01.0000] <annevk>
/me ponders

[03:47:02.0000] <zcorpan_>
dunno

[03:47:03.0000] <zcorpan_>
not sure what happens when you replace the root element either

[04:26:00.0000] <Lachy>
IMHO, deciding based upon the namespace is the most logical option.

[04:32:00.0000] <zcorpan_>
Lachy: what happens when you replace the root element?

[04:59:00.0000] <Lachy>
zcorpan_, it's not possible to replace the root element, since document.documentElement is read only. You'd have to create a new document.

[05:00:00.0000] <zcorpan_>
hmm

[05:01:00.0000] <Lachy>
I could be wrong, since I haven't tested it. I'm just going by what the DOM spec said

[05:02:00.0000] <zcorpan_>
iirc browsers allow it, but that might be a bug then

[05:02:01.0000] <Lachy>
do you have a test case?

[05:03:00.0000] <zcorpan_>
btw, documentElement being readonly doesn't imply that you can't remove it

[05:03:01.0000] <zcorpan_>
it just means that you can't do document.documentElement = somethingelse

[05:03:02.0000] <Lachy>
how else is it possible to change the root element?

[05:04:00.0000] <zcorpan_>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3Cscript%3Ewindow.onload%3Dfunction%28%29%7Bdocument.removeChild%28document.firstChild%29%7D%3C/script%3E

[05:04:01.0000] <zcorpan_>
with methods on the document

[05:07:00.0000] <Lachy>
ok

[05:08:00.0000] <Lachy>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3Cscript%3E%0Awindow.onload%3Dfunction%28%29%20%7B%0A%20%20document.removeChild%28document.firstChild%29%3B%0A%20%20document.appendChild%28document.createElementNS%28%22http%3A//www.w3.org/2000/svg%22%2C%20%22svg%22%29%29%3B%0A%20%20w%28document.documentElement%29%0A%7D%0A%3C/script%3E

[06:03:00.0000] <zcorpan_>
Lachy: w(document) says [HTMLDocument] in opera

[06:05:00.0000] <zcorpan_>
also when i try the same with an xml document. so we don't change the interface of the document in response to changes to the dom

[06:35:00.0000] <zcorpan_>
if the insertion mode is "after body"

[06:36:00.0000] <zcorpan_>
A character token that is one of one of U+0009 CHARACTER TABULATION, U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), or U+0020 SPACE

[06:36:01.0000] <zcorpan_>
Process the token as it would be processed if the insertion mode was "in body".

[06:36:02.0000] <zcorpan_>
would the character then be appended to the body element?

[06:37:00.0000] <zcorpan_>
since the current node is the body element

[06:38:00.0000] <zcorpan_>
e.g. "<body></body> </html>" is parsed as if it were "<body> </body></html>"

[06:43:00.0000] <zcorpan_>
correct?

[07:48:00.0000] <Whiskey_M>
'lo

[07:50:00.0000] <Philip`>
SyntaxError: EOL while scanning single-quoted string

[08:24:00.0000] <zcorpan_>
hmm. wonder if, when getting innerHTML in html, an element is in the html namespace but has a prefix, the prefix should be dropped

[08:24:01.0000] <zcorpan_>
so that it can be parsed into the same element again with the html parser

[08:25:00.0000] <zcorpan_>
though innerHTML in html wasn't designed to handle such cases anyway

[08:26:00.0000] <met_>
http://www.infoq.com/news/2007/08/hdflash

[08:26:01.0000] <met_>
I think it's the same format Apple want for <video> in Safari

[08:27:00.0000] <Whiskey_M>
zcorpan: would that have the knock on effect that when in an XHTML namespace it will pull the innerXML ?

[08:27:01.0000] <zcorpan_>
Whiskey_M: i don't understand the question

[08:28:00.0000] <zcorpan_>
what is innerXML?

[08:29:00.0000] <Whiskey_M>
I think I may have my knickers / DOMs in a twist (between the .net, asp and JS document models and the different calls for each I'm not suprised though)

[08:33:00.0000] <zcorpan_>
i mean, if you have an element with the following properties: prefix is "x", localName is "input", namespaceURI is "http://www.w3.org/1999/xhtml", and it is part of an html document. what should that element's parent node's innerHTML attribute return on getting?

[08:33:01.0000] <zcorpan_>
per html5, it should return "<x:br>"

[08:33:02.0000] <zcorpan_>
er

[08:33:03.0000] <zcorpan_>
"<x:input>"

[08:34:00.0000] <zcorpan_>
which won't be parsed into an html "input" element when parsed again

[09:27:00.0000] <zcorpan_>
hmmm... offsetLeft

[10:05:00.0000] <gsnedders>
Hixie: weren't you going to rewrite "The |m| element" to make it more distinct from |strong| and |em|? couldn't it do with a note of the issue?

[10:11:00.0000] <zcorpan_>
and rename it to "mark" so that it doesn't sound like "em"? or drop it altogether? :)

[10:12:00.0000] <gsnedders>
zcorpan_: I'd keep it, as it has its uses (like marking text that needs to be updated :P)

[10:20:00.0000] <gsnedders>
Hixie: At the end of #attributes (Attributes common to |ins| and |del| elements), there is a "element's >".  I assume this is a typo.

[10:32:00.0000] <Hixie>
gsnedders: send mail (about both)

[10:47:00.0000] <Philip`>
gsnedders: Maybe you should specify who "you" is referring to when emailing five hundred people :-)

[10:47:01.0000] <gsnedders>
Philip`: hmm, just maybe.

[10:48:00.0000] <Hixie>
jesus, i spend time writing a tool for people and all i get from the htmlwg are complaints

[10:48:01.0000] <gsnedders>
Philip`: it is rather implied by the point of rewriting part of the spec as to referring to the editors

[10:48:02.0000] <gsnedders>
Hixie: you suck.

[10:48:03.0000] <Hixie>
what happened to people pulling their own weight

[10:49:00.0000] <gsnedders>
Hixie: people have got too fat to do so.

[10:49:01.0000] <Philip`>
Maybe that's what happens when you discourage people from saying "+1" or similarly non-constructive agreements - only people with (hopefully-)constructive criticism(/complaints) reply :-)

[10:50:00.0000] <gsnedders>
/me waits for someone to take that completely out of context and threaten him at saying the group is fat

[11:08:00.0000] <Hixie>
ok that guy telling me i should make the page work with his broken tool really pissed me off

[11:09:00.0000] <zcorpan_>
Hixie: take a deep breath... :)

[11:10:00.0000] <Hixie>
it's amazing to me how much the philosophies of the whatwg and the htmlwg (ignoring people who came from the whatwg) differ

[11:10:01.0000] <Hixie>
you guys (the people who first got involved through the whatwg) always answer "ooh, great! let's add more!", the htmlwg-only guys usually answer "that wasn't enough!"

[11:11:00.0000] <Hixie>
grr. bbl.

[11:12:00.0000] <zcorpan_>
/me thinks he has sent <ol> feedback to whatwg before, but can't find it again

[16:29:00.0000] <jgraham>
Distinctly under tested, underoptimized, only html4 implemented (and with bugs, I'm sure): http://wordsandpictures.dyndns.org/table_inspector.html

[16:31:00.0000] <kingryan>
jgraham:  error: http://wordsandpictures.dyndns.org/cgi-bin/tables/table_inspector.py?uri=http%3A%2F%2Ftheryanking.com%2Fmicroformats%2Fwebzine.html&scope=1&headers=1

[16:31:01.0000] <jgraham>
That was quick :)

[16:31:02.0000] <kingryan>
first try :)

[16:52:00.0000] <zcorpan>
http://wordsandpictures.dyndns.org/cgi-bin/tables/table_inspector.py?uri=http%3A%2F%2Fsimon.html5.org%2Farticles%2Fmobile-results&scope=1&headers=1 hmm, the first row doesn't seem right


2007-08-22
[17:04:00.0000] <jgraham>
zcorpan: I'm not sure whether HTML4 expects th cells to have headers

[17:05:00.0000] <jgraham>
it doesn't make sense in lots of cases but I'm trying to do what it _says_ not what makes sense

[17:06:00.0000] <jgraham>
/me notes that there seems to be a bug in html5lib's etree treewalker which screws up the results for pages with multiple tables

[17:13:00.0000] <Hixie>
/me doesn't understand why he randomly has php5.cgi scripts running when he has no php on his site

[17:19:00.0000] <Hixie>
ohhhhh the whatwg forums are under my account

[17:19:01.0000] <Hixie>
that explains it

[17:56:00.0000] <webben_>
jgraham: I think it's pretty clear (when you read it carefully) that HTML4 does not expect th cells to have headers. The DTD notes say to use td when a cell is both data and header.

[17:56:01.0000] <webben_>
/me thinks for a moment

[17:56:02.0000] <webben_>
actually, maybe not so clear

[17:56:03.0000] <webben_>
i guess you can have header headers that aren't data

[20:36:00.0000] <Lachy>
I wonder what the objection is to having any tool hosted on whatwg.org instead of w3.org http://www.w3.org/mid/46CB1C8B.7010208⊙RAU

[20:36:01.0000] <Lachy>
it just seems that some people will find the smallest and most irrelevant issue to complain about

[20:37:00.0000] <Philip`>
whatwg.org involves insufficient bureaucracy

[20:40:00.0000] <Lachy>
some people just need to realise that the barrier between the 2 groups is very artificial, since we're working toegether on the spec

[20:42:00.0000] <Philip`>
/me hosts stuff on the old box on the floor next to him, through a residential ADSL connection, on someone else's domain name, since he isn't really into either bureaucracy or uptime and prefers to just stick stuff wherever is convenient

[20:43:00.0000] <Philip`>
Even when system administrators are being cooperative and helpful, it seems much easier to be your own instead of trying to organise deployment with someone else

[20:44:00.0000] <Lachy>
yeah, I'd set up and run my own server, though my connection is not suitable enough for it - it's far too slow

[20:45:00.0000] <Philip`>
I get 50KB/s upload speed, which seems plenty since I only have to access it over the local network and never notice the bandwidth limit :-)

[20:46:00.0000] <Lachy>
what's you're download speed?

[20:48:00.0000] <Philip`>
Theoretically it's 1MB/sec maximum; in practice I get up to about 500KB/s and more commonly below 250KB/s

[20:49:00.0000] <Philip`>
/me prefers being at university where the network bandwidth is greater than his hard disk bandwidth :-)

[20:49:01.0000] <Lachy>
so that's an 8 megabit connection, which is the theoretical maximum for ADSL

[20:49:02.0000] <Lachy>
(not ADSL2 or 2+)

[20:49:03.0000] <Philip`>
Yep

[20:50:00.0000] <Philip`>
(I've not seen either of those being available anywhere around here)

[20:51:00.0000] <Lachy>
I'm stuck on a slow 1500/512kbps

[20:52:00.0000] <Lachy>
I was supposed to be upgraded to ADSL2 soon, but I've just noticed that they've delayed the phone exchange upgrade for another 3 months :-|

[20:55:00.0000] <Philip`>
If I remember correctly, I had something like 512/256 at first, then it got upgraded to 2048/512, then later to 8192/512

[20:56:00.0000] <Philip`>
It's nice when all these free upgrades happen :-)

[20:56:01.0000] <Philip`>
(Well, "free" as in "still £30/month like it was before")

[20:56:02.0000] <Lachy>
at my previous place, last year, I was on 24000/1000, which was awesome!

[20:57:00.0000] <Lachy>
though, it practice I never got above about 8000

[20:59:00.0000] <Philip`>
Oops, I should go to sleep

[21:01:00.0000] <karlUshi>
Lachy: outside of the controversy of "ownership", there is a general policy of W3C of keeping the stuff online. It's why we usually encourage WGs to collect every documents and information on the W3C web site. To not have a 404 in 5 years from now. 

[21:02:00.0000] <Lachy>
yeah, the persistence policy is nice

[21:04:00.0000] <Lachy>
however, I just find it odd that complaints come as soon as anything is hosted on whatwg.org, but no-one has complained, for example, about these logs been on krijnhoetmer.nl

[21:04:01.0000] <karlUshi>
:))) I would complain but I don't have energy to do so ;) or let's say I prefer to put my energy elsewhere instead ;)

[21:04:02.0000] <karlUshi>
question of priorities

[21:55:00.0000] <Hixie>
yeah the persistence thing doesn't really apply here though since the whole point of the web app is to show what the status today is

[00:09:00.0000] <Hixie>
i wish roy would actually read the sniffing part of hte spec

[00:10:00.0000] <Hixie>
he seems to have this idea that the spec defines an ungodly sniffing algorithm

[00:10:01.0000] <Hixie>
when in fact it's barely enough to make browser vendors happy (all of them have complained about some part of it)

[00:10:02.0000] <Hixie>
(though ironically not the same part, which amuses me)

[00:21:00.0000] <Hixie>
I hope othermaciej and anne have fun with the task force

[00:21:01.0000] <othermaciej>
"fun"

[00:21:02.0000] <othermaciej>
that is totally what I'm expecting

[00:22:00.0000] <Hixie>
i've subscribed to the list already

[00:22:01.0000] <othermaciej>
there's a list already?

[00:22:02.0000] <Lachy>
what's the new list?

[00:23:00.0000] <Hixie>
i look forward to reading it avidly and not being allowed to post

[00:23:01.0000] <Hixie>
the list was created months ago

[00:23:02.0000] <Hixie>
public-forms.tg

[00:23:03.0000] <Hixie>
er

[00:23:04.0000] <Hixie>
public-forms-tf

[00:24:00.0000] <othermaciej>
I will be impressed if the task force can even reach a consensus decision on what its actual job is

[00:24:01.0000] <Hixie>
me too!

[00:24:02.0000] <Lachy>
it doesn't have subscribe/unsubscribe links next to it. Is anyone allowed to subscribe?

[00:24:03.0000] <Hixie>
it's a public list

[00:24:04.0000] <Hixie>
anyone can subscribe to any w3c public list

[00:24:05.0000] <Lachy>
ok

[00:25:00.0000] <Hixie>
from a productivity point of view i am really happy i'm not going to be allowed to be involved in those discussions

[00:25:01.0000] <Hixie>
from the point of view of my argumentative self, though, i'm jealous :-P

[00:26:00.0000] <Lachy>
are only the 6 TF members allowed to discuss on that list?

[00:26:01.0000] <Hixie>
i think it would be pointless to have officially appointed members if other people are allowed to take part in the discussion

[00:27:00.0000] <Hixie>
but i guess i'm just assuming we're not allowed to post

[00:27:01.0000] <Hixie>
seems like it would be rude to post

[00:27:02.0000] <Hixie>
for us

[00:29:00.0000] <Lachy>
oops, I sent my subscription request to public-forms-tf instead of public-forms-tf-request by mistake, though it still worked :-)

[00:29:01.0000] <Hixie>
heh

[00:33:00.0000] <Hixie>
http://www.w3.org/mid/46CB83B1.7040506⊙rca is an interesting mail to see cross-posted to public-html and wai-xtech

[00:33:01.0000] <Lachy>
Hixie, on the top issues list, could you add a vote count for the issues beside the names, so one doesn't have to open them up to see

[00:34:00.0000] <Lachy>
someone should advise him that he should be using the title attribute

[00:43:00.0000] <othermaciej>
karlUshi: should I find a new IRC channel where I am free to express my opinions honestly and have a sense of humor?

[00:44:00.0000] <Hixie>
Lachy: done

[00:44:01.0000] <Hixie>
whatwg is such a channel, if you feel otherwise please let me know by private IM and i will deal with the problem

[00:44:02.0000] <Lachy>
Hixie, thanks

[00:45:00.0000] <Lachy>
othermaciej, what did karlUshi complain about?

[00:45:01.0000] <Hixie>
lachy: (i'd already written all the backend code for that, just had to code in the client side)

[00:46:00.0000] <othermaciej>
Lachy: about my comments above circa 11:30 PM

[00:47:00.0000] <Lachy>
the "fun" and "that is totally what I'm expecting" comments?

[00:47:01.0000] <othermaciej>
and "I will be impressed if the task force can even reach a consensus decision on what its actual job is"

[00:50:00.0000] <Lachy>
Hixie, if I open http://www.whatwg.org/issues/listen directly in Firefox, it freezes and crashes. I wonder if that's a bug in FF or a problem with one of my extensions.

[00:51:00.0000] <karlUshi>
othermaciej: I have no issues with what you are saying for yourself. I have issues with WG loosing time on arguments about what you are saying. And as long as you are free to express yourself in public, you have to accept in return that people are free to express negative comments about your statements. :) that is the price of your free expression. It's not one way.

[00:52:00.0000] <othermaciej>
karlUshi: you are free to express anything you want, but knowing that you may forward any random thing I say here to the HTML Working Group chairs with a complaint, and without even talking to me about it first, makes this not a comfortable place to converse

[00:53:00.0000] <karlUshi>
it /is/ a public place. it /is/ logged online. it /is not/ a comfortable place by its nature.  Wake up :)

[00:53:01.0000] <Lachy>
othermaciej, indeed. I've had karlUshi and DanC both do that with previous things I've said

[00:53:02.0000] <karlUshi>
/me is always amazed at the lack of maturity of people with regards to public expression and its consequences

[00:54:00.0000] <othermaciej>
karlUshi: well, my alternative is to find a place of private expression where you are not invited

[00:55:00.0000] <othermaciej>
karlUshi: but I would instead prefer to converse with relative comfort and in semi-obscurity, as is the norm on most public IRC channels

[00:55:01.0000] <Hixie>
Lachy: could be a problem accessing window.parent or something

[00:55:02.0000] <karlUshi>
semi-obscurity doesn't exist. 

[00:55:03.0000] <karlUshi>
specifically when the logs are available

[00:56:00.0000] <karlUshi>
note that it is not about me.

[00:56:01.0000] <othermaciej>
it is the social norm of most IRC channels to not make a big deal of things said in the channel outside the channel (unless there is an unresolvable personal conflict)

[00:56:02.0000] <karlUshi>
it is about expressing one-self in a public environment which is recorded. Requesting free expression and not giving the right to other to react is... very surprising

[00:56:03.0000] <othermaciej>
while I can't force you to respect this social norm, I think it is rude not to do so

[00:57:00.0000] <Lachy>
karlUshi, others do have the right to react, but they should take it up with the person directly before involving anyone else

[00:57:01.0000] <othermaciej>
I'm not denying you any rights

[00:58:00.0000] <karlUshi>
Lachy: this specific channels is big wide open -> http://krijnhoetmer.nl/irc-logs/

[00:58:01.0000] <Lachy>
karlUshi, so?

[00:58:02.0000] <karlUshi>
/me remembers suddenly the discussion about old usenet suddenly archived by Google 

[00:58:03.0000] <Lachy>
karlUshi, that doesn't mean that one should involve higher authorities as soon as someone takes offense by what was said, without first discussing it with the offender

[00:59:00.0000] <karlUshi>
people willing to fight for the semi-obscurity

[00:59:01.0000] <karlUshi>
Lachy: I didn't take offense 

[00:59:02.0000] <karlUshi>
read my comment above

[00:59:03.0000] <karlUshi>
"I have issues with WG loosing time on arguments about what you are saying."

[01:00:00.0000] <Hixie>
could someone clarify exactly what it is that is being complained about?

[01:01:00.0000] <othermaciej>
by me or by Karl?

[01:01:01.0000] <Hixie>
both

[01:01:02.0000] <karlUshi>
me: http://krijnhoetmer.nl/irc-logs/whatwg/20070822#l-132

[01:01:03.0000] <karlUshi>
maciej: that I sent the sentence to danc and chris and maciej to say that it was not appropriate

[01:02:00.0000] <Lachy>
karlUshi, I don't think it was appropriate either

[01:02:01.0000] <othermaciej>
I'm complaining that Karl quoted a snippet of our conversation and sent it to the HTML working Group chairs with his comment that he thinks it makes me unsuitable as a Forms Task Force participant

[01:03:00.0000] <karlUshi>
not what I said othermaciej 

[01:03:01.0000] <othermaciej>
which, personally, I think is nonsense; being pessimistic about the outcome doesn't in any way mean that I won't participate in good faith

[01:03:02.0000] <karlUshi>
"I'm note sure that is a positive statement on your will to achieve work in this task force. I find it quite sad, personally. Either you participate to something because you think it will be fruitful, either you don't participate."

[01:04:00.0000] <Hixie>
ok.

[01:04:01.0000] <Hixie>
karlUshi: please resist forwarding conversations from this channel to the chairs without discussing it with the people you are quoting first.

[01:04:02.0000] <othermaciej>
I participate in web standards because it is part of my job to make them good and to make the process succeed, not because it necessarily fills my heart with joy

[01:04:03.0000] <karlUshi>
hixie: I'm willing to do that. 

[01:05:00.0000] <Hixie>
thank you.

[01:05:01.0000] <Lachy>
karlUshi, thanks

[01:05:02.0000] <othermaciej>
thanks, karlUshi

[01:05:03.0000] <karlUshi>
the funny thing

[01:05:04.0000] <Lachy>
does that apply to #html-wg too?

[01:05:05.0000] <karlUshi>
is that at first I sent it in private specifically to make a recorded mess online

[01:05:06.0000] <karlUshi>
which maciej just did ;)

[01:05:07.0000] <Hixie>
#html-wg is danc's problem

[01:06:00.0000] <karlUshi>
to not make

[01:06:01.0000] <Hixie>
you know, public-html is amusing

[01:06:02.0000] <Whiskey_M>
'lo

[01:06:03.0000] <othermaciej>
karlUshi: well, if you asked me about my comments here in the first place, then the public record would have our conversation, not just my out-of-context remarks

[01:07:00.0000] <Hixie>
many threads start with one message that i save because it has good feedback in it, followed by a lot of messages that really don't help much either way

[01:07:01.0000] <Hixie>
as an example:

[01:07:02.0000] <Hixie>
http://www.w3.org/mid/E99FD39C-9AE3-491C-B5EE-1405AF70A7A7⊙gc - useful feedback

[01:07:03.0000] <Hixie>
http://www.w3.org/mid/46CB1808.1010701⊙RAU - quibbling feedback

[01:07:04.0000] <Hixie>
http://www.w3.org/mid/op.txe6ftgzidj3kv@hp-a0a83fcd39d2 - feedback repeating the first point for the benefit of the second poster

[01:07:05.0000] <othermaciej>
karlUshi: I stand behind them, but I will note that I participate in good faith and make a strong effort in many endeavors despite pessimism about the outcome

[01:08:00.0000] <othermaciej>
the HTML Working Group itself is one such example

[01:08:01.0000] <othermaciej>
but I would like to be free to voice my cynicism once in a while

[01:11:00.0000] <Lachy>
hmm. I still haven't received the confirmation email to confirm my subscription to public-forms-tf was successful.

[01:11:01.0000] <Hixie>
maybe we're on a waiting list

[01:12:00.0000] <karlUshi>
I'm not the moderator

[01:13:00.0000] <othermaciej>
good night everyone

[01:16:00.0000] <Hixie>
nn

[01:17:00.0000] <MikeSmith>
Hixie - does your admonishment about forwarding links to comments in the public log for this channel apply only to forwarding such links to chairs, or to anybody?

[01:23:00.0000] <Hixie>
MikeSmith: it was merely a request, to karl, that he not forward conversations from this channel to the chairs without discussing them with the people involved first

[01:24:00.0000] <Hixie>
it is naturally common courtesey to speak to someone before you quote them in any case

[01:26:00.0000] <MikeSmith>
I agree completely.

[01:26:01.0000] <MikeSmith>
It's just that this is first time I can every remember seeing you make any request from anybody on this channel about what their behavior should be.

[01:27:00.0000] <Hixie>
it's the first time i've had someone say they were starting to feel uncomfortable talking openly in this channel

[02:41:00.0000] <krijnh>
http://krijnhoetmer.nl/irc-logs/whatwg/20070822#l-97

[02:41:01.0000] <krijnh>
After this Karl has sent me a mail to ask if the irc logs could be moved to w3.org

[02:42:00.0000] <krijnh>
Good idea? :)

[02:42:01.0000] <zcorpan>
/me couldn't care less

[02:46:00.0000] <krijnh>
Me neither

[02:46:01.0000] <krijnh>
I'm not about to take my site down any time soon

[02:50:00.0000] <Hixie>
the w3c is certainly very welcome to host irc logs

[02:50:01.0000] <Hixie>
as is everyone else

[02:51:00.0000] <Hixie>
at least as far as this channel goes

[02:51:01.0000] <Hixie>
now if they want to argue that they don't want you logging their network's channels, that's between them and you on their network :-)

[02:51:02.0000] <krijnh>
Of course

[02:52:00.0000] <krijnh>
But they could just let me know, if they don't want me to do that :)

[02:59:00.0000] <zcorpan>
ok. <object>. time for me to create tests on that i think

[02:59:01.0000] <krijnh>
Didn't Robert Burns make those already? :)

[03:00:00.0000] <zcorpan>
they weren't tests

[03:00:01.0000] <Lachy>
his tests weren't good

[03:00:02.0000] <krijnh>
I'm running almost 1337 mails behind, so I wouldn't know :(

[03:00:03.0000] <zcorpan>
1337 :)

[03:00:04.0000] <krijnh>
1332 tbe :)

[03:53:00.0000] <Lachy>
I'm surprised by how many people have objected to "Solve Real Problems"

[04:23:00.0000] <zcorpan>
Hixie: http://krijnhoetmer.nl/irc-logs/whatwg/20070821#l-666

[05:32:00.0000] <zcorpan>
hmm, why does http://validator.nu/ give me the About page?

[05:49:00.0000] <zcorpan>
Hixie: "Whenever the data attribute changes" -- does this also apply when the data attribute is removed?

[05:49:01.0000] <zcorpan>
on <object>

[05:50:00.0000] <zcorpan>
http://simon.html5.org/test/html/semantics/object/dynamic/001.htm

[07:53:00.0000] <Lachy>
what does this mean from Karl against Well-Defined Behaviour: "Dangerous. Typical example the search form with the magnifier of Safari, which led people to create invalid markup." http://www.w3.org/2002/09/wbs/40318/dprv/results#xwdb

[07:54:00.0000] <Lachy>
how can one blame the search form for the invalid markup on the web?

[07:56:00.0000] <Lachy>
I don't understand Sander's response either: "Given that the last sentence lists specific exceptions, this Design Principle appears to close the door for a UA to for instance render tables with fixed thead and tfoot and scrollable tbody; respect Content-Type; display equivalents side by side; not support javascript; etc."

[07:57:00.0000] <zcorpan>
i thought karl was in favor of well-defined behavior: http://www.w3.org/QA/2007/07/why-html-5-matters.html

[07:58:00.0000] <Lachy>
well, they're the only 2 that disagreed, and since their responses don't make sense, they don't really count.

[08:05:00.0000] <Philip`>
It seems unfair to say they don't count - it's more useful to say that they don't make sense and should clarify their position

[08:06:00.0000] <Lachy>
oh, I just meant for my purposes. I'm going through the entire results, looking for why people do and don't agree, in order to write my own responses

[08:07:00.0000] <Philip`>
Ah, okay

[08:31:00.0000] <zcorpan>
http://forums.whatwg.org/viewtopic.php?t=66#400 -- spam in sig?

[08:33:00.0000] <hasather>
zcorpan: yes

[08:36:00.0000] <zcorpan>
user now-or-never begone

[09:03:00.0000] <aja>
did conformance checker move, or just temporarily unavailable?

[09:04:00.0000] <zcorpan>
aja: i would guess a mistake on hsivonen's part; you could email him and ask

[09:05:00.0000] <aja>
/me was hoping to catch him here

[09:06:00.0000] <Philip`>
The site was offline yesterday for maintenance (at least I think that's what it said), so it looks like some glitch when it came back up

[09:07:00.0000] <aja>
Philip`: saw that yesterday

[09:08:00.0000] <aja>
/me doesn't speak any Nordic languages, either :)

[09:08:01.0000] <Philip`>
/me wonders how hard it is to run a mirror of the validator

[16:00:00.0000] <KevinMarks>
just found a browser layout oddity that seems HTML5ish

[16:01:00.0000] <KevinMarks>
http://meganmcardle.theatlantic.com/archives/2007/08/did_911_ensure_republican_hege.php has an invalid blockquote p nesting, that behaves differently in firefox in the feed compared to the page

[16:34:00.0000] <Philip`>
/me watches Googlebot and Yahoo! Slurp race through the static HTML issues list

[16:34:01.0000] <Philip`>
They're currently neck and neck, each having found 4 out of 1536 emails

[16:35:00.0000] <takkaria>
I'd never have thought that was a spectator sport

[16:35:01.0000] <Philip`>
Actually, they each found exactly the same four messages

[16:35:02.0000] <Philip`>
(Maybe they're just following links from the IRC logs?)


2007-08-23
[17:13:00.0000] <Hixie>
Philip`: so i dunno what happened but i now have 3346 e-mails in the database

[17:13:01.0000] <Hixie>
i'm thinking maybe the script died when it last ran

[17:14:00.0000] <Hixie>
do you want a dump again?

[17:14:01.0000] <Hixie>
(if so, what's your e-mail again?)

[17:16:00.0000] <Hixie>
in other news, who's in charge of the wiki again?

[17:16:01.0000] <Hixie>
i think we might want to do something about the spam, as suggested on http://wiki.whatwg.org/wiki/User_talk:Hixie, though i don't know what

[17:17:00.0000] <Philip`>
Hixie: When I download all the folder lists from you, that only gives 1515 messages

[17:17:01.0000] <Hixie>
it should have changed recently

[17:17:02.0000] <Hixie>
i reran the algorithm

[17:17:03.0000] <Hixie>
the script, rather

[17:17:04.0000] <Philip`>
Ah, okay

[17:17:05.0000] <Philip`>
The dump would be helpful so I don't have to download the rest, then

[17:18:00.0000] <Philip`>
(philip⊙zdcu)

[17:18:01.0000] <Hixie>
sent

[17:20:00.0000] <Philip`>
Received

[17:20:01.0000] <Philip`>
(Thanks!)

[17:20:02.0000] <Philip`>
/me tries to remember where he put his scripts

[17:24:00.0000] <Philip`>
Seems to be working

[17:26:00.0000] <Philip`>
The list of all emails is 7MB now

[17:26:01.0000] <Philip`>
and I see 3298 messages in it

[17:26:02.0000] <Philip`>
which is close enough to 3346

[17:26:03.0000] <Philip`>
(I guess some messages are not in any folder?)

[17:31:00.0000] <Hixie>
you should have exactly 3346 in that csv dump

[17:31:01.0000] <Hixie>
the folders probably changed since yesterday

[17:31:02.0000] <Hixie>
if that matters

[17:33:00.0000] <Philip`>
I get 3346 from the CSV, but downloading the 'folders' list then each 'folder ...' list gives (as of ten minutes ago) 3298

[17:33:01.0000] <Hixie>
odd

[17:34:00.0000] <Hixie>
i'd have expected the opposite

[17:34:01.0000] <Hixie>
that is, some of the e-mails are in more than one folder

[17:40:00.0000] <Philip`>
The CSV contains some message IDs like "<3019.217.124.88.212.1149790503.squirrel⊙wc"

[17:40:01.0000] <Hixie>
i bet the IDs are truncated to 64 bytes

[17:41:00.0000] <Hixie>
but that's 56

[17:41:01.0000] <Hixie>
not 64

[17:41:02.0000] <Hixie>
so nevermind

[17:41:03.0000] <Hixie>
i'm confused

[17:41:04.0000] <Philip`>
Of the messages that are in the CSV but not in the folders list, almost all are truncated like that

[17:42:00.0000] <Hixie>
weird

[17:42:01.0000] <Philip`>
except for <<<<f0a054a12911d5a153482846f135f669⊙1l>>>>

[17:42:02.0000] <Philip`>
Oh, I think I see that one in the online bit too

[17:43:00.0000] <Philip`>
*online folders list

[17:43:01.0000] <Philip`>
so ignore that

[17:43:02.0000] <Hixie>
it's in the database

[17:43:03.0000] <Hixie>
though why, i don't know

[17:44:00.0000] <Philip`>
I see 54 truncated names in the CSV, which is 3346 - 3298

[17:44:01.0000] <Philip`>
Uh, not it isn't

[17:44:02.0000] <Philip`>
*no

[17:45:00.0000] <Hixie>
do you have a preferred username?

