00:38
<takkaria>
gah, it's too warm to sleep
08:00
<hsivonen>
http://stackoverflow.com/questions/1038191/google-homepage-not-valid-html/1044778#1044778
08:33
<Lachy>
depending on how the page is built, I'm quite sure the performance of google could actually be improved by adopting a more standards based approach
08:33
<othermaciej>
I feel guilty for posting on the codec thread
08:33
<othermaciej>
I should have known better
08:34
<MikeSmith>
othermaciej: I can't see what part of your message somebody could reasonably take issue with.
08:34
<MikeSmith>
I think the hardware concern is a major concern
08:35
<othermaciej>
MikeSmith: I was just hoping that I wouldn't extend the flamewar
08:42
<MikeSmith>
othermaciej: if you didn't post on the thread at all, your silence would be interpreted as a clear indication that you are indeed part of the conspiracy
08:42
<othermaciej>
obviously
08:43
<MikeSmith>
I think it's high time I added some H.264 and Theora stuff to my wisdom generator
08:47
<MikeSmith>
heh. looking at it now, I see that I already had the "the MP/LA Expert Group" in there
08:47
<MikeSmith>
under the category "seemingly-benign-pawns"
08:56
<MikeSmith>
ok, now with even more wisdom: http://logopoeia.com/wisdom/
08:56
<MikeSmith>
well, slightly more
08:58
<annevk2>
first hit: "the divine principle of time inequality is encoded in Anne van Kesteren's weblog."
08:58
<annevk2>
though reading further it also says "the divine principle of time inequality is encoded in the Nag Hammadi Library"
08:59
<annevk2>
"Eminent thinkers throughout history (Roy Fielding, etc.) agree that the divine principle of time inequality is the transcendental principle behind all creation" :p
09:10
<hsivonen>
is any spec responsible for XMLSerializer?
09:13
<annevk2>
nope
09:15
<hsivonen>
:-( there should be
09:16
<hsivonen>
It turns out that MySpace uses it on HTML DOMs and makes expectations
09:16
<othermaciej>
maybe XMLSerializer and DOMParser can be snuck into Web DOM Core
09:16
<othermaciej>
what expectations do they have?
09:18
<hsivonen>
othermaciej: they expect no generated namespace declarations
09:18
<othermaciej>
where do they depend on that?
09:19
<othermaciej>
I think Safari would generate namespace declarations and so violate their expectation
09:19
<hsivonen>
they except h=h.replace(/<html>|<\/html>|<html\/>/gi,"") to zap the root element tags
09:19
<hsivonen>
when updating status/mood
09:21
<othermaciej>
you're making it sound like their way to do status updates is a Rube Goldberg device
09:25
<hsivonen>
othermaciej: it is
10:10
<roc>
if we want an good baseline video codec
10:10
<roc>
and the barrier is the choices made by device manufacturers and their hardware suppliers
10:10
<roc>
and their decisions are purely economic
10:11
<hsivonen>
how varied are the DSPs used in contemporary handhelds?
10:11
<roc>
then the only practical way forward is to encourage the proliferation of content in an unencumbered codec, to tip that economic balance
10:12
<roc>
so it seems to me that everyone who's interested in having a good baseline video codec should be encouraging that
10:13
<roc>
including by supporting Theora wherever the device constraints do not apply
10:15
<roc>
hsivonen: a lot are using these OMAP chips, but a lot aren't
10:16
<hsivonen>
just wondering if writing for a Nokia shipped TI chip results in any portability to whatever Apple uses in iPhone
10:16
<roc>
I believe the answer is no
10:17
<hsivonen>
ok :-(
10:42
<annevk2>
oh hey, gsnedders just invaded my inbox via our bug tracking software
10:42
<annevk2>
welcome!
10:45
<roc>
annevk2: I already made your point about vendor consensus!
10:46
<roc>
therefore it's a good one :-)
10:46
<hsivonen>
is Web DOM Core a W3C deliverable already?
10:47
<gsnedders>
annevk2: mwhaha?
10:49
<annevk2>
roc, teaches me to read the whole thread first (over and over again...) :)
10:49
<roc>
this is why gmail is superior
10:52
hsivonen
hasn't read the whole thread but points out that Nokia ships Flash on Maemo
10:52
hsivonen
wonders what *that* does to battery life when watching YouTube
10:52
<hsivonen>
I guess I should measure
11:21
<annevk2>
roc, congrats btw on Firefox 3.5, haven't tried the official release yet, but it sure seems like a great product
11:31
<roc>
thanks
11:34
<roc>
but 3.6prealpha1 is what the cool kids are using these days :-)
11:35
<annevk2>
so I hear, with a certain pref turned on rather than off :)
11:36
<takkaria>
roc: what exciting stuff is planned for firefox.next?
11:38
<roc>
Henri's HTML5 parser is on trunk, I assume that's what Anne was referring to
11:38
<roc>
hopefully we'll turn it on for FF.next
11:39
<roc>
well, I'm sure we'll turn it on, the question is whether we'll turn it off after we see what it does to the Web :-)
11:39
<takkaria>
hehe
11:39
<roc>
multiple backgrounds is already on trunk
11:39
<takkaria>
is there anyone looking at webforms stuff?
