07:05
<MikeSmith>
http://code.gregjopa.com/html5/audio/tabplayer/ is pretty cool
07:05
<MikeSmith>
http://hacks.mozilla.org/2011/01/html5guitar/
07:07
<paul_irish>
that's hot
07:09
<MikeSmith>
oh wow
07:09
<MikeSmith>
little keyboard thing in Google Translate
07:10
<MikeSmith>
where did that come from
07:10
<MikeSmith>
don't remember it being there til today
07:10
<MikeSmith>
maybe I just didn't notice
07:10
MikeSmith
tries the Korean one
07:12
<MikeSmith>
no Japanese one yet
07:12
<MikeSmith>
but then there's always http://inputking.com/
07:13
<MikeSmith>
which works just the same way a native platform/OS Japanese IME works
07:19
MikeSmith
finds http://code.google.com/apis/language/virtualkeyboard/overview.html
07:19
<MikeSmith>
http://code.google.com/apis/language/virtualkeyboard/v1/getting_started.html
07:20
<MikeSmith>
"With the Virtual Keyboard API, you can enable an onscreen keyboard on any text field or text area on your webpage"
09:47
<benschwarz>
Hixie: !ping
09:50
<Hixie>
vaguely here
09:50
<Hixie>
about to head to bed
10:31
<MikeSmith>
hsivonen: did you have a chance to look at that script-checker patch?
10:52
<benschwarz>
Damnit, missed Hixie's messages
10:52
<benschwarz>
oh well.
10:53
<benschwarz>
just wanted you to check the progress and intro text :)
10:58
<annevk>
@WHATWG still works, yay
10:58
<annevk>
I was wondering whether excluding editorial edits was actually necessary...
10:59
<MikeSmith>
I think some of supposedly editorial edits are often things other people might actually like to know about
11:00
<annevk>
Yeah...
11:00
<annevk>
Anyone else?
11:01
<MikeSmith>
for example, http://html5.org/r/5751
11:01
<MikeSmith>
"Add best-practices sections for media elements"
11:02
<MikeSmith>
maybe Hixie needs to add another flag to distinguish potentially interesting editorial changes from mundane ones
11:02
<Ms2ger>
Wasn't that @html5?
11:02
<Ms2ger>
http://twitter.com/html5/status/23525623414652928
11:03
<annevk>
@html5 has a lot of stuff
11:03
<annevk>
also it does not seem to filter Hixie's checkin messages
11:03
<MikeSmith>
yeah, and I'm considering turning of the notifications about spec changes getting retweeted there
11:04
<MikeSmith>
now that @whatwg is working again
11:04
<MikeSmith>
*turning off
11:05
<benschwarz>
MikeSmith: is the link that I used on the intro okay?
11:05
MikeSmith
looks
11:06
Ms2ger
likes
11:07
<MikeSmith>
benschwarz: I usually use http://people.w3.org/mike/ these days
11:07
benschwarz
is talking about http://developers.whatwg.org/
11:07
<MikeSmith>
but that's fine too
11:07
<benschwarz>
MikeSmith: I'll change it over?
11:07
<MikeSmith>
yeah, or I guess I should blog more
11:08
<benschwarz>
deploying…
11:08
<benschwarz>
done
11:08
<MikeSmith>
man the last time I wrote an actual blog entry at sideshowbarker.net was more than a year ago
11:09
<MikeSmith>
maybe I can hire somebody to write blog entries for me
11:09
<benschwarz>
make sure they're cheap
11:10
<benschwarz>
then I can hire them too
11:10
<benschwarz>
haha
11:10
<benschwarz>
my last was in may (the w3c one)
11:10
<benschwarz>
I'd drafted one when I was in japan, but never released
11:10
<benschwarz>
boo
11:11
<Ms2ger>
Might help skimming if the headings were pulled into the left margin, like the notes
11:11
<benschwarz>
Ms2ger: major headings are pulled left
11:11
<Rik`>
benschwarz: links in TOC are missing a "call to action sign" (no underline, no hover difference)
11:11
<benschwarz>
but I do need to properly design something. I've been just pushing the styles around that we had previously and working out how to do useful things to the splitters
11:12
<Ms2ger>
There's usually only one or two of those, though
11:13
<benschwarz>
Ms2ger: thats true, I debated about having the numbered sections, but I hid them for now
11:13
<benschwarz>
I might pull them back out
11:14
<Ms2ger>
And maybe style ul.brief differently
11:15
<benschwarz>
Ms2ger: where is it seen?