[17:45:01.0000] <Philip`>
I see 54 truncated names, plus 6 messages in two folders, and 3298 = 3346 - 54 + 6

[17:45:02.0000] <Philip`>
Depends on the context :-)

[17:46:00.0000] <Philip`>
(Usually something like "philip" works alright)

[17:46:01.0000] <Philip`>
(except when someone else has stolen that name, and I have to add random punctuation characters)

[17:47:00.0000] <Philip`>
((Actually, 5 messages in two folders and 1 message in three folders))

[17:53:00.0000] <Hixie>
ok so how did these get truncated

[17:53:01.0000] <Hixie>
wtf

[17:54:00.0000] <Hixie>
btw the way i update the tables each night is to lock them, drop them, then recreate them entirely

[17:54:01.0000] <Hixie>
which is why the number can change dramatically (if the script dies while it's updating)

[17:54:02.0000] <Philip`>
When you measured 56 bytes, did you happen to do it with something like perl -e'print length("<3019...⊙w.")' ?

[17:55:00.0000] <Hixie>
yes, did i screw that up? :-)

[17:56:00.0000] <Philip`>
Try the -w flag :-)

[17:56:01.0000] <Hixie>
d'oh

[17:56:02.0000] <Hixie>
i suck

[17:56:03.0000] <Philip`>
/me made precisely the same mistake

[17:56:04.0000] <Hixie>
ok so i need to make it longer

[17:56:05.0000] <Hixie>
that means shrinking the other index column

[17:58:00.0000] <Philip`>
Why does it affect indexes?

[17:58:01.0000] <Hixie>
cos i use this as port of one of my primary indexes

[17:58:02.0000] <Hixie>
in the 'folders' table iirc

[18:01:00.0000] <Philip`>
/me doesn't see why expanding one would require shrinking another

[18:05:00.0000] <Hixie>
because i'm at the limit of the index key size

[18:06:00.0000] <Philip`>
The index can just be on a short prefix of the column and it should work exactly the same, I thought

[18:06:01.0000] <Philip`>
But maybe that doesn't work so well when it's meant to be a unique index

[18:13:00.0000] <Hixie>
ok regenning the db

[18:14:00.0000] <Hixie>
the reason they were cropped in the dump and not on the site is that the site was using the folders table instead of the emails table and the two tables had different lengths for the ids!

[18:14:01.0000] <Hixie>
<- moron

[18:17:00.0000] <Philip`>
Okay - once it's done, I can just download the 54 extra emails through the API

[18:18:00.0000] <Hixie>
done

[18:18:01.0000] <Hixie>
(you can also dump it straight from the database)

[18:23:00.0000] <Philip`>
Hmm, it looks like I got the same folders list as before

[18:24:00.0000] <Philip`>
(unless I messed up somewhere)

[18:24:01.0000] <Philip`>
with only the 3298 messages

[18:29:00.0000] <Hixie>
there's something really weird going on

[18:32:00.0000] <Hixie>
there are 3,353 entries in the 'emails' table

[18:32:01.0000] <Hixie>
there are 3,359 in the 'folders' table (which includes dupes)

[18:33:00.0000] <Philip`>
There are 3,353 distinct 'message' values in the 'folders' table, and they all correspond to an 'id' in 'emails'

[18:33:01.0000] <Hixie>
right

[18:33:02.0000] <Philip`>
so that all seems to be correct

[18:33:03.0000] <Hixie>
(i just checked)

[18:33:04.0000] <Hixie>
so why do you only get 3298 if you use the api

[18:33:05.0000] <Philip`>
Is there any caching somewhere?

[18:33:06.0000] <Hixie>
oh, maybe, yeah

[18:33:07.0000] <Philip`>
(I deleted my own cache before downloading the folders list again)

[18:34:00.0000] <Hixie>
i just killed my server, try again

[18:35:00.0000] <Philip`>
/me tries

[18:38:00.0000] <Philip`>
3359

[18:38:01.0000] <Philip`>
Perfect :-)

[18:38:02.0000] <Hixie>
phew

[18:38:03.0000] <Hixie>
good call on the caching

[18:38:04.0000] <Hixie>
i've added a scary line to my regen script that kills the server

[18:40:00.0000] <Philip`>
Is there not a less drastic way to clear the cache?

[18:41:00.0000] <Philip`>
/me 's usual approach to caching is "nobody uses my stuff anyway, so it doesn't matter if it's slow and resource-intensive"

[18:41:01.0000] <Hixie>
Philip`: that was my approach, and then you started fetching all this data... ;-)

[18:41:02.0000] <Hixie>
Philip`: there's probably a better way, but *shrug*

[18:42:00.0000] <Hixie>
the server knows how to handle it and the CGI scripts that talk to it know how to handle the server script going away

[18:42:01.0000] <Hixie>
(the server i'm talking about isn't the database, it's a cgi script that talks a little tcp protocol to my other cgi scrits)

[18:42:02.0000] <Hixie>
scripts

[18:43:00.0000] <Philip`>
Hmm, good point - I'll make sure I don't offer any APIs and tempt people to use them :-)

[18:44:00.0000] <Philip`>
(Ah, okay, that's less drastic than killing MySQL and Apache and whatever)

[18:44:01.0000] <Hixie>
yeah

[18:44:02.0000] <Hixie>
i really wish we had TCPConnection

[18:44:03.0000] <Hixie>
it would be so much better than having this mess

[18:44:04.0000] <Hixie>
and then i could do away with everything except the server script

[18:50:00.0000] <Hixie>
hey anyone recall the uris of the studies similar to the webstats one i did?

[18:50:01.0000] <Hixie>
there was one that had a flower on the front page iirc

[18:51:00.0000] <Philip`>
http://triin.net/2006/06/12/Coding_practices_of_web_pages ?

[18:51:01.0000] <Hixie>
yes!

[18:51:02.0000] <Hixie>
thanks!

[19:01:00.0000] <Philip`>
http://triin.net/archive/kool/webstat/figure-33.png - you can tell the data set had a very high proportion of CNN pages

[19:02:00.0000] <Hixie>
you should see the number of <nyt_copyright> elements in the studies i do :-P

[19:02:01.0000] <Hixie>
not enough to come in the top 100 or anything, but still significant

[19:02:02.0000] <Philip`>
cnn.com was 5% of the URLs on dmoz.org when I last counted

[19:03:00.0000] <Hixie>
yeah

[19:03:01.0000] <Hixie>
cnn.com has a lot of pages

[19:03:02.0000] <Philip`>
I've no idea who keeps submitting them all to dmoz.org

[19:03:03.0000] <Hixie>
cnn, maybe

[19:03:04.0000] <Philip`>
Oh, that would make sense

[19:04:00.0000] <Hixie>
comparing http://triin.net/archive/kool/webstat/figure-12.png to http://code.google.com/webstats/2005-12/charts/unique-elements-per-page.svg is interesting

[19:05:00.0000] <Hixie>
it's clearly not a perfect normal distribution

[19:05:01.0000] <Hixie>
there's a very clear bump around 5-10 in both graphs

[19:05:02.0000] <Hixie>
actually i guess i'd be approximately a poisson, not a normal, distribution.

[19:07:00.0000] <Philip`>
It seems everyone has <html>, <head>, <title>, <body>, and then I guess you don't need more than a few of the next ones (<meta>, <a>, <img>, <br>, <table>, etc) to make up the rest of your page

[19:07:01.0000] <Lachy>
Hixie, I'm unable to login to the wiki server using SFTP. It responds with unable to authenticate.

[19:07:02.0000] <Hixie>
oh it was probably one of the accounts that dreamhost changed the passwd on

[19:07:03.0000] <Lachy>
same with the blog

[19:07:04.0000] <Hixie>
let me fix it

[19:08:00.0000] <Lachy>
whatwikiuser and lhunt were the usernames

[19:08:01.0000] <Philip`>
The spike at 6 on triin.net looks odd, since I can't imagine people doing much with just two extra tags

[19:08:02.0000] <Hixie>
Philip`: the spike is a bit later on the google one but yeah

[19:08:03.0000] <Hixie>
there's a prike

[19:08:04.0000] <Hixie>
spike

[19:09:00.0000] <Philip`>
/me wonders if he can get that graph from his own data

[19:09:01.0000] <Hixie>
what's the uri to yours again?

[19:09:02.0000] <Philip`>
http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/index

[19:14:00.0000] <Philip`>
http://canvex.lazyilluminati.com/survey/2007-07-17/tagcount.html

[19:15:00.0000] <Lachy>
/me installs the captcha extension

[19:16:00.0000] <Lachy>
can someone with an ordinary user account on the wiki attempt to make an edit to see if it works? (sysops won't see it)

[19:16:01.0000] <Philip`>
/me still likes the "[x] Are you a spambot?" approach to blocking spam

[19:16:02.0000] <Lachy>
wiki.whatwg.org

[19:16:03.0000] <Philip`>
(Doesn't work on human spammers, but computers never answer the question right)

[19:16:04.0000] <Hixie>
Philip`: see! see! your data has the same thing!

[19:17:00.0000] <Philip`>
My data comes from exactly the same source as the triin.net stuff, so that's not surprising :-)

[19:17:01.0000] <Philip`>
with the same 5% cnn.com, and quite a lot of weather.com

[19:17:02.0000] <Hixie>
heh

[19:17:03.0000] <Hixie>
pity :-P

[19:17:04.0000] <Philip`>
(except a much smaller sample)

[19:18:00.0000] <Philip`>
Lachy: It let me save a change without asking any hard questions

[19:19:00.0000] <Lachy>
oops, It hadn't uploaded LocalSettings.php

[19:19:01.0000] <Lachy>
try again

[19:19:02.0000] <Lachy>
aargh! fatal error.

[19:21:00.0000] <Hixie>
i assume triin.net is Pene Saarsoo's site

[19:21:01.0000] <Hixie>
Rene

[19:21:02.0000] <Philip`>
/me looks at the spikey bits in more detail

[19:22:00.0000] <Lachy>
Philip`, fixed the error, can you try again?

[19:26:00.0000] <Philip`>
Lachy: Is it meant to show on the editing page, or only after saving?

[19:26:01.0000] <Lachy>
I believe it's supposed to be after attempting to save

[19:27:00.0000] <Lachy>
these are the instructions http://www.mediawiki.org/wiki/Extension:ConfirmEdit

[19:27:01.0000] <Lachy>
it doesn't tell me much

[19:27:02.0000] <Philip`>
It let me save again without anything

[19:28:00.0000] <Philip`>
Hixie: Of 114 the pages with 6 tag names, 61 have <frameset> and <frame>

[19:28:01.0000] <Philip`>
(and most have <meta>, <title>, <html>, <head>)

[19:28:02.0000] <Philip`>
which explains the spike at 6

[19:28:03.0000] <Philip`>
(since they're all just frameset pages)

[19:29:00.0000] <Hixie>
oooh

[19:29:01.0000] <Hixie>
interesting

[19:29:02.0000] <Philip`>
Of the 208 pages with 8, 137 have <noframes>, 149 have <frameset> and <frame>, and most have meta/body/title/head/html

[19:30:00.0000] <Philip`>
so it seems they're all just frameset pages again

[19:30:01.0000] <Hixie>
aah

[19:30:02.0000] <Hixie>
interesting stuff!

[19:30:03.0000] <Philip`>
Same for 9 tags (244 pages, 168 have frames, 159 have noframes, 141 have p)

[19:31:00.0000] <Lachy>
Philip`, it will if you attempt to add a URL or attempt to create an account

[19:31:01.0000] <Lachy>
it asks the user a simple math questions like: 44 - 5 = ?

[19:32:00.0000] <Philip`>
Lachy: Aha, that works

[19:32:01.0000] <Philip`>
"69 - 5 ="

[19:32:02.0000] <Philip`>
/me thinks a bit

[19:32:03.0000] <Philip`>
Oh, it wasn't 61

[19:33:00.0000] <Philip`>
The "Create account" page says "82 + 6 ="

[19:33:01.0000] <Philip`>
so that seems to be working

[19:46:00.0000] <Philip`>
Hixie: http://canvex.lazyilluminati.com/survey/2007-07-17/tagcount.html

[19:46:01.0000] <Philip`>
now with details about what tags the various groups of pages use

[19:46:02.0000] <Hixie>
neat

[19:47:00.0000] <Philip`>
Hmm, the page using 41 tags looks quite legitimate

[19:48:00.0000] <Hixie>
heh

[19:48:01.0000] <Philip`>
(http://community.webshots.com/user/JulioUU)

[19:49:00.0000] <Hixie>
woot! you have some nyt_copyright elements too!

[19:49:01.0000] <Philip`>
http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/tag/nyt_copyright

[19:49:02.0000] <Philip`>
Only one :-(

[19:50:00.0000] <Hixie>
i got over a million of those in my sample last year

[19:50:01.0000] <Hixie>
(don't recall what the recent numbers are)

[19:50:02.0000] <Hixie>
er, not last year. year before.

[19:50:03.0000] <Hixie>
whenever my first sample was.

[19:53:00.0000] <Philip`>
11 pages with nothing but <meta> - looks like those are redirects

[19:53:01.0000] <Hixie>
yeah

[19:59:00.0000] <Philip`>
(It's nice having data about eight thousand pages rather than eight billion, because I can write hopelessly inefficient SQL queries and still get information back in a few seconds ;-) )

[20:32:00.0000] <Lachy>
why do people keep overreacting and bringing up the headers issue all the time?! http://lists.w3.org/Archives/Public/public-html/2007Aug/0926.html

[21:07:00.0000] <Hixie>
"headers= are allready counted by our editor as insignificant"

[21:07:01.0000] <Hixie>
they are? i thought i'd not yet looked at them.

[21:14:00.0000] <Lachy>
well, apparently Leif knows something about you that you don't :-)

[21:15:00.0000] <Hixie>
good to know

[21:15:01.0000] <Hixie>
i better update my opinions to think that headers="" is insignificant then

[21:31:00.0000] <Lachy>
I'm trying to think of a way to explain that the significance of a feature isn't dependent upon stats alone and give some examples of other factors that could be considered. any suggestions?

[21:32:00.0000] <othermaciej>
there's a lot of things to consider

[21:33:00.0000] <othermaciej>
- does the feature fulfill a needed use case?

[21:33:01.0000] <othermaciej>
- is it possible to do the same things as well without the feature?

[21:33:02.0000] <othermaciej>
- which existing user agents, if any, implement the feature?

[21:33:03.0000] <othermaciej>
- is the feature used widely in existing content?

[21:34:00.0000] <Hixie>
- have existing user agents invented similar proprietary features to address the use case

[21:34:01.0000] <Hixie>
- have libraries (e.g. dojo) implemented work arounds for the lack of the feature?

[21:34:02.0000] <Lachy>
- benefits to users and authors

[21:34:03.0000] <othermaciej>
- is most existing use (if there is any) such that it would be beneficial or hamful to the goal of the feature to support it?

[21:35:00.0000] <othermaciej>
for example if a feature is not implemented by mainstream browsers but is widely used, it becomes likely that at least some content using it will depend on it being ignored

[21:36:00.0000] <othermaciej>
there's all sorts of things to think about

[21:36:01.0000] <Lachy>
thanks, that's a good list

[21:37:00.0000] <othermaciej>
and another thing to consider is that different criteria may apply to whether a feature is required for implementations, and to whether it is valid for content

[21:37:01.0000] <othermaciej>
a feature that is considered to have no valid use cases and to violate various important principles may nontheless be required for implementations

[23:10:00.0000] <Hixie>
so i'm examining some of my data collected over teh past few months with an eye towards what can be published

[23:10:01.0000] <Hixie>
looking at the longdesc="" data closer, it seems the situation is even more dire than i thought

[23:11:00.0000] <Hixie>
only about 0.6% of <img> elements with a longdesc="" attribute have a useful value, as far as i can tell

[23:11:01.0000] <Hixie>
which make it about 0.0007% of <img> elements

[23:11:02.0000] <othermaciej>
what are some of the kinds of values that are non-useful?

[23:12:00.0000] <Hixie>
all the wikipedia <img> elements have bogus values that point to non-descriptions

[23:12:01.0000] <Hixie>
there are also many <img> elements that have longdesc="" attributes with values that are redundant with an ancestor <a> element's href="" attribute

[23:12:02.0000] <Hixie>
or that point to the root of another domain

[23:13:00.0000] <othermaciej>
wow

[23:14:00.0000] <Hixie>
hm, i didn't check for identity with the alt="" attribute

[23:14:01.0000] <Hixie>
i should have

[23:16:00.0000] <Lachy>
Hixie, there's a bug in your issues list http://www.whatwg.org/issues/top - see the top issue that says 3 votes, yet there's only 1

[23:17:00.0000] <Hixie>
woah, clicking it changes the result

[23:17:01.0000] <Hixie>
freaky

[23:17:02.0000] <Hixie>
noted

[23:17:03.0000] <Hixie>
thanks

[23:18:00.0000] <Lachy>
what evidence was presented in support of longdesc?

[23:18:01.0000] <Hixie>
dunno, haven't studied it beyond looking at this data so far

[23:19:00.0000] <Lachy>
/me checks the wiki

[23:19:01.0000] <Lachy>
hmm. some people want longdesc on iframe too

[23:20:00.0000] <othermaciej>
that sounds amazingly redundant

[23:20:01.0000] <othermaciej>
the iframe itself would in general contain markup

[23:20:02.0000] <othermaciej>
why would you need a markup alternative to it?

[23:21:00.0000] <Hixie>
where else can spammers put their links? we've blocked off most of the other places

[23:25:00.0000] <Lachy>
Hixie, can you publish a list of URIs for sites that look like they're using longdesc legitimately?

[23:27:00.0000] <Hixie>
http://junkyard.damowmow.com/292

[23:27:01.0000] <Lachy>
is that all of them, or just a sample?

[23:27:02.0000] <Hixie>
sample of 100

[23:28:00.0000] <Hixie>
there were millions that didn't hit any of my heuristics

[23:28:01.0000] <Hixie>
(about 6% of <img> elements with a longdesc="")

[23:28:02.0000] <Lachy>
This is not legitmate: <a href="http://www.google.co.jp/"><img src="http://blog2.fc2.com/2/20century/file/Logo_20s.gif"  alt="Google" height="75" width="143" longdesc="http://www.google.co.jp/logos.html" /></a>

[23:28:03.0000] <Hixie>
indeed not

[23:29:00.0000] <Hixie>
there are many that are not

[23:29:01.0000] <Hixie>
about 90% by the sample i looked at

[23:29:02.0000] <Lachy>
I'll let you know if I find one that is

[23:32:00.0000] <Lachy>
it's interesting how the entire list of use cases given in the wiki, doesn't actually include any use cases.  Just a list of different disabilities that might benefit from it

[23:39:00.0000] <Hixie>
k well the score doesn't change when you open one anymore

[23:39:01.0000] <Hixie>
but i don't see why the score of the top one is wrong

[23:40:00.0000] <Lachy>
maybe it thinks I'm so important that my vote is worth 2? :-)

[23:41:00.0000] <Hixie>
seems unlikely! ;-)

[23:47:00.0000] <Hixie>
either i don't understand GROUP BY / COUNT() or there's a bug in mysql

[23:56:00.0000] <Hixie>
oh!

[23:56:01.0000] <Hixie>
the message is in two folders!

[00:01:00.0000] <Hixie>
ok fixed

[00:01:01.0000] <Hixie>
/me changed COUNT(...) to COUNT(DISTINCT ...)

[00:58:00.0000] <Lachy>
Hixie, one attempted legitimate use of longdesc, but it's still questionable http://www.tcfp.state.tx.us/standards/standards_manual/standards_manual.asp?chapter=431429421 (see the state of texas logo at the bottom)

[00:58:01.0000] <Lachy>
links to http://www.tcfp.state.tx.us/image_description.txt

[01:20:00.0000] <jruderman>
.txt!

[01:21:00.0000] <othermaciej>
that could totally be alt text

[01:21:01.0000] <othermaciej>
the best part is that the plaintext contains a URL

[01:34:00.0000] <Lachy>
I updated the wiki to add some *real* use cases to replace the list of purported beneficiaries that was there http://esw.w3.org/topic/HTML/LongdescRetention

[02:06:00.0000] <Dashiva>
Lachy: Sorry if I double-posted you yesterday, had some trouble finding the right account for w3c posting

[02:06:01.0000] <Dashiva>
/me dreams about a mailing list with useful reply-to

[02:06:02.0000] <Lachy>
double posted me? which message?

[02:07:00.0000] <Lachy>
which name do you use in emails?

[02:07:01.0000] <Dashiva>
magnusrk+something@

[02:07:02.0000] <Lachy>
BTW, munging reply-to headers is considered bad practice and I'm glad the W3C doesn't do it

[02:09:00.0000] <Dashiva>
I've yet to see an argument against it that wasn't based on happening 20 years ago

[02:09:01.0000] <Lachy>
ah, right. I did get multiple copies of that one, but it doesn't bother me since I just deleted it along with the other dupes

[02:10:00.0000] <Lachy>
I've yet to see a convincing argument for it that isn't based on a users inability to use their mail client properly

[02:10:01.0000] <takkaria>
given that HTML5 is based in part around users' inability to write valid markup, it seems like that might be a pretty good reason. :)

[02:11:00.0000] <Lachy>
just press Reply All (or equivalent).  It's usually right next to the Reply button!

[02:11:01.0000] <Dashiva>
That's the problem

[02:11:02.0000] <takkaria>
Lachy: oh, I agree, I just enjoy the mild irony. :)

[02:11:03.0000] <Dashiva>
I got a mail from Robert Burns today that didn't contain a single word I've said because i've been passed along by reply all for who knows how long

[02:12:00.0000] <Lachy>
people need to learn to trim the recipient list too!

[02:12:01.0000] <Dashiva>
User education doesn't work, you should know that ;)

[02:12:02.0000] <takkaria>
Lachy: btw, rob moved your use cases to "authoring cases" and moved what were "use cases" before back there

[02:13:00.0000] <Lachy>
WTF???

[02:13:01.0000] <Lachy>
They're not use cases!

[02:14:00.0000] <takkaria>
I tried to clean up the <img>fallback</img> section a while ago to remove duplicates, and he just reverted it, so I gave up on it

[02:16:00.0000] <zcorpan>
amusing

[02:17:00.0000] <Lachy>
/me reverted rob's change

[02:17:01.0000] <Dashiva>
revert war, I choose you

[02:18:00.0000] <zcorpan>
/me escapes the expected bashing

[02:21:00.0000] <takkaria>
/me wonders if the Design Principles aren't better changed to "How the HTML WG should work"

[02:25:00.0000] <takkaria>
Lachy: btw, are you going to look through all of Hixie's longdesc urls?

[02:25:01.0000] <Lachy>
not all of them.  I looked through about a dozen randomly selected

[02:27:00.0000] <zcorpan>
random selection, eh? in the statistical sense? not biased in some way? ;)

[02:28:00.0000] <Lachy>
well, not entirely random.  I just picked a couple from the top, scrolled down a bit and picked a few others, then repeated.

[02:28:01.0000] <Lachy>
it was sort of biased by ignoring the wikipedia entires that I knew were bogus

[02:29:00.0000] <takkaria>
I assume someone's already argued with the wikimedia people about how they're misuing longdesc?

[02:30:00.0000] <Lachy>
don't know, probably.

[02:32:00.0000] <takkaria>
/me adds that to his todolist

[02:43:00.0000] <Lachy>
Rob isn't listening again, I give up

[02:45:00.0000] <takkaria>
Lachy: apologies; I've sent you two duplicate replies to your longdesc post, both off-list

[02:45:01.0000] <takkaria>
I confuse myself by having too many email accounts and never hitting reply-to-all

[02:47:00.0000] <Lachy>
takkaria, just resend it to the list, it doesn't matter if I get dupes

[02:47:01.0000] <takkaria>
/me has done so

[03:29:00.0000] <zcorpan>
um

[03:29:01.0000] <takkaria>
/me sighs at Rob for another contentless post

[03:29:02.0000] <zcorpan>
what should happen if you serve a gif as image/png and use it in an <object>?

[03:30:00.0000] <zcorpan>
browsers just decode it as a gif

[03:52:00.0000] <Lachy>
zcorpan, browsers don't care what image MIME is use, they just invoke the appropriate image library by sniffing the first few bytes of the file, looking for the signature

[03:56:00.0000] <Whiskey_M>
'lo

[04:29:00.0000] <zcorpan>
Lachy: right... so image/png, image/gif, etc, all go through "Content-Type sniffing: image"

[05:03:00.0000] <Lachy>
zcorpan, yeah, something like that

[06:42:00.0000] <zcorpan>
hmm, wonder if browsers look at the file extension in their sniffing code

[06:44:00.0000] <Whiskey_M>
if it helps I have noted that for sites that serve content (normally avi's ) with incorrect headers IE will normally serve it with the application based upon extension, rather than FF which tends to serve based on the header

[06:46:00.0000] <zcorpan>
interesting

[06:48:00.0000] <Philip`>
Is that for content that the browser displays itself, or for content where it either saves to disk or asks the OS to find some external program to open it?

[09:26:00.0000] <Lachy>
http://blog.whatwg.org/omit-alt

[09:29:00.0000] <Lachy>
jgraham, that's a good explanation of the flag example you posted.  Thanks, now I don't have to respond. :-)

[11:56:00.0000] <gsnedders>
who actually owns html5.org?

[11:57:00.0000] <Lachy>
gsnedders, anne

[11:57:01.0000] <gsnedders>
ah

[15:18:00.0000] <Hixie>
btw Lachy if you can suggest some page-only heuristics (i.e. not involving the network) for detecting bogus longdesc=""s that would have caught them in the URLs i mentioned, it would be useful

[15:18:01.0000] <Hixie>
i'd be able to rerun the study excluding those

[15:33:00.0000] <othermaciej>
I looked at a few of those and they generally didn't look detectable from the page context alone

[15:36:00.0000] <Hixie>
yeah that was my conclusion too

[15:37:00.0000] <Hixie>
so i haven't yet looked at the headers="" stuff in detail

[15:37:01.0000] <Hixie>
but i wonder

[15:38:00.0000] <Hixie>
if i look at it, and find that imho we shouldn't include headers="" (which could be the case, just like it could be the case that we _should_ include it)

[15:38:01.0000] <Hixie>
will the situation become worse or better?

[15:38:02.0000] <Hixie>
in other words, is it better for the working group for me to address the issue at its proper time, or should i prioritise it given that the risk of that is that the attribute does in fact not get kept?

[15:42:00.0000] <othermaciej>
whether it has a positive effect on attitudes depends on the outcome

[15:42:01.0000] <jgraham>
Hixie: The problem is that we are unlikely to ever reach consensus on a spec that does not include headers irrespective of its technical merit

[15:42:02.0000] <othermaciej>
doing enough work to predict the outcome but not actually make any changes would not really be acting in good faith though, I think

[15:43:00.0000] <Hixie>
jgraham: we'll never reach consensus on the spec anyway

[15:43:01.0000] <jgraham>
Hixie: Fair point.

[15:43:02.0000] <Hixie>
othermaciej: yeah i wouldn't do that

[15:43:03.0000] <othermaciej>
I think there is a sense from some quarters that any feature that is nominally "for accessibility" is absolutely needed, regardless of whether it does anything to help accessibility in practice

[15:43:04.0000] <Hixie>
right

[15:44:00.0000] <othermaciej>
and that even trying to study the question of whether it helps accessibility in practice is somehow illegitimate

[15:45:00.0000] <othermaciej>
notwithstanding what I think of specific features, it is hard to make a good spec when there are sacred cows that may not be examined critically, so I don't know how much we can accomodate that point of view

[15:45:01.0000] <Hixie>
i'm not considering any cows sacred

[15:45:02.0000] <billmason>
I would suggest that if headers got some kind of priority and a determination made, even if the determination was "keep it", there would just be a move to argue the next accessibilty issue in dispute immediately.  I would not make it a special priority just to give those evangelizing for it a more immediate answer.

[15:46:00.0000] <Hixie>
i'm not really concerned about appeasing people, i want to make the spec the best html spec possible, including addressing the goal of making documents that use the language universally accessible.

[15:46:01.0000] <othermaciej>
I know you don't - I'm just saying it's hard to collaborate with someone who does have sacred cows, even if you happen to agree with them on one point

[15:46:02.0000] <jgraham>
But it makes everyone's life miserable when some people consider cows to be sacred and they are taken to slaughter (so to speak)

[15:46:03.0000] <Hixie>
othermaciej: yeah

[15:47:00.0000] <Hixie>
billmason: yeah. so that argues for just treating it like any other issue, and going with the FIFO principle, right?

[15:47:01.0000] <billmason>
That would be my take on it, yes.

[15:47:02.0000] <Hixie>
i think that's probably the best course of action

[15:47:03.0000] <billmason>
And for parsing logs later to write emails, I say that as a person with a general accessibility interest/focus/etc.

[15:47:04.0000] <Hixie>
i expect Dan to start telling me to prioritise public-html wg feedback over existing whatwg feedback though.

[15:48:00.0000] <Hixie>
he seems to have interpreted my "a few months" as "3 or so months" not "20 or so months" which is probably closer to realistic.

[15:48:01.0000] <othermaciej>
I do think Chaals made a good point that accessibility features might deserve special status even if only a subset of especially accessible sites use them (properly), since those might be the only sites that someone with a given disability can use at all

[15:48:02.0000] <Hixie>
not sure what to do about that if he does ask me to do so

[15:48:03.0000] <othermaciej>
but I don't know if the evidence bears that out as a justification for headers="" or the like

[15:49:00.0000] <jgraham>
Hixie: Two stacks, one issue from the top of each in turn?

[15:49:01.0000] <othermaciej>
it would require examples of actual particularly accessible sites that consistently use it in a way that is beneficial

[15:49:02.0000] <othermaciej>
not just a theory that there might be some

[15:49:03.0000] <Hixie>
othermaciej: well, there are two things, right, there's the features that are rarely used but when used make the page better, and there are the features that are very widely misused to the point where they actively hurt the user's experience on most sites

[15:50:00.0000] <Hixie>
longdesc clearly falls into the latter category according to all the studies i've done or seen

[15:50:01.0000] <Hixie>
as in, exposing longdesc will mostly expose the user to spam or useless content, even on supposedly accessibility-aware pages

[15:51:00.0000] <zcorpan>
/me would think that the way tabindex is implemented, it actively hurts the user experience too. treating all positive numbers as 0 for tabindex would, i think, have better user experience

[15:52:00.0000] <Hixie>
that's possible too, i haven't even looked at tabindex yet other than adding the negative thing

[15:53:00.0000] <zcorpan>
yep. :) i just suddenly came to think of tabindex when reading the above

[15:53:01.0000] <Hixie>
one thing i really don't know how to fix is accesskey=""

[15:54:00.0000] <zcorpan>
dunno either

[15:54:01.0000] <jgraham>
That isn't going to be fun.

