00:04
<hober>
Hixie: I'm not sure what this is
00:07
<bholley>
hober: this is a meetup to spec out the corner-case behavior of cross-origin Location objects
00:11
<Hixie>
hober: https://www.w3.org/Bugs/Public/show_bug.cgi?id=20701
00:12
<Hixie>
i'm going offline til tomorrow. later all.
05:32
<MikeSmith>
astearns: I suppose I should have realize that. Dunno why he bothers, since he's not going to get what he wants, if that's what he wants. Unless he wants to start writing ones himself in the old-school style to replace them
05:33
<MikeSmith>
some people really just love formalisms I guess
06:14
<MikeSmith>
hsivonen: http://bugzilla.validator.nu/ keeps requiring me to log back in during the same session, even though my IP is not changing and anyway I have "Restrict this session to this IP address". (And in both Firefox and Chromium)
06:50
<hsivonen>
MikeSmith: I checked the relevant Bugzilla settings and didn't see anything obviously wrong.
06:51
<hsivonen>
MikeSmith: I'm not a Bugzilla expert, so I really don't know what might be causing the problem
06:51
<hsivonen>
MikeSmith: do you know what I should be looking for?
06:53
<hsivonen>
annevk: regarding glibc and gb18030: a) not suprised at maintainer attitude b) gb18030 was supposed to be a UTF, so why does there exist a 2005 update?
06:53
<hsivonen>
*surprised
07:00
<hsivonen>
annevk: I find the use of PUA in GBK decoders (per http://coq.no/character-tables/chinese-simplified/en) troubling
07:13
<hsivonen>
annevk: considering that IE doesn't support (according to coq.no) ISO-8859-14 and -16, did you try to find out if there exists non-test case Web content in those encodings?
07:25
<MikeSmith>
hsivonen: my Bugzilla knowledge is pretty limited. Don't know what to suggest you should look for. But if you try it yourself I'd think you'd be able to reproduce it.
07:47
<zcorpan>
no patches, no record :-(
07:50
<zcorpan>
Hixie: the application-name edit seems like it selects no meta if no language has been declared. shouldn't it just fall back to the first element instead of none?
08:26
<MikeSmith>
trying figure out of marcosc is getting trolled in http://lists.w3.org/Archives/Public/uri/2013Dec/0007.html or whether marcosc is actually the one doing the trolling
08:40
<hsivonen>
MikeSmith: indeed hard to say. marcosc's position looks rather extreme. what John Cowan says makes sense if he executes English as program code.
08:50
<zcorpan>
testing resolving url in a cache manifest was actually straight-forward, to my surprise
09:02
<zcorpan>
wonder if i should also test "fallback" and "online whitelist", that seems slightly more annoying
09:09
<MikeSmith>
hsivonen: claiming a spec that uses (English) prose pseudo-code is a "reference implementation" is like saying a recipe for a cake is the same thing as a cake
09:41
<marcosc>
what?
09:41
<marcosc>
me troll?
09:42
<marcosc>
hsivonen, MikeSmith - what he was implying doesn't make sense.
09:42
<marcosc>
English is not code
09:43
<marcosc>
And a spec not written in a form that can be implemented doesn't make sense
09:43
<marcosc>
also, I was basically giving the definition of a standard as defined by ISO, I think... let check on that
09:48
<tantek>
marcosc, standards that you have to pay for to get a copy of a PDF?
09:52
<zcorpan>
nice, i put \u00E5 in the query string in a cache manifest, presto fetches %C3%83%C2%A5
10:02
<marcosc>
tantek: isn't that what makes a standard? :)
10:04
<marcosc>
"This might be very useful for people writing implementation code, butit's not so helpful for people *using* the feature (like authoring fileURIs), or people trying to understand why something works the way it works."
10:04
<hsivonen>
marcosc: well, suggesting a browser is a reference implementation is pretty extreme even for the WHATWG and doesn't make sense without saying *which* browser
10:04
<marcosc>
hsivonen: I don't follow?
10:05
<hsivonen>
marcosc: John is right about the meaning of "reference implementation"
10:05
<hsivonen>
marcosc: which means a piece of software whose behavior constitutes the authoritative standard
10:06
<marcosc>
I don't agree. Or at least, and implementation to me is something else.
