03:45
<MikeSmith>
http://mpeg.chiariglione.org/working_documents/mpeg-04/part29/WVC.zip
03:47
<MikeSmith>
the "WebVC" profile of H.264 that Apple and Cisco and others want to make royalty-free
03:53
<zewt>
zero confidence in anything "free" related to the MPEG-LA
03:55
<zewt>
the *most* benign motive i can think of for them doing anything like that is keeping more resources from going into genuinely-free competing tech
04:18
<MikeSmith>
zewt: yeah
04:19
<MikeSmith>
as Rob Glidden has put it, "It seems every few months a patent pool or some one manages to spin up wishful thinking that there will soon be a “new deal” that will answer the needs of royalty free communities, only to cleverly leave the hopeful twisting in the wind."
04:21
<MikeSmith>
but WebVC is a bit different in that it's not an effort from MPEG overall, but instead a proposal from Apple and others that they are trying to convince the other MPEG members to make royalty free
04:21
<MikeSmith>
so it seems good that they are trying at least
04:47
<zewt>
doesn't seem terribly good that they're trying to arrange for the web to use a format that's "free" in some particular (presumably limited profile) state, whose other features and future development have no promise of following suit
04:48
<zewt>
(call me cynical; with anything involving organizations like the MPEG-LA, it's usually the safest default)
10:43
<MikeSmith>
anybody aware of Selectors API test suites other than http://dev.w3.org/2006/webapi/selectors-api-testsuite/ and http://ejohn.org/apps/selectortest/ ?
10:43
<annevk>
Gecko removed support for UTF-32 right? I wonder why it's still in http://mxr.mozilla.org/mozilla/source/intl/uconv/src/charsetalias.properties
10:45
<annevk>
Quite a few cases in http://mxr.mozilla.org/mozilla/source/intl/uconv/src/charsetalias.properties have uppercase characters in the key
10:45
<annevk>
e.g. "csMacintosh" and "csIBM864"...
10:46
<Ms2ger>
Those may be bugs, actually
10:47
<Ms2ger>
Or not
10:47
<annevk>
At some point bugs become features :)
10:48
<annevk>
Gecko also still has the three Thai encodings as distinct
10:49
<Ms2ger>
I thought I remembered that those weren't picked up, but I don't find code to back that up
10:49
<annevk>
I believe that's true
10:50
<annevk>
aah https://bugzilla.mozilla.org/show_bug.cgi?id=235090
10:51
<annevk>
UTF-16 is as recent as 2004
10:51
<annevk>
well part of it
10:51
<annevk>
I guess back then there still was a chance to get rid of it
10:53
<MikeSmith>
annevk: do you know of XHR test suite other than http://tc.labs.opera.com/apis/XMLHttpRequest/ ?
10:54
<annevk>
the CORS tests uploaded recently have XHR tests in them
10:54
<MikeSmith>
uploaded to where?
10:54
<MikeSmith>
dvcs?
10:54
<annevk>
http://dvcs.w3.org/hg/webappsec/
10:55
<MikeSmith>
hai
10:55
<MikeSmith>
thanks
11:56
<annevk>
never really thought about this http://www.reddit.com/r/IAmA/comments/nqerr/i_am_a_totally_blind_redditer/c3b6z7a that's kind of a major fail
12:00
<MikeSmith>
annevk: does anybody support the From-Origin header yet?
12:00
<annevk>
don't think so
12:00
<MikeSmith>
k
12:01
<MikeSmith>
know if there are any test suites for data URL support?
12:02
<annevk>
http://hixie.ch/tests/adhoc/data/
12:02
<MikeSmith>
thanks
14:56
<MikeSmith>
hmm, seeing some odd behavior with http://platform.html5.org/ in Opera
14:57
<MikeSmith>
try any of the links after the Server-Sent Events ones
15:05
<annevk>
works here...
15:06
<annevk>
ooh
15:06
<annevk>
a bit further on I do notice something odd
15:06
<smaug____>
Hixie: ping
17:58
<zewt>
wow. firefox with the single worst browser slogan I've ever seen: "Different by Design"
18:12
<zewt>
heh ff9 fullscreen api is seriously buggy ... no wonder it's disabled by default (but they shouldn't have mentioned it in the changelog)
19:59
<Yuhong>
I wonder about how much of the CSS2 recommendation from 1998 turned out to be implementable.
20:00
<gsnedders>
I don't there was that much that was, if only because it is so vague.
20:00
<gsnedders>
The vagueness leads to contradictions in edge-cases, though, to be fair.
20:01
<Yuhong>
Can anyone estimate a percentage?
20:35
<tantek>
Yuhong, Gsnedders, most of it was implementable.
20:36
<tantek>
even if vague
20:37
<Yuhong>
tantek: Yea, text-shadow in particular was I read implemented by Tasman, while it is only IE10 that finally implemented it in Trident.