[15:54:02.0000] <zcorpan>
although opera's implementation is somewhat useful, or at least not harming the user experience

[15:55:00.0000] <Hixie>
opera's implementation is unintuitive

[15:55:01.0000] <jgraham>
From what I remember Mike Smith saying it sounds like different solutions are appropriate on different devices

[15:55:02.0000] <Hixie>
and doesn't actually solve the problem of how  to make it device independent

[15:55:03.0000] <Hixie>
yeah

[15:55:04.0000] <Hixie>
maybe it really is a stylistic thing

[15:55:05.0000] <Hixie>
we do have key-equivalent in CSS iirc

[15:56:00.0000] <Philip`>
Hixie: I would assume users wouldn't see the spam or useless content since they wouldn't be explicitly asking their tool to give them the longdesc, except in the cases where they have a good expectation that it's going to be useful, so misuse wouldn't hurt the user much

[15:56:01.0000] <Hixie>
Philip`: really? how would they know to ask?

[15:57:00.0000] <Hixie>
for longdesc it really seems to me that the only cases i've seen where the longdesc was actually useful and wasn't something you could have just stuffed into alt="", it was actually useful to sighted users too

[15:57:01.0000] <Hixie>
and could have just been included on the page or in a link from the page

[15:57:02.0000] <Philip`>
I'd assume the tool would indicate in some quick way that there is a longdesc attached to the image, similarly to how it must quickly indicate wherever there is a link

[15:58:00.0000] <Philip`>
but I have precisely no experience of how relevant tools work in practice

[15:58:01.0000] <zcorpan>
jaws says "press enter for long description" after reading the image alt, and if you press enter it will open the url in a new window

[15:58:02.0000] <zcorpan>
iirc

[15:58:03.0000] <Hixie>
right but how do you know it's appropriate?

[15:59:00.0000] <Philip`>
(At least from what I've heard about table headers, they're not read out by default - you press some key when you've got the right cell selected, and if you see one cell has rubbish headers then you won't bother checking every single other cell - so the harm caused by misuse is similarly minimised)

[16:01:00.0000] <Philip`>
Hixie: By considering the context, like whether other images on the page have useful longdescs - e.g. you'd know it's worthwhile reading all the longdescs in the CSS spec after you've seen the first few

[16:02:00.0000] <Philip`>
(and on sites which misuse it, you'd quickly realise you should just ignore all the rest)

[16:02:01.0000] <Hixie>
that doesn't seem like the optimal user experience

[16:02:02.0000] <Hixie>
it's like reminding the user continually that the page wasn't designed for them but there's this secondary set of content they can access

[16:03:00.0000] <Hixie>
seems like it would make one bitter

[16:04:00.0000] <Hixie>
/me tries to build up a list of requirements for offline web apps

[16:06:00.0000] <Philip`>
It seems a less significant negative point than with e.g. <input usemap> (where if your browser supports it, there are features of some sites that just don't work at all) - it might waste a bit of time to read out useless longdescs, but it's not preventing the user from using the site

[16:08:00.0000] <Philip`>
A closer-to-optimal user experience would be good, though I don't have any ideas for that :-)

[16:08:01.0000] <Hixie>
yeah the main argument for me against longdesc is that it's not useful at all, except in rare cases where frankly sighted users would benefit too, and therefore you're better off putting the content on the page itself

[16:08:02.0000] <othermaciej>
Hixie: one thing that might work is to limit the number of access keys and make it a set that UAs can map in a natural way on each platform to something w/ no conflicts

[16:08:03.0000] <Hixie>
it's hard to say since i've seen so few useful uses of it

[16:08:04.0000] <Hixie>
othermaciej: yeah that's been suggested

[16:08:05.0000] <Hixie>
othermaciej: but then the UI becomes unclear

[16:09:00.0000] <Hixie>
othermaciej: i.e. discoverability drops through the floor

[16:12:00.0000] <kingryan>
Hixie: having CSS-useable hooks could help with styleability

[16:12:01.0000] <Hixie>
yeah

[16:12:02.0000] <kingryan>
s/styleability/discoverability/

[16:13:00.0000] <Hixie>
like  ...:after { content: ' (' copy-key ')'; }

[16:13:01.0000] <Hixie>
      ... { key-equivalent: copy-key; }

[16:13:02.0000] <kingryan>
yeah

[16:13:03.0000] <Hixie>
or    ... { key-equivalent: copy-key; } ...::after { content: ' (' key-equivalent ')'; }

[16:13:04.0000] <Hixie>
...to have resilience against the cascade

[16:13:05.0000] <kingryan>
....:access-key('n'):after { content: 'meta-key N'}

[16:14:00.0000] <kingryan>
yeah

[16:14:01.0000] <Hixie>
i was thinking of also having othermaciej's idea since that solves the device problem too

[16:14:02.0000] <Philip`>
Subtitles on TV shows would benefit sighted users too (e.g. when they get distracted and miss a couple of words, or can't understand someone's accent), but usually they aren't displayed along with the content since they're ugly and distracting and not sufficiently useful to be shown to all users; people may have similar reasons for not wanting to put image descriptions in the normal page content, and hiding it being [D] links or londesc

[16:15:00.0000] <kingryan>
Hixie: is om's idea of having a limited set?

[16:15:01.0000] <Philip`>
s/being/behind/

[16:15:02.0000] <Hixie>
yeah but subtitles aren't only accessible to blind users or hidden behind long properties pages, they're one-button accessible

[16:15:03.0000] <Hixie>
like an <a> link would be

[16:15:04.0000] <Hixie>
kingryan: yeah see above

[16:16:00.0000] <Philip`>
That just seems like a browser UI issue

[16:16:01.0000] <kingryan>
yeah, me wishes browsers couldn't remap stuff that the browser already handles

[16:16:02.0000] <kingryan>
or allow me to re-remap it

[16:16:03.0000] <Hixie>
Philip`: i don't think we should hide longdesc behind a context menu or something. i'm saying the link should be right there in the content just like for everything else.

[16:19:00.0000] <Philip`>
That sounds like D-links - I think the only argument I've heard against them is they don't look very nice (but there are quite possibly other arguments I haven't heard)

[16:19:01.0000] <othermaciej>
Hixie: if the set was 0-9 for instance, it could be unmodified 0-9 on phones with a keypad, Cmd-0 - Cmd-9 on Macs, Ctrl-0 - Ctrl-9 on windows, etc

[16:19:02.0000] <othermaciej>
(assuming those are actually free)

[16:20:00.0000] <Hixie>
Philip`: yeah, though i wouldn't use [D].

[16:20:01.0000] <othermaciej>
but I'm not sure how you address discoverability

[16:20:02.0000] <Philip`>
(http://www.w3.org/TR/WCAG10-HTML-TECHS/#long-descriptions says "Invisible d-links thus provide a (temporary) solution for designers who wish to avoid visible d-links for stylistic reasons.")

[16:20:03.0000] <othermaciej>
I think the page has to provide info about shortcuts

[16:20:04.0000] <Hixie>
othermaciej: i think kingryan's idea helps with that

[16:20:05.0000] <billmason>
I think there's some kind of study that says numeric shortcuts aren't really free, though.

[16:20:06.0000] <othermaciej>
that's how keyboard shortcuts work in native apps

[16:20:07.0000] <othermaciej>
they are listed in the menu bar

[16:21:00.0000] <othermaciej>
having a shortcut list with labels somewhere in the page or accessible from the chrome would be the natural analogy

[16:21:01.0000] <othermaciej>
I'm not sure if styling them to mention they key helps

[16:21:02.0000] <kingryan>
yeah, having it in the chrome seems reasonable

[16:21:03.0000] <othermaciej>
thee UA could do a better job if it is the only thing that knows the concrete key mapping, but you need a label to go next to the shortcut

[16:22:00.0000] <othermaciej>
*the UA

[16:22:01.0000] <othermaciej>
it could even be a menu or submenu in the menu bar

[16:22:02.0000] <Hixie>
hm yeah, for <command>s and other Command elements you could just have the UA create a menu somewhere with the key equivs

[16:22:03.0000] <othermaciej>
"Page Shorcuts"

[16:22:04.0000] <kingryan>
it'd be nice in interactive browsers if you could could style the access-key elements with the modifier key is pressed

[16:22:05.0000] <Hixie>
they could do that now with :active

[16:23:00.0000] <othermaciej>
kingryan: if the modifier is a commonly used one on the OS, that could be distracting

[16:23:01.0000] <kingryan>
othermaciej: indeed it could

[16:23:02.0000] <othermaciej>
something based on <command> seems like a good basic approach

[16:23:03.0000] <Hixie>
oh i misread what kingryan said

[16:23:04.0000] <othermaciej>
the idea is that a shorcut key actually activates a command, for which there might also be one or more UI elements

[16:23:05.0000] <kingryan>
there are some places, like dialogue windows in os x, where this is done already

[16:24:00.0000] <othermaciej>
it makes more sense to associate it with a command, which can then have an appropriate label

[16:24:01.0000] <othermaciej>
kingryan: example?

[16:24:02.0000] <othermaciej>
(I don't know of dialog windows that have keyboard shortcuts beyond the standard tab/enter/esc and such)

[16:25:00.0000] <kingryan>
I remember something adding "cmd-foo" in a button when cmd is pressed down

[16:25:01.0000] <kingryan>
I'm not sure where that is

[16:25:02.0000] <othermaciej>
I don't think that is standard

[16:25:03.0000] <kingryan>
it's not

[16:25:04.0000] <kingryan>
I'm just seeing that I've seen it done before and I helped with discoverability

[16:28:00.0000] <othermaciej>
usually in OS X all the keyboard shorcuts also have a menu item, so you can look in the menu system to see the shortcuts

[16:31:00.0000] <kingryan>
othermaciej: true

[16:33:00.0000] <othermaciej>
Windows is different since it uses the underline system sometimes, and that may apply both to menus and items in a dialog


2007-08-24
[17:14:00.0000] <Hixie>
ok

[17:14:01.0000] <Hixie>
/me has now read all feedback on offline storage and is coming up with ideas

[17:15:00.0000] <Hixie>
Please Stand By... Processing Feedback... 12% [##          ]

[18:11:00.0000] <zcorpan>
error reporting in the content-type sniffing algorithms might not be a bad idea, actually

[18:12:00.0000] <Philip`>
Does the error reporting need to be more complex than just complaining if the sniffed content type != the HTTP Content-Type?

[18:12:01.0000] <zcorpan>
don't think so

[18:13:00.0000] <zcorpan>
the html spec might not need to say anything about it, since it isn't really in scope for html

[18:13:01.0000] <zcorpan>
but we could still log such cases in the error console

[19:35:00.0000] <Hixie>
woot

[19:35:01.0000] <Hixie>
/me adds support for fragment identifiers to his issues list

[19:50:00.0000] <Philip`>
Hixie: With your idea, is there anything to stop people getting inconsistent caches if they're already in the middle of an update when the web server starts serving a newer version of the application?

[20:42:00.0000] <Hixie>
Philip`: no, that's the same as the status quo, effectively

[21:00:00.0000] <Lachy>
Hixie, re: "btw Lachy if you can suggest some page-only heuristics (i.e. not involving the network) for detecting bogus longdesc=""s that would have caught them in the URLs i mentioned, it would be useful"

[21:01:00.0000] <Lachy>
the only other heurisitic I thought of what checking the URL to see if it ended in .jpg, .png, .gif, etc. since linking a longdesc to an image is pointless. But that would have only caught one example that I found

[21:02:00.0000] <Lachy>
re [D] links and longdesc, http://joeclark.org/book/sashay/serialization/Chapter06.html#d-links

[21:02:01.0000] <Lachy>
that can be summarised as: "[D]" links look awful, difficult to associate a link directly with the image, longdesc doesn't cause catastrophic bugs in browsers, and "[D]" links are as bad as "click here"

[21:04:00.0000] <Lachy>
I think the first issue can be solved by using different link text or hiding it with CSS,

[21:04:01.0000] <Lachy>
the next could possibly be solved using <figure><img><a rel=longdesc/></figure>,

[21:05:00.0000] <Lachy>
and the third is countered by the argument that existing use of longdesc is more harmful

[21:05:01.0000] <Hixie>
Lachy: i actually saw some .png/.jpg/.gif links that were real description of the images

[21:05:02.0000] <Lachy>
oh, ok.

[21:05:03.0000] <Hixie>
but yeah

[21:06:00.0000] <Hixie>
i wanted to do that originally

[21:06:01.0000] <Hixie>
i think i saw someone had ...descriptions...?foo.gif

[21:06:02.0000] <Hixie>
or some such

[21:06:03.0000] <Hixie>
heh, maciej ignored my request at the top of the e-mail :-P

[21:06:04.0000] <Lachy>
oh, well you'd have to ignore the query string, but I suppose that's still no guarantee

[21:06:05.0000] <Hixie>
othermaciej: which list should i reply to you on

[21:07:00.0000] <othermaciej>
Hixie: what was your request?

[21:07:01.0000] <Hixie>
to only reply to one list :-)

[21:07:02.0000] <othermaciej>
oh, you cross-posted

[21:07:03.0000] <othermaciej>
didn't even notice that

[21:07:04.0000] <othermaciej>
mea culpa

[21:07:05.0000] <Hixie>
hehe

[21:07:06.0000] <Hixie>
no worries

[21:07:07.0000] <Hixie>
i'll just post to the htmlwg one i guess

[21:07:08.0000] <othermaciej>
I'm actually not sure which list contains more of the interested parties

[21:07:09.0000] <Hixie>
yeah me neither

[21:07:10.0000] <Hixie>
but i'm sure i'll get flamed if i use the whatwg list

[21:08:00.0000] <Hixie>
actually i really want to hit the scour people with this

[21:08:01.0000] <Hixie>
and i don't think they're on public-html

[21:08:02.0000] <Hixie>
so screw it

[21:08:03.0000] <Hixie>
whatwg it is

[21:12:00.0000] <Hixie>
hey Lachy?

[21:12:01.0000] <Lachy>
yo

[21:12:02.0000] <Hixie>
Lachy: i didn't get mail from public-forms-tf, should i have?

[21:13:00.0000] <Lachy>
the list still isn't showing up for me in the maintenance page, so I can't fix it

[21:13:01.0000] <Hixie>
ah ok

[21:13:02.0000] <Hixie>
danc told me you were in charge now, so i'd assumed it'd gone through

[21:14:00.0000] <Lachy>
I mailed sysreq yesterday and still haven't heard back

[21:14:01.0000] <Lachy>
well, sysreq initially responded and said it was all done.  it's hasn't and I'm waiting to hear back.  I'll check with danc later

[21:15:00.0000] <Hixie>
k

[21:18:00.0000] <Lachy>
http://blog.whatwg.org/omit-alt#comment-7669 he seems to ignored everything I wrote

[21:19:00.0000] <Lachy>
I wonder if he actually read the spec?

[21:21:00.0000] <Hixie>
unlikely

[21:21:01.0000] <Hixie>
but i wouldn't worry about it

[00:31:00.0000] <Lachy>
Hixie, http://canvex.lazyilluminati.com/misc/cgi/issues.cgi/message/%3Ca9699fd20704042335r113597a2u343d8134cb28e0f4%40mail.gmail.com%3E :-)

[00:34:00.0000] <Lachy>
I think it's base64 encoded, so it's unreadable

[00:36:00.0000] <Lachy>
same thing has happened to quite a few messages

[01:22:00.0000] <Lachy>
cool, I just noticed there's a new French translation on the blog http://blog.whatwg.org/pourquoi-le-texte-alternatif-peut-etre-omis-french - it was translated within 3.5 hours :-)

[01:42:00.0000] <MikeSmith>
I wasn't aware the it had been decided that alternate text was the first principle of accessibility.

[01:43:00.0000] <MikeSmith>
So, hey, I learned something today.

[01:47:00.0000] <MikeSmith>
It's like I asked Pope Ratzinger: Dude, a lot of that stuff you talk about all the time -- that you claim is gospel truth -- well, I can't actually find mention of a lot of that stuff in the Bible.

[01:47:01.0000] <MikeSmith>
But at least the Catholic church calls it exactly what it is, which is Dogma.

[01:49:00.0000] <MikeSmith>
http://en.wikipedia.org/wiki/Dogma

[01:49:01.0000] <MikeSmith>
[[

[01:49:02.0000] <MikeSmith>
established belief or doctrine held by a religion, ideology or any kind of organization, thought to be authoritative and not to be disputed or doubted.

[01:49:03.0000] <MikeSmith>
]]

[01:50:00.0000] <MikeSmith>
"Rejection of dogma is considered heresy in certain religions, and may lead to expulsion from the religious group."

[01:50:01.0000] <Lachy>
so you're saying we shouldn't be questioning the wisdom of the accessibility community, and just do what they say?

[01:50:02.0000] <krijnh>
Hmm, what's http://html4all.org/ ?

[01:54:00.0000] <MikeSmith>
Lachy - I can't answer that question for you. I guess it depends on whether you put your trust in Faith or Reason.

[01:56:00.0000] <Lachy>
well, I've never been one to accept dogma of any kind, so at the risk of being hated by the accessibility community, I choose reason and logic

[01:57:00.0000] <Lachy>
krijnh, looks like John Foliot owns that wiki

[01:57:01.0000] <krijnh>
Lachy: Ah

[01:57:02.0000] <Lachy>
http://www.dnsstuff.com/tools/whois.ch?ip=html4all.org

[01:59:00.0000] <MikeSmith>
Lachy - I think it might not be all the productive to accept the usage of the term "accessibility community"

[01:59:01.0000] <Lachy>
well, how else should I refer to the group of people who obviously dislike me and are typically involved with accessibility?

[02:00:00.0000] <MikeSmith>
Some of those objecting to the current HTML5 spec makes it sound like there's this unified monolithic group with a single set principles that they all agree on.

[02:00:01.0000] <krijnh>
The anti-Lachsessibility people

[02:00:02.0000] <MikeSmith>
And that they are speaking for that community.

[02:01:00.0000] <Lachy>
well, I know they don't speak for the whole community, but they're certainly the most vocal

[02:01:01.0000] <MikeSmith>
like the Moral Majority

[02:01:02.0000] <krijnh>
Same with some HTML5 people

[07:06:00.0000] <kjetilkWork>
virtuelv: I just assigned to give a talk about XML parsing vs. JSON parsing too, and why JSON is interesting

[07:06:01.0000] <kjetilkWork>
of course, I have no clue about that, got any good pointers? :-)

[07:07:00.0000] <kjetilkWork>
got the Opera static build running on amd64, BTW

[07:22:00.0000] <virtuelv>
kjetilkWork: json.org?

[07:24:00.0000] <kjetilkWork>
virtuelv: I was thinking independent speed benchmarking and that kind of stuff

[07:24:01.0000] <virtuelv>
no clue

[07:25:00.0000] <virtuelv>
I fear we're running wildly off-topic, though

[07:25:01.0000] <kjetilkWork>
ok

[10:02:00.0000] <MikeSmith>
krijnh - true

[10:03:00.0000] <MikeSmith>
The lesson from it I guess it that everybody ought to be continuously checking/questioning their own assumptions.

[10:04:00.0000] <MikeSmith>
It's like that moment in the Matrix, blue pill or red pill

[10:05:00.0000] <MikeSmith>
Do you want to see things as they really are, even if it's ugly and scary? Or do you want to keep seeing things as you've been comfortable seeing them and assumed they really are?

[10:07:00.0000] <Lachy>
I already accepted the ugly and scary reality 3 years ago

[10:09:00.0000] <MikeSmith>
Well, I think you're going to continue to have people strolling in, looking at selective parts of the HTML5 work from their own blinkered perspective, and saying, You guys are really making a big mess of things. When the reality is of course, No, the mess is already all around you, and has been for a long time.

[10:14:00.0000] <krijnh>
Ping

[10:14:01.0000] <Hixie>
at the risk of offending people, from what i can tell from speaking to people who really are part of the "accessibility comunity", most of the vocal complainers about accessibility in the htmlwg aren't actually really part of the "real" accessibility community

[10:15:00.0000] <krijnh>
Sorry, my connection dropped :/

[10:17:00.0000] <MikeSmith>
Hixie - I can imagine that in some cases, they could in fact be individuals who have actually had similar combative discussions with people in the WAI WGs

[10:17:01.0000] <MikeSmith>
Just hypothetically, I'm imagining.

[10:19:00.0000] <MikeSmith>
Anyway, it's really hard at times to see how to engage in discussion with them productively.

[10:19:01.0000] <MikeSmith>
If they are not willing to consider questioning whether some of their views are not in fact just dogma.

[10:20:00.0000] <MikeSmith>
like that "alternate text is the first principle of accessibility" statement in the comments to Lachy's post about Alt

[10:20:01.0000] <Lachy>
Hixie, do you mean there are others more involved with accessibility who actually agree with your controversial decisions?

[10:20:02.0000] <MikeSmith>
as if that is self-evident or axiomatic or something

[10:21:00.0000] <MikeSmith>
/me notes that my last statement was continuation of my previous one, not a response to what Lachy just said

[10:22:00.0000] <Hixie>
Lachy: i wouldn't necessarily go that far :-P

[10:24:00.0000] <MikeSmith>
/me needs to wander off for a bit

[10:24:01.0000] <MikeSmith>
back on later

[10:25:00.0000] <Hixie>
i guess tomorrow unless someone sees something wrong with it i'll write up the offline storage stuff

[10:41:00.0000] <Dashiva>
Hixie: At the risk of being quoted again, it's missing a longapplication attribute ;)

[10:41:01.0000] <Hixie>
no comment! :-P

[10:42:00.0000] <Dashiva>
One thing I thought about was the ability to run more than one copy of the same app, but I suppose that could be handled in the app itself

[10:44:00.0000] <zcorpan>
wonder if we only want to do fancy html attribute processing in quirks mode

[10:44:01.0000] <zcorpan>
color attribute processing that is

[10:45:00.0000] <MikeSmith>
Hixie, Lachy - regarding the earlier discussion, I wonder why you guys might think of the idea of sort of agreeing to declare a moratorium on discussion of document conformance issues on public-html.

[10:45:00.0000] <zcorpan>
ie7 does it in standards mode as well, but opera, safari and firefox limit it to quirks mode

[10:46:00.0000] <MikeSmith>
What I mean is, no discussions on public-html at this point about whether a particular element or attribute should/will be dropped, whether or not is should be conformant or non-conformant.

[10:46:01.0000] <Lachy>
so we can focus more on processing requirements?

[10:47:00.0000] <MikeSmith>
Exactly

[10:47:01.0000] <MikeSmith>
on interoperability issues

[10:47:02.0000] <MikeSmith>
On the conformance criteria for implementations, not for documents.

[10:47:03.0000] <Hixie>
MikeSmith: that's up to DanC

[10:47:04.0000] <Hixie>
MikeSmith: on whatwg, i welcome all feedback, on any topic

[10:47:05.0000] <Lachy>
some people might object to that. There are people who think we should only define processing requirements for conforming documents

[10:48:00.0000] <MikeSmith>
Lachy - I don't think that the charter would justify restricting discussion of processing requirements to only conforming documents.

[10:48:01.0000] <Lachy>
indeed. I'm just saying why some people would still object

[10:49:00.0000] <MikeSmith>
I know

[10:49:01.0000] <Hixie>
people will object whatever we do

[10:49:02.0000] <Hixie>
i don't think trying to reduce objections will be very productive to be honest

[10:49:03.0000] <MikeSmith>
Hixie - why?

[10:49:04.0000] <Hixie>
i think our time is better spent sifting through the feedback and summarising it into problem statements

[10:49:05.0000] <Lachy>
well, something needs to be done with public-html. I'm just not sure what would help

[10:50:00.0000] <Hixie>
MikeSmith: why to which?

[10:50:01.0000] <Hixie>
the only thing that imho needs to be done to whatwg is the stopping of people's rants being cross-posted to whatwg

[10:50:02.0000] <Hixie>
i was unhappy when i saw john's rant

[10:51:00.0000] <MikeSmith>
Why you don't think it would productive to try to reduce objections, I meant. But anyway, my point was not about trying to reduce objections, it was about trying to get people's attention on the parts of the work that are a much bigger priority.

[10:52:00.0000] <MikeSmith>
I would say the same thing about such rants being cross-posted to public-html

[10:52:01.0000] <Hixie>
oh, because i don't think it's possible. i think you could reduce the number of vocalised objections, but bottling them up is only going to cause more pain later.

[10:52:02.0000] <Lachy>
such rants shouldn't be posted anywhere

[10:53:00.0000] <Hixie>
Lachy: no argument from me there, but that's up to the chairs of each group. i don't find crap like that on whatwg acceptable behaviour.

[10:53:01.0000] <Hixie>
as i think i made clear to him

[10:53:02.0000] <Hixie>
we'll see

[10:53:03.0000] <MikeSmith>
I seriously wonder if we should perhaps start another list, public-html-conformance or something

[10:54:00.0000] <Lachy>
we've already got 3 lists for the HTMLWG

[10:54:01.0000] <Whiskey_M>
Good morning one and all :-)

[10:54:02.0000] <Lachy>
I'm not sure we need a 4th

[10:54:03.0000] <Hixie>
more lists doesn't help me at all, since i just end up getting the same amount of mail, except i need to work out which one to post to

[10:55:00.0000] <MikeSmith>
Hixie - what would you think about freezing the document-conformance parts of the spec for now? Not making any further changes with regard to whether a particular element or attribute is conformant or not?

[10:55:01.0000] <Lachy>
we've got public-html, public-html-wg-announce, public-html-comments, whatwg, whatwg-implementers, whatwg-help and now public-forms-tf. 7 lists directly related to the work on the spec is enough

[10:56:00.0000] <MikeSmith>
Lachy - I guess I suggested it just as a way to reduce traffic on public-html

[10:56:01.0000] <Hixie>
MikeSmith: i'm just gonna be going down the list of feedback

[10:56:02.0000] <Hixie>
MikeSmith: if it involves document conformance, sobeit

[10:56:03.0000] <MikeSmith>
to focus discussion on public-html, for the time being, around spec'ing implementation behavior

[10:57:00.0000] <Hixie>
you won't reduce traffic to public-html

[10:57:01.0000] <Hixie>
there are 400 people subscribed to it

[10:58:00.0000] <Hixie>
(hmm, whatwg is up to 804)

[10:58:01.0000] <MikeSmith>
Hixie - You don't think that would at least help to reduce some of the unproductive, contentious discussion?

[10:59:00.0000] <MikeSmith>
I mean, how unproductive or contentious could a discussion about the tokenizer, parser, DOM behavior etc., be?

[11:00:00.0000] <Lachy>
probably not, I doubt the contentious issues are still going to come up in comments until they get their own way (note, even in the design principles survey, people used it as a platform to voice their concern about various contentious issues)

[11:00:01.0000] <Hixie>
MikeSmith: the topic of discussion is not why the discussion is unproductive

[11:00:02.0000] <Lachy>
s/doubt/think/

[11:00:03.0000] <MikeSmith>
Hixie - what is the reason, then?

[11:02:00.0000] <MikeSmith>
Hixie - in going through the list of feedback, you're of course not under any obligation to do it in serial fashion. You can prioritize what to focus on, and I'm sure you do already, anyway.

[11:02:01.0000] <MikeSmith>
You can choose to make responding to feedback about conformance a lower priority if you want.

[11:03:00.0000] <Hixie>
actually my priorities tend to be strongly influenced by what browser vendors are working on

[11:03:01.0000] <Hixie>
i regularly get private requests to fast track a particular feature area because they are going to fix bugs in that area shortly

[11:03:02.0000] <Hixie>
it's one of the best ways to ensure that the browsers actually follow the spec

[11:04:00.0000] <MikeSmith>
Hixie - right, which would seem to make me think most of the conformance criteria is not a priority at all

[11:04:01.0000] <Hixie>
(and one of the vendors in question is henri, who's writing a conformance checker, so conformance is a priority sometimes)

[11:05:00.0000] <MikeSmith>
So Henri's conformance checker is if equal priority to you as browser implementations are?

[11:07:00.0000] <Hixie>
Henri's goodwill is of high priority to me, yes

[11:07:01.0000] <Hixie>
he has been a huge help to the spec

[11:08:00.0000] <MikeSmith>
I recognize that of course

[11:09:00.0000] <MikeSmith>
But the issue is not Henri personally, it's whether at this point the needs of having a spec that's useful for building a conformance checker are of the same priority as having a spec that's useful for solving browser interoperability problems

[11:10:00.0000] <MikeSmith>
Or the relative market demand for a interoperable browsers vs. conformance checkers

[11:11:00.0000] <Hixie>
it's all important

[11:20:00.0000] <Hixie>
nn

[11:20:01.0000] <Lachy>
bye hixie

[11:25:00.0000] <MikeSmith>
so, about some of the most recent messages to public-html on the Offline Web Apps thread: If, hypothetically, you see somebody posting a massive number of long messages to mailing list over a long period of time, and as far as you can tell that person's day job -- professional/educational responsibilities, whatever -- doesn't seem to related at all to the HTML work or even to Web technologies in general...

[11:26:00.0000] <MikeSmith>
...it sorta might make one wonder if they are actually doing the job that their employer/advisor/whoever is actually paying/expecting them to be doing right now

[11:27:00.0000] <MikeSmith>
And if so, where they actually find the time to do that, between the sending of huge volumes of e-mail

[11:29:00.0000] <MikeSmith>
But I guess it would be none of my business, anyway, to question such a person. If such were to exist, hypothetically

[12:00:00.0000] <nickshanks>
how are things in here? sorry i've not been around for a while, have been burdened by getting a job

[12:31:00.0000] <Hixie>
John just replied to the thread I told him to not forward to whatwg.

[12:31:01.0000] <Hixie>
/me goes to cycle to work to cool off and not explode at him

[12:33:00.0000] <nickshanks>
molly was talking about lots of angst on her blog too. what happened?

[12:34:00.0000] <takkaria>
nickshanks: lots and lots of misunderstanding and not listening

[13:23:00.0000] <Dashiva>
So, where do we get our "Member of the IRC cabal" t-shirts?

[13:24:00.0000] <Philip`>
Was http://www.spreadshirt.com/shop.php?sid=108788 popular enough to justify a new design?

[13:25:00.0000] <othermaciej>
I just kind of exploded at him

[13:26:00.0000] <Dashiva>
I never liked that one, Philip`. It's too obviously true for peopel to ponder the implications

[13:27:00.0000] <Philip`>
It's so obviously true that people will ponder the implications of why you felt the need to express that statement on your clothing

[13:28:00.0000] <jgraham>
I hope John realises he is not doing the term "accessibility advocate" any favours

[13:28:01.0000] <hober>
I get asked about it every single time I wear it.

[13:28:02.0000] <gsnedders>
/me still hasn't ordered one

[13:28:03.0000] <hober>
"So what's the deal with your shirt? Of course 5 is greater than 2."

[13:29:00.0000] <gsnedders>
I'd just laugh.

[13:29:01.0000] <gsnedders>
as myself and othermaciej argued, the joke is lost if you actually give it any context, and it becomes too geeky

[13:30:00.0000] <Dashiva>
jgraham: As indicated earlier, I really like "IRC cabal" however :)

[13:31:00.0000] <Philip`>
Is it really a cabal if it publishes logs of everything that goes on?

[13:32:00.0000] <gsnedders>
Philip`: we're totally deniable.

