00:03
<Hixie>
well we have a quirks spec now so they're no longer really quirks :-)
00:06
<tantek>
I'm sure it'll evolve into the "no doctype" mode over time for ultraminimalists.
01:08
<hober>
is there a w3c bugzilla component for the fullscreen spec?
05:04
<hsivonen>
Wilto: interesting. not at all who I expected. thanks.
05:05
<hsivonen>
looks like my guess that hober was on track to editing HTML5 was right but not exclusively right
05:06
<hsivonen>
Wilto: anyway, to get adaptive images to implementation, I expect working through the WHATWG Community Group to be more productive than working through the HTML WG
05:07
<hsivonen>
Wilto: if you try to get it into HTML5, it will go into the Process bureaucracy and get rejected
05:07
<hsivonen>
Wilto: if you try to get it into HTML.next via the HTML WG, it will get delayed due to the focus on HTML5
05:07
<hsivonen>
Wilto: while the actual ongoing work on HTML happens at the WHATWG
05:09
<hsivonen>
congrats to the new editors
05:55
<MikeSmith>
hober: https://www.w3.org/Bugs/Public/buglist.cgi?product=WebAppsWG&component=Fullscreen&resolution=---
05:58
<Hixie>
how is that in WebApps
05:58
<Hixie>
isn't that a CSS thing?
06:00
<roc_>
arguably, but it's more DOM API than anything else
06:00
<Hixie>
i meant CSSWG thing
06:01
<Hixie>
agreed entirely that it's an API
06:02
<roc_>
I guess
06:02
<Hixie>
i shouldn't try to understand w3c stuff
06:04
<Hixie>
hey so i'm marking all the things in the spec that lead to fingerprinting
06:04
<Hixie>
anyone think of any i might have missed?
06:16
<benvie>
in js you can identify the browser and often js engine revision by identifying the MAX_STACK count though that's not a whatwg thing
06:17
<Hixie>
as in function foo() { count+=1; foo() } var count = 0; try { foo() } catch (e) { alert(count) } ?
06:17
<benvie>
yeah
06:17
<benvie>
in v8 you can pin it to a specific revision
06:18
<benvie>
since for some reason it varies each release, but is static for that release
06:18
<Hixie>
huh
06:18
<Hixie>
heh, i guess firefox implements tail recursion optimisations because it never stopped for me :-)
06:19
<benvie>
yeah it's weird
06:19
<benvie>
it worse in spidermonkey shell, but not firefox
06:19
<benvie>
it works in any webkit and it works in IE
06:19
<AryehGregor>
Hixie, why are Latin and CJK the only scripts singled out in inputmode? Is <input inputtype=latin> really meant to mean "whatever the usual script is for the current language"? If so, why is it called "Latin"?
06:19
<benvie>
so jsc and v8 are similar
06:20
<benvie>
but in spidermonkey shell it always returned 4000 so it was useless except to identify generally that it was firefox
06:20
<benvie>
er spidermonkey
06:23
<benvie>
http://code.google.com/p/chromium/issues/detail?id=117531 is the issue I filed with Chromium since it's so specific as to almost be a security issue
06:53
<Hixie>
ok, i've added notes in the spec for fingerprinting risk
06:54
<Hixie>
AryehGregor: because nobody from any cultures other than the japanese responded to me six zillion requests for information, and i only knew about western (latin) input mode requirements.
06:54
<Hixie>
s/me/my/
06:55
<Hixie>
AryehGregor: so i finally said screw it, i'll give what the japaneses and the latin-script-using-west need, and if anyone else wants anything, they can get it when they provide me with the info i need
07:00
<hsivonen>
Hixie: yay for "latin" instead of "en", etc.
07:00
<hsivonen>
It's super-annoying when software is designed with the assumption that people who write in two languages write in two scripts
07:01
<hsivonen>
maybe once Spanish usage goes high enough in the U.S., more software designers realize that people input multiple languages with one Latin layout
07:02
hsivonen
has to turn off input prediction on Android, because it doesn't support predicting both English and Finnish without being explicitly told which one to predict at a given time
07:04
<hsivonen>
(on the bright side, Android support putting the U.S. English speech regornition button on a Finnish keyboard)
07:04
<hsivonen>
*supports
07:05
<hsivonen>
Hixie: btw, any reason not to reuse 4-letter script codes from the language tag RFC?
