00:08
<AryehGregor>
I like the phrase "ipso facto" a lot, but I've discovered it doesn't really mean anything, so I never get a chance to use it. This makes me sad.
00:10
<TabAtkins>
Assign a meaning to it and use it consistently.
00:10
<zewt>
give ipso facto a de facto meaning :|
00:11
<TabAtkins>
But it does mean something. You can usually substitute "thus" for it, and so you can probably replace some of your "thus"es with "ipso facto".
04:04
<Hixie>
ok whatwg.org and hixie.ch are now IPv6-enabled
06:37
<annevk>
sure felt like six months :p
06:37
<Hixie>
can't argue with that :-)
07:04
<annevk>
AryehGregor, are your specs in some bitbucket or other repository somewhere?
07:08
<Hixie>
is order of enumeration of JS objects defined at all?
07:09
<Hixie>
it seems to be order of creation but deleting then resetting a property doesn't seem to move it to the end
07:10
<annevk>
it's defined to be unordered but implementations have to do at least creation order to not break web pages
07:11
<Hixie>
is there a way to move a property to the end of the order?
07:11
<Hixie>
short of creating a whole new object...
07:11
<annevk>
http://wiki.whatwg.org/wiki/Web_ECMAScript#Property_Enumeration
07:11
<annevk>
no idea :/
07:12
<Hixie>
oh well
07:12
<Hixie>
thanks anyway
07:12
Hixie
updates that wiki page with what he's found about "delete"
07:28
<annevk>
http://tools.ietf.org/html/draft-hodges-websec-framework-reqs
07:32
<annevk>
the introduction especially is an interesting read
07:33
<Hixie>
i have a page which consists primarily of a two-cell css table fixed positioned to the viewport, where the right cell is shrink-to-fit and the left cell fills the rest of the available space
07:33
<annevk>
security seems to be getting impossibly hard
07:33
<Hixie>
i'd like the left cell to be overflow:auto in case its contents are too big to fit the available space in the viewport
07:33
<Hixie>
is there a way to do that?
07:33
<Hixie>
everything i've thought of has the table cell grow to fit the contents
07:34
<Hixie>
which defeats the point (it means the whole viewport scrolls and i lose the right cell)
07:34
<annevk>
table-layout:fixed was sort of designed for it I think but never really implemented well?
07:34
<annevk>
I'm not really sure
07:34
<Hixie>
table-layout:fixed requires that i know the right cell's width ahead of time, i think
07:34
<Hixie>
i hope TabAtkins' drafts solve this
07:35
<annevk>
there's a new value for width that means shrink-to-fit
07:35
<annevk>
not sure if that works
07:35
<annevk>
well it does in Mozilla, but in combination with table-layout:fixed...
07:36
<hsivonen>
I'm losing track of goings on at the IETF. What's the story behind there being an informational Internet-Draft and a standards track ID about Do Not Track by different authors?
07:36
<annevk>
it's a bit sad that to date nobody has figured out table layouts
07:36
<annevk>
and written it down
07:36
<annevk>
hsivonen, didn't follow that one
07:38
<Hixie>
annevk: the problem is my shrink-to-fit cell comes on the same row as my as-big-as-possible cell
07:38
<Hixie>
annevk: so i don't think table-layout:fixed can solve it
07:38
<Hixie>
even if implemented per spec
07:38
<annevk>
karlcow, would be nice if there was a dedicated category for http://www.w3.org/QA/2011/03/open-web-weekly-summary.html
07:45
<hsivonen>
karlcow: nice to see that they let you link to lists.whatwg.org from the W3C blog
07:45
<annevk>
I think that may be a first
07:47
<annevk>
karlcow, also, it needs a more compelling title
07:47
<annevk>
karlcow, extremely long name followed by two digitized dates just confirms the W3C is boring
07:48
<annevk>
so we still haven't announced developers.whatwg.org on the blog
07:48
<annevk>
I tried getting hold of ben to do it, but that seems difficult :)
07:49
<annevk>
anyone minds if I go ahead and write a short note?
07:51
<annevk>
(no response in an hour means I'll go ahead)
07:54
<karlcow>
for title, I adopted the same model than http://my.opera.com/karlcow/blog/index.dml/tag/web%20standards%20links :) It was my decision. I accept my boringness
07:55
<karlcow>
for category, yes good suggestion. I will put that on the todo list.
08:12
<annevk>
I need a todo list
08:16
<karlcow>
nah you do not need, you are younger than me.