20:37
<tantek>
gsnedders, on the contrary, vagueness leads to being able to resolve apparent contradictions by selective interpretation of said vagueness.
20:37
<tantek>
It's excessive precision (duplicated across sections/chapters) that lead to most contradictions.
20:37
<gsnedders>
tantek: Yeah, I know
20:37
<Yuhong>
tantek: I bitched in Chris Wilson's blog about Tasman not being ported to Windows. You worked at MS doing it, right?
20:38
<tantek>
Yuhong, yes, I was the dev lead for Tasman
20:38
<gsnedders>
I was under the impression that Tasman was ported to Windows, albeit never shipped
20:38
<tantek>
gsnedders, that's right.
20:38
<annevk>
what's the difference between http://mxr.mozilla.org/mozilla-central/source/intl/uconv/src/ and http://mxr.mozilla.org/mozilla/source/intl/uconv/src/
20:38
<gsnedders>
tantek: Not for IE, though, right?
20:38
<tantek>
and ported to numerous other OS's/processors (mostly embedded systems like set top boxes etc.)
20:39
<gsnedders>
tantek: Or was it intented that IE would use it eventually?
20:39
<tantek>
gsnedders - there was a Win32 Tasman Shell that ran the Tasman engine.
20:39
<tantek>
There was a possibility of using Tasman to do XHTML rendering.
20:39
<tantek>
But then it turned out that XHTML didn't matter.
20:39
<Yuhong>
tantek: Has hooking Tasman via X-UA-Compatible been considered?
20:40
<tantek>
Yuhong, since the latter years of the 2000s, browser engines long since eclipsed Tasman.
20:41
<Yuhong>
tantek: But it might have a good starting point. I know it is too late for it now though.
20:41
<tantek>
except perhaps in size/speed (we got it down to being smaller than Opera, less memory requirements, and much faster)
20:42
<Ms2ger`>
annevk, the latter is Firefox 3.0
20:42
<gsnedders>
tantek: How many people did you have working on it the time of IE/Mac's cancelation?
20:43
<Yuhong>
tantek: But what about back in 2007-2008?
20:43
<Yuhong>
tantek: When X-UA-Compatible was being considered during IE8 development.
20:44
<gsnedders>
I doubt Tasman would be a better base than Trident as-of-IE8 for what has been implemented CSS-wise in IE9/10.
20:45
<gsnedders>
IE8 gave them a layout engine that's probably a decent advantage over everyone else for a lot of the more radical CSS3 modules.
20:45
<Yuhong>
gsnedders: I know by the time of IE9 it was too late, but I am talking about during *IE8* development.
20:46
<gsnedders>
Yuhong: I expect regardless of choice of Trident/Tasman you're gonna have to rewrite a lot of it for make up for years of non-development.
20:46
<gsnedders>
When you're rewriting it, it won't make much difference as to which you choose.
20:51
<Yuhong>
gsnedders: but note for example Tasman already implemented XHTML, while IE8 didn't.
20:52
<Yuhong>
gsnedders: And IE6/7 lack of support for XHTML was already bad enough.
20:53
<gsnedders>
True, but plenty of people don't view lack of XHTML support as a major concern.
20:53
<gsnedders>
(Though I was led to believe Trident had major architectural issues that made it difficult)
20:57
<annevk>
Ms2ger`: interesting, so where did charsetalias.properties move?
21:00
<annevk>
Ms2ger`: is that http://mxr.mozilla.org/mozilla-central/source/intl/uconv/src/nsUConvModule.cpp now?
21:00
<Ms2ger`>
intl/locale/src/charsetalias.properties
21:00
<Ms2ger`>
(Bug 637599)
21:01
<Yuhong>
So can anyone estimate the percentage of CSS2 recommendation that was implementable?
21:01
<Yuhong>
So can anyone estimate the percentage of the CSS2 recommendation from 1998 that was implementable?
21:01
<Ms2ger`>
Sure
21:01
<Ms2ger`>
About 142%, I'd say
21:05
<annevk>
thanks Ms2ger`
21:06
<annevk>
Ms2ger`: that file looks more reasonable than the one I have been looking at
21:06
<Ms2ger`>
Np
21:09
<Yuhong>
gsnedders: Yea, IE9 had to rewrite the XML support.
21:09
<Yuhong>
gsnedders: Before IE9, XML parsing in IE used the external MSXML library.
21:10
<Yuhong>
(in IE6 and later, MSXML3)
21:35
<Yuhong>
I wonder if there is a way to fool sites into thinking that document.all exists.
21:35
<Yuhong>
I wonder if there is a way to fool sites into thinking that document.all exists in Mozilla.
21:37
<Yuhong>
(I want to play with this: http://grox.net/doc/web/javascript/dynduo/)
21:38
<Yuhong>
Funny thing that Mozilla only now is completing their support for the IE4 DOM.
23:51
<Yuhong>
Trivia: You can set contentEditable in IE8 and later's Developer Tools.
23:51
<Yuhong>
I am using that to play around.