07:11
<Hixie>
hsivonen: on android you can make the space bar support switching languages, which is what i do for english vs french
07:12
<annevk>
Hixie: Fullscreen is in both WebApps and CSS :(
07:12
<Hixie>
hsivonen: but yeah, same keyboard, different language, that's a separate axis than inputmode
07:12
<Hixie>
hsivonen: reuse the script codes for what?
07:12
<annevk>
Hixie: And the CSS part is problematic with their sync decision making and mailing list I'm not subscribed to
07:12
<hsivonen>
Hixie: keyboard category
07:13
<Hixie>
annevk: why not just do it like your other specs?
07:13
<hsivonen>
Hixie: I don't want to tell Android whether I'm writing Finnish or English
07:13
<Hixie>
hsivonen: oh because i don't expect the script labels to match well to what keyboards inputmode="" needs
07:13
<annevk>
Hixie: The editor's draft is the WHATWG draft
07:13
<hsivonen>
Hixie: surely Google's mighty algorithms should be able to figure out which language of the two I'm writing after a word or so
07:13
<hsivonen>
Hixie: ok
07:14
<Hixie>
annevk: fun
07:14
<annevk>
so much
07:15
<Hixie>
hsivonen: it's pretty hard to distinguish poor typing from another language, probably. still, long-tap space bar, tap language, it's not exactly a burden to switch back and forth
07:16
<Hixie>
annevk: just FSA it and then you don't have to worry every again about the webapps/css versions :-)
07:23
<Hixie>
nn
07:24
<annevk>
Hixie: did you mark the Navigator object's friends?
07:25
<annevk>
looks like
07:25
<annevk>
that is one ugly icon btw
07:26
<annevk>
:)
07:46
<AryehGregor>
Hixie, what info do you need?
07:47
<AryehGregor>
I'd think most scripts would work the same as Latin, basically.
07:48
<AryehGregor>
What's the difference meant to be between "verbatim", "latin", and "latin-prose"? What are the use-cases for the distinction?
08:18
<AryehGregor>
I can fill you in on any Hebrew-specific issues, if there are any.
08:22
<jgraham>
So, what did I miss?
08:31
<liebrand>
hi there… i've been trying to unsubscribe from the whatwg mailing list, but although it claims it has sent me a confirmation email to unsubscribe, I never receive that email.. i've tried a few times now and checked my spam folder as well.. is there any other way to unsubscribe?
08:32
<AryehGregor>
liebrand, are you sure you're unsubscribing the right e-mail address?
08:32
<liebrand>
yeah, 99% sure
08:34
<liebrand>
is there someone one here that can verify my email address?
08:35
<AryehGregor>
Well, Hixie can. Dunno if anyone else has admin privileges to the list.
08:37
<liebrand>
ok thanks
08:59
<annevk>
liebrand: email ian⊙hc if all else fails
09:09
<Smylers>
Apologies for the duplicate mails to the list.
09:10
<Smylers>
Apparently my mailer was picking up both the CC: whatwg⊙wo and the List-Post: <mailto:whatwg⊙lwo> headers of the mail I was replying to.
09:10
<Smylers>
I'll try to watch out for that in future.
09:17
<jgraham>
So, uh, why did all my bugs get cloned?
09:18
<odinho>
jgraham: Because whatwg got its own HTML bugzilla place :)
09:18
<odinho>
jgraham: Read Administrativa email to the WHATWG-list.
09:19
<liebrand>
annevk: thanks!
09:21
<annevk>
Smylers: Gmail copes
09:22
<jgraham>
odinho: That email told me to stop reading after the first paragraph :p
09:22
<odinho>
jgraham: lol, that doesn't mean you.
09:22
<odinho>
"If you've been happily ignoring the W3C's involvement with HTML these past
09:22
<odinho>
few years, you can stop reading now. If you got a bunch of bugmail
09:22
<odinho>
recently and want to know why, the explanation is below."