08:17
<annevk>
Maybe, but I'm so unprepared for the next three months... And behind on silly things like administration.
08:18
<annevk>
Immediate future (i.e. until this weekend) is pretty much sorted though :)
08:21
<karlcow>
I'm looking forward the reports of your trips when you are back. Pretty exciting.
09:09
<abarth>
annevk: i've failed at every todo list system i've tried
09:16
<annevk>
:)
09:16
<annevk>
me too
09:16
<annevk>
my current approach is memory, sometimes notes in TextWrangler since I'm using that a lot, and notes in my calendar
09:17
<annevk>
and putting objects like letters in obvious places :)
09:41
<karlcow>
http://labs.adobe.com/technologies/wallaby/
09:41
<karlcow>
"Convert Adobe Flash FLA files into HTML and reach more devices"
10:28
<annevk>
i have a very short blog post now
10:28
<annevk>
anyone wants to review before it goes live?
10:31
<annevk>
http://blog.whatwg.org/html5-for-web-developers
10:33
<aho>
looks fine
10:49
<annevk>
per twitter it seems quite a few people hadn't seen it yet
11:29
<annevk>
oh wow, our type=color implementation makes use of <datalist>?
11:29
<annevk>
I did not know
11:55
<annevk>
hmm, updating event-source tests has to wait a little
11:56
<annevk>
EventSource, even
11:57
<karlcow>
http://www.bbc.co.uk/news/technology-12668552
11:57
<karlcow>
From 25 May, European laws dictate that "explicit consent" must be gathered from web users who are being tracked via text files called "cookies".
12:43
<zewt>
always nice when politicians stick their heads into stuff without even trying to understand it
13:04
<aho>
"Currently, most browsers support low-complexity AAC and MP3 audio [...]" <- yea right, apple :v
13:05
<Lachy>
aho, where's that quote from?
13:05
<aho>
http://developer.apple.com/library/safari/documentation/AudioVideo/Conceptual/Using_HTML5_Audio_Video/AudioandVideoTagBasics/AudioandVideoTagBasics.html
13:06
<zcorpan>
does chrome support AAC?
13:07
<Lachy>
if they're counting Safari, current versions of Chrome and IE9, then perhaps 3 out of 5 is technically "most"
13:07
<zcorpan>
or was it thrown out together with h.264?
13:07
<jgraham>
Lachy: Weird calculus you have there
13:07
<Lachy>
it's supported in MP4 along with h.264, and support should disappear when they remove it
13:08
<Lachy>
jgraham, I'm not doing calculus.
13:08
<zcorpan>
didn't they remove it already?
13:08
<aho>
Safari on the desktop (Mac OS X and Windows) supports all media supported by the installed version of QuickTime <- this includes he-aac, right?
13:08
<jgraham>
Lachy: You have a system for calculating
13:08
<Lachy>
I thought they were planning to do it in 5 months
13:08
<zcorpan>
oh
13:08
<aho>
afaict i need ogg/vorbis and (he?)-aac for ie9 and safari
13:10
<aho>
and i gotta use the mp4 container for that, right?
13:11
<aho>
3gp should theoretically work, too
13:11
<aho>
*scratches head*
13:11
zcorpan
thought 3gp was the same thing as mp4
13:11
<jgraham>
Lachy: (fwiw see meaning 2 under http://oxforddictionaries.com/view/entry/m_en_gb0116730#m_en_gb0116730 )
13:11
<aho>
3GP (3GPP file format) is a multimedia container format defined by the Third Generation Partnership Project (3GPP) for 3G UMTS multimedia services.
13:12
<aho>
MPEG-4 Part 14 or MP4 file format, formally ISO/IEC 14496-14:2003, is a multimedia container format standard specified as a part of MPEG-4.
13:12
<aho>
oh and there is mov, too
13:12
<Lachy>
jgraham, ok. I just assumed you meant meaning 1, since that's the only usage I've heard before
13:12
<aho>
iirc flash supported these 3, but i don't have a clue about safari or ie9
13:13
<jgraham>
Lachy: Think also "lambda calculus"
13:13
<Lachy>
jgraham, I have no idea what "lambda calculus" means
13:13
<aho>
guess mp4 is the safest bet
13:13
<jgraham>
http://en.wikipedia.org/wiki/Lambda_calculus
13:15
<Lachy>
zcorpan, "These changes will occur in the next couple months but we are announcing them now to give content publishers and developers using HTML <video> an opportunity to make any necessary changes to their sites."