[13:33:00.0000] <gavin_>
Philip`: are you looking to start log-editing conspiracies? ;)

[13:36:00.0000] <gsnedders>
meh. even the small 5 > 2 shirt might be too big

[13:36:01.0000] <Philip`>
It seems quite helpful to have a group of people who tend to think along similar lines and are happy to communicate with each other frequently, in terms of making progress in some direction

[13:37:00.0000] <Philip`>
(It's much less good if that's a bad direction; but I'd think it's still better than making no progress at all)

[13:47:00.0000] <jgraham>
Philip`: The htmlwg is making the best attempt it can at making no progress at all

[13:48:00.0000] <hober>
not to promote my own joke too much, but we could always make "vast, browser-wing conspiracy" t-shirts (http://www.mail-archive.com/www-archive⊙wo/msg00554.html)

[13:48:01.0000] <hober>
maybe that's too US-centric?

[13:49:00.0000] <gavin_>
heh

[14:08:00.0000] <Dashiva>
The implementati

[14:35:00.0000] <Hixie>
/me sighs deeply

[14:36:00.0000] <Hixie>
i didn't want to do that

[14:36:01.0000] <Hixie>
i wish people would not ignore my requests and not insult the whatwg community

[14:37:00.0000] <Hixie>
on an unrelated note, gsnedders: if you leave the htmlwg please do remain in the whatwg, your feedback is very useful

[14:37:01.0000] <gsnedders>
Hixie: I have no intention of leaving WHATWG

[14:37:02.0000] <Hixie>
cool

[14:38:00.0000] <gsnedders>
I would not be amazed if the WG needs its charter re-examined due to implementers pulling out

[14:38:01.0000] <Hixie>
that would be so sad

[14:39:00.0000] <gsnedders>
On an unrelated note, I've started work on an I-D regarding HTTP request/response parsing

[14:39:01.0000] <Hixie>
interesting

[14:39:02.0000] <Hixie>
good luck with that

[14:40:00.0000] <Hixie>
the http community is not one that really believes in defined error handling

[14:40:01.0000] <Hixie>
or in caring about browsers

[14:40:02.0000] <gsnedders>
look at RFC2616, which already has a section on tolerant applications

[14:40:03.0000] <Hixie>
(see, e.g. the cool reception that the suggestion to remove Content-Location received)

[14:40:04.0000] <Hixie>
yeah

[14:40:05.0000] <Hixie>
but they don't like defining what that means, in my experience :-)

[14:41:00.0000] <othermaciej>
gsnedders: it would be sad if a minority of abusive people could drive out useful contributors; I hope it doesn't come to that

[14:41:01.0000] <gsnedders>
I've put various things like that (cont-location) in an informative appendix, as it is really out of scope

[14:41:02.0000] <othermaciej>
what's the difference between Location and Content-Location supposed to be?

[14:41:03.0000] <Hixie>
Content-Location is like <base href>

[14:41:04.0000] <Hixie>
except a huge number of servers send out bogus values

[14:41:05.0000] <Hixie>
so implementing it per spec breaks sites

[14:41:06.0000] <Hixie>
a lot of them

[14:41:07.0000] <gsnedders>
IIS bug. documented by MS.

[14:42:00.0000] <othermaciej>
oh

[14:42:01.0000] <Hixie>
iirc it's not just IIS, but yeah

[14:42:02.0000] <gsnedders>
Hixie: I looked through various things before pulling support for it myself, I never saw anything that wasn't IIS with the bug

[14:42:03.0000] <Hixie>
hm ok

[14:42:04.0000] <Hixie>
what did you pull support from?

[14:43:00.0000] <Hixie>
/me goes to get food and will then write up the offline web apps idea in the spec

[14:43:01.0000] <gsnedders>
a _really_ shit implementation of HTTP/1.0 thrown together really quickly for a feed library, namely SimplePie

[14:43:02.0000] <Hixie>
so if you have comments on that, especially "it doesn't work", mention them soon :-)

[14:43:03.0000] <Hixie>
gsnedders: aah

[14:43:04.0000] <G0k>
Hixie: i had a thought about the offline web apps thing. For multi-page apps, could you use <link> elements? like <link rel="application resource" href="someotherpage.html" />

[14:43:05.0000] <othermaciej>
Hixie: I think there's lots of smaller issues besides the pretty big one (IMO) of not really working for multi-page apps

[14:43:06.0000] <othermaciej>
Hixie: I don't think I'll have time to review it in detail today though

[14:44:00.0000] <Hixie>
i think aaron's idea (to just have an API to initiate the caching of another top-level page app) solves the multipage issue

[14:44:01.0000] <G0k>
like a DOM API?

[14:44:02.0000] <othermaciej>
Hixie: you can't cache the other page correctly without executing all scripts in it

[14:44:03.0000] <Hixie>
G0k: yeah, though we could have a declarative thing too

[14:44:04.0000] <othermaciej>
that seems obviously unacceptable

[14:44:05.0000] <Hixie>
othermaciej: sure, you just use the background window thing just like the update process

[14:44:06.0000] <gsnedders>
the bug is fixed in IIS7, from what I can see on the web

[14:44:07.0000] <Hixie>
what is obviously unacceptable?

[14:45:00.0000] <othermaciej>
Hixie: how does the background window avoid executing scripts?

[14:45:01.0000] <Hixie>
othermaciej: why would it need to avoid executing scripts?

[14:45:02.0000] <G0k>
i feel like anything that depends on scripting being executed makes this overly complicated

[14:45:03.0000] <Hixie>
othermaciej: you want it to run scripts.

[14:45:04.0000] <G0k>
what if you want a web app that's static?

[14:45:05.0000] <Hixie>
not much of an app, then

[14:45:06.0000] <Hixie>
normal http caching already solves the "offline web page" problem

[14:45:07.0000] <othermaciej>
Hixie: executing all scripts on a 50-page site just to cache all 50 pages seems like a large and probably unacceptable performance cost

[14:46:00.0000] <Hixie>
othermaciej: what web app uses 50 pages?

[14:46:01.0000] <G0k>
right but http caching doesn't do multi-page stuff

[14:46:02.0000] <Hixie>
G0k: sure, just crawl the site. even IE4 can do that.

[14:46:03.0000] <othermaciej>
Hixie: something that used history state with URI updating could easily use a lot more than 50 (where nearly all resources are actually shared)

[14:46:04.0000] <othermaciej>
Hixie: I like the seductive simplicity of your idea

[14:47:00.0000] <othermaciej>
Hixie: but I think many more complex cases are much better addressed with an explicit manifest of some kind

[14:47:01.0000] <othermaciej>
the manifest could still be in the markup

[14:47:02.0000] <othermaciej>
and you could recognize that multiple pages should use the same cache by the fact that they link the same manifest

[14:47:03.0000] <othermaciej>
then you can pre-cache a bunch of stuff without the need for redundant script execution or HTML parsing

[14:47:04.0000] <Hixie>
othermaciej: none of the applications i looked at (admittedly mostly google ones) actually need multipage caching, i'm somewhat reluctant to support that without some pretty concrete examples.

[14:48:00.0000] <Hixie>
especially given that it makes it more complex by an order of magnitude for pretty much everyone

[14:48:01.0000] <Hixie>
brb fod

[14:48:02.0000] <Hixie>
fod

[14:48:03.0000] <G0k>
yeah i mean i feel like we should have a real semantic way of saying "I want you to cache this other page"

[14:48:04.0000] <Hixie>
foOd

[14:48:05.0000] <G0k>
or maybe more simply

[14:48:06.0000] <othermaciej>
there's lots of multi-page web apps out there

[14:48:07.0000] <othermaciej>
I'm not sure how many would benefit strongly from an offline mode

[14:48:08.0000] <G0k>
"This other page should have the same caching policy as me"

[14:48:09.0000] <othermaciej>
flickr and upcoming are two very popular and obvious examples

[14:49:00.0000] <G0k>
how about word processor?

[14:49:01.0000] <G0k>
google write or whatever

[14:49:02.0000] <othermaciej>
digg and reddit would also be examples of multi-page web apps

[14:49:03.0000] <G0k>
why would you want digg offline?

[14:50:00.0000] <gsnedders>
Basecamp?

[14:50:01.0000] <othermaciej>
G0k: maybe I visit it when offline, and then I want to be able to read offline at any time and see more than the front page

[14:50:02.0000] <gsnedders>
A lot of 37 Signals stuff could do with being usable offline

[14:51:00.0000] <G0k>
why not just have the server properly give cache control instructions to the UA so that it does that?

[14:52:00.0000] <othermaciej>
you mean http cache-control?

[14:53:00.0000] <G0k>
yeah

[14:54:00.0000] <othermaciej>
cache-control doesn't give a way to get the UA to use possibly-stale versions of the content when offline

[14:54:01.0000] <othermaciej>
nor to update a group of resources atomically

[14:54:02.0000] <G0k>
doesn't that what offline mode does?

[14:55:00.0000] <G0k>
*isn't

[14:55:01.0000] <G0k>
hm

[14:56:00.0000] <G0k>
i guess cache-control doesn't have a way to say "cache me indefinitely"

[14:57:00.0000] <G0k>
although you could say....make it not expire for 1000 years

[14:57:01.0000] <G0k>
but then even then, the UA could decide not to cache anyway

[14:57:02.0000] <Hixie>
that'd be bad, the client might never check that the file had changed :-)

[14:58:00.0000] <G0k>
well i feel like people should start changing the entire model of their app designs

[14:58:01.0000] <G0k>
rather than sending a huge chunk of content + design

[14:58:02.0000] <G0k>
just send an "empty" shell of the app

[14:58:03.0000] <G0k>
then fetch content seperately

[14:58:04.0000] <G0k>
thus the shell part never expires

[14:59:00.0000] <Hixie>
/me looks at flickr, upcoming, digg, reddit

[14:59:01.0000] <G0k>
and doesn't have to keep getting fetched

[14:59:02.0000] <G0k>
maybe use event-source to get new stories. :)

[14:59:03.0000] <Hixie>
heh

[14:59:04.0000] <takkaria>
I don't see that flickr is a useful thing to use offline

[14:59:05.0000] <Hixie>
G0k: http://www.digg.com/spy

[15:00:00.0000] <G0k>
hixie: yeah i think they using polling

[15:00:01.0000] <G0k>
or the long download technique

[15:01:00.0000] <Hixie>
that's what whatwg.org/issues/top uses

[15:01:01.0000] <Hixie>
i hate it

[15:01:02.0000] <Hixie>
bring on TCPConnection

[15:01:03.0000] <G0k>
ok that's a question i have too

[15:01:04.0000] <G0k>
what's the point of making TCPConnection use that weird protocol

[15:01:05.0000] <G0k>
?

[15:02:00.0000] <G0k>
why not make it a real raw TCP connection?

[15:02:01.0000] <takkaria>
security risks?

[15:02:02.0000] <G0k>
such as?

[15:02:03.0000] <Hixie>
yeah, you can't possibly have raw sockets flying about

[15:03:00.0000] <takkaria>
well, what's the point in having cross-domain security restrictions if you can just write your own HTTP client in JS with none of them?

[15:03:01.0000] <G0k>
well not a raw socket, just a pure tcp connection

[15:03:02.0000] <Hixie>
what takkaria said

[15:03:03.0000] <Hixie>
G0k: that's what i meant

[15:03:04.0000] <G0k>
the TCP connections could still have the same domain security policy stuff

[15:03:05.0000] <Hixie>
how?

[15:04:00.0000] <Hixie>
TCP connections are IP-bound, not domain-bound

[15:04:01.0000] <Hixie>
hixie.ch and whatwg.org and damowmow.com are all at the same IP

[15:04:02.0000] <G0k>
well ok, IP security at least

[15:04:03.0000] <G0k>
so then what?

[15:05:00.0000] <gsnedders>
see y'all tomorrow (and we'll see whether I'll still be in the WG then)

[15:05:01.0000] <othermaciej>
same-domain isn't good enough since it still allows cross-protocol attacks

[15:05:02.0000] <othermaciej>
the only way to offer raw TCP sockets that I've heard which is at all workable is to have a central control file served by HTTP from the server

[15:06:00.0000] <G0k>
an evil person could hijack hixie.ch and then send TCP requests to some other service on that server?

[15:06:01.0000] <othermaciej>
which would control access

[15:06:02.0000] <othermaciej>
G0k: currently http servers running on two different ports on the same system are considered separate security domains

[15:07:00.0000] <othermaciej>
takkaria: subsets of flickr could be useful offline - viewing photos I've viewed before, queuing new photos, descriptions and comments for upload on next connection, etc

[15:07:01.0000] <G0k>
well this certainly wouldn't be any less secure than Java or Flash, which already allow you to connect to arbitrary services on the document's origin IP

[15:07:02.0000] <othermaciej>
takkaria: it has basically the same offline usefulness as gmail

[15:08:00.0000] <takkaria>
mm, OK

[15:08:01.0000] <takkaria>
I use a dedicated photo upload tool so I don't get that so much

[15:08:02.0000] <G0k>
i mean inventing this new protocol dramatically limits the usefulness of this thing

[15:09:00.0000] <G0k>
plus it's a complete misnomer

[15:09:01.0000] <G0k>
it's not a TCPConnection

[15:09:02.0000] <Hixie>
yeah, the name should change

[15:09:03.0000] <Hixie>
the great thing about the way it works now is that we can allow arbitrary cross-domain connections

[15:09:04.0000] <takkaria>
I had visions of implementing an IRC client when I first read that name. :)

[15:09:05.0000] <G0k>
well that may be a useful feature too

[15:09:06.0000] <othermaciej>
well, if you have a local photo store and a native photo upload tool, then obviously you have less need for your web-based photo gallery to work offline

[15:09:07.0000] <G0k>
but a real TCPConnection  would be neat too

[15:10:00.0000] <othermaciej>
but if you want to use your photo sharing site as the *only* place you manage your photos, then you do want it to work offline

[15:11:00.0000] <takkaria>
othermaciej: yeah.  fair enough. :)

[15:12:00.0000] <G0k>
but yeah i mean that dom server side events thing was like...how i originally found WHAT WG

[15:12:01.0000] <G0k>
working on an app which needs something just like that

[15:13:00.0000] <aa>
with implicit gathering of resources for the offline application, how do apps request resources that they want to come from the server while loading?

[15:13:01.0000] <aa>
as a simple example lots of offline apps request an image from the server during load as a health check

[15:16:00.0000] <Hixie>
aa: you mean as opposed to the scriptable API for adding a file to the cache?

[15:17:00.0000] <G0k>
has that already been specified

[15:17:01.0000] <G0k>
?

[15:17:02.0000] <Hixie>
no

[15:17:03.0000] <aa>
I mean the opposite: you don't want something to go to the cache

[15:17:04.0000] <aa>
which is a GET

[15:17:05.0000] <G0k>
you use no-cache in the HTTP headers for that?

[15:17:06.0000] <Hixie>
ah

[15:18:00.0000] <Hixie>
yeah, there's no way to do a GET that doesn't hit the cache in the current idea

[15:18:01.0000] <Hixie>
you can do a POST

[15:18:02.0000] <aa>
seems heavy handed

[15:18:03.0000] <G0k>
maybe you could add a delete-from-cache API?

[15:18:04.0000] <Hixie>
aa: well, what's the use case?

[15:18:05.0000] <aa>
plus it means you can't use ye old new Image() trick

[15:18:06.0000] <Hixie>
if you want to ping the server, getting an image seems like not the right thing to do :-)

[15:19:00.0000] <G0k>
should add an ICMPConnection. what could posssibly go wrong?

[15:19:01.0000] <aa>
the use case is to determine whether your server is reachable

[15:19:02.0000] <aa>
what reader and gearpad do is request x.gif?r=<rand>

[15:19:03.0000] <aa>
probably that is Bad, but there it is

[15:19:04.0000] <Hixie>
aa: well we can have an explicit API for that

[15:19:05.0000] <Hixie>
aa: that seems like something you'd want anyway

[15:20:00.0000] <G0k>
i think that's a spendid idea

[15:20:01.0000] <aa>
Hixie: there are other use cases besides that

[15:20:02.0000] <G0k>
*splendid

[15:20:03.0000] <Hixie>
aa: like what?

[15:20:04.0000] <aa>
Hixie: What if you want to talk to your server to synchronize during load.

[15:21:00.0000] <aa>
not wait until onload

[15:21:01.0000] <aa>
you have to use POST?

[15:23:00.0000] <Hixie>
aa: like to get the data?

[15:24:00.0000] <aa>
yes

[15:24:01.0000] <Hixie>
aa: wouldn't you just do a normal GET, and not care if you're offline or not?

[15:25:00.0000] <Hixie>
hm

[15:25:01.0000] <aa>
So the use case I'm imagining is using xmlhttprequest

[15:25:02.0000] <Hixie>
though i see what you mean

[15:25:03.0000] <aa>
maybe it could have a property to bypass the offline cache

[15:25:04.0000] <Hixie>
you might want to have the data not cached

[15:25:05.0000] <Hixie>
yeah

[15:25:06.0000] <aa>
i think that would be enough

[15:25:07.0000] <Hixie>
hmm

[15:25:08.0000] <Hixie>
yeah

[15:25:09.0000] <Hixie>
maybe that's the way to do it

[15:26:00.0000] <aa>
another idea: maybe requests originating from javascript don't count for the auto-gathering

[15:26:01.0000] <aa>
only html tags literally in the source

[15:27:00.0000] <aa>
that seems weird too though :-)

[15:27:01.0000] <Hixie>
yeah i don't think i like that

[15:31:00.0000] <othermaciej>
it seems like when you consider all the details, implicit gathering might not be any easier to use than an explicit manifest

[15:32:00.0000] <Hixie>
it's conceptually easy, you just pass everything to the app cache instead of your main cache

[15:33:00.0000] <othermaciej>
I mean easier for the web app author - if the set of things they want cached needs to include items that aren't loaded and exclude items that are, then the implicit gathering plus API scattered around their app code might not be easier to use than a separate manifest file

[15:34:00.0000] <othermaciej>
the manifest file can also allow low-cost checking for updates

[15:34:01.0000] <aa>
FWIW, we have gotten pretty strong feedback that people don't like the manifest.

[15:34:02.0000] <othermaciej>
how do they want it to work? all automatic?

[15:35:00.0000] <othermaciej>
there's no way an all-automatic system can check for an update with a single conditional GET

[15:35:01.0000] <othermaciej>
afaict

[15:35:02.0000] <aa>
I think part of the problem is that the manifest is confusing during development because everything just works so long as you have a connection or you have the resources cached.

[15:35:03.0000] <aa>
you don't find out that you forgot something until later.

[15:35:04.0000] <aa>
i had proposed a mode where requests that are *not* in the manifest fail, but that of course brings up other questions.

[15:36:00.0000] <othermaciej>
requests that are not in the manifest should fail when offline

[15:36:01.0000] <othermaciej>
that makes it easy to test offline mode by actually going offline

[15:38:00.0000] <aa>
it's not easy because most of the time during development you are not offline. also you have to be careful to clear your cache otherwise you still don't see it.

[15:39:00.0000] <othermaciej>
yes, I'm saying you should not be able to get stuff from the cache when offline (at least when using a web app with a manifest)

[15:39:01.0000] <aa>
what does "offline" mean?

[15:39:02.0000] <aa>
file > offline?

[15:40:00.0000] <aa>
maybe this is just a developer feature ... to not have the regular browser cache be consulted

[15:40:01.0000] <othermaciej>
I don't understand your question

[15:41:00.0000] <othermaciej>
also I need to step our for a bit, brb

[15:41:01.0000] <aa>
at least in gears there is no concept of offline. so I dont know what you mean when you say the browser should change behavior when offline.

[15:41:02.0000] <othermaciej>
Safari already won't load anything from the cache if your network is disconnected (maybe not the best behavior but we have not really gotten complaints)

[15:42:00.0000] <kingryan>
othermaciej: I'd like to complain

[15:42:01.0000] <kingryan>
:)

[15:42:02.0000] <aa>
k, we can continue this later.

[15:42:03.0000] <aa>
i don't want to keep you.

[15:43:00.0000] <othermaciej>
one random parting thought before I step out

[15:43:01.0000] <Hixie>
note that "offline" is somewhat nebulous -- you can be "online" (from the browser's point of view) but really your network is gonna be returning 302s everywhere (captive portal) or 500s (server down) or not resolving dns (wireless down) etc

[15:43:02.0000] <Hixie>
i tried to handle that case too

[15:43:03.0000] <aa>
Hixie: I think that is really important. Thanks.

[15:44:00.0000] <aa>
(as a frequent user of unreliable wireless)

[15:44:01.0000] <othermaciej>
maybe it can be an option for the manifest to say all resources from the current page load should be included, but list exceptions and additions

[15:44:02.0000] <Hixie>
aa: yeah, from a personal perspective i know the feeling. :-)

[15:44:03.0000] <Hixie>
othermaciej: i certainly don't mind there being a way to list resources declaratively (as well as through an API)

[15:44:04.0000] <othermaciej>
Hixie: normal operating mode for the user handling that case is key, but I don't think that has to influence what happens when you pull your ethernet cable to test offline behavior

[15:45:00.0000] <Hixie>
othermaciej: but i think that requiring a manifest is different from supporting a manifest as a secondary resource that causes loads

[15:45:01.0000] <othermaciej>
Hixie: my brief mailing list proposal was to have an optional manifest (that would in practice be needed for multipage web apps)

[15:46:00.0000] <othermaciej>
w/ the additional semantic that sharing a manifest makes you share an offline cache

[15:46:01.0000] <Hixie>
yeah i'm still studying the multi-page case

[15:46:02.0000] <Hixie>
i'm not sure i like the idea of having more than one "top-level page" per cache, but i'm still thinking about it

[15:47:00.0000] <othermaciej>
I'm also not sure if in your proposal "subresources of the page" only includes things loaded up to onload time, or also things loaded later

[15:47:01.0000] <othermaciej>
(either answer would raise issues I think)

[15:47:02.0000] <othermaciej>
anyway

[15:47:03.0000] <othermaciej>
bbl

[15:48:00.0000] <Hixie>
ttyl

[15:48:01.0000] <Hixie>
it's supposed to include anything you hit while online, the load event is only used when checking for an update to prime the new cache

[15:58:00.0000] <Lachy>
good morning

[15:59:00.0000] <Hixie>
hey Lachy 

[16:00:00.0000] <gsnedders>
hi Lachy 

[16:01:00.0000] <Lachy>
Hixie, that's a nice way to deal with trolling on the whatwg list :-)

[16:02:00.0000] <Hixie>
he ignored my request

[16:02:01.0000] <Lachy>
yeah, I noticed

[16:02:02.0000] <Hixie>
so yeah

[16:02:03.0000] <Hixie>
/me gets good advice on how to handle trolls from his teammates at work

[16:02:04.0000] <Hixie>
(i work in the open source group, so they're all experts at handling communites)

[16:03:00.0000] <Hixie>
communities, even

[16:03:01.0000] <Hixie>
http://video.google.com/videoplay?docid=-4216011961522818645 being a good example

[16:04:00.0000] <Hixie>
come to think of it i think i was in that room when that video was taped

[16:04:01.0000] <Lachy>
I'm wondering if I should respond to Joe's off list email to us re our IRC commetns, and suggest that if he has any alternative solutions (instead of just complaints), that he should send them.

[16:05:00.0000] <Hixie>
did i get that e-mail? i forget

[16:05:01.0000] <Lachy>
hmm. is hixie @ google.com a valid email address for you? I wasn't aware of it, but that's where joe sent it

[16:06:00.0000] <Hixie>
no, that won't work

[16:06:01.0000] <Hixie>
so i guess i didn't

[16:06:02.0000] <Hixie>
ian⊙hc is my address

[16:06:03.0000] <Lachy>
I just saw hixie in the CC list and assumed it got to you. I'll forward it

[16:07:00.0000] <Hixie>
no thanks!

[16:07:01.0000] <Lachy>
oh, too late

[16:07:02.0000] <Hixie>
oh well :-)

[16:07:03.0000] <Hixie>
and there i was with an excuse

[16:09:00.0000] <Hixie>
i see he didn't mail mike

[16:10:00.0000] <Hixie>
which would have made him realise that what he was calling the whatwg actually was the w3c in that case :-)

[16:20:00.0000] <Hixie>
ok. since clearly the offline web apps stuff is still being discussed, let's do the datatemplate stuff hyatt and i came up with

[16:22:00.0000] <Lachy>
what's datatemplate?

[16:23:00.0000] <Hixie>
the replacement for repetition blocks

[16:24:00.0000] <othermaciej>
replacement for WF2 repetition model but based on XUL templates so it provides much more of the general capabilities that XForms repetition does

[16:24:01.0000] <Hixie>
yeah

[16:24:02.0000] <othermaciej>
inspired by comment from John Boyer

[16:24:03.0000] <Hixie>
(though it requires external xml files sadly, unless you use xhtml, or unless you are very careful with html)

[16:25:00.0000] <othermaciej>
I am shocked to see this sort of paying attention to feedback

[16:25:01.0000] <Lachy>
wow

[16:25:02.0000] <othermaciej>
we'll have to discuss this at the next cabal meeting

[16:25:03.0000] <Hixie>
dude i have to pay attention to _hyatt_

[16:25:04.0000] <Hixie>
he's one of the _editors_

[16:25:05.0000] <Hixie>
he's got a jet pack and everything!

[16:26:00.0000] <Lachy>
why haven't you dismissed that feedback like we do with everything else? ;-)

[16:26:01.0000] <takkaria>
othermaciej: oh, didn't you hear?  #whatwg is always a cabal meeting

[16:26:02.0000] <othermaciej>
takkaria: yes, that's what we want you to... oops, I've said to much

[16:27:00.0000] <Hixie>
( http://ajaxian.com/archives/proposal-for-the-w3c-to-adopt-html-5#comment-249313 )

[16:27:01.0000] <takkaria>
othermaciej: oh, so there's another backroom irc channel around? :P

[16:27:02.0000] <Hixie>
amusingly ajaxian using <dt> and <dd> backwards

[16:27:03.0000] <Hixie>
i wish

[16:27:04.0000] <Hixie>
if there was we wouldn't fear people quoting us all the time :-P

[16:36:00.0000] <Lachy>
I like how we get accused of having a secret cabal (in a publicly logged IRC channel), yet it's ok for there to be a clear conspiracy among some others to file provocative bug reports and complain

[16:37:00.0000] <jgraham>
Lachy: I think there's not really a conspiracy to complain but a set of individuals who are motivated to do negative things for whatever reason

[16:38:00.0000] <takkaria>
I think they just haven't seen the light yet...

[16:38:01.0000] <Lachy>
yeah, well, they appear to be working well together to achieve their common goals

[16:40:00.0000] <Hixie>
i very highly doubt there's a conspiracy there

[16:40:01.0000] <jgraham>
If their common goal is "disrupt the WG" then yeah. But I don't think there's a real agenda, except perhaps a common feeling that they want to exert more influence

[16:40:02.0000] <jgraham>
(that is not supposed to imply that they do not exert influence, only that they feel they do not)

[16:41:00.0000] <Hixie>
they clearly exert a lot of influence, they've been the topic of this channel for days :-)

[16:41:01.0000] <Hixie>
much more so than any technical issue has

[16:41:02.0000] <Philip`>
Lachy: Why is that a clear conspiracy? I assume they just read the IRC logs and find links to the existing bug reports and then maybe add their own comments

[16:43:00.0000] <Lachy>
I was just basing my conspiracy theory on othermaciej's email, I wasn't entirely serious.

[16:43:01.0000] <Lachy>
this one http://www.w3.org/mid/DA234C3B-59E7-4E9E-A779-00181EDF06BE⊙ac

[16:44:00.0000] <Hixie>
i think he was applying reducto ad absurdum

[16:44:01.0000] <Hixie>
not suggesting there actually was a conspiracy :-)

[16:44:02.0000] <othermaciej>
I was trying to

[16:44:03.0000] <othermaciej>
but apparently I was more right than I thought :-)

[16:45:00.0000] <Hixie>
oh?

[16:45:01.0000] <Hixie>
why?

[16:45:02.0000] <othermaciej>
well I don't think the whole thing was staged

[16:46:00.0000] <othermaciej>
but Rob did privately complain about my remark to John and the chairs, without even Cc'ing me

[16:46:01.0000] <Hixie>
aah

[16:46:02.0000] <Hixie>
so many private complaints

[16:46:03.0000] <Hixie>
how are we supposed to follow the gossip if they don't cc www-archive

[16:46:04.0000] <Dashiva>
They should take a hint from us and do their secret scheming in public logs

[16:47:00.0000] <jgraham>
/me waves at all the log readers

[16:47:01.0000] <takkaria>
I don't see how people can complain about transparency and make private complaints at the same time

[16:47:02.0000] <Lachy>
it was suggested to me once by someone a few days ago that we set up a special IRC channel for such people :-)

[16:48:00.0000] <Dashiva>
I recall one such channel

[16:49:00.0000] <Dashiva>
#whatwg-secret-tree-house-no-patents-allowed

[16:49:01.0000] <othermaciej>
if the private complaints at least were used instead of private complaints, that might be a net benefit

[16:49:02.0000] <othermaciej>
*instead of public complaints

[16:49:03.0000] <Hixie>
hah

[16:49:04.0000] <takkaria>
/me grins

[16:49:05.0000] <othermaciej>
having both is the worst of both worlds

[16:51:00.0000] <Dashiva>
It's too bad this is actually serious business. Could have a lot of fun setting up a fake-logged cabal channel

[16:53:00.0000] <Hixie>
/me updates himself with the controversy on www-archive

[16:53:01.0000] <Hixie>
you know what bugs me more than whiners?

[16:54:00.0000] <Hixie>
whiners who complain about things that are actually non-issues (e.g. nobody has actually complained that they are forced to use an AT tool that doesn't work with the issues page)

[16:54:01.0000] <Hixie>
and even more than that

[16:54:02.0000] <Hixie>
whiners who complain about things yet are not willing to step up to the plate and actually address them

[16:55:00.0000] <takkaria>
/me avoids commenting at all for that last reason. :)

[16:55:01.0000] <Hixie>
heh

[16:55:02.0000] <Hixie>
i don't mind people making suggestions

[16:55:03.0000] <Hixie>
"it would be better if it could support IE"

[16:55:04.0000] <Hixie>
it's people who whine that i find annoying

[16:55:05.0000] <takkaria>
yeah

[16:55:06.0000] <Hixie>
"it sucks because it doesn't support IE"

[16:56:00.0000] <takkaria>
I have nothing to add that other people don't mention at some point, anyway.  I don't even do much web-based stuff anymore

[16:56:01.0000] <Hixie>
heh

[16:56:02.0000] <takkaria>
though I quite enjoyed looking through longdesc stuff the other day

[16:56:03.0000] <takkaria>
I signed up to look at spec development, not politics.  it's sad that things are this bad, really

[16:57:00.0000] <Hixie>
yeah

[16:58:00.0000] <Philip`>
Steve Faulkner indicated in one email that he would like to help fix the tool (though he couldn't do it easily since he didn't know how), which sounds positive

[16:58:01.0000] <Hixie>
yeah, that went nowhere quickly


2007-08-25
[17:02:00.0000] <zcorpan>
/me wonders if the tool would work better in ie if some script from dean edwards was applied

[17:03:00.0000] <zcorpan>
like "ie7" or base2 or something -- i haven't looked closely at either script nor the tool

[17:04:00.0000] <Hixie>
given that as far as i can tell it's not inconveniencing any actual users, i don't see much point in making the page less standards compliant

[17:06:00.0000] <Hixie>
/me gens a version of the spec with three extra elements (without defining them yet)

[17:07:00.0000] <takkaria>
/me sighs at twitter for being crap

[17:07:01.0000] <Hixie>
twitter's so unreliable

[17:07:02.0000] <Hixie>
is there another system that also supports IM, sms, and rest api interaction?