11:16
<Ms2ger>
http://developers.whatwg.org/elements.html#global-attributes
11:18
<jgraham>
Ms2ger: Does Web DOM Core have tests for not case folding in XHTML for e.g. createElement, getAttribute, tagName, etc.
11:19
<gsnedders>
jgraham: Not in the tests I wrote, at least
11:19
<benschwarz>
Rik`: I added a hover, for now
11:19
<Rik`>
benschwarz: thanks
11:21
<benschwarz>
I'm out for tonight. Queries? Log an issue :)
11:21
<benschwarz>
laters!
11:22
jgraham
is unhappy to see that many Web DOM Core tests abuse testharness.js, adding weight to the arguments of those who have complained about the design :(
11:26
<gsnedders>
jgraham: Abuse it in what way?
11:29
<jgraham>
gsnedders: Lots of independent tests in a single test()
11:30
<Ms2ger>
Blame Anne ;)
11:30
<Ms2ger>
And doesn't look like it
11:31
<jgraham>
What doesn't look like it?
11:31
<jgraham>
https://bitbucket.org/ms2ger/web-dom-core/src/c7135c7ce9f9/tests/Element-setAttribute.html
11:31
<jgraham>
Oh, you mean the original question
11:32
<Ms2ger>
Yeah
11:36
<bga_>
lol compare http://www.w3.org/TR/DOM-Level-3-Events/#events-KeyboardEvent-initKeyboardEvent and https://developer.mozilla.org/en/DOM/event.initKeyEvent
11:37
bga_
confused
11:42
<bga_>
hm http://docs.blackberry.com/en/developers/deliverables/11849/KeyboardEvent_569351_11.jsp#KeyboardEvent_initKeyboardEvent_569356_11
11:48
<bga_>
also there is tupo in spec
11:49
<bga_>
in header initKeyboardEvent with `repeat` arg
11:49
<bga_>
but in details nothing `repeat` arg
12:09
<smaug____>
bga_: what is strange in that comparison? Gecko implements an old version DOM0 key events, DOM 3 Event (which isn't yet a recommendation) defines key events in a different way.
12:14
<bga_>
smaug____ i see. branching code hell.
12:14
<bga_>
w3c + gecko + ie
12:15
<smaug____>
+webkit + opera
12:15
<bga_>
does webkit has own implementation of KeyboardEvent interface ?
12:16
<smaug____>
I don't know what kind of interface it has
12:16
<bga_>
i see w3c's version in ch10
12:16
<smaug____>
Safari 5 might have something else
12:16
<bga_>
:(
12:16
<smaug____>
and DOM 3 Events may still change
12:17
<annevk>
DOM Core tests are not original testharness.js tests
12:17
<annevk>
they are ported to that format
12:17
<annevk>
fwiw
12:17
<Philip`>
bga_: Why would you expect technology as advanced and complex and keyboard input to be interoperably supported already?
12:18
<Philip`>
We've barely had a dozen years to understand it
12:18
<annevk>
I'm gonna update WebAppsTweet to also tweet for editorial changes now
12:19
<MikeSmith>
cool
12:26
<bga_>
Philip` i`m just trying switch keyboard leds using dispatchEvent :)
12:27
<annevk>
http://wiki.whatwg.org/wiki/WebAppsTweet documents source code for WebAppsTweet
12:36
<Philip`>
bga_: Uh, that doesn't really sound like it'll ever work :-p
12:36
<Philip`>
since keyboard events come much later in the chain of processing than LED changes
12:37
<MikeSmith>
http://vl.kamibu.com/2011/01/why-webgl/
12:42
<annevk>
heh, MikeSmith, so friendfeed still has a non-OAuth API?
12:42
annevk
was looking at how @html5 solved the problem
12:49
<MikeSmith>
annevk: friendfeed uses OAuth for Twitter
13:05
<matjas>
How about this piece of markup? Price: €99 <small>(VAT included)</small>
13:06
<matjas>
Thoughts?