10:06
<marcosc>
an spec cannot be an implementation of itself
10:06
<marcosc>
as an implementation cannot pass a test suite
10:07
<marcosc>
im·ple·men·ta·tion "the process of putting a decision or plan into effect; execution."
10:07
<hsivonen>
marcosc: John is right about what "reference implementation" means. When he says WHATWG specs are reference implementations, he probably considers WHATWG spec language style as algorithmic enough English to be like code
10:08
<marcosc>
even if that were true (which, is demonstratively not the case as English is pretty bad at executing), his definition of a standard would still be flawed
10:09
<marcosc>
as what he is suggesting for a standard is something that would not follow these rules and hence one could not implement reliably?
10:09
<hsivonen>
dunno how he defines "standard"
10:09
<marcosc>
well, he clearly excludes the WHATWG specs
10:09
<marcosc>
which makes me very suspicious and leads me to conclude that he is confused about what a standard is.
10:09
<hsivonen>
marcosc: maybe RFCs are something that one could not implement reliably, and, therefore, count as standards. ;-)
10:10
<marcosc>
heh
10:10
<marcosc>
could be :)
10:10
<marcosc>
I guess the same applies to W3C specs?
10:10
<hsivonen>
at least the ones that aren't just copies of WHATWG specs. :-)
10:28
<Ms2ger>
Nobody mentioned http://blog.chromium.org/2013/12/ecma-forms-tc52-for-dart-standardization.html yet?
10:44
<jgraham>
Ms2ger: zgrep
10:45
<jgraham>
Oh, I was way into the backscroll
10:45
<Ms2ger>
jgraham, yep, I found that, thanks :)
10:45
<jgraham>
Sorry
11:24
<darobin>
I had no idea that Ecma standardised Eiffel
11:26
<jgraham>
I hear it was a towering achievement
11:26
jgraham
gets his coat
11:28
<darobin>
badum *tsch*
11:32
<gsnedders>
11:36
<marcosc>
You know ECMA is the real deal because they publish using Word!... though those in-spec-algorithms my not cut the "standards" mustard.
11:39
<Ms2ger>
Pff, tc39 only does reference implementations
11:42
<jgraham>
They should have written the dart standard in js and the js standard in C# and the C# standard in Eiffel. Then it really would be an Eiffel tower.
11:43
<tantek>
jgraham++
11:44
<gsnedders>
jgraham: Don't say that. People might get ideas.
11:45
<Ms2ger>
Did gsnedders just get ideas for a thesis?
11:46
<jgraham>
gsnedders: It's fine. At the end they would publish the whole lot in Open Office XML transcoded into JSON and distribute it on HVD-ROM cartridges
11:52
<annevk>
hsivonen: I think gb18030 had an update to note more code points have a mapping, not sure exactly what the purpose was; it might be that the standard also has font requirements and such
11:52
<zcorpan>
Hixie: why the readyState check for video resize event?
11:52
<annevk>
hsivonen: I have not tried to find out anything for -14 and -16, if only one browser did not support an encoding my assumption was that I should include it
11:56
<annevk>
hsivonen: so it seems gb18030 and gbk in the Encoding Standard are basically wrong
11:57
<annevk>
hsivonen: I will figure out how to fix that I suppose :/
12:05
<zcorpan>
Hixie: nm
12:23
<zcorpan>
css 'binding' is basically obsolete, right?
12:33
<annevk>
marcosc: oh man, that thread
12:49
<gsnedders>
darobin: I'm not sure I want to accept your pull requests. I'll merge them if it seems like people are actually going to implement them in browsers.
12:53
<hsivonen>
annevk: looking forward to finding out if we can get away with aliasing gbk to gb18030
12:53
<hsivonen>
Hixie: when you put gb18030 instead of gbk in the HTML spec, did you do research or did you just make an optimistic guess that that would work?
12:54
<hsivonen>
(in the HTML spec as the Simplified Chinese fallback encoding that is)
12:54
<annevk>
hsivonen: so I think that alias is not feasible unless those 25 code points turn out not to matter
12:54
<annevk>
hsivonen: well actually, given what Opera did for a long time maybe it is...