[17:07:03.0000] <Hixie>
meeting. bbl.

[17:08:00.0000] <takkaria>
for free?  I doubt it

[17:09:00.0000] <Philip`>
From a position of knowing exactly nothing about it, <nest> sounds like a fairly ambiguous name

[17:24:00.0000] <Dashiva>
So is it only IEx that doesn't support the issues page, or are other browsers (possibly older versions) also unable to use it?

[17:27:00.0000] <zcorpan>
jgraham: nice with the table inspector. i might look into writing test pages with tables that we like the algorithm to work with -- based on ben's research -- so we can do rapid prototyping of alternative algorithms

[17:33:00.0000] <Philip`>
Dashiva: The voting doesn't really work in Konqueror

[17:33:01.0000] <Philip`>
(It does count the vote, but doesn't update the screen after voting)

[17:38:00.0000] <Philip`>
(So, uh, I suppose that means the voting does really work, but it just has cosmetic issues)

[17:48:00.0000] <Dashiva>
Gonna have to set up a local server proxy for the listen and whatnot, I suppose... where are those cross-domain auth specs when you need them

[22:30:00.0000] <MikeSmith>
wondering if anybody may have looked at the OpenAjax "Hub" spec and have any thoughts on it

[22:30:01.0000] <MikeSmith>
http://www.openajax.org/OpenAjax%20Hub.html

[22:31:00.0000] <MikeSmith>
in particular:

[22:31:01.0000] <MikeSmith>
http://www.openajax.org/OpenAjax%20Hub.html#Example

[22:32:00.0000] <MikeSmith>
[[

[22:32:01.0000] <MikeSmith>
To make the application work, the developer registers a callback function that is invoked whenever the user selects a new date in the calendar widget. This callback function then broadcasts the new date event using the OpenAjax Hub's publish() function:

[22:32:02.0000] <MikeSmith>
]]

[22:32:03.0000] <MikeSmith>
[[

[22:32:04.0000] <MikeSmith>
Then the developer includes code such that each chart widget and data grid widget subscribes to the new date event and provides a callback function. The various callback functions then update the given visualization widget appropriately:

[22:32:05.0000] <MikeSmith>
]]

[22:33:00.0000] <MikeSmith>
[[

[22:33:01.0000] <MikeSmith>
The Hub's key feature is its publish/subscribe manager (the "pub/sub manager"). The pub/sub manager allows one part of a mashup to broadcast an event to which other application components subscribe. 

[22:33:02.0000] <MikeSmith>
]]

[12:36:00.0000] <Dashiva>
/me ponders the "structured options" email.

[12:44:00.0000] <Philip`>
/me wonders why white-space:pre doesn't work in <select>s in any browser

[12:44:01.0000] <Philip`>
(Is that a bug?)

[12:47:00.0000] <zcorpan>
a lot of stuff don't work with <select>s or form controls in general

[12:47:01.0000] <Philip`>
(Thinking of http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%0A%3Cselect%20style%3D%22font-family%3A%20monospace%3B%20white-space%3A%20pre%22%3E%0A%3Coption%20value%3D%221%22%3E4302WW%20%7C%20Red%20%20%20%20%20%20%20%7C%20%A310.99%0A%3Coption%20value%3D%222%22%3E2915II%20%7C%20Turquoise%20%7C%20%A310.99%0A%3C/select%3E in particular)

[12:48:00.0000] <Philip`>
Um, if anyone knows why that causes "Service Temporarily Unavailable", I'd be happy to find out

[12:51:00.0000] <Philip`>
Hixie: http://software.hixie.ch/utilities/js/live-dom-viewer/?%7C%20%20%7C says "Service Temporarily Unavailable" which is peculiar

[12:51:01.0000] <Philip`>
Anyway, I was thinking of http://canvex.lazyilluminati.com/misc/dom-viewer/?%3C%21DOCTYPE%20html%3E%0D%0A%3Cselect%20style%3D%22font-family%3A%20monospace%3B%20white-space%3A%20pre%22%3E%0D%0A%3Coption%20value%3D%221%22%3E4302WW%20%7C%20Red%20%20%20%20%20%20%20%7C%20%A310.99%0D%0A%3Coption%20value%3D%222%22%3E2915II%20%7C%20Turquoise%20%7C%20%A310.99%0D%0A%3C/select%3E

[12:53:00.0000] <zcorpan>
Philip`: encoding problem with the dead viewer?

[12:54:00.0000] <zcorpan>
http://canvex.lazyilluminati.com/misc/dom-viewer/reflect.cgi?%u263A for ☺

[12:55:00.0000] <Philip`>
Yes, but I don't know where since I tried to make everything just use UTF-8

[12:55:01.0000] <Philip`>
Oh, that won't work because the CGI program only understands %xx values

[12:55:02.0000] <zcorpan>
indeed, but that's what the js sends

[12:56:00.0000] <zcorpan>
it should send %E2%98%BA instead

[12:56:01.0000] <Philip`>
In FF3 it sends %A3

[12:56:02.0000] <Philip`>
(as far as I can see)

[12:56:03.0000] <Philip`>
(Uh, for £ symbols at least)

[12:57:00.0000] <zcorpan>
yeah, opera too

[12:58:00.0000] <zcorpan>
it's escape()

[12:59:00.0000] <zcorpan>
use encodeURIComponent() instead

[13:00:00.0000] <Philip`>
Oh, never heard of that before

[13:00:01.0000] <zcorpan>
and decodeURIComponent() instead of unescape()

[13:03:00.0000] <Philip`>
http://canvex.lazyilluminati.com/misc/dom-viewer/?%C3%A6%E2%80%9D%C2%A2%C3%B0e%C4%91%C5%8B%C4%A7%E2%86%92j%C4%B8%C5%82%C2%B5n%C3%B8%C3%BE%40%C2%B6%C3%9F%C5%A7%E2%86%93%E2%80%9C%C5%82%C2%BB%E2%86%90%C2%AB

[13:03:01.0000] <Philip`>
Seems to work - thanks :-)

[13:04:00.0000] <Philip`>
http://canvex.lazyilluminati.com/misc/dom-viewer/x.html?%3Chtml%20xmlns%3D%22http%3A%2F%2Fwww.w3.org%2F1999%2Fxhtml%22%3E%C3%A6%E2%80%9D%C2%A2%C3%B0e%C4%91%C5%8B%C4%A7%E2%86%92j%C4%B8%C5%82%C2%B5n%C3%B8%C3%BE%40%C2%B6%C3%9F%C5%A7%E2%86%93%E2%80%9C%C5%82%C2%BB%E2%86%90%C2%AB%3C%2Fhtml%3E - seems to work too

[13:16:00.0000] <gsnedders>
so the majority of HTML WG email was technical email (7/9ths of it was), so I stay

[13:31:00.0000] <Retrieving>
/me #whatwg modes...

[13:34:00.0000] <Philip`>
Do any browsers sniff text/plain content differently to unlabelled content? (I'm wondering how many web sites would break when upgrading to an Apache that doesn't send Content-Type: text/plain by default)

[14:08:00.0000] <Dashiva>
Philip`: I wouldn't expect any intentional differences, but perhaps that some forget to do parts of the sniffing because there hasn't been a need yet

[14:18:00.0000] <gsnedders_>
/me wonders how risky it'd be to ship the HTML 5 type sniffing


2007-08-26
[21:53:00.0000] <Lachy>
does anyone know if Alan Flavell's resources from http://ppewww.ph.gla.ac.uk/~flavell/ have been mirrored anywhere since they disappeared? Sadly, they appear to have been deleted from the internet archive as well http://web.archive.org/web/*/http://ppewww.physics.gla.ac.uk/~flavell/

[22:01:00.0000] <Lachy>
never mind, found them http://web.archive.org/web/*/ppewww.ph.gla.ac.uk/~flavell/

[22:11:00.0000] <Lachy>
hey, this page uses longdesc http://web.archive.org/web/20060903033519/http://ppewww.ph.gla.ac.uk/~flavell/alt/alt-text.html

[22:12:00.0000] <Lachy>
pointing to http://web.archive.org/web/20000814074740/ppewww.ph.gla.ac.uk/~flavell/alt/round-tuit.txt

[22:12:01.0000] <Lachy>
though, not even that looks like it's a particularly good use of longdesc

[22:39:00.0000] <Lachy>
this looks like a useful resource http://html.cita.uiuc.edu/text/

[22:43:00.0000] <Lachy>
woot! found an excellent example of using longdesc! http://www.webcredible.co.uk/webcreds/episode8.shtml :-)

[02:03:00.0000] <Hixie>
Lachy: no! it's not an excellent use! because there's a link on that page to the exact same page!!

[02:03:01.0000] <Hixie>
which is what i've been saying, if it's useful for longdesc, it should (and would) just be available to everyone in a normal link

[02:03:02.0000] <Lachy>
Hixie, yeah, I know it does, but it's still a good use of a long description

[02:04:00.0000] <Lachy>
the other example I included in my email to public-html only uses longdesc

[02:04:01.0000] <Hixie>
http://issues.apache.org/bugzilla/show_bug.cgi?id=13986

[02:04:02.0000] <Lachy>
this one http://www.dizabled.com/comics/stairs/

[02:05:00.0000] <Hixie>
cool

[02:05:01.0000] <Hixie>
if you're sending it to public-html make sure to put it in the wiki too :-)

[02:06:00.0000] <Lachy>
I knew I should have sent to whatwg! :-)

[02:07:00.0000] <Lachy>
(but I figured it would be better to lead by example and show others on public-html the kind of evidence and constructive emails to post)

[02:07:01.0000] <Hixie>
hehe

[02:15:00.0000] <Lachy>
Hixie, in that other link I provided above <http://html.cita.uiuc.edu/text/>, it lists a whole heap of image categories.  It might be useful if spec included examples of advertising banners and CAPTCHA images

[02:16:00.0000] <Lachy>
the spec seems to cover all the other relevant categories already

[02:19:00.0000] <Hixie>
cool

[02:19:01.0000] <Hixie>
did you mail that anywhere?

[02:19:02.0000] <Lachy>
no, but I will to whatwg

[02:19:03.0000] <Hixie>
cool thanks

[02:22:00.0000] <Hixie>
btw, i don't really ever use proposed replacement texts, because i usually find that to make them fit my style i have to make almost as many changes as just writing it from scratch based on the description of the proposla

[02:23:00.0000] <Lachy>
I tried to make it fit your style (I even used some parts of the existing text)

[02:24:00.0000] <Hixie>
yeah, but i suffer from NWH syndrome :-) (not-written-here)

[02:24:01.0000] <Lachy>
LOL! (me too) :-)

[02:25:00.0000] <Hixie>
hm, the feedback on progress events was sent to whatwg instead of webapi

[02:31:00.0000] <Lachy>
glad you noticed. I'm on so many lists, I barely notice which email gets sent to where

[02:32:00.0000] <Lachy>
though I did notice Dmitry Turin is sending more HTML feedback to www-style :-)

[02:43:00.0000] <Lachy>
the comments on the blog have turned into little more than ad hominem attacks http://blog.whatwg.org/omit-alt

[02:55:00.0000] <Lachy>
Hixie, Steve Faulkner mentioned this awesome list of resources about alt text. I know you've probably seen many of them already, but there's a few that I hadn't. http://www.d.umn.edu/itss/support/Training/Online/webdesign/accessibility.html#alt

[04:28:00.0000] <zcorpan>
/me wonders if the alt section should s/empty value/empty or placeholder value/

[05:07:00.0000] <Lachy>
zcorpan, what do you mean by placeholder value?  something like alt="photo" or "unkown"?

[05:14:00.0000] <zcorpan>
Lachy: yeah, dummy text

[05:19:00.0000] <webben_>
Is there any research into how hard it would be to explain to ordinary users "how or why to provide alternate text"? While professionals disagree about what provides the perfect alt text, it does not seem difficult to explain how to provide text that would be an improvement over nothing.

[05:21:00.0000] <webben_>
It seems to me the main obstacle for a site like Flickr to deal with (and this is actually basically limited to photo sharing sites) is how to handle /bulk/ uploads (the problem Maciej keeps mentioning).

[05:21:01.0000] <webben_>
My suspicion is that this is basically a technical problem, rather than a human problem, to be solved.

[05:22:00.0000] <Lachy>
webben_, it's a human problem. You try explaining to the millions of typical users, regardless of if they upload in bulk or individually, that they're expected to provide alternate text.

[05:22:01.0000] <Lachy>
and also teaching them about how to write good quality text.

[05:23:00.0000] <webben_>
Lachy: No I mean /bulk/ is a technical problem. Alt text is a human problem, but it doesn't seem a massively difficult one. (It's not more difficult to explain to a million users than a 100 users. Or even to 10 users.)

[05:23:01.0000] <Lachy>
it takes long enough to teach web developers how to write good alt text (many still get it wrong), let alone someone who just wants to upload some happy snaps of their kids birthday

[05:24:00.0000] <Lachy>
brb

[05:24:01.0000] <webben_>
I don't think writing alt text is intrinsically easier for your average webdev.

[05:25:00.0000] <webben_>
(In fact, in so far as webdev's may tend to have backgrounds in scientific or visual arts disciplines rather than highly verbal ones, it may be /harder/ for your average webdev than many other professionals.)

[05:27:00.0000] <webben_>
Also, webdevs are more likely to be confused by other groups by having read incorrect information about alt, or by being introduced to it primarily as a tool for SEO.

[05:28:00.0000] <webben_>
the text "my kid's birthday" is itself halfway usable alt text (it's not /perfect/ alt text, but it's usable)

[05:38:00.0000] <Lachy>
considering the title would probably be something like "Jack blowing out his candles" (assuming the user even bothered to provide that), I don't see how such alt text would be useful

[06:00:00.0000] <webben_>
Lachy: Well, first, I don't limit the question of alt text to the alt attribute (I'm also considering things like the alt element). Second, repeated text that uniquely identifies an image is preferable to failing to identify it. Third, such alt text is useful when photos are displayed without their titles (as is common on Flickr, e.g. for the photostream list on the right at http://www.flickr.com/photos/minouzers/1230100121/).

[06:02:00.0000] <webben_>
(Repeated text is better than none because you can select and manipulate the image from a list of items on the page, distinguishing it from the other images on the same page.)

[06:08:00.0000] <webben_>
alt="" is best reserved for items that should simply be omitted from the page.

[06:09:00.0000] <webben_>
(and this will never work for widget or link complete content)

[06:09:01.0000] <Lachy>
why would a user who couldn't see the image and is thus seeing the alternate text want to "select and manipulate the image from a list of items on the page"?

[06:10:00.0000] <webben_>
Lachy: Because blind users regularly want to share images with sighted peers.

[06:10:01.0000] <webben_>
e.g. simple use case: Jack's visually impaired grandma wants an image off Flickr to email to her friends.

[06:11:00.0000] <Lachy>
yeah, saving an image is one thing.  But it's the list of items that I'm not so sure about.

[06:11:01.0000] <webben_>
Lachy: Oh. Lists of things is a common assistive technology technique (providing an alternative for serial processing of the entire page.)

[06:11:02.0000] <webben_>
e.g. lists of links, lists of images, lists of headings

[06:11:03.0000] <webben_>
VoiceOver does a sort of list of "items"

[06:12:00.0000] <webben_>
but it's the same problem if the page is processed serially

[06:12:01.0000] <Lachy>
hmm. it's just sounding more and more like the "list of links" argument used for using unique link text

[06:12:02.0000] <webben_>
you can't necessarily distinguish one photo from another without some alt value

[06:13:00.0000] <webben_>
Lachy: Forget the list. That will just be part of the reality of how users process the page (whether people think it's "reasonable" or not). It doesn't change the problem.

[06:13:01.0000] <Lachy>
you can distinguish between them when read in context

[06:13:02.0000] <webben_>
not in the photostream

[06:13:03.0000] <webben_>
not if you jump around the page

[06:14:00.0000] <webben_>
not if the developer isn't massively careful when it comes to distinguishing avatar images (in the comments, for example) from "the" photo

[06:15:00.0000] <Lachy>
well, I'd like some sort of usability study to demonstrate that.  In fact, I asked for that a few weeks ago. http://lists.w3.org/Archives/Public/public-html/2007Aug/0577.html

[06:15:01.0000] <webben_>
Lachy: that being which in particular?

[06:16:00.0000] <Lachy>
your hypothesis that repeating text in the alt attribute is better than omitting it or a blank value

[06:17:00.0000] <webben_>
Ah. Yeah. I'd be interested in such a usability study too. It will be difficult to construct effectively I suspect.

[06:17:01.0000] <Lachy>
I did ask Joshue O Connor in that email if he could help out, since he offered his services for such things, but never heard back about it

[06:17:02.0000] <webben_>
for one thing, you'd need to give people some tasks where it might help (like the sharing example I gave above)

[06:18:00.0000] <webben_>
If I had the testing resources at my disposal, I'd led them. But I don't. :(

[06:18:01.0000] <webben_>
*lend

[06:31:00.0000] <zcorpan>
Lachy: could you commit your changes to the status script to google code, please? :)

[06:31:01.0000] <Lachy>
this one http://status.whatwg.org/annotate-web-apps.js ?

[06:32:00.0000] <zcorpan>
yeah

[06:32:01.0000] <zcorpan>
http://html5.googlecode.com/svn/trunk/status/

[06:32:02.0000] <Lachy>
I got charlvn to add those changes for me, I'll ask him to do it

[06:32:03.0000] <zcorpan>
ok

[06:36:00.0000] <Lachy>
I've messaged him, but he appears to be away. You could check it in yourself if you like, though. The only changes were adding everything from the comment with my name in it, getElementsByClassName() and below

[06:41:00.0000] <Lachy>
he said he'd do it shortly

[10:29:00.0000] <zcorpan>
ok

[10:30:00.0000] <zcorpan>
i've written a front-end interface for the status marker updater thing

[10:36:00.0000] <Lachy>
zcorpan, where is it and how does it work?

[10:43:00.0000] <zcorpan>
http://html5.googlecode.com/svn/trunk/status/

[10:43:01.0000] <zcorpan>
it adds a form to the TOC

[10:43:02.0000] <zcorpan>
and lots of radio buttons

[10:49:00.0000] <Lachy>
I don't see a form when I look at the spec

[11:00:00.0000] <zcorpan>
indeed, it hasn't been moved to status.whatwg.org yet

[11:01:00.0000] <Lachy>
yeah, I just realised that

[11:01:01.0000] <Lachy>
is there an easy way to test it out?

[11:02:00.0000] <zcorpan>
i'll upload a demo, hold on

[11:07:00.0000] <zcorpan>
hmm, can't access simon.html5.org by ftp atm :(

[11:09:00.0000] <Lachy>
where's update-markers.php? I don't see it in SVN or on stats.whatwg.org, but it's referenced in the script

[11:09:01.0000] <zcorpan>
it doesn't exist yet

[11:09:02.0000] <zcorpan>
someone needs to write it :)

[11:09:03.0000] <Lachy>
ok, I'm too impatient :-)

[11:10:00.0000] <markp>
Lachy: ping

[11:10:01.0000] <Lachy>
yo

[11:10:02.0000] <markp>
got a minute to talk about NOALT?

[11:10:03.0000] <Lachy>
sure

[11:10:04.0000] <markp>
i was intrigued by paul's proposal

[11:10:05.0000] <markp>
on the blog thread that is now useless

[11:11:00.0000] <markp>
you or someone mentioned that NOALT had been considered but rejected for lack of a convincing use case?

[11:11:01.0000] <Lachy>
indeed, it was one of the more constructive posts :-)

[11:11:02.0000] <Lachy>
yeah, it's not clear why <img noalt> is better than <img>

[11:11:03.0000] <markp>
can you point me to some background discussion of that?

[11:12:00.0000] <markp>
well, <img noalt> is explicit, and <img> is implicit

[11:12:01.0000] <Dashiva>
noalt is worse, because it introduces ambuigity if both noalt and alt, or neither, are present

[11:12:02.0000] <Lachy>
I thought someone had given a more thorough reasoning on public-html, but I couldn't find it earlier.  I'll see what I can find now

[11:13:00.0000] <markp>
i don't follow; but the addition of noalt would allow the spec to require either alt or noalt (but not both, and not neither)

[11:13:01.0000] <markp>
a missing alt would be a mistake, and could be flagged as such

[11:13:02.0000] <Dashiva>
It's easy to say something is not conforming, but we also have to specify how it works when it doesn't conform

[11:14:00.0000] <markp>
ok, we'll get there

[11:14:01.0000] <markp>
just brainstorming the idea atm

[11:14:02.0000] <markp>
i'm just thinking about the case of mass uploads to flickr, as has been discussed

[11:15:00.0000] <Lachy>
the end of this post from Hixie mentions it briefly http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2007-August/012378.html

[11:15:01.0000] <markp>
no alternate text is available, we know that at time of publication, this would let us explicitly state that

[11:15:02.0000] <Lachy>
the comparison with nohref is a good reason, since that attribute is totally useless

[11:15:03.0000] <Philip`>
Requiring an explicit <img noalt> makes the no-alt case harder (rather than like <img> which is easier to write than the with-useful-alt case), which sounds like the wrong way to motivate people to add alt text

[11:16:00.0000] <Dashiva>
I could've missed an argument, but I still don't have any clear distinction in use between "we know there's no alt" and "there's no alt"

[11:16:01.0000] <markp>
well, making alt required for the past 10 years hasn't done much to motivate people

[11:16:02.0000] <markp>
dashiva: think about the difference between http 404 and 410

[11:16:03.0000] <markp>
410 is intentional

[11:17:00.0000] <Dashiva>
But the use is the same - you have no page to show

[11:17:01.0000] <Lachy>
noalt doesn't really solve any accessibility issue. Even if it does catch on, the validation-badge hunters are just going to stick alt="" or noalt in there to keep the validator happy

[11:17:02.0000] <Philip`>
I saw alt on two thirds of the imgs I looked at a while ago, so something has made people use it a lot, though that's independent of whether they're actually adding useful alt text

[11:17:03.0000] <markp>
you're right, noalt doesn't solve any accessibility issue

[11:18:00.0000] <markp>
an image that should have alternate text, but doesn't, is inaccessible regardless of the exact markup

[11:18:01.0000] <Philip`>
(Er, I meant half the imgs, or two thirds of the pages with at least one img)

[11:18:02.0000] <markp>
but i think noalt is a good alternative solution to making alt optional

[11:19:00.0000] <markp>
the spec could require either alt or noalt

[11:19:01.0000] <Lachy>
and I personally don't see any practical difference between a validator always issuing an error for missing alt, or issuing a warning to those who request it.

[11:19:02.0000] <markp>
and a missing alt attribute would be a sign that you forgot something and need to fix it

[11:19:03.0000] <markp>
if alt is optional, the validator can't flag that

[11:20:00.0000] <markp>
and people who genuinely forgot it (and would add it if reminded) might miss it

[11:20:01.0000] <markp>
that seems like an obvious benefit to me, but i can tell i'm not winning any converts yet

[11:20:02.0000] <Dashiva>
A missing alt would be a problem regardless

[11:21:00.0000] <markp>
A missing alt would be an accessibility problem regardless

[11:21:01.0000] <markp>
there, i fixed it for you

[11:21:02.0000] <markp>
i dislike how html 4 has conflated validation with accessibility

[11:21:03.0000] <Lachy>
I would expect good validators and authoring tools to be able to notify users of missing alt attributes, regardless of its requirement in the spec

[11:22:00.0000] <markp>
it's led to the problem we're now faced with, valid markup which is actively harmful to accessibility (because software is auto-filling harmful alt text)

[11:22:01.0000] <markp>
lachy: the problem with that is that it will add to the "noise" of the validation output

[11:23:00.0000] <markp>
like if your publishing platform doesn't escape ampersands in urls and you can't do anything about it

[11:23:01.0000] <markp>
then you go to validate and you can't find the real problems that you can fix, because there's too much noise

[11:23:02.0000] <Dashiva>
Just because you can't fix it doesn't make it any less a real problem, though

[11:24:00.0000] <markp>
sigh

[11:24:01.0000] <Lachy>
it would depend on the validators UI

[11:24:02.0000] <markp>
i'm not getting anywhere

[11:24:03.0000] <zcorpan>
markp: validators can still warn about missing alt, and you can have options in the validator to suppress certain warnings or errors

[11:24:04.0000] <markp>
yeah, ok

[11:25:00.0000] <Philip`>
Might validation-badge hunters choose to use the most lenient conformance checker that does the minimum amount of required testing before giving them badges?

[11:26:00.0000] <Lachy>
this is so annoying since I have conflicting opinions

[11:27:00.0000] <Philip`>
(in which case there is benefit in having the spec require things, rather than hoping conformance checkers will choose to provide warnings, to raise the minimum quality level of conformant conformance checkers)

[11:27:01.0000] <Dashiva>
Badge hunters tend to mess it up anyhow, I'd say

[11:27:02.0000] <zcorpan>
Philip`: badge hunters will just use the w3c validator, because other validators aren't w3c approved... :)

[11:27:03.0000] <Dashiva>
Even if you make them conform to all machine-checkable criteria, they can still play ball with the rest

[11:28:00.0000] <Philip`>
Badge hunters post badges on their web sites, which gives the badge-providing validator high visibility and high search-engine-findingness and everything, so it is likely to become the most popular validator, so well-meaning people who want to properly validate their pages to find problems will end up using that one too

[11:29:00.0000] <zcorpan>
Philip`: require things doesn't necessarily improve the badge hunters' quality of their markup

[11:29:01.0000] <Philip`>
(Or at least that's why I've only ever used the W3C validator, because I don't hear about any probably-better validators and there aren't loads of badges pointing me at them)

[11:30:00.0000] <Lachy>
so Henri should issue badges for everything, like different grades depending on the number of errors :-)

[11:30:01.0000] <Dashiva>
A+++ would validate again

[11:30:02.0000] <gsnedders>
badges!

[11:30:03.0000] <zcorpan>
"This page is Invalid! Look for yourself!"

[11:31:00.0000] <Lachy>
zcorpan, there's already a badge for that :-)

[11:31:01.0000] <zcorpan>
oh snap

[11:31:02.0000] <Dashiva>
Oh yeah

[11:31:03.0000] <Dashiva>
/me goes to read that article again

[11:31:04.0000] <Philip`>
http://www.mikeindustries.com/blog/archive/2004/06/march-to-your-own-standard ?

[11:31:05.0000] <Lachy>
that's it! I was looking for that

[11:32:00.0000] <Philip`>
(found via http://blogs.msdn.com/michkap/default.aspx )

[11:33:00.0000] <zcorpan>
but where's the badge?

[11:33:01.0000] <Lachy>
it seems to be gone

[11:34:00.0000] <Lachy>
http://www.mikeindustries.com/blog/images/validatethis.gif

[11:34:01.0000] <Philip`>
The blogs.msdn.com page does have the badge, just under the (Tick!) Unicode ENCODED

[11:34:02.0000] <Lachy>
he used to have it in his sidebar http://web.archive.org/web/20050123150650/http://www.mikeindustries.com/blog/archive/2004/06/march-to-your-own-standard

[11:35:00.0000] <Dashiva>
He sold out!

[11:35:01.0000] <Lachy>
his page is still invalid, even without the badge

[11:42:00.0000] <Lachy>
apparently UAAG actually requires UAs to treat <img> and <img alt=""> differently. http://www.w3.org/TR/UAAG10/guidelines.html#tech-missing-alt

[11:48:00.0000] <Lachy>
markp, http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2007-April/010995.html has more reasoning.

[11:48:01.0000] <Lachy>
and all the other mails about noalt if you're interested http://lists.whatwg.org/mmsearch.cgi/whatwg-whatwg.org?config=whatwg-whatwg.org&restrict=&exclude=&method=and&format=short&sort=score&words=noalt

[11:52:00.0000] <Lachy>
"In SMIL 1.0 [SMIL], on the other hand, alt is not required on media objects." -- http://www.w3.org/TR/UAAG10-TECHS/guidelines.html#tech-missing-alt

[11:53:00.0000] <Lachy>
I wonder if that lack of requirement has harmed the accessibility of SMIL in any way

[11:54:00.0000] <takkaria>
is SMIL actually used anywhere?

[11:54:01.0000] <Lachy>
I don't know

[12:30:00.0000] <webben_>
takkaria: I suspect in most cases where people have thought about using SMIL, they've ended up using Flash. Although SMIL actually has widespread support in some form (e.g. Internet Explorer, RealPlayer. QuickTime), Flash content can be played in-page in Gecko and WebKit. Plus, there are far more people who already know how to author in Flash. (Given the prohibition on using the Flash spec to create new plugins and players, this isn't necessarily

[12:31:00.0000] <takkaria>
you got cut off at "necessarily"

[12:31:01.0000] <webben_>
sorry.

[12:31:02.0000] <webben_>
necessarily ... a great idea, but you can see the market forces driving it

[12:32:00.0000] <takkaria>
right

[12:32:01.0000] <takkaria>
/me nods

[12:33:00.0000] <webben_>
There are plenty of experiments with SMIL (e.g. http://www.domsmith.co.uk/computing/bbc-backstage/smil-news/ ), but I suppose broadcasters tend to be more interested in getting a big audience right now than picking the best technology for maximising audience over the long-term. If SMIL has a current home, I suspect it would be for DVD presentation-type things.

[12:35:00.0000] <webben_>
In theory, there's no big barrier to global SMIL adoption. There's an open source player and plugin, Ambulant, which seems to implement the spec a lot better than the closed-source alternatives.

[12:35:01.0000] <webben_>
http://sourceforge.net/projects/ambulant/

[12:36:00.0000] <webben_>
takkaria: Oh, actually, one actual widespread use for SMIL is DAISY: http://www.daisy.org/z3986/

[12:37:00.0000] <webben_>
but you don't see much DAISY content on the web (because the content created in DAISY is usually limited to access by registered people with visual impairments)

[12:37:01.0000] <webben_>
I think DAISY (or at least some forms of DAISY) use a SMIL wrapper.

[12:38:00.0000] <webben_>
here's the spec: http://www.daisy.org/z3986/2005/z3986-2005.html

[12:38:01.0000] <takkaria>
ta

[12:38:02.0000] <takkaria>
interesting. :)

[12:50:00.0000] <webben_>
takkaria: hmm also interesting: http://www.kaourantin.net/2007/08/what-just-happened-to-video-on-web_20.html "You can easily write your own SMIL parser in ActionScript though."

[13:04:00.0000] <takkaria>
well, it's XML, so it's not that surprising

[13:06:00.0000] <webben_>
guess not

[13:06:01.0000] <zcorpan>
what is "SMIL parser"?

[13:07:00.0000] <webben_>
zcorpan: I guess it would turn SMIL into a SWF animation.

[13:07:01.0000] <webben_>
(that is to say, a SMIL parser wouldn't be much use without rendering)

[13:07:02.0000] <zcorpan>
so an implementation of SMIL

[13:07:03.0000] <webben_>
yeah

[13:08:00.0000] <zcorpan>
/me would like to study SMIL implementations sometime

[13:08:01.0000] <zcorpan>
in particular how they deal with namespaces

[13:10:00.0000] <webben_>
well the ambulant implementation is supposed to be "namespace-based": http://lists.w3.org/Archives/Public/public-evangelist/2006Aug/0003.html

[13:11:00.0000] <webben_>
whatever that means

[14:03:00.0000] <Dashiva>
IE7 doesn't seem to want to update its attribute selector selection

[14:06:00.0000] <Dashiva>
/me hates forcing reflows

[14:22:00.0000] <Dashiva>
Summary of adapting whatwg/issues for IE7: helper function for textContent, helper function for hasAttribute, force reflow on 'irrelevant' attribute toggling, workaround for missing white-space:pre-wrap support.