13:07
<annevk>
MikeSmith, oh okay
13:07
<matjas>
okay, annevk didn’t call me out so I’m gonna assume this is a valid use case for <small>
13:07
<annevk>
heh, looks fine to me :)
13:08
<annevk>
matjas, could file a bug to get it added to the spec if you want
13:08
<matjas>
will do
13:21
<bga_>
heh only ff returns currect Function#length for host functions
13:23
<schalkn>
Hi all
13:24
<schalkn>
A belated but Happy New Year to everyone
14:10
<beowulf>
is Toby Inkster about?
14:47
<hsivonen>
MikeSmith: I sent you email about code review
14:48
<MikeSmith>
thanks
14:48
<hsivonen>
MikeSmith: sorry for disappearing for the weekend without saying anything
14:48
<MikeSmith>
no problem man
14:48
<MikeSmith>
there's no rush on this stuff
14:54
<MikeSmith>
david_carlisle: just sent a reply to your e-mail message about the MathML schema
14:55
<MikeSmith>
I don't know why the Openmath stuff is there -- need hsivonen to comment on that
15:07
<david_carlisle>
MikeSmith: thanks (although I saw the email before i saw you telling me I had an email, ii only sporadically watch the IRC still:-)
15:07
<MikeSmith>
OK
15:19
<jgraham>
Hmm, removeAttribute seems to be case-insensitive in gecko/Opera
15:23
<Philip`>
"The best solution for HTML5 (via tooling, education, author guidance) is to not fake text entry via canvas elements – much better native mechanisms for text editing already exist in HTML5." - hmm, seems Microsoft opinions are starting to align more and more with the cabal
15:24
<Workshiva>
You mean the cabal is becoming embraced and extended by Microsoft? :P
15:26
<hsivonen>
so I know next to nothing about scripting Flash Player...
15:26
<hsivonen>
Are there some JS methods that Flash Player always exposes on object elements that have loaded .swf?
15:26
<hsivonen>
regardless of what the .swf is?
15:42
hsivonen
discovers GetVariable
16:00
<hsivonen>
Wow. Arista and Transmageddon are so different from every other video compression app I've ever used.
16:00
<hsivonen>
once you choose WebM, there are no addititional tweakable settings
16:01
<hsivonen>
it just does something with setting the developers have deemed to be appropriate
16:01
<hsivonen>
which is very easy for the user
16:01
<hsivonen>
takes away the trouble of how to tweak the settings
16:02
<Rik`>
hsivonen: Miro convertor does that too ?
16:02
<Philip`>
That's no fun
16:03
<Philip`>
What's the point in encoding videos if it doesn't give you an opportunity to play with exciting strings like -vcodec libx264 -vpre slow -bf 0 -crf 25 -acodec libfaac -ac 2 -ar 48000 -ab 192k which you don't understand and have to randomly tweak until they look like they're probably working?
18:24
<paul_irish>
is anyone familiar with the `target` style property? apparent use: a { target: new front; }
18:24
<TabAtkins>
Defined in the link behavior draft (I think that's the right name...). No one implemented it, it's been dead for years.
18:24
<paul_irish>
thx tab.
18:26
<Ms2ger>
Not even Opera?
18:26
<TabAtkins>
Don't think so.
18:27
<paul_irish>
i learned about it from w3schools http://www.w3schools.com/css3/css3_pr_target.asp ;)
18:28
<TabAtkins>
Ah, found the original spec: http://www.w3.org/TR/css3-hyperlinks/
18:32
<Ms2ger>
Doesn't look like it, indeed
18:35
<Ms2ger>
Also, doesn't seem like anybody whined about Opera implementing -apple-dashboard-region
18:35
<TabAtkins>
...
18:35
<TabAtkins>
Can I whine about it now?
18:35
<Ms2ger>
Certainly
18:36
<Peter`>
I whined about WebKit implementing -apple-dashboard-region :/
19:02
<othermaciej>
why in the world would Opera implement -apple-dashboard-region?