09:23
<odinho>
Second part of that is absolutely you.
09:23
<Smylers>
annevk: In that case, apologies for bothering you here with an unnecessary apology!
09:23
<jgraham>
odinho: Then Hixie should reorder his if clauses
09:23
<jgraham>
Is there somewhere I can report a bug on the email?
09:23
<Smylers>
jgraham: You've been happily ignoring the W3C, while using their Bugzilla to report spec bugs?
09:24
<odinho>
jgraham: I think this channel is it.
09:26
<jgraham>
Smylers: It wouldn't make much difference to me if the bug tracker were run by the US president or Tibetan monks
09:27
<Smylers>
jgraham: OK, that covers the ‘ignoring’. Can you prove the ‘happily’?
09:29
<jgraham>
Hmm, so if the spec is Just Wrong, I now have to file two bugs rather than one?
09:30
<Smylers>
jgraham: Only if you care about the W3C's edition of the spec.
09:30
<Smylers>
jgraham: If you continue ignoring it — happily or otherwise — then you don't have to report it there.
09:30
<jgraham>
Or is the assumption that being wrong ("stable") is a desired property of the W3C spec?
09:31
<Smylers>
jgraham: Also, Hixie said he'll keep an eye on bugs reported on the W3C spec to see if they apply to the WhatWG spec too. Possibly the new W3C editors will do likewise in reverse.
09:32
<jgraham>
Smylers: It seems to me that the new system requires me to pay more attention to the W3C than the old system. Previously Hixie ensured that there were few differences in areas that are actually interoperability concerns
09:33
<jgraham>
Now everyone has to do this for themselves
09:36
<odinho>
jgraham: What with the pessimism? Lets try it out :P
09:36
<jgraham>
This wouldn't be a problem but people continue to percieve the W3C version of the spec as the "definitive" one that they should refer to from other specs
09:36
<jgraham>
odinho: I don't understand your negative attitude toward pessimism
09:36
<odinho>
:P
09:37
<Stevef_>
jgraham: suppose the assumption is that implementation bugs in HTMl5 will be fixed by the editorial team up until time when it gets frozen, then bugs will get fixed in htmlx
09:39
jgraham
retreats to the relative safety of public-webapps
09:40
<Stevef_>
jgraham: there are a number of people on the team that are switched on to the importance of keeping the specs aligned on the brwoser implementation side so the hope is...
09:41
<jgraham>
Stevef_: If people are going to put in the effort to keep the specs in alignment as far as UA conformance criteria go, that is great, but I fear the task will be Herculean
09:42
<Smylers>
jgraham: If the W3C spec suffers from bugs, because they've only been fixed in the WhatWG spec, that may lead to a change in the perception of which one is seen as definitive.
09:42
<jgraham>
Smylers: Ah, is that what optimisim sounds like?
09:43
<Stevef_>
jgraham: understand the concern, but there are 4 editors + their will be a full time technical editor, so they may have some chnace in keeping up with hixie. Also the canvas stuff will have a seperate team
09:46
<jgraham>
I suppose that Hixie should be flattered that he is being replaced by > 5 people
10:06
<AryehGregor>
It's not really a fair comparison, because Hixie is like the only full-time web spec editor in the universe.
10:08
<Ms2ger>
Presumably whoever the W3C ends up hiring will be the second, then
10:09
<AryehGregor>
Oh, I didn't realize there was a full-time job.
10:09
<AryehGregor>
Interesting.
10:11
<Stevef_>
aryehgregor: http://www.w3.org/Consortium/Recruitment/#techeditor
10:12
<Stevef_>
its still listed, but get the feeling they alreaedy have a candidate
10:13
<AryehGregor>
Interesting.
10:14
<AryehGregor>
I wonder if they'll get someone good. The obvious candidates are all already employed.
10:14
<Ms2ger>
Will it be hober?