[14:25:00.0000] <zcorpan>
Dashiva: fun :)

[14:27:00.0000] <Philip`>
/me hopes there aren't people with AT that only supports IE6, not IE7

[14:28:00.0000] <Dashiva>
Supporting IE6 would probably be easier, since they don't support attribute selector to begin with, so you don't need to reflow to show content :)

[14:30:00.0000] <Philip`>
Wouldn't it be harder since you'd have to emulate attribute selectors via some other method? :-)

[14:31:00.0000] <Dashiva>
Not really. It's just [irrelevant]{display:none;} so it could just as easily be replaced by .style.display = 'none'; now that I have hte helper functions in place

[14:32:00.0000] <Dashiva>
Didn't hixie mention something about the API being public and/or documented and/or otherwise usable?

[14:33:00.0000] <webben_>
Philip`: Probably (based on reading user mailing lists) a slim majority of AT users have AT that either only supports IE6 or works better with IE6 than IE7.

[14:33:01.0000] <webben_>
Philip`: In addition, many users downgrade from IE7 they hate it so much.

[14:36:00.0000] <Philip`>
Dashiva: Do you mean http://www.whatwg.org/issues/API ?

[14:37:00.0000] <Dashiva>
I probably do

[14:38:00.0000] <Philip`>
It does seem to be public and documented and usable, and it doesn't even generate ill-formed XML any more

[14:41:00.0000] <Dashiva>
I don't particularily like the listen stream format, but one makes do I suppose

[14:41:01.0000] <Dashiva>
There would be even more screaming if it required support for event-source :D

[14:43:00.0000] <Philip`>
Even more if it used TCPConnection - it would still be perfectly standards-based, but also perfectly useless

[14:44:00.0000] <Hixie>
there's actually another API below this one

[14:44:01.0000] <Hixie>
which doesn't use HTTP

[14:45:00.0000] <Dashiva>
By the way Hixie, there's still no action when you click the vote/remove vote button

[14:46:00.0000] <Dashiva>
You have to reload for the button text to change

[14:46:01.0000] <Hixie>
in what browser?

[14:46:02.0000] <Dashiva>
I'd say all of them based on the code, but I've only tried opera and IE so far

[14:47:00.0000] <Hixie>
works fine in firefox, opera, and safari for me

[14:47:01.0000] <Philip`>
It worked when I tried it in Firefox

[14:47:02.0000] <Dashiva>
Ah right, it's being set after the listen echo...

[14:48:00.0000] <Dashiva>
I haven't got that one to proxy properly yet

[14:49:00.0000] <Dashiva>
(Another reason I would prefer a pull option, with XML to avoid string parsing)

[14:50:00.0000] <Hixie>
yeah i can actually give you another protocol for that

[14:50:01.0000] <Hixie>
hold on

[14:50:02.0000] <Dashiva>
No hurry, I'm finished for today :)

[14:58:00.0000] <mpt_>
zcorpan, "validators can still warn about missing alt, and you can have options in the validator to suppress certain warnings or errors" is like those badly-designed GUIs that have "Are you sure?" alerts with "Do not warn me again" checkboxes

[14:58:01.0000] <Hixie>
i've start updating API, i'll let you know when it's done

[14:59:00.0000] <Hixie>
can't do it now

[14:59:01.0000] <mpt_>
zcorpan, in both cases, the problem is not that some people always want the warnings and some people never want them, but that for *everyone* the warning is useful *sometimes*

[15:00:00.0000] <mpt_>
And the solution is to prevent the warning from being necessary in the first place.

[15:01:00.0000] <mpt_>
If I'm a conscientious Web application author, noalt would let me distinguish between images that had no alt text because they were user-supplied, and images that had no alt text because I'd accidentally left it out.

[15:02:00.0000] <mpt_>
Just omitting alt= doesn't let me do that, it leaves me having to choose between zero validation warnings and too many validation warnings.

[15:03:00.0000] <zcorpan>
mpt_: the case markp described was actually in specific cases -- "like if your publishing platform doesn't escape ampersands in urls and you can't do anything about it ... then you go to validate and you can't find the real problems that you can fix, because there's too much noise"

[15:03:01.0000] <mpt_>
I don't understand that analogy, but I agree with the rest of markp's point.

[15:05:00.0000] <zcorpan>
if a validator complains about something you can't fix, then the straightforward solution to get rid of the noise is to suppress the warning/error in the validator

[15:05:01.0000] <zcorpan>
imho

[15:07:00.0000] <mpt_>
zcorpan, absolutely, but it's better to be able to do that for individual images or types of image that you can't provide alt for ahead of time (using noalt) than for every image in the document (using the validator option).

[15:08:00.0000] <mpt_>
Otherwise you can't test for mistakenly missed alt= on any page where people can upload their own images.

[15:08:01.0000] <zcorpan>
however i realise that adding alt="" as a way to suppress warnings might be more appealing than changing options in the validator, which is an argument for noalt

[15:09:00.0000] <zcorpan>
mpt: true

[15:14:00.0000] <webben_>
I'm not really sure what weight this "validator" stuff holds when most of the web consists of invalid HTML.

[15:14:01.0000] <webben_>
(and people concerned with accessibility "validation" tend to use dedicated checkers)

[15:15:00.0000] <zcorpan>
indeed

[15:15:01.0000] <webben_>
and the sites like Flickr that we appear to be worrying about don't publish valid HTML

[15:15:02.0000] <webben_>
it seems like a bit of a red herring, i dunno

[15:16:00.0000] <webben_>
(well, red herring is an exaggeration, perhaps more of a debate about the number of angels on a pinhead ;) )

[15:17:00.0000] <mpt>
I don't think the existence of dedicated accessibility checkers should allow the HTML specification to make some things inherently uncheckable

[15:17:01.0000] <webben_>
mpt: I don't follow. Can you please rephrase?

[15:18:00.0000] <mpt>
Sure. How do you propose that *the accessibility checker itself* do the checking, if it can't tell the difference between "this image has no alternate text because it's user-supplied" and "this image has no alternate text because we forgot it"?

[15:21:00.0000] <webben_>
mpt: Oh. I wasn't making an argument against noalt or whatever. I was just saying I think the discussions around alt are probably placing more emphasis on validation and the effects of people validating and creating fake alt text to validate then they actually deserve.

[15:22:00.0000] <webben_>
Especially as replacing loads of fake alt text with no alt text or noalt does not make for a massively more accessible document.

[15:22:01.0000] <webben_>
noalt might offer marginal improvements

[15:22:02.0000] <webben_>
e.g. in Flickr you could distinguish content images from decorative images, maybe.

[15:22:03.0000] <mpt>
Given the expected lifespan of HTML5, I think the discussions about it are giving remarkably little emphasis to *everything* :-)

[15:23:00.0000] <webben_>
mpt: I guess I mean, too much weight in comparison to other factors. :)

[15:24:00.0000] <mpt>
I think the argument about <img src="foo" noalt> being no more accessible than <img src="foo">, *that* is the red herring

[15:24:01.0000] <mpt>
This isn't about the final document, it's about the authoring process.

[15:24:02.0000] <zcorpan>
from a UA's point of view, how do you handle an <img> that has neither alt nor noalt? same as alt=""? same as noalt?

[15:24:03.0000] <webben_>
zcorpan: It depends on context.

[15:24:04.0000] <mpt>
Same as noalt, I think

[15:25:00.0000] <zcorpan>
webben_: what, quirks mode/standards mode? :)

[15:25:01.0000] <mpt>
but again, that's not the issue here

[15:25:02.0000] <webben_>
zcorpan: e.g. <a href="foo"><img alt=""></a> is very different to bla bla bla<img alt="">bla bla bla.

[15:25:03.0000] <Philip`>
Use UA-dependent heuristics to guess what would be the most useful replacement text

[15:26:00.0000] <webben_>
zcorpan: and bla bla bla<img>bla bla bla is different again

[15:26:01.0000] <Philip`>
(because then it's not our problem :-) )

[15:26:02.0000] <Philip`>
(and it's presumably what people do already when alt is missing)

[15:26:03.0000] <zcorpan>
webben_: right

[15:26:04.0000] <webben_>
Philip`: Well it would be useful to try and work out if there's a good common algorithm (a la the work that's going on with th/td associations)

[15:31:00.0000] <webben_>
likewise it would be useful to have ways of associating text elsewhere with the image (figure/legend, the proposed alt element, perhaps an altfor attribute)

[15:46:00.0000] <Philip`>
/me needs to try re-running his web survey thing to get more data (like common values of certain attributes, and maybe content-types of linked scripts/stylesheets/images/etc to see how bad that is, and I can't think of much else yet - suggestions would be appreciated :-) )

[15:46:01.0000] <Philip`>
(But I need to remember how to write Java first, so I can use hsivonen's parser)

[15:47:00.0000] <webben_>
Philip`: maybe create a suggestion page or something for the next iteration?

[15:47:01.0000] <webben_>
hmm ... maybe that would end up duplicating that wikipage Rob Burns was creating on Research needing to be done.

[15:49:00.0000] <webben_>
Philip`: Did you collect longdesc data already btw (and does your sample include lots of .gov and .edu sites, which seem more likely to use longdesc)?

[15:51:00.0000] <Philip`>
webben_: Not specifically - all I have is the general attribute-occurrence information at http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/attr/longdesc

[15:52:00.0000] <webben_>
oh

[15:52:01.0000] <webben_>
Philip`: it would be interesting (well to me anyhow) to see a table with img in one column and a longdesc link in the next column

[15:53:00.0000] <webben_>
I've found plenty of vaguely appropriate longdesc's but it would be nice to try and see more examples.

[15:54:00.0000] <Philip`>
Out of 8192, I had 31 .gov and 175 .edu

[15:54:01.0000] <Philip`>
(*numbers of pages)

[15:54:02.0000] <webben_>
well that's some

[15:54:03.0000] <Philip`>
(This is a random sample from dmoz.org's list)

[15:55:00.0000] <webben_>
It might help at some point to do some sort of clumping differentiation based on what sort of site given markup is on.

[15:55:01.0000] <Philip`>
(It'd be easy to do a sample of e.g. all .gov sites extracted from dmoz.org's list)

[15:55:02.0000] <webben_>
cool

[15:55:03.0000] <Philip`>
Comparing the dmoz.org data vs the Alexa top 500 shows some very significant differences

[15:56:00.0000] <webben_>
I wonder how representative or not dmoz is generally.

[15:56:01.0000] <Philip`>
(http://canvex.lazyilluminati.com/misc/stats/analyse.cgi/index vs http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/index)

[15:57:00.0000] <mpt>
When I spent some time reading through dmoz-related discussions a few years ago I got the impression that it was a pile of fiefdoms

[15:57:01.0000] <Philip`>
(<script> and <form> were the most obvious differences that I noticed)

[15:58:00.0000] <webben_>
Philip`: Have you considering pushing beyond home pages?

[16:00:00.0000] <webben_>
e.g. for stuff like longdesc, you wouldn't normally expect to find that on a homepage.

[16:07:00.0000] <Philip`>
webben_: I've not really thought in any detail how to find a good set of pages to analyse

[16:08:00.0000] <Philip`>
Some kind of crawler would be good, but I don't know how to make it find interesting pages and to not get stuck in a few deep sites

[16:13:00.0000] <webben_>
I've often wondered whether it would be possible to use the Internet Archive's database. But I dunno how deep that really goes either.

[16:15:00.0000] <Philip`>
I think working out which set of pages to examine is a more difficult problem than actually downloading the pages, and I imagine the IA archive doesn't help with the first problem

[16:15:01.0000] <webben_>
I think these trawlings are more helpful for finding examples of markup in use, then for the statistics (because the sampling is inevitably going to be kinda skewey unless you take samples of very specific things, e.g. how is markup used on Wikipedia or Blogger or something).

[16:16:00.0000] <webben_>
I guess I was thinking the IA Archive is going to be a different set then DMOZ.org. But actually I have no real idea how IA is collected anyhow.

[16:16:01.0000] <webben_>
might be through dmoz ;)

[16:18:00.0000] <webben_>
oh i see, it's like Alexa Top 200, but everything: http://www.archive.org/about/faqs.php

[16:18:01.0000] <webben_>
I guess even then there might be issues like: have academic collections (which might feature longdesc more, or whatever) asked to be removed from WayBack.

[16:19:00.0000] <Philip`>
(I think the mean HTML page size is about 30KB (or was it 60KB?) so you can look at tens of thousands in a few hours on a relatively normal internet connection)

[16:19:01.0000] <webben_>
that's pretty good

[16:21:00.0000] <Philip`>
It'd be nice if Google web search had a regexp feature like Google code search :-)

[16:21:01.0000] <webben_>
It certainly would.

[16:21:02.0000] <webben_>
You could construct some very interesting web services around that.

[16:22:00.0000] <webben_>
e.g. find microformated content for X

[16:23:00.0000] <webben_>
or find images with alt="*kitten*" ;)

[16:31:00.0000] <zcorpan>
hmm, safari doesn't reflow on className change :(


2007-08-27
[20:38:00.0000] <G0k>
any good fights today?

[21:00:00.0000] <Lachy>
G0k, there's a bit of fighting in the comments http://blog.whatwg.org/omit-alt :-)

[21:01:00.0000] <Lachy>
hi MikeSmith

[21:02:00.0000] <MikeSmith>
Lachy - hei

[21:02:01.0000] <Lachy>
did you ever receive a response to this http://lists.w3.org/Archives/Public/www-archive/2007Aug/0046.html ?

[22:17:00.0000] <jwalden>
Hixie: ping

[22:33:00.0000] <Hixie>
hey

[22:39:00.0000] <jwalden>
Hixie: regarding postMessage, did you intend for the event to be dispatched at the target window's document?  I'd thought, when I requested it be moved to Window, that the dispatch target would change too

[22:39:01.0000] <MikeSmith>
Lachy - sorry, was having some machine troubles, then just gave up and went to lunch

[22:39:02.0000] <jwalden>
Hixie: I can send an email about it if you think it's reasonable, just wanted to get an opinion first

[22:41:00.0000] <Hixie>
jwalden: i didn't consider them related

[22:41:01.0000] <Hixie>
jwalden: in fact thinking about it maybe it should fire at the body element

[22:41:02.0000] <Hixie>
jwalden: feel free to send mail, not sure we'd want it to fire at Window though

[22:42:00.0000] <jwalden>
hrm

[22:43:00.0000] <G0k>
Hixie: hate to keep bugging about it but i was wonder if you wanted to post that last email i sent you about networking-event-source to http://www.whatwg.org/issues/

[22:45:00.0000] <jwalden>
I guess I'll change my implementation to fire at document; if you thought a change to window might happen, I'd propose it, but I don't really like firing at document.body ;-)

[22:48:00.0000] <Hixie>
G0k: http://www.whatwg.org/issues/ automatically updates from my imap folders around 5am every day

[22:48:01.0000] <Hixie>
jwalden: heh

[22:49:00.0000] <G0k>
Hixie: er...i sent that email from 2 weeks ago but it's not there

[22:49:01.0000] <Hixie>
G0k: uri?

[22:49:02.0000] <G0k>
i never posted it to the main list, i sent it straight to you

[22:52:00.0000] <Hixie>
hm

[22:52:01.0000] <Hixie>
subjet line?

[22:52:02.0000] <G0k>
Re: about event-source

[22:52:03.0000] <G0k>
talks about true/false and some other stuff

[22:52:04.0000] <Hixie>
oh

[22:53:00.0000] <Hixie>
you sent it straight to me

[22:53:01.0000] <Hixie>
so it'll never appear on the list

[22:53:02.0000] <Hixie>
i hide e-mails that weren't cc'ed publically

[22:53:03.0000] <Hixie>
(i get a lot of confidential e-mail about the spec that ends up in my folders)

[22:54:00.0000] <G0k>
ah

[22:55:00.0000] <G0k>
so should i reformulate and echo it to the list?

[22:55:01.0000] <Hixie>
nah it's still on my list, it just doesn't show up on the site

[22:55:02.0000] <G0k>
ah

[22:56:00.0000] <G0k>
oh and uh anyone interested in my webkit implementation can play with it http://bugs.webkit.org/show_bug.cgi?id=14997

[22:56:01.0000] <G0k>
so...so there

[22:57:00.0000] <Hixie>
:-)

[22:57:01.0000] <G0k>
what do we want next? video elements?

[22:58:00.0000] <Hixie>
in webkit?

[22:58:01.0000] <G0k>
well i dunno i guess i kinda feel like just trying to implement stuff is a good way to find which parts of the spec are confused

[23:00:00.0000] <G0k>
but hey, maybe the "finalize first, implement later" method is better. :)

[23:11:00.0000] <Hixie>
G0k: no, implemention in tandem with spec development is definitely the way we want it :-)

[23:11:01.0000] <G0k>
yeah so. :)

[23:12:00.0000] <G0k>
actually i think it might be cool to make some kind of more general reference implementation just for this purpose

[23:12:01.0000] <G0k>
maybe built on html5lib?

[23:12:02.0000] <Hixie>
i wouldn't recommend building a reference implementation browser

[23:12:03.0000] <Hixie>
you'll be here for decades

[23:13:00.0000] <G0k>
it doesn't need to work for real pages though

[23:13:01.0000] <G0k>
i feel like my concern is that if you start implementing in an existing thing

[23:14:00.0000] <G0k>
then you might accidentally create an installed base for a feature which will have to change

[23:32:00.0000] <jwalden>
work in a non-release branch :-)

[23:33:00.0000] <Hixie>
G0k: eh, it's part of hte game

[23:34:00.0000] <G0k>
yeah so....get these specs finalized faster people! :)

[23:34:01.0000] <jwalden>
better > faster

[23:35:00.0000] <G0k>
has anyone tried implementing any of WF2?

[23:35:01.0000] <G0k>
or...3 i guess

[23:36:00.0000] <Hixie>
opera has most of wf2 done

[23:36:01.0000] <Hixie>
and shipped, even

[23:36:02.0000] <G0k>
k. sounds like a place to start/continue

[23:37:00.0000] <Hixie>
see also the whatwg blog about a week ago or so

[23:37:01.0000] <Hixie>
someone's done a js version

[23:41:00.0000] <G0k>
input type="search" never made it into a spec?

[01:25:00.0000] <gsnedders>
/me had a dream of Hixie's deayh :\

[01:44:00.0000] <Dashiva>
deayh?

[02:05:00.0000] <mpt>
Don't ask :-)

[02:46:00.0000] <Lachy>
gsnedders, did you mean s/deayh/death/?

[02:47:00.0000] <zcorpan>
gsnedders: you're still on the html wg?

[02:50:00.0000] <Lachy>
zcorpan, was he thinking about leaving?

[02:54:00.0000] <zcorpan>
Lachy: http://krijnhoetmer.nl/irc-logs/html-wg/20070824#l-457

[02:55:00.0000] <Lachy>
wow, looks like I have a lot of logs to read to find out why

[02:57:00.0000] <zcorpan>
/me unsubscribed from www-style...again

[02:58:00.0000] <Lachy>
I haven't been reading www-style lately. what's happening on it?

[02:59:00.0000] <zcorpan>
not much interesting (to me anyway)

[03:37:00.0000] <Lachy>
hmm. those IRC logs are depressing to read

[04:23:00.0000] <zcorpan>
indeed

[05:51:00.0000] <zcorpan>
http://simon.html5.org/temp/annotate.html

[05:52:00.0000] <zcorpan>
demo for the status updater js

[05:59:00.0000] <zcorpan>
wonder if getting innerHTML in html should serialize U+00A0 characters as &nbsp;

[06:03:00.0000] <zcorpan>
ie7 and firefox do, safari and opera don't

[06:21:00.0000] <Lachy>
zcorpan, is it a working demo? can I submit the form?

[06:23:00.0000] <Lachy>
it didn't work

[06:47:00.0000] <gsnedders>
Lachy: yes

[06:47:01.0000] <gsnedders>
zcorpan: no

[06:48:00.0000] <Lachy>
gsnedders, I can't remember what my question was

[06:48:01.0000] <gsnedders>
Lachy: "gsnedders, did you mean s/deayh/death/?"

[06:48:02.0000] <Lachy>
oh right

[06:49:00.0000] <gsnedders>
_really_ odd dream.

[06:49:01.0000] <Lachy>
gsnedders, was the "no" to zcorpan in response to you being on the HTMLWG?

[06:49:02.0000] <gsnedders>
Lachy: not having left the WG

[06:50:00.0000] <Lachy>
right, so you're staying for now. cool

[06:50:01.0000] <Lachy>
so how did Hixie die?

[06:50:02.0000] <gsnedders>
I dunno. I spent the whole dream trying to find that out.

[06:50:03.0000] <Lachy>
that's weird

[06:53:00.0000] <gsnedders>
tantek told me he had died in here.

[06:53:01.0000] <gsnedders>
:\

[06:53:02.0000] <Lachy>
what? do you mean he told you in your dream?

[06:54:00.0000] <gsnedders>
yeah

[06:54:01.0000] <gsnedders>
dreaming about IRC :\

[06:54:02.0000] <Lachy>
that's just sad

[06:54:03.0000] <gsnedders>
totally.

[06:54:04.0000] <Lachy>
... the fact that you're dreaming about IRC

[06:54:05.0000] <gsnedders>
I've only once dreamed about _anything_ related to computers before

[06:56:00.0000] <Lachy>
Did I make an appearance in your IRC dream?

[06:57:00.0000] <gsnedders>
hmmm…

[06:57:01.0000] <gsnedders>
I think so

[08:11:00.0000] <Lachy>
zcorpan, yt?

[08:22:00.0000] <Lachy>
zcorpan, I created a basic form handler for the script, though it doesn't yet handle DB updating http://html5.googlecode.com/svn/trunk/status/update-markers.php

[08:23:00.0000] <zcorpan>
Lachy: nice!

[08:23:01.0000] <Lachy>
test it here http://lachy.id.au/temp/html5/annotate.html

[08:23:02.0000] <Lachy>
if you forget to enter an email and/or rationale, it will prompt you for it

[08:24:00.0000] <zcorpan>
Lachy: i think the idea is for the script to send an email to Hixie or someone who can approve the change before it updates the database

[08:25:00.0000] <Lachy>
ok, so it would have to somehow preserve the data temporarily somewhere while it awaits approval

[08:25:01.0000] <zcorpan>
yeah

[08:25:02.0000] <zcorpan>
perhaps also in the database but another table

[08:25:03.0000] <zcorpan>
or in the email

[08:26:00.0000] <Lachy>
it could check which fields have changed, and then include those fields in the query string

[08:26:01.0000] <zcorpan>
yeah

[08:27:00.0000] <zcorpan>
wonder what would happen if you make another change before the first was approved

[08:28:00.0000] <Lachy>
that wouldn't matter, they wouldn't get committed until the approval.  So it would depend on the order in which Hixie approved them

[08:28:01.0000] <zcorpan>
i guess

[08:29:00.0000] <Lachy>
maybe it should have some authentication system so Hixie doesn't have to approve all changes.  Authorised users could just commit directly

[08:29:01.0000] <Lachy>
though we'd need that anyway, so it couldn't be hacked

[08:30:00.0000] <zcorpan>
perhaps the js could figure out what the changes were so that the whole shebang isn't sent back to the php script

[08:30:01.0000] <zcorpan>
whatever is simplest

[08:32:00.0000] <Lachy>
easy way would be to include an empty value for the intial value so it doesn't get submitted if selected: <label class=initial><input ... name=foo value="" checked> TBW</label>

[08:34:00.0000] <zcorpan>
aha

[08:34:01.0000] <zcorpan>
radio buttons with empty values aren't submitted?

[08:36:00.0000] <zcorpan>
hmm, seems they are, just with no value... :)

[08:37:00.0000] <Lachy>
no, I was wrong.  They are submitted

[08:37:01.0000] <zcorpan>
but the php script doesn't need to figure out what the changes were

[08:37:02.0000] <zcorpan>
just ignore the ones with no value

[08:37:03.0000] <Lachy>
yeah, that makes it easy

[08:38:00.0000] <zcorpan>
ok, i'll implement that in the js

[08:38:01.0000] <Lachy>
I'll update the PHP to check for empty values

[08:39:00.0000] <zcorpan>
the value "none" means that the initial was something else and the marker is to be removed

[08:43:00.0000] <Lachy>
I've updated the PHP

[08:53:00.0000] <zcorpan>
js updated too

[08:53:01.0000] <zcorpan>
i have a send email script in php somewhere

[08:54:00.0000] <zcorpan>
that sends emails encoded as utf-8 and works correctly with funny characters in headers

[08:58:00.0000] <zcorpan>
http://simon.html5.org/temp/mail-utf8.txt

[08:59:00.0000] <zcorpan>
although i guess the email address itself might well contain non-ascii characters too nowadays

[08:59:01.0000] <Lachy>
I deployed the new script on my server, it seems to work

[09:00:00.0000] <zcorpan>
nice

[09:00:01.0000] <Lachy>
I'm not sure how non-ascii email addresses work or if PHP would support them

[09:00:02.0000] <zcorpan>
probably not worth worrying about :)

[09:01:00.0000] <zcorpan>
i guess very few systems allow non-ascii in email addresses

[09:06:00.0000] <Lachy>
should we have this email comit-watchers as well?

[09:06:01.0000] <Lachy>
after it's confirmed

[09:07:00.0000] <zcorpan>
yeah that would be good

[09:07:01.0000] <zcorpan>
twitter?

[09:07:02.0000] <Lachy>
yes, that too

[09:07:03.0000] <Lachy>
I'm just implementing the email script now and to have it email me for testing

[09:09:00.0000] <zcorpan>
ok

[09:10:00.0000] <Lachy>
your email script was missing the format=flowed parameter

[09:13:00.0000] <zcorpan>
hmm, do you do wrapping of the body?

[09:14:00.0000] <zcorpan>
or does php do that?

[09:15:00.0000] <zcorpan>
$message = wordwrap($message, 72); ?

[09:18:00.0000] <zcorpan>
linebreaks in headers should be \r\n also

[09:18:01.0000] <zcorpan>
between headers that is

[09:19:00.0000] <zcorpan>
not sure if php wordwrap() is RFC 2646 compliant though :)

[09:22:00.0000] <zcorpan>
it's not

[09:23:00.0000] <zcorpan>
words shorter than 998 shouldn't be wrapped

[09:25:00.0000] <zcorpan>
soft linebreaks should also be SP CRLF

[09:30:00.0000] <zcorpan>
/me finds http://drupal.org/node/154218 which might be interesting

[09:33:00.0000] <Lachy>
probably doesn't matter for this, since I can control the line lengths manually.  does the php mail() funciton replace \n with \r\n properly?

[09:33:01.0000] <zcorpan>
dunno

[09:38:00.0000] <zcorpan>
php should just do this boring processing by default :)

[09:39:00.0000] <zcorpan>
with options about encoding, format, and some others, and then Do The Right Thing

[09:40:00.0000] <webben>
PHP is not exactly the DTRTiest language ;)

[09:40:01.0000] <zcorpan>
indeed

[09:45:00.0000] <Lachy>
I suppose we should add some check to make sure only the values TBW, WIP, SCS and none will be accepted. At the moment, it doesn't do much input validation, which could be a problem

[09:46:00.0000] <zcorpan>
yeah

[09:46:01.0000] <Lachy>
should we restrict the length of the rationale, especially if we're going to twitter it?

[09:46:02.0000] <zcorpan>
seems reasonable

[09:46:03.0000] <zcorpan>
what does twitter accept?

[09:46:04.0000] <Lachy>
140 chars total

[09:47:00.0000] <zcorpan>
we might want some context text in the twitter messages

[09:47:01.0000] <Lachy>
so if we send the twitter message as: "Status Update: $rationale", that leaves 125 chars for the rationale

[09:47:02.0000] <zcorpan>
yeah

[09:48:00.0000] <Lachy>
I'll add a comment to the PHP, update the JS to make the rationale input maxlength=125

[09:48:01.0000] <Dashiva>
Is twitter 140 chars or 140 bytes, though?

[09:51:00.0000] <gsnedders>
chars

[09:54:00.0000] <zcorpan>
Lachy: i already updated the js

[09:55:00.0000] <zcorpan>
http://krijnhoetmer.nl/irc-logs/whatwg/20070524#l-461

[10:03:00.0000] <Lachy>
I've committed the current PHP versions, feel free to fix some of the bugs, it's getting a bit late for me

[10:03:01.0000] <Lachy>
there's a TODO list at the top of the file

[10:03:02.0000] <zcorpan>
ok

[10:09:00.0000] <Lachy>
we'll need to wordwrap the rationale in the email.  Just do wordwrap($rationale,72) and then fix then append a space before the new lines (if any)

[10:15:00.0000] <Lachy>
use wordwrap("Rationale: $rationale", 72, '\n', false);

[10:17:00.0000] <Lachy>
then, if you want to be really cautious, do wordwrap($body, 998, '\n', true); to ensure no line is longer than that which could happen with a long URL

[10:17:01.0000] <zcorpan>
not wordwrap("Rationale: $rationale", 72, ' \r\n', false); ?

[10:18:00.0000] <zcorpan>
maxlength is 125 already :)

[10:18:01.0000] <zcorpan>
we could cut at 125 before wrapping if it's longer

[10:18:02.0000] <Lachy>
yeah, I know, I just left it there as a reminder for when we do the twitter messages, and we need to do input validation for it

[10:19:00.0000] <Lachy>
probably where I check for empty $email and $rationale, we should also check for strlen($rationale) < 125 and check for a valid email address

[10:20:00.0000] <zcorpan>
yeah

[10:20:01.0000] <othermaciej>
good morning everyone

[10:21:00.0000] <zcorpan>
morning othermaciej

[10:21:01.0000] <Lachy>
should probably use mb_strlen also

[10:21:02.0000] <Lachy>
hi othermaciej 

[10:21:03.0000] <othermaciej>
I've been away all weekend - any exciting flamewars?

[10:21:04.0000] <Lachy>
no, not since your last one with ROb

[10:21:05.0000] <Lachy>
:-)

[10:23:00.0000] <Lachy>
oh, but there's nearly a flamewar in the whatwg blog comments

[10:44:00.0000] <Lachy>
zcorpan, we should probably use this UTF-8 wordwrap function instead http://www.php.net/manual/en/function.wordwrap.php#57090

[10:45:00.0000] <zcorpan>
Lachy: aha, cheers

[10:46:00.0000] <Lachy>
and it would probably be easier if all the word wrapping was handled in the sendMail function in mail-utf8.php

[10:46:01.0000] <Lachy>
that way, it doesn't need to be handled individually for each email we generate

[10:47:00.0000] <zcorpan>
ok

[10:54:00.0000] <zcorpan>
though utf8_wordwrap doesn't have a cut argument

[10:55:00.0000] <Lachy>
oh, well, it shouldn't be too hard to modify it

[10:55:01.0000] <Lachy>
I can probably do that tomorrow if you like

[11:00:00.0000] <zcorpan>
i'll look into it for a bit later today

[11:03:00.0000] <Lachy>
hmm. This comment suggests that wordwrap() doesn't preserve existing line breaks, but I just tested it and it seems to preserve them http://www.php.net/manual/en/function.wordwrap.php#53403

[12:56:00.0000] <zcorpan>
Hixie: you have escape() on 2 more places... :)

[12:59:00.0000] <Hixie>
but those are for data: URIs

[12:59:01.0000] <Hixie>
aren't they ok?

[12:59:02.0000] <Hixie>
btw i updated the API docs for /issues/ to include the lower-level protocol

[13:01:00.0000] <markp>
Hixie: are you actively involved in html5lib coding?