19:02
<othermaciej>
it does nothing in Safari
19:05
<gsnedders>
othermaciej: It only does something in the Widget Runtime
19:05
<othermaciej>
does the Widget Runtime attempt to run Dashboard widgets?
19:05
<othermaciej>
(as opposed to W3C widgets)
19:08
<gsnedders>
AFAIK no.
19:21
<othermaciej>
then I can't imagine a reason to copy the vendor prefix or even the syntax...
19:34
AryehGregor
stops arguing with Bjartur
19:48
<AryehGregor>
Does anyone know of a way to get Live DOM Viewer to work in IE9?
19:48
<AryehGregor>
The XSS protection blocks saved links.
19:49
<AryehGregor>
Actually, it completely disables the functionality.
20:04
<AryehGregor>
Ms2ger, is there some way to create a Range object other than selection.getRangeAt() or range.cloneRange()?
20:05
<Ms2ger>
document.createRange?
20:05
<AryehGregor>
Ah, thanks.
20:16
<Hixie>
btw if anyone sees issues in developers.whatwg.org, ben's asked us to file them in https://github.com/benschwarz/developers.whatwg.org/issues
20:19
<annevk>
http://twitter.com/WHATWG/status/24561930723131393
20:19
<annevk>
and with that I'm off to bed
20:26
<foolip_>
Hixie, could you add a redirect from http://www.whatwg.org/specs/web-apps/current-work/multipage/microdata.html to http://www.whatwg.org/specs/web-apps/current-work/multipage/links.html#microdata ? My links are broken :(
20:28
<beowulf>
Hixie: shouldn't the nav be in the header?
20:44
<AryehGregor>
Ms2ger, if you're familiar with Firefox's Selection implementation, could you comment on this? http://www.w3.org/Bugs/Public/show_bug.cgi?id=10624
20:44
<AryehGregor>
I'm not sure what Firefox is doing here.
20:44
<Ms2ger>
Not at all, but I can read the source
20:44
<AryehGregor>
Ah.
20:48
<AryehGregor>
Do you know who I could ask?
20:49
<Rik`>
anyone know the status of user-select: none ?
20:50
<Rik`>
i see it in a previous draft of css3-ui but nothing in the latest one
20:50
<Ms2ger>
ehsan on Mozilla IRC, looks like
20:52
<Ms2ger>
Oh, and Gecko indeed has a direction member
20:52
<TabAtkins>
Rik`: I'd ask tantek.
20:53
<Hixie>
foolip_: i think Philip` is the one who can help with that
20:53
<foolip_>
Philip`, ^ ?
20:55
<Ms2ger>
And extend() looks like a whole new world of hurt
20:56
<Hixie>
beowulf: ben's the one who is setting it up
20:57
<Philip`>
foolip_: http://www.whatwg.org/specs/web-apps/current-work/multipage/microdata.html does redirect for me
20:57
<Philip`>
after a few seconds
20:58
<Philip`>
(The 404 page has scripts to do that)
20:58
<foolip_>
Philip`, but first you see 404 for a while
20:58
<beowulf>
Hixie: 'k, ta
20:58
<Philip`>
foolip_: Yeah, I think anything else would need server-side support, and that would be Hixie's responsibility ;-)
20:58
<Hixie>
foolip_: it can't easily be done at the server level because the server doesn't have the fragment identifier
20:59
<foolip_>
Hixie, oh, didn't know you had redirects like that too
20:59
<Hixie>
Philip`: if you have a list of permanent redirects for page-level updates, stick them in a file in the tarball and i'll merge it into the .htaccess file (the html files still do the fragid redirects right?)
21:00
<foolip_>
hrm, I don't suppose the script could be run earlier in <head> so that one doesn't get to see the 404 message? I instantly close 404 pages without waiting for the second or whatever this took
21:00
<Hixie>
i can make the 404 page not look like a 404 page
21:00
<Hixie>
file a bug
21:00
<jgraham>
1
21:01
<jgraham>
Erm...
21:01
<Ms2ger>
2
21:01
<Hixie>
2?
21:01
<Philip`>
Hixie: The list is http://www.whatwg.org/specs/web-apps/current-work/multipage/fragment-links.js
21:01
<Ms2ger>
Great minds think alike...
21:01
<Ms2ger>
And sick minds, too?