11:40
<roc>
someone is looking at file upload
11:40
<roc>
I'm not sure what that plan is
11:41
<roc>
Jonas is working on XBL2, I suspect we may try to build new web forms stuff on top of that
11:41
<roc>
<video poster> is on trunk
11:41
<hsivonen>
fun. text-indent makes WebKit indent the video controller
11:41
<hsivonen>
in both Chrome and Safari
11:42
<roc>
lots of other stuff is going on
11:44
<hsivonen>
hmm. looks like Cortado doesn't work right in IE with two instances on the same page :-(
11:44
<takkaria>
roc: sounds cool, I should start reading the mozilla planet again sometime
11:45
<hsivonen>
hmm. maybe I should try to get my blog feed on planet so that I could advertise the HTML5 parser once the Windows crasher patch and comment end bang state are in
11:49
<hsivonen>
how are Xiph Cortado and Wikimedia Cortado related? I suspect the Wikimedia Cortado worked better than the Xiph one in IE.
11:50
<hsivonen>
I think it's important to promote Ogg-only video also for IE, so that people don't need to use royalty-encumbered fallback formats, but it seems Cortado isn't quite ready for prime time
12:26
<Lachy>
has anyone researched the h.264 patents to find out when they're all due to expire, which would make h.264 royalty free?
12:26
<Lachy>
I checked the MPEG-LA patent list earlier, but they only mention the expiration dates for the few that already expired. The others just list patent numbers
12:28
<gsnedders>
Seeming the first draft was 1999, I would guess no earlier than 17 years after that
12:28
<gsnedders>
The final draft was 2003, so potentially 2020
12:28
<hsivonen>
hey, that's before 2022 :-)
12:28
<gsnedders>
:)
12:29
<roc>
the problem is that there will be an H.265 or whatever
12:29
<roc>
hsivonen: yes, you should definitely be on planet.mozilla.org
12:30
<Lachy>
roc, are you just speculating about h.265, or is there already work in progress?
12:30
<roc>
I'm speculating
12:30
<gsnedders>
Lachy: "under study for potential development" according to wikipedia
12:30
<hsivonen>
has On2 said anything about why they made the VP3 stuff free but not VP6?
12:30
<roc>
why wouldn't there be?
12:31
hsivonen
wonders if On2 will provide nicely packaged Theora encoding solutions
12:31
<Lachy>
http://en.wikipedia.org/wiki/H.265
12:32
<Lachy>
gsnedders, I thought the patent term was 20 years, not 17
12:32
<gsnedders>
Lachy: It depends on country. Don't expect me to remember the max :)
12:34
<Lachy>
"Patents must be granted in all "fields of technology," although exceptions for certain public interests are allowed (Art. 27.2 and 27.3 [1]) and must be enforceable for at least 20 years (Art 33)." - http://en.wikipedia.org/wiki/TRIPS#The_requirements_of_TRIPS
12:34
<Lachy>
so that applies to all countries that have agreed with TRIPS
12:55
<hsivonen>
quick black-box poking suggests the XMLSerializer in WebKit is already an HTML serializer when applied to HTML
12:57
<hsivonen>
what's the right mailing list for XMLSerializer? www-dom?
12:59
<Lachy>
I don't see how www-dom would be relevant
12:59
<hsivonen>
what ARIA cases is chaals referring to here: http://lists.w3.org/Archives/Public/www-dom/2009AprJun/0009.html ?
13:01
<Lachy>
no idea. Unfortunately, he's not around for me to ask him
13:01
<MikeSmith>
hsivonen: about XMLSerializer, lacking any better place, I'd say public-webapps,
13:02
<hsivonen>
MikeSmith: ok.
13:02
<hsivonen>
MikeSmith: btw, I ran the V.nu deployment script yesterday
13:03
<MikeSmith>
hsivonen: cool
13:03
<MikeSmith>
I'll see if I can get the W3C instance synced up too
13:48
<Lachy>
Hixie, is QUOTA_EXCEEDED_ERR used in webstorage defined as a constant anywhere for DOMException?
13:57
<Lachy>
does anyone here working on WebKit know if you enforce any disk quota limits for localStorage and sessionStorage?
13:58
<Lachy>
I just tried testing it with an infinite loop that keeps storing data until it throws an exception, but it just kept going
14:00
<Rik|work>
Lachy: I've seen bugs to fix that
14:00
<Rik|work>
don't know if they're closed or not
14:02
<Rik|work>
Lachy: https://bugs.webkit.org/show_bug.cgi?id=22765
14:02
<Rik|work>
and https://bugs.webkit.org/show_bug.cgi?id=26155
14:03
<Lachy>
Rik|work, ok. I'd recommend also setting a quota for sessionStorage
14:35
<gsnedders>
annevk2: yt?
14:35
<gsnedders>
annevk2: What's "ouch" about http://www.w3.org/TR/2004/REC-DOM-Level-3-Core-20040407/core.html#baseURIs-Considerations?