13:08
<hsivonen>
So compared to ISO Greek, Windows Greek just bikesheds the position of Ά? why oh why?
13:09
<annevk>
What came first? ISO or Windows?
13:09
<gsnedders>
hsivonen: Swapping it with what? An undefined character?
13:10
<annevk>
Might as well have been the standardization committee that had a hatred for Microsoft
13:11
<hsivonen>
"this difference was instrumented in order to rectify the appearance of an early version of Microsoft Word for Windows in which the end-of-section symbol clashed with the "Greek capital alpha with acute" glyph"
13:14
<hsivonen>
gsnedders: Windows Greek has a paragraph symbol where ISO Greek has capital alpha with acute
13:15
<hsivonen>
annevk: ISO Greek is 1987 and a copy of a Greek standard from 1986
13:15
<hsivonen>
annevk: dunno when Windows got Greek support
13:17
<annevk>
hsivonen: seems like Microsoft hated ISO then :/
13:19
<darobin>
gsnedders: I don't expect you to merge them before you have more evidence
13:20
<gsnedders>
darobin: Okay, good to know we're on the same page :)
13:20
<darobin>
I mean, feel free to if you want to, but there's a reason I marked them as "merge with caution"
13:20
<darobin>
gsnedders: I'm landing this because input from implementers is that this is what they plan to do, but as always I'll believe that when I've seen it :)
13:21
<gsnedders>
darobin: Are there any bugs anywhere for it?
13:23
<darobin>
gsnedders: not on the parsing changes but the use cases were brought up from https://www.w3.org/Bugs/Public/show_bug.cgi?id=20115 and https://www.w3.org/Bugs/Public/show_bug.cgi?id=20114
13:24
<gsnedders>
darobin: I meant for implementors
13:24
<darobin>
ah, sorry
13:25
<darobin>
gsnedders: there's https://bugzilla.mozilla.org/show_bug.cgi?id=33339 if you want to read long history; I'm not aware of others
13:25
<gsnedders>
(I don't care so much about why it's there as whether I should implement it :P)
13:25
<darobin>
gsnedders: my understanding was that WebKit wanted a bug when this came into the spec, so as to have a link
13:25
<darobin>
I'll be making sure that happens
13:26
<darobin>
and I can't give you a bug for IE though the input was positive :)
13:26
<darobin>
I don't know about Blink
13:26
<darobin>
gsnedders: I wouldn't bother taking the PRs yet, just leave them open and apply when something happens
13:26
<gsnedders>
Wow, I actually remember my login details for Bugzilla. :)
13:27
<darobin>
*normally* the changes I made to html5lib ought to be enough
13:27
<gsnedders>
darobin: BTW, you might want to add Ruby to the sanitizer whitelists — it should be safe, no?
13:27
<darobin>
mmm, that's a good point
13:27
<gsnedders>
On the other hand, my near total rewrite of the sanitizer will probably land before this :P
13:28
<gsnedders>
darobin: Can you comment as much on the bug, and explicitly say about waiting for impls for html5lib-tests (not everyone with reviewer access there so closely follows both WHATWG/W3C specs).
13:29
<gsnedders>
(I have an essay to somehow finish in the next 30 minutes or so.)
13:29
<darobin>
I was thinking of my own sanitiser anyway ;)
13:29
<darobin>
go finish your essay, I'll add the comments
13:29
<gsnedders>
(I'm not going to finish it. I've been ill for weeks, been granted maximum extension, and I'm still nearly 1000 words under min word count)
13:30
<darobin>
sorry to hear you've been ill
13:30
<darobin>
if you close IRC that just a little bit over a word every two seconds...
13:31
<gsnedders>
:)
13:31
<gsnedders>
I'm not sure I can write more without reading more papers to actually be able to cite anything more.
13:36
<darobin>
thanks, hober, really, for bikeshedding straight from the top of a thread
13:45
<annevk>
hsivonen: so it turns out they actually changed the mapping in gb18030 around 2005
13:46
<annevk>
hsivonen: ftp://ftp.oreilly.com/pub/examples/nutshell/cjkv/pdf/GB18030_Summary.pdf has some background story
13:48
<annevk>
Pretty interesting: "We will later see that both the code space and the character repertoire of GBK create the foundation for the almost completely compatible code space of GB 18030. In fact, there is text in the standard indicating that GB 18030 (the “standard”) is meant to replace (代替 daiti) GBK (the “specification”)."