21:01
<Philip`>
Hixie: and the 404 page for foo.html looks up foo in that list, and goes to whatever.html#foo
21:01
<Philip`>
Hixie: for any foo (they're not known in advance)
21:02
<Hixie>
Philip`: ah
21:02
<Hixie>
k well then that leaves making the 404 page less of a 404-looking page
21:02
<Hixie>
i can do that
21:02
<Hixie>
file a bug if you want it
21:02
<Hixie>
bbiab, lunch
21:07
<foolip_>
very exciting bug filed: http://www.w3.org/Bugs/Public/show_bug.cgi?id=11722
21:08
<foolip_>
before I go on a coding rampage, does anyone have tools to notify you of changes to particular sections of the spec? For example, I'd like to be notified automatically when the <video> section changes, among others
21:17
<jgraham>
foolip_: You plan to implement that? I would really like it :)
21:18
<foolip_>
jgraham, well, I really want it, but I was hoping someone else would have it
21:19
<foolip_>
jgraham, one simple way to do it is to remove the section you want to follow in git and keep moving that patch along until it conflicts, for example
21:20
<foolip_>
s/git/your favorite VCS/
21:21
<cying>
if i want to start manipulating a DOM, do i need to wait for the DOMContentLoaded event?
21:21
<jgraham>
foolip_: That sounds like a clever hack but not quite like an elegant solution :)
21:21
<jgraham>
cying: No
21:21
<gsnedders>
cying: No.
21:21
<cying>
(if the DOM content is inlined)
21:21
<Philip`>
foolip_: That sounds kind of indirect - why not just cut out the section you want using regexps, and diff it each time to see what changed?
21:21
<jgraham>
gsnedders: That extra period cost you
21:22
<cying>
jgraham: gsnedders: so i can just start manipulating it in javascript without waiting for anything?
21:22
<jgraham>
cying: Well it has to be loaded
21:22
<foolip_>
Philip`, something like that was my second idea
21:22
<foolip_>
for some reason my first idea usually involves git
21:22
<Philip`>
foolip_: As a bonus, it doesn't involve git
21:23
<jgraham>
<!-- initial dom --><script><!-- I can manipulate anything in initial dom --></script><!-- final dom -->
21:23
<cying>
jgraham: is that an event? or do you just mean that the DOM comes from inlined markup before the script?
21:23
<cying>
jgraham: ahhh
21:23
<jgraham>
cying: The lattter
21:23
<cying>
ahhh
21:23
<cying>
cool!
21:23
<gsnedders>
cying: You can rely upon everything up to and including the script element being executed being in the DOM, but nothing more
21:24
<cying>
gsnedders: ah! that must be written somewhere i bet
21:24
<jgraham>
Yes, that's true, you can manipulate the script itself if you are evil
21:24
<gsnedders>
jgraham: That won't cause it to run again, though, no?
21:24
<jgraham>
No
21:25
<jgraham>
Unless changing the src has that effect?
21:25
<jgraham>
(for an external script)
21:25
<cying>
fascinating!
21:25
<jgraham>
But I think it doesn't
21:51
<Hixie>
foolip_: i don't think we have any section-specific following stuff, but it would be interesting if someone set that up.
21:51
<Hixie>
foolip_: part of the problem is knowing how to define a section, since i often move stuff around
22:16
<jgraham>
me, I might try using an html parser and the document structure to find my way around. Not that easy to get from a line-based diff to "did this section change" of course, but possible
22:20
<jgraham>
Also, what is the disadvantage of replacing <hgroup> with <h1>Main heading <subhead>Subheading</subhead></h1>
22:21
<jgraham>
?
22:21
<TabAtkins>
That feels weird.
22:22
<jgraham>
TabAtkins: It seems to be equivalent to me, albeit with slighly different legacy fallback semantics
22:23
<jgraham>
Maybe it is bad if we ever get a :heading selector though
22:23
<jgraham>
Since we would want that to not match the subheading (presumably)
22:23
<TabAtkins>
Actually, I think we would want it to match.
22:24
<TabAtkins>
The subheading is still part of the heading.
22:24
<jgraham>
So you would do :heading(n) and :heading(n) > subhead ?