10:14
<Ms2ger>
Maybe Marcosc
10:39
<annevk>
Marcos is too radical for them
10:39
<annevk>
they want someone they can control
10:40
<annevk>
and hober is already happily employed
11:12
<Stevef_>
well hober is one of the HTML5 editors now
11:13
<Stevef_>
along with sylvia pfeiffer
12:01
<sedovsek>
There's still no support for break-before, break-after, etc. for Multiple Columns in any modern browser? :/
13:41
<annevk>
MikeSmith: hsivonen: I just approved this comment: http://blog.whatwg.org/vnu-parser-1-4#comment-43879
13:43
<annevk>
anyone know who Erika Doyle Navara is btw?
13:43
<Ms2ger>
Oh, that's me
13:45
<annevk>
of course you are
13:47
<odinho>
hehe
14:32
<david_carlisle>
annevk: don't know, but linkedin has an Erika Doyle at "Senior Writer at Microsoft Internet Explorer Developer Content Team" ?
14:33
<david_carlisle>
annevk: oh and an Erika Navara at the same place:-)
14:34
<david_carlisle>
http://www.linkedin.com/pub/erika-navara/1/524/385
14:35
<annevk>
just says "Greater Seattle Area Computer Software" for me
14:37
<david_carlisle>
annevk: I think you get more text if you log in.
14:37
<annevk>
ah okay, don't have an account
14:37
<david_carlisle>
annevk: very wise:-)
14:39
<annevk>
maybe we should start spelling W3C W3 C or some such
14:39
<annevk>
http://www.w3.org/QA/2012/07/html5_and_htmlnext.html how hard can it be to get a name right
14:41
<annevk>
(personal pet peeve, misspellings of names annoys me)
14:41
<david_carlisle>
It's as easy to get a name right as it is to guess gender fro the name, isn't it ?
14:41
<david_carlisle>
from
14:42
<annevk>
getting the name right is one Google search
14:42
<annevk>
getting the gender right is one Google image search typically
14:42
<david_carlisle>
annevk: well you show up in the odd google search as well:-)
16:20
<dglazkov>
good morning, Whatwg!
18:06
<jarek>
Hi
18:06
<jarek>
it looks like window.getComputedStyle completely ignores styles set with SVG presentational attributes
18:07
<jarek>
is this by design? If so, it doesn't make sense because presentational attributes are part of the cascade
18:25
<jarek>
nvm, it actually works as expected
18:49
<hober>
this is *screaming* to be meme-ified, but i fear the super cool ski instructor has made too many appearances lately. http://hueniverse.com/2012/07/oauth-2-0-and-the-road-to-hell/
18:57
<Ms2ger>
The question is, is dbaron's desk at Harvard OAuth 2.0 compliant?
19:00
<Hixie>
man, he's come a long way
19:00
<Hixie>
he was instrumental in bringing websocket to the ietf
20:34
<annevk>
hober: could use the hyatt meme
20:34
<annevk>
hober: where everything blows up
20:35
<annevk>
hober: caption: "Just use OAuth"
21:42
<hober>
hmm.
21:42
<Hixie>
hmm?
21:43
<hober>
contemplating annevk's meme suggestion
21:43
<matjas>
astral Unicode symbols in <input type=password> show up as two asterisks: http://software.hixie.ch/utilities/js/live-dom-viewer/saved/1684
21:43
<Ms2ger>
Neat
21:43
<matjas>
in all browsers I tested so far… so i guess it’s standard
21:44
<matjas>
(OS X 10.8 seems to do the same for all password fields)
23:26
<zewt>
matjas: heh yeah, the hope of web developers ever getting non-BMP characters right is pretty remote when browsers don't even get it right
23:27
<zewt>
matjas: ie9 shows a single *, fyi
23:30
<zewt>
opera shows two dots but it edits like a single character (better than ff/chrome, anyway)
23:38
<Hixie>
i swear, i'm _this_ close to dropping the authorship information in the references section
23:52
<WeirdAl>
Hixie - tired of staring at a thousand names?
23:58
<gsnedders>
zewt: Opera on what platform? This is probably different between platforms.
23:58
<gsnedders>
Well, the impl is. The behaviour might be.
23:58
<zewt>
win7
23:59
<zewt>
haha context menu in opera on password inputs has "check spelling"
23:59
<zewt>
that might possibly be among the most useless context menu items ever