13:51
<annevk>
"Update: Actually, a third large gap can be found in the re-released mapping data now that is caused by the removal of the Unicode Surrogate code points (U+d800 through U+dfff) from the four-byte mapping area."
13:51
<annevk>
Whoa, so initially gb18030 covered more than Unicode scalar values
13:53
<annevk>
Kinda of like UCS-4, but more restricted in range...
13:54
<annevk>
http://en.wikipedia.org/wiki/UTF-32 does not even mention it excludes surrogates...
14:00
<annevk>
http://shop.oreilly.com/product/9781565922242.do needs a second edition to tell the story of gb18030
15:17
<annevk>
The gb18030 situation might be less worse than expected. Extreme Decemberfest might hold me back in fixing it today though
15:17
<annevk>
s/worse/bad/
15:17
<jgraham>
Ms2ger: -webkit-? Really
15:17
jgraham
is disappointed :p
15:18
<annevk>
jgraham: in Ms2ger or Chinese encodings?
15:18
<jgraham>
Both?
15:19
<Ms2ger>
jgraham, I did what now?
15:19
<jgraham>
meter::-webkit-meter-horizontal-bar and more gunk
15:19
<Ms2ger>
Oh, that
15:19
<Ms2ger>
I was young :)
15:20
<annevk>
Ms2ger: that phrase is reserved for us
15:35
MikeSmith
hums "whatever happened to the teenage dream?"
15:56
<gsnedders>
annevk: Bah, you're old!
16:57
<paxcoder>
keygen tag ooooh yeaaah
17:09
<Hixie>
hsivonen: re gb18030, the comment in the spec for that line says "Windows Vista, Chrome, and Firefox agreed", so it was part of the research you had me do with the big table from vista and chrome and firefox data
17:09
<Hixie>
hsivonen: oh, nevermind
17:10
<Hixie>
hsivonen: there's a source comment at the top of that section that says "[assumption:] Windows-936 is a basically a subset of GBK which is basically a subset of GB18030 (supported by wikipedia)"
17:52
<paxcoder>
can someone explain what the challenge (attribute) is in the keygen tag?
18:00
<Hixie>
paxcoder: the sum total of my understanding of keygen is in the spec :-(
18:00
<Hixie>
paxcoder: i recommend looking up SignedPublicKeyAndChallenge in [X690]
18:02
<Ms2ger>
paxcoder, you realize that this is something netscape came up with in the nineties? :)
18:03
<paxcoder>
Hixie, thanks... i'm still clueless though :-P
18:03
<paxcoder>
Ms2ger, i don't, but it rocks
18:04
<paxcoder>
though i guess it doesn't prevent MITM so it doesn't rock that much.
18:04
<Hixie>
paxcoder: what's the underlying problem you're trying to solve?
18:05
<paxcoder>
Hixie, nothing, i'm just going over HTML5 tags I don't know, and I'm wondering what a challenge is. But my hope is to get rid of CAs
18:05
<paxcoder>
(though I'm not sure if that's realistic)
18:05
<Hixie>
aah
18:06
<Hixie>
well if you manage to find a way to get rid of CAs, lots of people would be very happy (and lots very unhappy, but that's another story)
18:06
<paxcoder>
who'd be unhappy?
18:06
<Hixie>
the CAs, mainly
18:06
<paxcoder>
doesn't diffie-hellman get rid of them though?
18:07
<Hixie>
still need some way to prove identity
18:07
<Hixie>
and the CAs are the only way we have to do that so far
18:07
<paxcoder>
darnit
18:11
<Hixie>
https://www.w3.org/Bugs/Public/show_bug.cgi?id=24094 wtf
18:11
<Hixie>
someone went to the multipage spec's elements.html page
18:11
<Hixie>
then went to the namespaces.html page
18:11
<Hixie>
then clicked on the svg namespace
18:12
<Hixie>
then submitted the comment "i like may you install for the video html"
18:12
<Hixie>
WHAT ARE YOU DOING
18:13
<Hixie>
it's an African IP. So on the one hand, yay, we're getting feedback from Africa, I keep saying that using e-mail and Web is the way to do that and that f2fs and telecons disenfranchise many parts of the world including probably much of africa.