22:25
<jgraham>
To style the two parts
22:25
<TabAtkins>
Yeah, with that structure. Similar with <hgroup>.
22:25
<jgraham>
It's harder with hgroup, I think
22:25
<TabAtkins>
Hm. Benefit of <subhead> in that context is that it's a single element, so you don't need to remember what precise heading level was used.
22:25
<jgraham>
Yeah
22:25
<oojacoboo>
hixie, like the new spec look :)
22:26
<Hixie>
you mean developers.whatwg.org?
22:26
<Hixie>
that's all ben's work
22:26
<oojacoboo>
ah
22:26
<oojacoboo>
http://simon.html5.org/html5-elements
22:26
<oojacoboo>
I use that, so, I guess he switched it over to the developers.whatwg.org domain
22:26
<Hixie>
yeah
22:26
<jgraham>
And it is more obvious what :heading should match (no question if it is the <hgroup> or the actual heading)
22:26
<TabAtkins>
jgraham: Indeed. Hm, okay, I'm liking your idea better now.
22:27
<Hixie>
jgraham: i still occasionally get 500s when genning the spec with the w3c annotations
22:28
<Hixie>
jgraham: fwiw
22:28
<Hixie>
jgraham: dunno if you care (i just restart my script and that seems to fix it usually)
22:28
<TabAtkins>
<3 @WHATWG being back up.
22:28
<jamesr_>
TabAtkins: i'm working on something like https://developer.mozilla.org/en/DOM/window.mozRequestAnimationFrame for webkit and will want to move towards standardizing it. where would something like this be defined?
22:29
<Hixie>
TabAtkins: all annevk's work :-)
22:29
<TabAtkins>
jamesr_: You mean which standards body?
22:29
<Hixie>
jamesr_: sounds like a cssom thing
22:29
<Hixie>
jamesr_: annevk's baby
22:30
<jgraham>
Hixie: I expect it times out sometimes
22:30
<jamesr_>
TabAtkins: more wondering what it would fall under
22:30
<jgraham>
But I could check the logs I guess
22:30
<jamesr_>
there's nothing css-specific about it really
22:30
<Hixie>
jamesr_: CSSOM isn't CSS-specific
22:30
<TabAtkins>
jamesr_: That's a hard one.
22:30
<jamesr_>
Hixie: o.O
22:31
<Hixie>
jamesr_: it's just where media-dependent DOM features go
22:32
<jamesr_>
why's it called CSSOM then?
22:33
<Hixie>
historical reasons
22:33
<Hixie>
same reason Window is specced in a spec that claims to be HTML
22:44
<roc>
jamesr_: heycam already wrote something BTW
22:46
jgraham
wonders if we can have pretty styles for the implementor version of the spec
22:47
<jgraham>
Although not with the serif typefaces
22:48
<TabAtkins>
Boo at serifs outside of headings.
22:51
<oojacoboo>
what's the thought behind not allowing form elements to be styled in any consistent manner
22:51
<TabAtkins>
The thought is that tantek is working on it.
22:51
<TabAtkins>
Also: form elements are displayed differently in different browsers/OSes.
22:51
<paul_irish>
https://wiki.mozilla.org/Tantek-Mozilla-projects#Styling_HTML5_UI_elements
22:51
<TabAtkins>
Frex, check out a <select multiple> in Opera Mobile.
22:51
<oojacoboo>
TabAtkins: that doesn't have to be the case
22:52
<Hixie>
jgraham: got to be honest, personally i think the developers edition styles are unusable and love the whatwg spec styles :-)
22:52
<Hixie>
jgraham: but if you have any suggestions as to what would make the spec more usable, let me know
22:52
<TabAtkins>
oojacoboo: But... it *is* the case.
22:53
<oojacoboo>
TabAtkins: if that's the attitude everyone took, where will we be tomorrow?
22:53
<TabAtkins>
The question is where we are *today*. And today we are in a world where form elements are displayed differently in different browsers and OSes.
22:54
<TabAtkins>
And today I'm willing to let tantek solve that problem.
22:54
<oojacoboo>
so, the answer is, that someone is working on a consistent styling spec for form elements?