[13:01:01.0000] <Hixie>
no

[13:02:00.0000] <markp>
who is?

[13:02:01.0000] <Hixie>
http://code.google.com/p/html5lib/

[13:02:02.0000] <Hixie>
see "project owners"

[13:06:00.0000] <zcorpan>
Hixie: no, they're not ok. e.g. ☺ becomes data:text/html,%u263A instead of data:text/html,%E2%98%BA (you might even want to do data:text/html;charset=utf-8,... )

[13:07:00.0000] <Hixie>
aah

[13:07:01.0000] <Hixie>
wtf is %u1234

[13:07:02.0000] <zcorpan>
it's what escape() does

[13:07:03.0000] <zcorpan>
it replaces \ with % i think

[13:07:04.0000] <kingryan>
markp: you have a question about html5lib?

[13:08:00.0000] <Hixie>
zcorpan: changed

[13:08:01.0000] <Hixie>
man we need to kill escape()/unescape() then

[13:08:02.0000] <zcorpan>
/me is happy now :)

[13:08:03.0000] <Hixie>
or at least escape()

[13:08:04.0000] <Hixie>
i guess unescape() is fine

[13:08:05.0000] <zcorpan>
it's not

[13:09:00.0000] <Hixie>
does it not do +s or something?

[13:09:01.0000] <zcorpan>
%E2%98%Ba becomes \E2\89\BA

[13:09:02.0000] <Hixie>
oh it interprets as latin1?

[13:09:03.0000] <Hixie>
interesting

[13:09:04.0000] <zcorpan>
yeah, basically

[13:16:00.0000] <Hixie>
afk

[13:21:00.0000] <markp>
kingryan: yeah, i was talking with sam ruby about adding some test cases

[13:21:01.0000] <markp>
and enhancing its functionality as a validator

[13:21:02.0000] <markp>
for example, teaching it which attributes are allowed/required in which elements

[13:21:03.0000] <kingryan>
sounds good to me

[13:22:00.0000] <kingryan>
if you add test cases I'll update the ruby version :)

[13:22:01.0000] <markp>
and then later maybe some stuff around inline content vs. strictly inline content, etc.

[13:23:00.0000] <markp>
to start with, i'd like to refactor the parserError logging

[13:23:01.0000] <markp>
to use constants enumerated in constants.py

[13:23:02.0000] <markp>
we did that in feedvalidator and later used the constant names to link to documentation pages

[13:23:03.0000] <markp>
very useful

[13:24:00.0000] <kingryan>
nice

[13:24:01.0000] <markp>
any objection?

[13:24:02.0000] <kingryan>
not from me

[13:24:03.0000] <kingryan>
I work solely with the ruby port, though

[13:24:04.0000] <markp>
i see

[14:33:00.0000] <Dashiva>
/me ponders javascript sockets

[14:48:00.0000] <zcorpan>
per rfc2646, what do you do with words that are longer than 998 characters? insert a soft linebreak?

[15:38:00.0000] <Esine>
I read slashdot daily and I've seen AKAImBatman (Intelligent Blogger) post often. His sig says "Before you adopt Silverlight, read the WHATWG specs [whatwg.org]. Article coming soon...". Are you here AKAImBatman? :) What I came here to ask was about your thought about Silverlight.

[15:39:00.0000] <Esine>
I was on a computer event and there was a Microsoft stand where some microsoftie demoed Silverlight to me

[15:39:01.0000] <Dashiva>
I'd venture to guess the thoughts are related to avoiding single-vendor lockin :)

[15:40:00.0000] <Esine>
It really seemed quite nice and much better than Flash. I'm no lover (not hater) of Microsoft or proprietary technology, but still Silverlight seemed to be quite open. If the Mono guys can really make a good silverlight plugin  for browsers and such I don't see much problem with it

[15:40:01.0000] <zcorpan>
how is it open?

[15:41:00.0000] <Esine>
I code websites a lot and I code to HTML 4.0 Strict (and some WHATWG extensions) and don't use Flash or anything, but.. really, Silverlight doesn't seem too bad if I can run it on every platform and I can develop software for it on every platform

[15:41:01.0000] <Esine>
well I *think* they have documented it all in Microsoft Developer Network (msdn.microsoft.com), but I'm not sure. I mean how else would the Mono guys do it?

[15:42:00.0000] <Esine>
So, I need you to tell me why is it bad, if it is open enough and most likely 90% people in 5 years will have it installed on their computers (because it will ship with a windows update most likely)

[15:42:01.0000] <kingryan>
Esine: can anyone else build silverlight implementations?

[15:42:02.0000] <Esine>
kingryan, yes.

[15:43:00.0000] <Esine>
See the moonlight website: http://www.mono-project.com/Moonlight

[15:43:01.0000] <Esine>
it's Silverlight implementation on Mono (C# interpreter for UNIX)

[15:43:02.0000] <kingryan>
so anyone = anyone using .Net or a .Net clone

[15:43:03.0000] <Esine>
and a browser plugin for it. It's sort of unstable right now but they're improving at rapid rate. I tried it out a month ago and only half of the examples on microsoft's site worked but..

[15:44:00.0000] <Esine>
yes I understand that's not such a nice thing, but I understand you don't NEED to code in .Net to develop on it

[15:45:00.0000] <Esine>
You can call the functions in pure JavaScript if you wish, but .Net bytecode is for performance. I personally don't intend to code in .Net or any other managed toy language like that, but I can't see it as such a bad thing if it really and truly works under all platforms with no problems

[15:46:00.0000] <kingryan>
"working on all platoforms" is not the same as "being open"

[15:46:01.0000] <Esine>
very true indeed

[15:46:02.0000] <Esine>
but still it doesn't seem to matter in real life (see Flash for example)

[15:47:00.0000] <Esine>
I'm not sure how open really it is though

[15:47:01.0000] <Esine>
since they say everything is documented on MSDN

[15:47:02.0000] <kingryan>
flash has succeeded because it can do things that other technologies can't (right now)

[15:47:03.0000] <Esine>
what we're missing right now is a high performance technology we can use on Web

[15:47:04.0000] <kingryan>
flash is useful for online video because browsers don't have a compatible way to do it today, but they hopefully will in the future

[15:48:00.0000] <Esine>
you say SVG or Canvas, well guess what, they're slow as fsck (at least the current implementations on Konqueror, Opera and Mozilla)

[15:48:01.0000] <Esine>
and they can't do everything Flash (or Silverlight for that matter) can

[15:48:02.0000] <kingryan>
there's also http://www.whatwg.org/specs/web-apps/current-work/#video

[15:48:03.0000] <kingryan>
<video>

[15:48:04.0000] <Esine>
does WHATWG have a better solution than those?

[15:48:05.0000] <Esine>
I'll check that

[15:48:06.0000] <kingryan>
and <audio>

[15:48:07.0000] <zcorpan>
3d canvas

[15:49:00.0000] <kingryan>
and sure, the implementations are currently slow, but that doesn't mean they can't be made to be fast

[15:49:01.0000] <kingryan>
at this point interoperability would be more important

[15:49:02.0000] <Esine>
I understand that but you can't get anyone to use them if they are slow

[15:49:03.0000] <webben>
kingryan: Flash popularity has more to do with an install base than theoretical interoperability.

[15:50:00.0000] <Esine>
since they offer nothing really exciting to the Web compared to other technologies, even though the other ones may not be as open

[15:50:01.0000] <kingryan>
webben: doesn't a large install base lead to interop?

[15:50:02.0000] <webben>
kingryan: Well no, not necessarily.

[15:51:00.0000] <webben>
kingryan: it just lowers the usability barrier on key platforms

[15:51:01.0000] <kingryan>
it seems to have helped flash, at least for online video

[15:51:02.0000] <othermaciej>
Silverlight is not an open standard

[15:51:03.0000] <othermaciej>
Microsoft controls it

[15:51:04.0000] <othermaciej>
maybe other people can build implementations today

[15:51:05.0000] <othermaciej>
but Microsoft is free to change it at any time

[15:51:06.0000] <othermaciej>
once it is no longer in their interests to allow clones

[15:52:00.0000] <Esine>
true..

[15:52:01.0000] <webben>
othermaciej: Hmm. But presumably that would only apply to subsequent versions? (I haven't looked at the licence at all. I just know the Mono project has been developing a Linux version.)

[15:52:02.0000] <othermaciej>
<canvas> is fast enough to code simple 3D games with

[15:53:00.0000] <Esine>
othermaciej, actually no. I can't even run 20 _wireframe_ polygons on screen at a good FPS rate

[15:53:01.0000] <othermaciej>
Esine: http://canvex.lazyilluminati.com/

[15:53:02.0000] <Esine>
yes I've seen that

[15:54:00.0000] <Esine>
but that isn't really complex, is it?

[15:54:01.0000] <Esine>
it doesn't render many polygons per frame

[15:55:00.0000] <Esine>
you say it doesn't matter since no one is going to code a Quake 3 on <canvas>. That's true, but still there are a lot other things where you need to be able to draw graphics very fast

[15:55:01.0000] <zcorpan>
we will have 3d canvas in due course, based on open gl es

[15:55:02.0000] <othermaciej>
in some screens there's probably quite a few more polygons than that, and I'm getting 30-50 fps

[15:55:03.0000] <kingryan>
Esine: I don't think performance is a big deal at this point. canvas can be made fast later.

[15:56:00.0000] <webben>
I wonder hows fast 3D (http://www.web3d.org/ ) is.

[15:56:01.0000] <webben>
s/3D/X3D/

[15:56:02.0000] <othermaciej>
Safari's <canvas> impl is pretty fast, when doing 3D your bottleneck is more likely JS execution time doing the math

[15:56:03.0000] <othermaciej>
anyway

[15:56:04.0000] <Esine>
oh yes that too

[15:57:00.0000] <Esine>
JavaScript is very slow to parse and run at real time

[15:57:01.0000] <othermaciej>
open standards based web technology is indeed behind proprietary technologies in some areas

[15:57:02.0000] <Esine>
Flash and Silverlight run compiled bytecode which again is really fast

[15:57:03.0000] <othermaciej>
a big goal of HTML5 is to close the gap

[15:57:04.0000] <Esine>
so that's one problem too..

[15:57:05.0000] <othermaciej>
even if you might still need Flash or Silverlight for some things, more things will be doable in pure web tech

[15:57:06.0000] <othermaciej>
there's no fundamental reason precompiled bytecode needs to be faster than JS

[15:57:07.0000] <Hixie>
othermaciej: so i was looking at flickr, and it's not clear to me what it would mean to take it offline, as an app

[15:57:08.0000] <othermaciej>
and modern JS engines are pretty fast

[15:58:00.0000] <Hixie>
othermaciej: is there really a problem with requiring that an offline app be written to work with a single top-level URI?

[15:58:01.0000] <Philip`>
Esine: http://blog.vlad1.com/archives/2007/07/28/149/ is a SVG vs Silverlight demonstration, in case you haven't seen it before

[15:58:02.0000] <othermaciej>
Hixie: I'd like to be able to view my own flickr photos, add tags and captions, and "upload" new photos (scheduling for real upload next time I'm online) when I don't have a net connection

[15:58:03.0000] <webben>
Hixie: Wouldn't it be a sort of synced iPhoto?

[15:59:00.0000] <othermaciej>
Hixie: right now I use iPhoto for that a lot of the time, basically only because it works offline

[15:59:01.0000] <Esine>
Philip`, no I haven't thanks

[15:59:02.0000] <Hixie>
othermaciej: ok but why would you want multiple top-level URIs for that?

[15:59:03.0000] <Philip`>
(Canvex doesn't draw 3D polygons at all - it's basically 2D, but extended vertically a bit)

[15:59:04.0000] <kingryan>
Hixie: have you looked at the flickr organizr ?

[15:59:05.0000] <Hixie>
kingryan: uri?

[15:59:06.0000] <othermaciej>
Hixie: well, flickr's current UI uses lots of different URIs

[15:59:07.0000] <kingryan>
http://flickr.com/photos/organize/

[15:59:08.0000] <othermaciej>
Hixie: should we say they need to rearchitect their whole site to support offline mode?

[16:00:00.0000] <Philip`>
(I tried porting a JVM-based software 3D renderer to <canvas>, but JavaScript is just far too slow at doing all the matrix calculations to handle 3D polygons)

[16:00:01.0000] <kingryan>
it's all flash, but if you could take that offline it'd be really useful

[16:00:02.0000] <othermaciej>
anyway, I need to go get coffe

[16:00:03.0000] <othermaciej>
will be back

[16:00:04.0000] <webben>
Multiple URIs is good, since it allows bookmarking.

[16:00:05.0000] <kingryan>
Hixie:  you might need a flickr account, though

[16:00:06.0000] <Hixie>
othermaciej: they will have to rearchitect anyway, they'd have to have all kinds of offline-specific things like an offline database, coping with updates, synchronising with conflicts, etc

[16:01:00.0000] <Philip`>
webben: I don't think the performance of X3D is something that really exists - it's just a way of describing scenes, and the performance depends entirely on the implementation

[16:01:01.0000] <Hixie>
kingryan: it won't let me see it (i haven't uploaded any photos)

[16:01:02.0000] <Hixie>
kingryan: was there anything specific about it you wanted me to look at?

[16:02:00.0000] <webben>
Philip`: I see what you mean.

[16:02:01.0000] <kingryan>
Hixie: nothing in specific, but it is like an "online iphoto"

[16:02:02.0000] <Hixie>
kingryan: ah ok

[16:02:03.0000] <Esine>
Philip`, I did a JavaScript and <canvas> 3D renderer/engine that does solid lightning and animation, but as I said before I could only get like 30fps with a _five_ polygon object on screen

[16:02:04.0000] <Philip`>
(i.e. it's like talking about the performance of RTF files, or something, so it's not relevant except to the extent that it dictates the architecture of the implementations)

[16:02:05.0000] <kingryan>
if they could take that offline, do the oranizing, then resync, it could be a useful use case

[16:02:06.0000] <Hixie>
webben: i'm not sure what it really means to talk about bookmarking the state of an offline app

[16:02:07.0000] <Esine>
that's why I was so interested in Silverlight.. because it seemed so fast

[16:03:00.0000] <webben>
Hixie: I guess it depends on how weblike the offline app is. Bookmarking makes perfect sense with offline web pages.

[16:03:01.0000] <Philip`>
OpenGL <canvas> would be much better - I had hundreds of texture-mapped polygons with specular lighting and reflective water underneath, going at 60fps with no problem :-)

[16:04:00.0000] <Hixie>
webben: i dont really see what it would mean to e.g. bookmark the state of gmail

[16:04:01.0000] <Esine>
Philip`, you mean you had that on the opengl <canvas> or normal 2d <canvas>?

[16:04:02.0000] <webben>
That's because Gmail isn't a remotely REST-like app.

[16:05:00.0000] <webben>
very little state is stored in simple URIs.

[16:05:01.0000] <Philip`>
Esine: The OpenGL one

[16:05:02.0000] <Esine>
is there any patch  for adding opengl support for browsers?

[16:05:03.0000] <Esine>
Philip`, which browser?

[16:05:04.0000] <webben>
Flickr doesn't seem like that.

[16:05:05.0000] <Hixie>
webben: i don't see this as a problem...

[16:06:00.0000] <Philip`>
Esine: There's an experimental implementation in Mozilla, but I don't know if it still works, and you probably have to compile Firefox yourself to use it

[16:06:01.0000] <webben>
Hixie: I guess it's a problem if you're trying to maintain a consistent interface between offline and online.

[16:06:02.0000] <Esine>
Philip`, is it in the SVN trunk?

[16:06:03.0000] <Hixie>
webben: i would understand having the photo viewing side bookmarkable, since there you are viewing photos

[16:06:04.0000] <Philip`>
Esine: http://lxr.mozilla.org/seamonkey/source/extensions/canvas3d/

[16:06:05.0000] <Hixie>
webben: but that's usually someone else's photos

[16:06:06.0000] <Esine>
thanks

[16:07:00.0000] <Hixie>
webben: for the app part -- changing tags, etc -- i don't really see what you would bookmark

[16:08:00.0000] <webben>
Hixie: Well, I guess I'd assume you'd still be able to browse your collection offline.

[16:08:01.0000] <Hixie>
webben: that would be a LOT of data to download

[16:08:02.0000] <Philip`>
/me predicts that OpenGL <canvas> would be great fun for anyone who cares about security (either breaking it or fixing it) in browsers...

[16:09:00.0000] <Hixie>
webben: but hm

[16:09:01.0000] <webben>
Hixie: I was kind of envisaging permanent sync.

[16:09:02.0000] <Philip`>
(2D <canvas> is bad enough for security already)

[16:09:03.0000] <webben>
If it's all about upload, I'm not sure how much value it would add to the existing tools.

[16:10:00.0000] <Esine>
Philip`, oh? mind telling me about that?

[16:10:01.0000] <Esine>
security implications of <canvas>

[16:10:02.0000] <Dashiva>
Intranet images mainly

[16:11:00.0000] <Philip`>
Esine: It's not really <canvas>'s fault, it's just the implementations - they tend to corrupt memory and crash, or expose random chunks of browser memory to scripts, because they have bugs (or use libraries with bugs) or forget to deal with certain cases

[16:12:00.0000] <Hixie>
webben: interesting

[16:12:01.0000] <Esine>
oh..

[16:13:00.0000] <Philip`>
Esine: They're also big DOS targets - you can try creating massive bitmaps, or painting hugely complex shapes, and browsers tend to just freeze or crash instead of handling it gracefully

[16:16:00.0000] <webben>
Hixie: OTOH, I guess it would be free whereas the iPhoto plugin for Flickr actually costs money.

[16:19:00.0000] <othermaciej>
Hixie: assuming that flickr wanted back/forward to work, and wanted people to be able to send each other URLs to specific photos or photosets or profile pages etc, then the only way they could have a single top-level URI for offline mode is to use a single URI for the whole site

[16:19:01.0000] <othermaciej>
with fragment IDs

[16:19:02.0000] <othermaciej>
it seems very intrusive for offline support to require doing that to your site

[16:21:00.0000] <Philip`>
(Incidentally, I'd like to try implementing some subset of X3D if browsers provided a lower-level (OpenGL) rendering API - I don't really know anything about it, but it looks generally sensible and it'd be useful when you want convenience at the expense of flexibility)

[16:22:00.0000] <othermaciej>
so instead of http://flickr.com/photos/othermaciej/329780552/ you would have to have URLs like http://flickr.com/#photos/othermaciej/329780552

[16:22:01.0000] <othermaciej>
and the server would only ever serve one actual page per person

[16:22:02.0000] <othermaciej>
I can imagine all sorts of ways this could be bad for their caching strategies

[16:29:00.0000] <Philip`>
Esine: By the way, have you seen http://polyhedra.org/poly/ which does 3D polygons at a reasonable speed when it's not just showing PHP errors?

[16:29:01.0000] <Esine>
no I haven't..

[16:29:02.0000] <Esine>
I'll bookmark that page

[16:30:00.0000] <Hixie>
othermaciej: yeah

[16:30:01.0000] <Hixie>
othermaciej: i don't see how to do the offline stuff in a simple yet workable fashion without having a single top-level URL

[16:31:00.0000] <Hixie>
but clearly we'll have to do so

[16:31:01.0000] <Philip`>
Esine: (Also there's http://matt.west.co.tt/files/canvastastic/canvastastic_beta_1/eg/robot.html which is a less reasonable speed but not too awful)

[16:32:00.0000] <Esine>
yeah well that's around the same speed I was getting

[16:32:01.0000] <othermaciej>
Hixie: I will think on it more and reply to other aspects of your email, I've been asked to look at offline support more at work

[16:32:02.0000] <othermaciej>
the dancing robot doesn't look half bad in my Safari 3 build

[16:33:00.0000] <othermaciej>
I don't know what the FPS is

[16:33:01.0000] <kingryan>
the robot looks good in safari 3.0.3

[16:34:00.0000] <Philip`>
http://canvex.lazyilluminati.com/3dmodel/example0.html pushes Canvastastic harder, but unfortunately it very much doesn't work in anything except Firefox

[16:35:00.0000] <Philip`>
(since it uses E4X to parse Collada models, and some JS1.7 bits just for fun)

[16:35:01.0000] <Esine>
yeah can't see anything in Opera

[16:37:00.0000] <Philip`>
I get 1.5fps

[16:46:00.0000] <markp>
wow, i so totally called this 3 months ago: http://diveintomark.org/archives/2007/05/02/silly-season#comment-9288

[16:48:00.0000] <Dashiva>
Oh it's a markp

[16:49:00.0000] <Hixie>
i love the comment a bit lower down on that page

[16:49:01.0000] <Hixie>
saying "Matt Cutts : Google :: Silverlight : The Future"

[16:49:02.0000] <Hixie>
i can't tell if they're saying that silverlight is the best thing ever or if they're saying that silverlight is merely one tiny part of the future

[16:50:00.0000] <Dashiva>
Maybe it's SAT form. "Silverlight is to the future The Future as..."

[16:50:01.0000] <markp>
that's not the worst of it

[16:50:02.0000] <markp>
it's stuff like http://diveintomark.org/archives/2007/05/02/silly-season#comment-9310 that makes me realize how far away i am from most people

[16:51:00.0000] <Dashiva>
I recognize that comment, but I can't imagine how I managed to read through all the preceding ones without going mad

[16:53:00.0000] <markp>
it deteriorated rapidly after that

[16:54:00.0000] <markp>
as blogs are prone to do

[16:58:00.0000] <Dashiva>
I think the comment that blog comments have too low barrier of entry had something going for it

[16:59:00.0000] <Philip`>
Maybe that's why people started using CAPTCHAs like "7 - 2 = ?", to weed out the kind of people who leave comments at certain less-cultured sites


2007-08-28
[17:01:00.0000] <markp>
i once threatened to make captchas based on strunk and white

[17:01:01.0000] <Philip`>
You could extend it to something like "d^2x/dy^2 - 5 x = 0; y = ?" if you want to raise the barrier a little

[17:06:00.0000] <Dashiva>
Make them write alt text for the captcha

[17:08:00.0000] <Philip`>
Make them write alt text for one of the 3000 unlabelled holiday photos that the blog's owner uploaded to Flickr

[17:09:00.0000] <Dashiva>
I wonder if anyone has contacted Flickr for statistics :)

[17:09:01.0000] <markp>
actually i've thought about using mechanical turk for adding alt text

[17:10:00.0000] <markp>
it's a skill, it can be taught, individuals could get good at it and get paid for it

[17:10:01.0000] <webben>
One could certainly use something like Google Image Labeller as some sort of CAPTCHA. reCAPTCHA uses captcha responses to OCR books.

[17:10:02.0000] <markp>
recaptcha++

[17:11:00.0000] <Dashiva>
Any captcha can be broken by an ample supply of cheap human labor, after all

[17:11:01.0000] <webben>
throw Google Image Labeller together with something like http://research.microsoft.com/asirra/

[17:12:00.0000] <markp>
seriously though, imagine if flickr (yahoo) sponsored an ongoing project to add really good alt text to images as people uploaded them

[17:12:01.0000] <markp>
photo owners could be notified when their images got tagged and could accept/reject/edit them

[17:13:00.0000] <webben>
markp: Yep that's one (of many ways) Flickr could start producing alt text.

[17:13:01.0000] <Hixie>
do blind people care?

[17:14:00.0000] <webben>
Hixie: Some do. Some don't. Bit like other people really.

[17:14:01.0000] <markp>
there's more to flickr than images

[17:14:02.0000] <Hixie>
i don't understand why anyone would want photos described. they're boring enough when you can see them.

[17:14:03.0000] <webben>
flickr turns photos into social objects

[17:14:04.0000] <markp>
people add annotations, tags, comments

[17:15:00.0000] <markp>
all of which are text

[17:15:01.0000] <Hixie>
yeah i never understood the point of the tags and annotations either

[17:15:02.0000] <Hixie>
the captions and comments make sense

[17:15:03.0000] <Hixie>
if you're showing pictures to your friends

[17:15:04.0000] <Hixie>
/me shrugs

[17:16:00.0000] <markp>
anything that increases the amount of accessible content in the world is a good thing

[17:16:01.0000] <markp>
the street will find its own uses for it

[17:16:02.0000] <Hixie>
i don't really see how alt text increases the accessibility of a photo

[17:16:03.0000] <Hixie>
it's like lyrics for a song

[17:17:00.0000] <Hixie>
ok, so you can see the lyrics

[17:17:01.0000] <Hixie>
still doesn't give you the song

[17:17:02.0000] <webben>
I think people get too hung up on the irreducibility of experiences.

[17:17:03.0000] <kingryan>
Hixie: but if you can't hear the song, but everyone's talking about it, wouldn't it be nice to have the context of the lyrics?

[17:17:04.0000] <webben>
Often the song, the image are less important (or only as important) as they social meaning and context anyhow

[17:17:05.0000] <markp>
/me quietly hums "we didn't start the fire"

[17:18:00.0000] <Hixie>
kingryan: i have a "disability" in that i cannot recognise fashionable clothing from unfashionable clothing. My solution is not to have people describe to me what is fashionable and what isn't. My solution is to not care.

[17:18:01.0000] <Hixie>
/me shrugs again

[17:19:00.0000] <Philip`>
I would guess the solution of having a female decide what you should wear is a more common one for that problem :-)

[17:20:00.0000] <Hixie>
my girlfriend actually has the same "disability" as i do. our solution is to get her gay boyfriend to help us pick clothes when necessary. :-)

[17:20:01.0000] <othermaciej_>
tags are pretty good for photos, though they don't seem as nice for most other things

[17:21:00.0000] <Hixie>
othermaciej: oh i use tags a ton on iPhoto. I just don't see the point in tags for other people's photos.

[17:21:01.0000] <webben>
othermaciej: Do you not find tags useful for bookmarking?

[17:22:00.0000] <Hixie>
othermaciej: but then, i don't really care about photos that aren't relevant to me, so...

[17:22:01.0000] <mpt>
Hixie, do you ever use Google Image Search?

[17:23:00.0000] <Hixie>
mpt: yeah, that's a great example of where tagging isn't necessary.

[17:24:00.0000] <Hixie>
even in iPhoto, image tagging is really just a workaround for the computer's inability to recognise content in the photo

[17:24:01.0000] <Hixie>
i don't enjoy manually tagging photos, i shouldn't have to

[17:24:02.0000] <webben>
Good example of the blind caring about the social context of photography was a recent blindtech question: a couple of blind parents looking for a photo sharing site that didn't require a visual CAPTCHA to register, so they could upload some photos of their newborn

[17:24:03.0000] <Hixie>
i should just be able to ask the computer for every photo in the collection with a cat in it

[17:24:04.0000] <Hixie>
or every photo that has an element in focus

[17:24:05.0000] <Hixie>
(or doesn't)

[17:24:06.0000] <mpt>
How would you know that it has a cat in it, if "cat" is in the tags but not in the title/description?

[17:24:07.0000] <webben>
Hixie: you want Riya and Polar Rose ;)

[17:25:00.0000] <webben>
http://www.riya.com/ and http://www.polarrose.com/

[17:25:01.0000] <othermaciej>
Hixie: they are interesting in flickr because (a) it makes searching work better than just captions and titles would, given that AI image anaysis does not exist yet;

[17:26:00.0000] <othermaciej>
Hixie: and (b) because they provide a categorization scheme, so I can easily see "all of my friends' pictures of dogs" or the like

[17:26:01.0000] <Hixie>
webben: again, those are their own photos, and the alt text wouldn't really help insofar as i can tell -- how could text possibly convey anything useful?

[17:26:02.0000] <othermaciej>
webben: I don't find boomkarking useful, so I'm not a good person to ask

[17:26:03.0000] <Hixie>
mpt: image analysis

[17:26:04.0000] <webben>
Hixie: which photo is which, primarily.

[17:26:05.0000] <mpt>
ha ha

[17:26:06.0000] <Hixie>
mpt: (same way a human can tell)

[17:26:07.0000] <Hixie>
webben: that's a title

[17:26:08.0000] <webben>
othermaciej: ah okay :)

[17:27:00.0000] <Hixie>
othermaciej: again i'm just talking about my personal preference, but personally, i have never found search on flickr to be useful

[17:27:01.0000] <Hixie>
the image on my blog being a great example of how it's not useful :-)

[17:28:00.0000] <webben>
Hixie: Titles don't necessarily identify what images actually are. (They might be a pun etc... )

[17:28:01.0000] <webben>
also, you can display images without titles

[17:28:02.0000] <webben>
in which case you'd need alt to tell them apart

[17:28:03.0000] <Hixie>
mpt: but, text that identifies the image is useful to everyone, not just if you don't have images.

[17:28:04.0000] <Hixie>
er

[17:28:05.0000] <webben>
(e.g. Flickr photostreams)

[17:28:06.0000] <Hixie>
s/mpt/webben/

[17:29:00.0000] <Hixie>
note that i'm not arguing against including alt text here

[17:29:01.0000] <webben>
Hixie: It's more useful to people who can't see the image, since to some degree you can tell images apart just by looking at them.

[17:29:02.0000] <Hixie>
i'm just saying i don't understand why anyone would care for pure alt text (as opposed to text that is shown near the picture even when you do have images)

[17:29:03.0000] <Hixie>
webben: *shrug*

[17:29:04.0000] <mpt>
I thought we were talking about tags

[17:29:05.0000] <mpt>
My apologies if we weren't

[17:30:00.0000] <webben>
Hixie: Yeah. I can't really see why you wouldn't want an image described.

[17:30:01.0000] <mpt>
tags would be lousy alt= text

[17:30:02.0000] <Hixie>
mpt: regarding tags, all i was saying was that they are an annoying and hopefully temporary workaround for searching photos.

[17:30:03.0000] <webben>
Then again, my best friend recently finished writing about a third of the catalogue for an eighteenth-century art exhibition. So I've got well-used to reading long descriptions ;)

[17:31:00.0000] <othermaciej>
Hixie: I find it useful for ego search :-)

[17:31:01.0000] <Philip`>
Hixie: Where "temporary" means "until we've solved AI"? :-)

[17:31:02.0000] <Hixie>
webben: that's another example of how the alt text for a photo should be visible (and isn't "alt text")

[17:31:03.0000] <othermaciej>
Hixie: or finding pictures of specific other people

[17:31:04.0000] <webben>
(It's one of the reasons I think that Lady of Shalott example isn't that good. In practice, images of the Lady are all interestingly different.)

[17:31:05.0000] <Hixie>
Philip`: right

[17:31:06.0000] <Hixie>
Philip`: a specific form of ai, image analysis, which is actually quite advanced

[17:31:07.0000] <Hixie>
Philip`: i wouldn't call it AI really

[17:32:00.0000] <webben>
Hixie: Well, not really. It's just that analysis of an image mixes in description of it.

[17:32:01.0000] <Hixie>
webben: the example in the spec is purely decorative, imho

[17:32:02.0000] <webben>
Strictly, you'd need yet more description to understand the image.

[17:32:03.0000] <webben>
I suppose it's partly a matter of print vs online conventions. In print, decorative paintings tend to be credited.

[17:33:00.0000] <Hixie>
webben: i don't honestly see that any amount of description can explain a photo or painting.

[17:33:01.0000] <Philip`>
People seem to still have problems doing image analysis to detect a human face in an image, when it's not facing straight at the camera

[17:33:02.0000] <webben>
Hixie: I think that's like saying: I don't see how a photo can represent the Grand Canyon.