13:15
<Lachy>
http://blog.chromium.org/2011/01/html-video-codec-support-in-chrome.html
13:15
<Lachy>
so they don't say exactly when they will remove it
13:16
<zcorpan>
k
13:19
<aho>
http://html5doctor.com/native-audio-in-the-browser/ <- omits aac... meh :l
13:35
<hsivonen>
I guess different people have different idea of what counts as "removal". removal in nightly, dev, beta or stable
13:39
<zcorpan>
hsivonen: is it removed in some of them?
13:40
<hsivonen>
zcorpan: I think I read somewhere it's now removed in nightly and dev. I didn't verify.
13:42
<zcorpan>
canPlayType in dev chrome gives me '' for canPlayType('video/mpeg') but 'maybe' for audio/mpeg
13:42
<hsivonen>
what about video/mp4?
13:43
<zcorpan>
oh right, audio/mpeg is mp3
13:43
<zcorpan>
'maybe' for audio/mp4 and video/mp4
13:44
<hsivonen>
does it actually play an H.264 test file?
14:12
<zcorpan>
hsivonen: yes
14:13
<hsivonen>
zcorpan: ok. I guess my source or my recollection was wrong then.
14:27
<hsivonen>
Hmm. no mention of the DoJ/MPEG-LA story on Daring Fireball
14:31
<karlcow>
annevk: category done. http://www.w3.org/QA/archive/open_web/
14:35
<MikeSmith>
hsivonen: I notice that investigation seems to be limited to just the anti-competitive positioning against VP8
14:37
<MikeSmith>
and not the broader issue of whether the H.264 patent pool itself was accumulated in good faith according to this initial assertions they made about its purpose and scope
14:37
<MikeSmith>
remember there was an analysis that somebody did about that last year
14:38
<MikeSmith>
something like, their initial claim is that it would be limited to just a couple dozen patents -- less that 50 or whatever
14:38
<MikeSmith>
but despite that it ballooned into what it is now
14:38
<MikeSmith>
500+ patents or whatever
14:41
<MikeSmith>
hmm, more like 800, I guess
14:41
<MikeSmith>
the Nero lawsuit
14:42
<MikeSmith>
filed last May
14:42
<MikeSmith>
and don't remember hearing much about it since
14:42
<zewt>
i think the world needs more patent abuse at levels beyond absurdity
14:42
<zewt>
nothing less will get anything changed, heh
15:33
<erlehmann>
zewt, more abuse does not necessary yield revolution.
15:34
<zewt>
not sufficient, but possibly necessary
15:36
<erlehmann>
?
15:38
<zewt>
more abuse is not sufficient to trigger reform, but it does seem necessary for it to ever happen. not that I hold out much hope in any case
15:38
<erlehmann>
do we need more hashbangs to abolish them entirely?
15:39
<zewt>
i don't think there's much or any parallel :)
15:40
<zewt>
(and they'll never be abolished until the history API is universally available--so the ball's still in UAs' court on that one)
15:43
<zcorpan>
video implementors, see http://www.w3.org/Bugs/Public/show_bug.cgi?id=12267
15:48
<jgraham>
zcorpan: You don't derive humor from race conditions?
15:51
<Philip`>
What if it's a three-legged race condition?
15:54
<jgraham>
Today a pancake race condition would be more apropos
15:55
<jgraham>
(for i18n purposes I am required to link to http://en.wikipedia.org/wiki/Shrove_Tuesday#Festivities at this point)
15:56
<zcorpan>
i thought today was fettisdagen
15:56
<zcorpan>
so semla race condition
15:57
<jgraham>
Is that where someone offers you semla and you run away screaming?
15:58
<zcorpan>
it's where you line up 20 semlor and the first one to eat them all wins
15:58
<zcorpan>
(i.e. 20 semlor per contentant)
15:58
<jgraham>
I fear you might not be joking
15:58
<zcorpan>
i just made it up
15:59
<zcorpan>
but i'd love to watch such a race
15:59
<jgraham>
Still it sounds plausible
15:59
<jgraham>
I bet you could convince people it is a tradition
16:38
<virtuelv>
cute little replaceState() hack: http://grack.com/blog/2011/03/07/abusing-the-html5-history-api-for-fun-and-chaos/
16:46
jgraham
notes that abarth has committed to the html5lib repo 5 times with 5 different email addresses
17:09
<annevk>
lesson learned: even when WHATWG news seems spread all over, still post it to the blog
17:51
<annevk>
character encoding registry still malfunctioning
17:52
<annevk>
maybe once I'm back I'm interested in updating my research again and actually write a spec
18:22
<AryehGregor>
annevk, my specs are at aryeh.name/gitweb.cgi if I have nowhere else specific to put them.