22:55
<oojacoboo>
or that no one gives a shit, and the way it currently is, is the way it is and will be
22:55
<TabAtkins>
Yes. Tantek.
22:55
<jgraham>
Hixie: I think the problem is it depends on the use case. Sometime I would really like narrower columns for easier reading. Sometimes I want as much on the screen as possible
22:55
<jgraham>
Alternative style sheets are not a good solution here :)
22:58
heycam
moves requestAnimationFrame higher up his mental todo list
23:02
<Hixie>
jgraham: why would alt ss not be a good solution?
23:02
<Hixie>
jgraham: sounds like exactly the right solution
23:02
<Hixie>
jgraham: i agree that there are different use cases
23:02
<Hixie>
jgraham: i'm happy to add other style sheets or whatnot
23:03
<jgraham>
Hixie: Because I probably wouldn't know which I prefered until after I had done whatever it was
23:03
<Hixie>
jgraham: or to make the current one better in any way you can suggest
23:03
<beowulf>
jgraham: what about <h1><em>Main head</em>sub head</h1>
23:03
<jgraham>
and switching stylesheets probably works badly
23:03
<Hixie>
jgraham: switching stylesheets works fine.
23:03
<Hixie>
jgraham: the other problem seems to apply to any solution.
23:03
<jgraham>
Hixie: It retains the scroll position?
23:04
<jgraham>
even if the document is resized
23:04
<Hixie>
jgraham: wait you want to change the style sheet _while_ reading?
23:04
<jgraham>
and is almost instananeous?
23:04
<Hixie>
jgraham: dude, switching to another document isn't going to maintain your scroll position either and it's gonna be a hecka lot less instantaneoes
23:04
<Hixie>
instantaneous
23:04
<Hixie>
man that's hard to type or dvorak
23:05
<Hixie>
on
23:05
<jgraham>
Well the point is that I wouldn't realise that I was using the wrong stylesheet until I had already found my position in the document and so on
23:05
jgraham
notes that he would like to be asleep but there is supposed to be someone returning 1/2 bags that never arrived in England on 20/12/2010
23:06
<Hixie>
well i don't know how to solve that problem
23:07
<Hixie>
but i'm willing to solve any problem that has solutions :-)
23:23
<jgraham>
beowulf: Why would that be better? It would negate several of the advantages of the design. If you really wanted to reuse an existing element <h1>Main <small>sub</small></h1> would be better
23:23
<jgraham>
But I don't really see the point
23:23
<hober>
I don't really see the problem with <hgroup>
23:24
<TabAtkins>
Plus, those both prevent you from using <em>/<small> for their intended purposes in headings.
23:24
<beowulf>
just asking, looked similar to me
23:24
<TabAtkins>
hober: I don't have a problem with <hgroup>, but I see some advantages to jgraham's idea.
23:26
<jgraham>
hober: I think that <hgroup> does add some complexity since you have to care about the whole tree to determine if a given <hx> is actually a heading or not
23:26
<jgraham>
Well the whole parent heirachy at least
23:26
<TabAtkins>
Hm? Surely you only need to check the immediate parent.
23:30
<jgraham>
TabAtkins: even in the invalid case?
23:30
<TabAtkins>
Damn invalid cases.
23:38
hober
thinks Hixie should use a 241543903 image as an example in the spec
23:40
<Hixie>
hm?
23:41
<hober>
google it and you'll get the idea
23:41
<hober>
I don't know when this started, but it's quite odd and surprisingly funny
23:43
<jamesr_>
roc: oh? where is it?
23:44
<jamesr_>
roc: the major difference in the webkit impl is that we currently don't plan to support animationStartTime
23:44
<jamesr_>
roc: and i'd like to add an element parameter to .requestAnimationFrame() for authors to use if the animation is tied to some element (i.e. updating a <canvas>), so we can throttle offscreen ones
23:52
<Hixie>
hober: o_O
23:54
<hober>
Hixie: that's what I said :)
23:54
<Hixie>
to be honest i try to avoid memes like that because they have a pretty short shelf-life (though i guess this one might have a longer shelf-life than usual since freezers are used to-- never mind)
23:57
<hober>
I wonder if there are any Alcor photos with this meme involved...