18:13
<Hixie>
but on the other hand, what did they mean???
18:13
<TabAtkins>
That doesn't even English, so just close as invalid.
18:14
<TabAtkins>
Like, there's no way to interpret it into a meaningful sentence.
18:14
<Hixie>
i usually close those as NEEDSINFO
18:14
<Hixie>
but yeah
18:14
<Hixie>
(i reserve INVALID for things like actual obvious spam)
18:47
<TabAtkins>
Augh, good lord, talking on the SVG ml is almost impossible at times, with all the top-quoting and HTML-emailling. >_<
18:48
<Ms2ger>
At least it's not SVG-emailing
18:53
<TabAtkins>
Thank jeebus for that.
19:25
<jamesr__>
http://dev.w3.org/fxtf/compositing-1/#canvascompositingandblending - is that intended to add new values to 2d canvas?
19:26
<Ms2ger>
There's been some discussion on www-style
19:26
<Ms2ger>
Mostly coming down to "Are you high?"
19:28
<jamesr__>
yes, i've been posting a lot of that. i'm trying to figure out if this is different from the previous state or not
19:36
<Hixie>
jamesr__: i hope to defer to that spec eventually,
19:37
<Hixie>
jamesr__: but hopefully not until it's sane
19:41
<jamesr__>
it's not
19:42
<jamesr__>
most of the definitions are in sections marked non-normative
19:42
<jamesr__>
including all of the values referenced in the globalCompositeOperation section
19:44
<jamesr__>
even the css blend mode definitions refer to keywords defined in sections marked non-normative
19:45
jamesr__
is really curious how this thing made it almost to CR
19:45
<Ms2ger>
Nobody's watching?
19:46
<jamesr__>
s'pose
19:49
<Hixie>
CR doesn't mean much...
20:02
<paxcoder>
Hixie, I think the only thing we can say when faced with the problem of identity and central authorities is: WOT?
20:02
<paxcoder>
but heck, I'm skeptical about the Web of trust too.
20:04
<paxcoder>
Degrees of separation and chinese whispers.
20:04
<paxcoder>
Actually, first thing only. The whispers are the work of Mallroy.
20:05
<paxcoder>
i'm rambling, bye :-)
20:16
<TabAtkins>
Sigh, I knew last time I reviewed Compositing that it was unreadable. I need to rewrite it.
20:48
<zcorpan>
what does <input type=file> do if you delete the selected file?
21:19
<Hixie>
zcorpan: ideally, once you've selected it, the browser makes a copy of it, so that you can't delete it
21:20
<Hixie>
(a hardlink, obviously, not a real copy, that would be way too inefficient for big files)
21:21
<zcorpan>
Hixie: where does that leave Blob.close()? :-)
21:21
<zewt>
well, if the user selects a file and then deletes it, they may well expect that it means the browser can't access it either
21:22
<zewt>
the website, rather
21:23
<zewt>
it also depends on what the behavior is when you select a file and then modify it: if you revoke access if the mtime changes (eg. selecting a file means "you have access to the file in this state, but not future states") then that works, but if it does mean to give access to changes too, then a hardlink will sometimes break that
21:23
<zewt>
(when programs save over files by creating a new file then renaming over the old one)
21:48
<Ms2ger>
http://office.microsoft.com/en-us/outlook-help/print-emails-showing-names-on-the-bcc-line-HA102919777.aspx
21:57
<dekiss>
guys
21:57
<dekiss>
do you support webgl ?