18:52
<annevk>
yay more URLs
19:10
<Ms2ger>
We just need someone to put specs in bzr...
19:11
<AryehGregor>
Is there an established term for "a Node or its ancestor"? The original DOM Range spec uses "ancestor container".
19:11
<AryehGregor>
I guess I'll go with that for now.
19:18
<annevk>
ancestor or self
19:18
<annevk>
I think
19:18
<annevk>
or maybe that's just XPath
19:18
<AryehGregor>
"If foo is an ancestor or self of bar"?
19:18
<AryehGregor>
Doesn't read well.
19:20
<zewt>
"if foo is bar or an ancestor of bar"
19:20
<AryehGregor>
I was using that, but it gets awkward if you need it a lot.
19:20
<AryehGregor>
Old definition of "partially contained": A Node is partially contained in a range if either it is the same as or an ancestor of the range's start node but is neither equal to nor an ancestor of its end node, or is the same as or an ancestor of the range's end node but is neither equal to nor an ancestor of its start node.
19:20
<AryehGregor>
New definition: A Node is partially contained in a range if it is an ancestor container of the range's start but not its end, or vice versa.
19:21
<zewt>
strange term; "ancestor container" doesn't suggest "same node" at all to me
19:22
<AryehGregor>
I know, it's a bad name.
19:22
<AryehGregor>
I didn't make it up.
19:23
<Philip`>
AryehGregor: I have a book that says a node is its own ancestor
19:23
<AryehGregor>
It might help to know that "container" really means "node", in the original context.
19:23
<Philip`>
but not its own proper ancestor
19:23
<AryehGregor>
Philip`, yes, that would be the mathematical way to do it.
19:23
<AryehGregor>
It was my first thought, really.
19:23
<AryehGregor>
But it's confusing to non-mathematicians.
19:23
<AryehGregor>
"I'm my own grandpa!"
19:23
<zewt>
Philip`: i don't think using that definition of "ancestor" is a good idea--programmers don't think of the term that way
19:23
<Philip`>
To avoid ambiguity, call it an improper ancestor :-)
19:24
<zewt>
that's so much clearer :P
19:25
<AryehGregor>
An improper ancestor would be an ancestor that's not not proper, i.e., the node itself.
19:25
<AryehGregor>
It would be a possibly-improper ancestor.
20:29
<twisted_>
how come there's a huge difference in rendering (nowadays?) between gecko and webkit?
20:29
<twisted_>
I suddenly have a lot of inconsitencies...
20:30
<Ms2ger>
There usually isn't
20:30
<twisted_>
Ms2ger: I know but i got a certain design atm
20:30
<twisted_>
where it looks perfect in Webkit and totally fubar'd in firefox
20:30
<twisted_>
and really... confused
20:35
<AryehGregor>
twisted_, link?
20:35
<roc>
create a minimal testcase and post it here :-)
20:35
<twisted_>
sure wait a sec lemme upload
20:39
<aho>
note that firefox 3.x doesnt use the html5 parser. some valid constructs are completely b0rked. e.g. <a ...><div>...</div></a> will *randomly* break. it's a timing issue which depends on reflows. as such, things like gzip, length, latency, and the phase of the moon affect it :I
20:39
<twisted_>
AryehGregor: http://bastardoperatorfromhell.org/~jakoury/contact.html
20:40
<kbrosnan>
roc: ^
20:40
<twisted_>
oh and I get one interesting error which I understand but... thought I was allowed
20:41
<twisted_>
not allowed to use a <header> inside a <header> but I thought it's allowed if you start a new <section> or <article>
20:41
<aho>
twisted_, section and so forth need a display:block
20:41
<aho>
try eric's new reset
20:41
<aho>
it's a good starting point
20:41
<aho>
http://meyerweb.com/eric/tools/css/reset/
20:42
<aho>
(unknown elements default to display:inline)
20:42
<twisted_>
cool talking a look
20:43
<twisted_>
aho: small change but getting closer
20:43
<zcorpan>
twisted_: what's the use case for header in section or article in header?
20:44
<twisted_>
well I'm afraid indeed that I gotta rewrite that part... hehe... I started on index.html and worked from there but THERE the header made sense
20:44
<twisted_>
here it should just be a normal div I think
20:44
<Ms2ger>
zcorpan, s/header in section/section in header/?