14:51
<hsivonen>
hmm. looks like Daily Motion has some borked browser sniffing
14:52
<hsivonen>
Telling me to install Firefox in order to see video without proprietary plug-ins
14:52
<hsivonen>
when I'm using Firefox
14:52
<annevk2>
gsnedders, liveness
14:53
<hsivonen>
oops. I've overridden my User-Agent string
14:53
<gsnedders>
annevk2: Ah, OK. So nothing more.
14:53
<annevk2>
maybe
14:53
<gsnedders>
Heh. Opera's the only browser that ever treats it as live, though we have bugs.
14:56
<annevk2>
hsivonen, willing to write the spec for XMLSerializer and DOMParser?
14:57
<hsivonen>
annevk2: can't volunteer right now :-(
15:12
<gsnedders>
href in the DOM on HTMLAnchorElement doesn't give the absolute URL, right? The only way to get that is to recompose the decomposition attributes?
15:16
<jgraham>
gsnedders: It gives the absolute URL
15:19
<gsnedders>
The spec says that it reflects the content attribute though…
15:19
<gsnedders>
Hixie?
15:20
<annevk2>
gsnedders, yes, but it reflects a URL value
15:20
<annevk2>
gsnedders, see the definition of reflect
15:23
<gsnedders>
annevk2: Ah, right. I need to actually read beyond the note
15:23
<jgraham>
annevk2: The definition of reflect could be improved to note that there are special cases wwhere the behaviour is more complex
15:25
<annevk2>
hmm, that's just an introductory note, it doesn't define anything
15:26
<annevk2>
but euh, I'm not the editor :)
15:29
<MikeSmith>
attribute [ReflectURL] DOMString src;
15:29
<MikeSmith>
[[
15:29
<MikeSmith>
This indicates that the src DOM attribute reflects a content attribute
15:29
<MikeSmith>
named src, but that the result is a completed URL, not the raw
15:29
<MikeSmith>
attribute value:
15:29
<MikeSmith>
]]
15:29
<annevk2>
I'd think you want [Reflect=URL]
15:30
<annevk2>
'cause all "reflect" thingies have something in common, but some have specific types
15:31
<annevk2>
(e.g. they all create a new content attribute if none exists)
15:31
<MikeSmith>
annevk2: I guess they didn't do it that way maybe because of [Reflect=URL] would imply there's a content attribute named "URL"
15:32
<MikeSmith>
because they're also doing, attribute [Reflect=class] DOMString className;
15:39
<annevk2>
I see
15:40
<gsnedders>
annevk2: Is there any spec which says http://t/imported/annevk/xml/xml-base/004.xht should pass?
17:47
<Philip`>
Is Perl installed on OS X by default?
17:48
<Philip`>
Hmm, sounds like it is
17:50
Philip`
wonders why somebody gets a "perl: No match." error
17:51
<Philip`>
Oh, it's because they're using csh and writing "[...]" on the command line
17:55
<Hixie>
Lachy: web dom core iirc
18:17
<gsnedders>
MikeSmith: yt?
18:51
<MikeSmith>
gsnedders: I'm here now
18:52
<MikeSmith>
for a while at least
19:12
<jgraham>
MikeSmith and gsnedders are like star crossed lovers
19:14
<MikeSmith>
heh
19:54
gsnedders
wonders where he can buy a birthday card, and what he should get as a present.
19:56
<Philip`>
gsnedders: A birthday card shop
19:56
<gsnedders>
Philip`: Located where in Linköping, at what address?
19:56
<Philip`>
gsnedders: A tea towel
19:56
<takkaria>
gsnedders: make your own
19:56
<Philip`>
gsnedders: I don't know - walk around and look for one that has thousands of birthday cards in the front window
19:56
<gsnedders>
takkaria: Me!?
19:57
<gsnedders>
takkaria: Do I look creative?
19:57
<takkaria>
if you write code, you're at least somewhat creative
19:57
<gsnedders>
Do I look as if I have artist capabilities?
19:59
takkaria
shrugs
20:01
<takkaria>
looks can be and often are deceiving
20:16
<jgraham>
gsnBokia sells some cards if you don't fancy DIYing it
20:16
<jgraham>
Oh he left. That's why tab did nothing
21:46
<Hixie>
WebSocket ports and schemes; text/html and application/xhtml+xml; text/cache-manifest; text/event-stream; application/microdata+json and application/microdata+bibtex
21:46
<Hixie>
anything else need registering with IANA?
21:48
<annevk2>
text/javascript update?
21:50
<annevk2>
someone needs to finish http://tools.ietf.org/html/draft-hoehrmann-javascript-scheme
21:51
<annevk2>
ideally the character encoding stuff would be fixed at the IETF/IANA/Unicode-level
21:54
<annevk2>
rel values needs some more coordination prolly
21:56
<annevk2>
is text/uri-list defined?
21:57
<annevk2>
only as experimental in http://tools.ietf.org/html/rfc2483 afaict
21:59
<Hixie>
i think i'll leave those for now
21:59
<Hixie>
but yeah
21:59
<Hixie>
thanks
23:34
<sayrer>
lol, hixie sent "suitable" to whatwg and "consensus" to w3c
23:34
<sayrer>
hahaha