[17:33:03.0000] <Hixie>
webben: indeed

[17:33:04.0000] <webben>
experiences are not reducible in their entirety

[17:33:05.0000] <Hixie>
webben: i agree

[17:33:06.0000] <webben>
but they are reducible enough that the reduction may be useful

[17:34:00.0000] <Hixie>
webben: but that reduction need not be hidden from those who can see the unreduced form

[17:34:01.0000] <Hixie>
webben: (especially since if it's not hidden from them, they can improve it, wiki style)

[17:35:00.0000] <webben>
Hixie: Well that's true. (And a strong argument for allowing authors to label explicitly visible text as an alternative/complement.) But not all designers will want all that text.

[17:35:01.0000] <Hixie>
webben: not all designers of what?

[17:35:02.0000] <webben>
pages

[17:35:03.0000] <Hixie>
webben: i thought we were talking about photo gallery pages like flickr

[17:35:04.0000] <webben>
Also, there's no reason why sighted people shouldn't have access to long descriptions on another page

[17:36:00.0000] <Hixie>
(or at least, sites where the image is the point)

[17:36:01.0000] <webben>
Hixie: Oh. I thought we were talking about alt's for photos generally.

[17:36:02.0000] <Hixie>
webben: another page is equivalent to hiding the text in the alt attribute, which i content is bad.

[17:36:03.0000] <Hixie>
contend

[17:37:00.0000] <webben>
Hixie: Well, not hiding the text can also be bad, because in practice when you write text alongside an image you risk assuming the image.

[17:37:01.0000] <othermaciej>
I think in the case of photo galleries, most text that would be useful to someone who can't consume the image would be useful to those who can as well

[17:37:02.0000] <webben>
(and hence failing to reduce the experience at all)

[17:37:03.0000] <webben>
othermaciej: probably. But not when e.g. in a slideshow.

[17:38:00.0000] <othermaciej>
and offering the ability to add visible text will probably in practice lead to better text than the ability to add normally hidden text

[17:38:01.0000] <Hixie>
webben: the rusk that the text will assume the image is far less imho than the risk that the text will be omitted altogether.

[17:38:02.0000] <Hixie>
risk

[17:38:03.0000] <Hixie>
in a slideshow you wouldn't show the text at all

[17:38:04.0000] <webben>
Hixie: well exactly, but you would want the alternative text.

[17:38:05.0000] <Hixie>
you also wouldn't expect a blind person to use a slideshow

[17:38:06.0000] <webben>
i don't see why not

[17:39:00.0000] <webben>
it's just a sequential progress through images

[17:39:01.0000] <Hixie>
it's a specifically visual representation!

[17:39:02.0000] <othermaciej>
flickr's slideshow feature doesn't show any of the text

[17:39:03.0000] <webben>
it's a sequence is all

[17:39:04.0000] <othermaciej>
the photostream is also a sequence

[17:39:05.0000] <webben>
yep

[17:39:06.0000] <Hixie>
the semantic might be a sequence, but the presentation is a visual one

[17:40:00.0000] <webben>
partly

[17:40:01.0000] <Hixie>
now another stylesheet might be aural, sure

[17:40:02.0000] <Hixie>
but then it's not a slideshow.

[17:40:03.0000] <othermaciej>
the reason for a slideshow to exist is that showing the images in large format on a single page is not convenient for visual consumption

[17:40:04.0000] <webben>
e.g. a lot of slideshows have an audio component

[17:40:05.0000] <Hixie>
/me doesn't understand the point webben is trying to make

[17:40:06.0000] <othermaciej>
lots of text at once vs. one bit of text at a time does not have the same kind of difference in usability

[17:40:07.0000] <webben>
othermaciej: well it can also be about the order, depending on the sort of slideshow.

[17:41:00.0000] <Hixie>
anyway

[17:41:01.0000] <Hixie>
i'm surprised nobody has yet mailed public-html quoting this and telling DanC that i'm clearly showing lack of care for the disabled and should be banned from the group

[17:42:00.0000] <Hixie>
the complainers are clearly not running at their usual speed today

[17:44:00.0000] <kingryan>
where's andy mabbett when you need him?

[17:47:00.0000] <kingryan>
ok, maybe that's only funny in the microformats channel

[17:48:00.0000] <hober>
/me laughed

[18:04:00.0000] <Philip`>
Lachy: About "IE 5.x [is] so insignificant, [it's] not even worth looking at": http://www.thecounter.com/stats/2007/August/browser.php suggests it's about as (in)significant as all versions of Opera combined

[18:05:00.0000] <webben>
Philip`: One possible issue with that is that they may not be efficiently filtering out bots that claim to be IE 5.5.

[18:05:01.0000] <webben>
(never mind opera masquerading as other browsers)

[19:42:00.0000] <Lachy>
Philip`, IE5.x is also obsolete, Opera isn't.

[19:44:00.0000] <othermaciej>
I like to look at this chart: http://marketshare.hitslink.com/report.aspx?qprid=6

[19:44:01.0000] <othermaciej>
but there are other considerations like upward vs. downward trend, and share in particular markets

[19:45:00.0000] <othermaciej>
note, it shows Opera 9.x as significantly higher than IE 5.5

[19:46:00.0000] <othermaciej>
I find the "thecounter" stats dubious

[19:46:01.0000] <othermaciej>
their 10% "Netscape comp." category seems too large even for an "other" total, let alone an actual different browser

[19:49:00.0000] <othermaciej>
http://marketshare.hitslink.com/report.aspx?qprid=0 seems pretty in line with the general view of the top 4 browser engines

[19:53:00.0000] <othermaciej>
incidentally wikipedia seems to agree with me that this particular usage share survey is the most credible: http://en.wikipedia.org/wiki/Comparison_of_web_browsers

[19:57:00.0000] <Lachy>
zcorpan, see RFC 3676, which obsoletes 2646. I think the delsp parameter is what you need when splitting long words, though I particularly wouldn't worry about it for our needs

[20:05:00.0000] <MikeSmith>
othermaciej - interesting to see how high up the Nintendo Wii is in the "Operating System Market Share" chart there

[20:05:01.0000] <MikeSmith>
http://marketshare.hitslink.com/report.aspx?qprid=2

[20:07:00.0000] <MikeSmith>
and how up up the browser charts Opera Mini is, as well as other mobile browsers

[20:08:00.0000] <MikeSmith>
especially that they now have more share than particular versions of desktop browsers

[20:11:00.0000] <karlUshi>
http://www.xitimonitor.com/en-us/browsers-barometer/ie7-and-firefox-2-july-2007/index-1-2-3-103.html

[21:12:00.0000] <Lachy>
Hixie, do you have PHP4 or 5 running on status.whatwg.org?

[21:25:00.0000] <Hixie>
Lachy: yes

[21:25:01.0000] <Hixie>
(dunno which one)

[21:26:00.0000] <Lachy>
hmm. the response header tells me: Server: Apache/2.0.54 (Unix) PHP/4.4.7 mod_ssl/2.0.54 OpenSSL/0.9.7e mod_fastcgi/2.4.2 DAV/2 SVN/1.4.2

[02:09:00.0000] <Lachy>
zcorpan, yt?

[02:12:00.0000] <Lachy>
zcorpan, I completely rewrote the emailing code. made it object oriented, dealt with the word wrapping and fixed a bunch of other stuff

[03:30:00.0000] <jwalden>
there's a lunar eclipse right now if you're in CA

[03:31:00.0000] <Lachy>
also if you're in eastern Aus

[03:35:00.0000] <krijnh>
Also if you're in NL

[04:34:00.0000] <zcorpan>
Lachy: i did some changes to the php yesterday, but utf8_wordwrap() still needs a 4th argument

[04:34:01.0000] <Lachy>
utf8_wordwrap doesn't work well at all, I did some testing with it and it's not needed

[04:34:02.0000] <zcorpan>
aha

[04:34:03.0000] <zcorpan>
ok

[04:35:00.0000] <zcorpan>
wordwrap() is ok?

[04:35:01.0000] <Lachy>
the existing wordwrap function seems to work reasonably well

[04:35:02.0000] <zcorpan>
good then

[04:35:03.0000] <Lachy>
there's only one bug that I found, but it's never likely to occur

[04:35:04.0000] <zcorpan>
split inside a multibyte character?

[04:36:00.0000] <Lachy>
if you have a long line longer than 990 characters, without any white space, and the bytes 990-991 are part of a multibyte char

[04:36:01.0000] <zcorpan>
right

[04:36:02.0000] <zcorpan>
the url will be ascii only

[04:36:03.0000] <Lachy>
yeah

[04:36:04.0000] <zcorpan>
and the rest won't contain that long lines

[04:37:00.0000] <Lachy>
since it gets urlencoded

[04:37:01.0000] <zcorpan>
yeah even that

[04:37:02.0000] <zcorpan>
ok

[04:53:00.0000] <Lachy>
/me is going to write some twitter posting code

[06:56:00.0000] <Lachy>
PHP is annoying, at least the way it's configured on my server. I need to use constructors and destructors in PHP5, but the curl functions only work when I use PHP4 :-(

[14:53:00.0000] <zcorpan>
Lachy_: the DelSp parameter was not added to handle lines longer than 998 characters. i'm still not sure what generating agents are supposed to do with such lines

[15:09:00.0000] <markp>
kingryan: i've checked in revision 962 of html5lib

[15:09:01.0000] <kingryan>
markp: I see that

[15:10:00.0000] <kingryan>
I'll updated the ruby version this evening

[15:10:01.0000] <markp>
ok


2007-08-29
[20:57:00.0000] <markp>
jgraham: you around?

[20:57:01.0000] <markp>
or any other html5lib hackers?

[07:14:00.0000] <virtuelv>
is anyone going to be offended if I refer to the selectors API naming debate as a bikeshed problem?

[07:19:00.0000] <zcorpan>
not me

[07:19:01.0000] <zcorpan>
and i participated in it :(

[07:20:00.0000] <zcorpan>
/me will stay away from naming debates in the future

[07:20:01.0000] <Dashiva>
It was a bikeshed problem where all the colors were ugly shades of purple and beige

[07:22:00.0000] <virtuelv>
fwiw, http://programming.reddit.com/info/2jrrg/comments 

[07:27:00.0000] <zcorpan>
accessing radiobuttons with the [[Get]] method is funny -- if there's only one control that matches, it returns it directly, but if there are more it returns a nodelist

[07:27:01.0000] <zcorpan>
document.forms[0]["foo"]

[07:28:00.0000] <zcorpan>
and undefined if 0 match

[07:36:00.0000] <Dashiva>
Same with checkboxes

[07:37:00.0000] <zcorpan>
yeah

[07:42:00.0000] <Dashiva>
And we should do our best to prevent the same behavior from happening to responsebodies in multipart xhr responses :)

[08:59:00.0000] <gsnedders>
is there anyway to check WF2 support through JS?

[09:03:00.0000] <gsnedders>
document.implementation.hasFeature('WebForms', '2.0')?

[09:16:00.0000] <krijnh>
gsnedders: Yeah, that returns true in Opera

[09:18:00.0000] <zcorpan>
gsnedders: it's probably safer to check specific methods etc before using them

[09:18:01.0000] <gsnedders>
zcorpan: need some way to check if |input|@type=date has a native controller before falling back to a JavaScript one

[09:19:00.0000] <gsnedders>
(which will just be a text field on something with neither WF2 or JS)

[09:19:01.0000] <zcorpan>
.type == "date" ?

[09:19:02.0000] <gsnedders>
does that work?

[09:19:03.0000] <zcorpan>
think so

[09:19:04.0000] <gsnedders>
does that set itself to "text" in browsers that don't support it?

[09:20:00.0000] <zcorpan>
exactly

[09:20:01.0000] <zcorpan>
so you know if it's not supported

[09:20:02.0000] <zcorpan>
if (input.type != "date") { // not supported

[09:20:03.0000] <gsnedders>
falls back in Saf at least

[09:22:00.0000] <Dashiva>
falls back in ie and ff too

[09:22:01.0000] <Dashiva>
ff2, ie7

[09:22:02.0000] <zcorpan>
there you go

[09:23:00.0000] <gsnedders>
zcorpan: thanks

[09:23:01.0000] <zcorpan>
welcome :)

[11:14:00.0000] <Lachy>
http://ajaxian.com/archives/selectors-api-method-names-selectelement-and-selectallelements

[11:14:01.0000] <Lachy>
http://dean.edwards.name/weblog/2007/08/names/

[11:14:02.0000] <Lachy>
http://lachy.id.au/log/2007/06/naming-debate

[11:17:00.0000] <Lachy>
oops, wrong link. http://lachy.id.au/log/2007/08/naming-debate-revisited

[12:05:00.0000] <gsnedders>
kingryan: what did you do regarding text/plain sniffing?

[12:05:01.0000] <kingryan>
gsnedders: what do you mean?

[12:06:00.0000] <gsnedders>
kingryan: in your implementation of feed/html, did you just count every text/plain document as such, and not do what browsers do, and sniff it?

[12:06:01.0000] <kingryan>
the sniffing impl I've added to html5lib/ruby isn't actually integrated w/ the rest of the parser

[12:07:00.0000] <kingryan>
in other words, I haven't implemented the part of the decision tree that switches on content-type

[12:08:00.0000] <kingryan>
... if that answers your question, gsnedders 

[12:08:01.0000] <gsnedders>
ya

[12:08:02.0000] <kingryan>
I've also implemented it at technorati, where we don't really care about text/plain

[12:08:03.0000] <kingryan>
we only really care wether something is a feed or html

[12:08:04.0000] <kingryan>
or other

[12:09:00.0000] <gsnedders>
kingryan: but what if text/plain _is_ a feed or HTML?

[12:09:01.0000] <kingryan>
I wasn't clear... we don't care if people *say* its text/plain

[12:09:02.0000] <gsnedders>
you just sniff it regardless of the claim?

[12:10:00.0000] <kingryan>
yup

[12:11:00.0000] <kingryan>
but we're only indexing URLs that either 1) people pinged us with or 2) were discovered through link[@rel=~alternate]

[12:13:00.0000] <kingryan>
so, it works well enough for us

[12:13:01.0000] <gsnedders>
the feed/html algorithm currently concludes "otherwise return text/html". How do you tell apart HTML from the rest?

[12:14:00.0000] <kingryan>
because of where it sits in the decision tree, it doesn't matter

[12:14:01.0000] <markp>
/me goes off to start a text/plain blog

[12:14:02.0000] <kingryan>
if someone says "this is a feed", but its not, we give up on processing it

[12:14:03.0000] <markp>
does that happen a lot?

[12:14:04.0000] <kingryan>
the common case is an http 500 which has html

[12:15:00.0000] <markp>
heh

[12:15:01.0000] <kingryan>
or 503

[12:15:02.0000] <gsnedders>
503?

[12:15:03.0000] <kingryan>
503 = "temporarily unavailable"

[12:15:04.0000] <markp>
http error 503: my web hosting sucks

[12:15:05.0000] <gsnedders>
markp: I was wondering whether I'd get such a response from you :)

[12:15:06.0000] <kingryan>
the more common case is that someone says their feed is rss , but its really atom

[12:16:00.0000] <gsnedders>
I don't even differentiate between the two. Just check if we have a feed or not.

[12:16:01.0000] <markp>
kingryan: do you use feedparser to determine feed version?

[12:17:00.0000] <markp>
just curious

[12:19:00.0000] <kingryan>
markp: our old spider uses UFP, but is slowly being replaced by one written in ruby

[12:20:00.0000] <kingryan>
for which the parsing is mostly based on html5lib

[12:33:00.0000] <Philip`>
/me tries running the html5lib validator on his documentation examples

[12:34:00.0000] <Philip`>
http://james.html5.org/cgi-bin/parsetree/parsetree.py?source=%3C%21DOCTYPE+HTML%3E%0D%0A%3Ctitle%3E%3Cb%3E+%26amp%3B+%3Ci%3E%3C%2Ftitle%3E - how come that's sensible there, but gives different results in SVN html5lib?

[12:40:00.0000] <zcorpan>
Philip`: what does the latter give?

[12:41:00.0000] <zcorpan>
shouldn't the innerHTML view escape < and > in title, btw?

[12:41:01.0000] <zcorpan>
and &

[12:42:00.0000] <Philip`>
>>> print p.parse('<!doctype html><title><b> &amp; <i></title>').printTree()

[12:42:01.0000] <Philip`>
#document

[12:42:02.0000] <Philip`>
|  <!DOCTYPE html>

[12:43:00.0000] <Philip`>
|  <html>

[12:43:01.0000] <Philip`>
|    <head>

[12:43:02.0000] <Philip`>
|      <title>

[12:43:03.0000] <Philip`>
|        <body>

[12:43:04.0000] <Philip`>
|          <b>

[12:43:05.0000] <Philip`>
|            " & "

[12:43:06.0000] <Philip`>
|            <i>

[12:43:07.0000] <zcorpan>
aha.

[12:44:00.0000] <zcorpan>
regression then

[12:45:00.0000] <Philip`>
Looks like <title> is meant to be parsed as RCDATA

[12:46:00.0000] <zcorpan>
yes

[14:50:00.0000] <jgraham>
Oh attribute values

[14:50:01.0000] <jgraham>
I didn't do those

[14:50:02.0000] <zcorpan>
right

[14:52:00.0000] <jgraham>
That looks a bit more promising

[14:54:00.0000] <zcorpan>
jgraham: <plaintext> is also a cdata element

[14:54:01.0000] <zcorpan>
for this purpose anyway

[14:55:00.0000] <zcorpan>
otherwise looks right

[14:56:00.0000] <zcorpan>
hmm, http://james.html5.org/cgi-bin/parsetree/parsetree.py?source=%3Cnoscript%3E doesn't look like it's parsed correctly (or there's a bug in the spec)

[14:56:01.0000] <Philip`>
markp: Oh, I'm happy to slack - I just wanted to see if it worked already, and only found that one problem, so I'm content to wait until it works even better in the future ;-)

[15:00:00.0000] <zcorpan>
"<noscript>" is supposed to be parsed into <html><head><noscript></noscript></head><body></body></html> if i read the spec right

[15:01:00.0000] <jgraham>
zcorpan: That looks more plausible

[15:02:00.0000] <jgraham>
/me is wondering why html5lib has constants named rcdataElements and cdataElements that seem to be almost exactly back to front

[15:31:00.0000] <jgraham>
zcorpan: Where should the </noscript> be generated in the spec?

[15:33:00.0000] <zcorpan>
jgraham: i don't understand the question

[15:34:00.0000] <zcorpan>
jgraham: you mean as an implied token in the tree construction?

[15:34:01.0000] <jgraham>
Yeah

[15:34:02.0000] <zcorpan>
they aren't

[15:35:00.0000] <zcorpan>
the cdata parsing algorithm just looks until it finds a token that's not a character token

[15:35:01.0000] <zcorpan>
if that's the end tag it's ignored

[15:35:02.0000] <zcorpan>
otherwise you carry on

[15:35:03.0000] <zcorpan>
so say the tokens are: start tag "noscript", character "X", end-of-file

[15:36:00.0000] <jgraham>
Oh, I see

[15:36:01.0000] <jgraham>
We're mistakenly inserting the token for <noscript> into the stack of open elements

[15:36:02.0000] <zcorpan>
when you get to the end-of-file it's a parse error and you get back to "in head"

[15:37:00.0000] <zcorpan>
seems so

[15:39:00.0000] <zcorpan>
it should get the same treatment as <style> i think

[15:39:01.0000] <zcorpan>
basically

[15:40:00.0000] <zcorpan>
or exactly :)

[15:41:00.0000] <zcorpan>
if scripting is enabled

[15:41:01.0000] <jgraham>
zcorpan: I think we get <style> wrong too

[15:41:02.0000] <zcorpan>
http://james.html5.org/cgi-bin/parsetree/parsetree.py?source=%3Cstyle%3E

[15:41:03.0000] <jgraham>
Eh, that doesn't look like it should work

[15:51:00.0000] <jgraham>
zcorpan: OK I have a fix but I think this code needs a little more love (tomorrow :) )

[15:56:00.0000] <zcorpan>
jgraham: ok :)

[16:13:00.0000] <zcorpan>
wonder if looking at the links href="" would yield better results than the image's src="" for <a href=...><img src=...></a>

[16:13:01.0000] <zcorpan>
perhaps href more often contains things like query strings

[16:14:00.0000] <Philip`>
<a href="page18.html"><img src="/images/next.gif"></a> wouldn't do so well

[16:15:00.0000] <zcorpan>
indeed


2007-08-30
[17:45:00.0000] <markp>
woot, i've passed the 1000-test mark for the validator module

[17:45:01.0000] <markp>
god bless hixie for http://www.whatwg.org/specs/web-forms/current-work/#summary

[18:07:00.0000] <Nicklas18>
Why leave the sense of logic at the door?

[18:07:01.0000] <Nicklas18>
Does that mean that you are not logical?

[18:07:02.0000] <Nicklas18>
Shouldn't people in charge of making a standard have common sense?

[18:47:00.0000] <Nicklas18>
Hello?

[18:51:00.0000] <Nicklas18>
Fucking retards.

[18:51:01.0000] <Nicklas18>
Suck my huge cock.

[19:00:00.0000] <karlUshi>
the whatwg channel is becoming entertaining

[19:01:00.0000] <karlUshi>
maybe huge, but a bit too quick 

[20:28:00.0000] <mpt>
Philip`, the objective is not to maximize "the prevalence of alt", it is to maximize the average accessibility of an image on the Web.

[20:28:01.0000] <mpt>
And showing alt= as a tooltip has given many people the wrong idea about how to author it.

[21:12:00.0000] <markp>
Hixie: yt?

[06:24:00.0000] <Lachy>
There's some discussion of Steve's alt research on WSG as well http://www.mail-archive.com/wsg⊙wo/msg30500.html

[06:28:00.0000] <zcorpan>
jgraham: <pre> and <textarea> should have a LF added in the innerHTML view

[06:29:00.0000] <zcorpan>
"<pre>x" should become: "<html><head></head><body><pre>

[06:29:01.0000] <zcorpan>
x</pre></body></html>"

[06:34:00.0000] <zcorpan>
jgraham: also, the "View In Browser" paragraph should be moved to the Parse Tree section, since it will build the tree parsed by html5lib, not parse the serialized innerHTML

[06:35:00.0000] <zcorpan>
jgraham: finally :) it would be nice if Input Document was in a textarea with a submit button so you can modify the input without having to go back

[14:23:00.0000] <jgraham>
zcorpan: Try that


2007-08-31
[03:13:00.0000] <zcorpan>
jgraham: nice :)

[06:21:00.0000] <Lachy>
I just discovered that TimBL's original list of HTML tags from 1991 had elements for highlighting. I wonder if the intention was similar to the <m> element. http://www.w3.org/History/19921103-hypertext/hypertext/WWW/MarkUp/Tags.html#20

[06:54:00.0000] <zcorpan>
Lachy: also see html 3.0

[06:55:00.0000] <Lachy>
zcorpan, at what specifically?

[07:02:00.0000] <zcorpan>
hmm, i thought html 3.0 had elements for hightlight too, but can't find it

[07:05:00.0000] <zcorpan>
i might have mixed it up with html 1

[09:15:00.0000] <zcorpan>
hmm, i think annevk will need to update his smil namespace documentation for smil 3.0

[09:56:00.0000] <zcorpan>
http://lists.w3.org/Archives/Public/www-smil/2007JulSep/0060.html

[12:38:00.0000] <KevinMarks>
is it just me, or did firefox mac get all beachball-spinny very often with a recent update?

[13:45:00.0000] <jruderman>
KevinMarks: which recent update? 2.0.0.6? trunk nightly?

[14:00:00.0000] <KevinMarks>
2.0.0.6

[14:39:00.0000] <Philip`>
What's up with data:text/html,<p><![CDATA[ ]]]>123456789</p> in Opera 9.2? (It says "9")

[14:39:01.0000] <Philip`>
(It acts differently in the Live DOM Viewer)

[14:40:00.0000] <jruderman>
too many "]" characters?

[14:41:00.0000] <Philip`>
That's just Opera's normal crazy CDATA parsing - I'm wondering why it's discarding all but the last character in the text

[14:41:01.0000] <jruderman>
discarding entirely or not displaying?

[14:42:00.0000] <Dashiva>
If you remove all the digits, it displays the >

[14:42:01.0000] <Dashiva>
And so on until you're only left with <![CDATA[ ] displaying the ]

[14:42:02.0000] <Philip`>
The characters are not in the DOM at all

[14:42:03.0000] <Philip`>
http://canvex.lazyilluminati.com/misc/dom-viewer/?%3Cp%3E%3C!%5BCDATA%5B%20%5D%5D%5D%3E123456789%3C%2Fp%3E

[14:43:00.0000] <Philip`>
If I remove the </p> then it shows all of the characters (including the <![CDATA[... as plain text)

[14:44:00.0000] <Dashiva>
So the </p> triggers reparsing, then...

[14:44:01.0000] <Philip`>
It's the < rather than the </p>

[15:38:00.0000] <jgraham>
/me is so confused by Rob Burn's latest message

[16:11:00.0000] <Lachy>
/me wonders what Rob's criteria is for a page to be considered XHTML content?!

[16:12:00.0000] <Lachy>
he seems to think that anything that isn't well formed isn't XHTML, and so can be ignored in any study. So effectively, 100% of pages that he would classify as XHTML would be well formed

[16:19:00.0000] <jgraham>
Indeed. I don't think that's a sensible point of view but I'm not sure it's worth discussing

[16:20:00.0000] <Lachy>
of course it's not worth discussing, it's worth ignoring

[16:22:00.0000] <webben>
How broken can XHTML be and still actually be XHTML?

[16:23:00.0000] <jgraham>
webben: In the context of the discussion, it's about authorial intent.

[16:24:00.0000] <jgraham>
authors who use XHTML doctypes presumably think they are authoring XHTML

[16:24:01.0000] <webben>
jgraham: I doubt that's true of the majority of documents actually.

[16:25:00.0000] <jgraham>
Why are they labelling their documents as XHTML then?

[16:25:01.0000] <othermaciej>
it's hard to tell how to detect XHTML intent

[16:25:02.0000] <othermaciej>
but if it has an XHTML doctype, an XHTML namespace declaration and some elements with minimized syntax, odds are good there was some intent

[16:25:03.0000] <othermaciej>
or if it has a "valid xhtml!" button

[16:26:00.0000] <webben>
jgraham: you're assuming it's a deliberate decision rather than a) what their authoring tool does automatically or b) copy and paste or c) they have no idea that there's a difference between XHTML and HTML

[16:26:01.0000] <webben>
given the adoption of faux-XHTML by common tools like WordPress, I suspect case a) is probably the most widespread

[16:26:02.0000] <jgraham>
c is more-or-less a superset of a and b

[16:27:00.0000] <jgraham>
actually I guess not

[16:27:01.0000] <webben>
they may or may not go together.

[16:28:00.0000] <jgraham>
Even if the author has delegated responsibility to their tool author you have to wonder why tools would label documents as XHTML if they did not intend them to conform to XHTML

[16:29:00.0000] <webben>
jgraham: Sure. But that's a rather different thing to /authorial/ intent.

[16:31:00.0000] <jgraham>
OK, but the essential point is that at some stage people have made a choice to label the documents as XHTML even though they probably do not conform to the XHTML spec and cannot be processed by XML tools

[16:31:01.0000] <webben>
(And I think the answer boils down to a sort of marketing gimmick.)

[16:31:02.0000] <webben>
WordPress devs don't seem to spend a lot of effort on XHTML conformance.

[16:32:00.0000] <webben>
jgraham: Sure. But those people often aren't the authors of the documents.

[16:35:00.0000] <jgraham>
webben: Let me restate my position because I don't think I have made it clear. As far as I am concerned documents that are delivered in such a way that they are processed by the XML pipeline in browsers are XHTML. 

[16:35:01.0000] <jgraham>
Other documents are pseudo-XHTML

[16:36:00.0000] <jgraham>
However these are usually the interesting documents to find out what people who think they are producing XHTML do - 

[16:36:01.0000] <jgraham>
unlike Rob, I think Philip`looked at exactly the right thing

[16:38:00.0000] <webben>
jgraham: But do WordPress devs really think they are producing XHTML? And are the resultant documents, typically highly corrupted by plugins and commenters etc, really representative of what WordPress devs are aiming to do?

[16:39:00.0000] <webben>
And, perhaps more importantly, how much difference would it really have made if they'd chosen an HTML doctype?

[16:39:01.0000] <jgraham>
I don't know what the wordpress devs think they are doing. I do seem to remember that #some people tried to convince them to relabel their documents as HTML and they refused

[16:40:00.0000] <webben>
(yeah, I was one of them ;) ... one of the key defences is that their showcase blogs only failed to be XHTML because of their plugins ... but there aren't that many WP blogs without plugins)

[16:40:01.0000] <webben>
I think the conclusion is that WordPress uses the XHTML doctype as a fashion statement.

[16:41:00.0000] <jgraham>
to me? It makes no difference either way as long as they understand that for every difference that exists between XHTML-as-XML and HTML their "XHTML" will behave like HTML

[16:41:01.0000] <jgraham>
I think saying "XHTML can be served in such a way that it is processed as HTML" is harmful because it forces everyone to know that

[16:42:00.0000] <jgraham>
list of differences

[16:42:01.0000] <webben>
yep, it's highly problematic.

[16:47:00.0000] <webben>
But if pseudo-XHTML is largely adopted as a fashion statement by tool devs, then if you want to know what problems people are trying to solve with XHTML, XML-served XHTML is likely to be a more interesting sample.

[16:48:00.0000] <webben>
The one caveat there being that many of the languages one's supposed to only be able to use with XHTML can actually be served with text/html content (e.g. MathML, Ruby).

[16:51:00.0000] <jgraham>
The thread wasn't about the problems that XML can solve though, it was about whether to endorse XHTML as text/html

[16:52:00.0000] <othermaciej>
I think the spec should just say you MUST NOT send content that isn't conforming HTML5 (classic serialization), even if it is conforming XHTML5

[16:53:00.0000] <othermaciej>
as text/html that is

[16:53:01.0000] <webben>
Sorry, I haven't actually read the thread. I was just reacting to what was being said on here.

[16:53:02.0000] <webben>
What would "endorse" mean in this context.

[16:53:03.0000] <jgraham>
webben: I have great sympathy for not reading the thread :)

[16:54:00.0000] <jgraham>
"XHTML documents served as text/html result in interoperable behavior

[16:54:01.0000] <jgraham>
in typical cases, so that constraint [XHTML must be sent using an XML MIME type] is too strong. Please change

[16:54:02.0000] <jgraham>
it to "SHOULD be sent..." "

[16:54:03.0000] <jgraham>
is roughly what DanC said 

[16:55:00.0000] <webben>
othermaciej: What would that actually mean, as a conformance criterion though? That tools (e.g. Apache? WordPress) cannot serve HTML 4.01 /or/ HTML5 and be conforming?

[16:56:00.0000] <webben>
Or is it purely tautologous? That when serving HTML5 as text/html, tools must serve the text/html serialisation of HTML5?

[16:57:00.0000] <othermaciej>
webben: it would mean that a document is not conforming to the HTML5 specification if it is sent as text/html and is not conforming to the HTML serlialization of HTML5, even if it happens to be conforming to the XML serialization

[16:57:01.0000] <othermaciej>
webben: this would be a change from the previous registration of the text/html MIME type

[16:58:00.0000] <webben>
How? The existing registration doesn't mention HTML5 presumably?

[16:58:01.0000] <webben>
(The actual conformance criterion sounds okay, I just don't see how it's a change rather than an addition.)

[16:59:00.0000] <othermaciej>
the existing registration says that you may send HTML of any version up to 4.01 or XHTML 1.0 as text/html

[16:59:01.0000] <webben>
but that's consistent, since HTML5 is later than either.