20:44
<aho>
twisted_, add overflow:hidden to #top
20:45
<zcorpan>
Ms2ger: no
20:45
<zcorpan>
header in (section or article) in header
20:46
<Ms2ger>
That's not the canonical operator precedence :)
20:47
<zcorpan>
twisted_: i'm asking because maybe the spec should be changed if there is a use case. what did you have when it made sense?
20:48
<twisted_>
zcorpan: well, on the index page I had a <header> because it was the top part of the page (the head) and underneat it had the rest. but INSIDE that header I needed later to add some other thingies and that also needed a header inside...
20:49
<twisted_>
thing I'm missing the most for an <article> is something in between <header> and <footer> (like a <body>)
20:49
<twisted_>
cause explaining that <article> had a column-count of 2 makes everything go all whooey, but now I wrap the <p>'s inbetween <header> and <footer> in a <div>
20:50
<roc>
twisted_: looks like we don't support position:relative with 'right' and 'bottom'
20:51
<AryehGregor>
o_O
20:51
<AryehGregor>
Isn't that kind of basic?
20:51
<twisted_>
roc: it _suddenly_ works
20:51
<twisted_>
I refereshed and poof
20:51
<twisted_>
it looks the same
20:51
<roc>
AryehGregor: I don't think anyone ever uses anything but 'left' and 'top' for relative positioning
20:51
<roc>
until now!
20:51
<roc>
I didn't even know you *could* use right and bottom
20:52
<twisted_>
roc: well cause this crazy designer thought of aligning it to the right
20:52
<twisted_>
and then it's easier to measure it from there
20:52
<twisted_>
roc: used it before btw
20:52
<AryehGregor>
I guess for relative positioning, right and bottom are equivalent to the negative of left and top?
20:52
<roc>
yes
20:52
<twisted_>
checkout the prijzen.html
20:52
<AryehGregor>
For absolute positioning, obviously there's a really big difference, so you'd use all of them.
20:52
<roc>
sure
20:52
<roc>
we would have noticed by now if right and bottom didn't work for abs-pos :-)
20:52
<zewt>
heh
20:53
<zewt>
nothing's better than the things that are broken and seem like they can't possibly actually be broken
20:53
<twisted_>
\o/
20:53
<twisted_>
ok now off to fiddle with getting custom radiobuttons
20:54
<twisted_>
this designer wants to have a 3px solid #a40061 border around a circle
20:54
<twisted_>
instead of the normal thingies
20:54
<twisted_>
grrr
20:54
<roc>
wow, I think we don't even have a bug on file about this
20:55
<twisted_>
roc: refresh this: http://bastardoperatorfromhell.org/~jakoury/contact.html position problem has been solved
20:55
<twisted_>
but only thing I changed... was the overflow: none
20:55
<roc>
oh wait
20:55
<twisted_>
so... dunno
20:55
<roc>
we do have code for this
20:55
<roc>
so something else is broken
20:55
<roc>
I dunno
20:55
<roc>
need minimal testcase, etc
20:55
<roc>
at least file a bug
20:56
<roc>
please
20:56
<roc>
thanks
21:02
<twisted_>
roc: uhm... I'd love to but... not sure what just fixed it actually
21:02
<twisted_>
*magic*
21:05
<roc>
oh, that was my psychic powers
21:06
<roc>
many developers find that their presence makes users unable to reproduce bugs
21:22
<twisted_>
roc: it's the common law of computing
21:22
<twisted_>
users call for their sysadmin (me) and suddenly the problem they had is gone
21:23
<twisted_>
big chance that when I check tomorrow the bug is back ;) haha
21:33
<twisted_>
question maybe someone here can clarify: I understand that during development of new features there's a -webkit or -moz prefix to a css property but for example the border-radius and -moz-border-radius has (afaik) the same syntax. Why not then have it as just border-radius?
21:36
<bfrohs>
There are still bugs with -moz-border-radius
21:37
<bfrohs>
This way, you know -moz-border-radius will have the bugs, while border-radius in a future version of firefox will not
21:37
<bfrohs>
(at least not major bugs that they know about)
21:37
<twisted_>
;)
21:37
<twisted_>
hmm makes sense
21:43
<Ms2ger>
Also, they don't have the same syntax for the longhand, iirc
21:45
<twisted_>
this site is gonna be an all nighter I'm afraid
21:45
<twisted_>
pff
21:57
<jamesr>
-moz-border-radius is not the best example
22:01
<zcorpan>
twisted_: you can't know that border-radius is going to have the same syntax until the spec is finished