22:06
<Hixie>
zcorpan: dunno
22:06
<Hixie>
zcorpan: ping https://www.w3.org/Bugs/Public/show_bug.cgi?id=24089 btw
22:13
<zcorpan>
Hixie: see https://www.w3.org/Bugs/Public/show_bug.cgi?id=23968
22:14
<Hixie>
zcorpan: what CSS does is somewhat orthogonal, since it would override HTML on this
22:14
<Hixie>
the text in HTML is only really relevant in the hypothetical case of a language that doesn't specify what to do, yes supports <style>
22:14
<Hixie>
the only case i can think of is JSSS
22:14
<Hixie>
and nobody supports that anyway nowadays
22:15
<zcorpan>
Hixie: oh. well in that case i don't care
22:15
<zcorpan>
Hixie: i care about css :-)
22:15
<Hixie>
k :-)
22:15
<Hixie>
me too, i was a bit surprised by your bug :-)
22:16
<Hixie>
anyway, that css bug says it should likely not use utf-8 either
22:16
<Hixie>
(your last comment)
22:16
<Hixie>
the html spec as written says to use the doc encoding
22:21
<zcorpan>
yeah
22:30
<fischman>
good afternoon; is there policy about attributes vs. events? E.g. media elements have a durationchange event, but they do not have an ondurationchange attribute, which MediaController elements _do_ have. Is it generally expected that on each element that has an event foo there will be a corresponding onfoo attribute, or is it a case-by-case basis, according to what's in the spec? (concretely: I'm wondering whether my blink CL adding the
22:32
<zcorpan>
fischman: onfoo attributes in html are generally available on *all* html elements *and* on the document *and* on window
22:32
<zcorpan>
fischman: http://www.whatwg.org/specs/web-apps/current-work/multipage/webappapis.html#event-handlers-on-elements,-document-objects,-and-window-objects
22:34
<Hixie>
woah, onresize isn't on elements
22:34
<Hixie>
let me fix that
22:34
<zcorpan>
fischman: MediaController isn't an element
22:35
<dekiss>
is the ewebgl part of html5?
22:35
<dekiss>
webgl*
22:35
<Hixie>
dekiss: depends what you mean by that exactly
22:35
<dekiss>
if no why ou don't make it guys
22:35
<dekiss>
3d canvas
22:35
<dekiss>
webgl
22:36
<dekiss>
Vendors may define experimental contexts using the syntax vendorname-context, for example, moz-3d.
22:36
<dekiss>
† For example, the "webgl" value in the case of a user agent having exhausted the graphics hardware's abilities and having no software fallback implementation.
22:36
<dekiss>
ahh I see it is
22:37
<Hixie>
i have no idea what you are asking
22:37
<Hixie>
i actually have less of an idea now than i did when you first asked your question :-)
22:38
<fischman>
zcorpan: thanks. Is there a backstory for why this is so? I.e. what does it _mean_ for a <form> (say) element to have a non-null onseeking attribute set?
22:38
<Hixie>
fischman: just means it has an event handler
22:38
<fischman>
Hixie: that can never fire?
22:38
<Hixie>
it can fire whenever such an event is dispatched at the element in question
22:38
<Hixie>
or, for bubbling events, any of its descendants
22:39
<Hixie>
it's just like addEventListener(), more or less
22:39
<fischman>
Hixie: ah, bubbling is the thing I was forgetting.
22:39
<Hixie>
you can also dispatch events manually
22:39
<Hixie>
dispatchEvent()
22:39
<fischman>
Hixie: thanks for the clarifications (and for the event! :))
22:40
<zcorpan>
fischman: as for the why, i think it was because it's simpler to implement
22:41
<fischman>
zcorpan: yeah, makes sense in a world where events aren't tied to elements (as bubbling and generic dispatchEvent() are).
22:42
zcorpan
is low on battery. both in laptop and brain
22:42
<zcorpan>
nn
22:42
<smaug____>
events aren't tied to elements
22:45
<dekiss>
Hixie sorry
22:45
<dekiss>
is the webgl part of the html5 specifications ?
22:45
<dekiss>
or any official w3c or whatwg specification
22:46
<dekiss>
omg
22:46
<dekiss>
ignore this please :)))) <dekiss> † For example, the "webgl" value in the case of a user agent having exhausted the graphics hardware's abilities and having no software fallback implementation.
22:46
<dekiss>
this was wrong paste :)
22:46
<Hixie>
what do you mean by "part of"?
22:46
<dekiss>
well.......
22:46
<dekiss>
part of hhh
22:46
<dekiss>
is tehre anything about webgl I am seeing something but don't udnertand :S
22:46
<Hixie>
do you mean, "is the specification of the webgl api published in the same document as the specification of the HTMLCanvasElement API?"
22:46
<dekiss>
let me check again
22:47
<dekiss>
no no
22:47
<dekiss>
I see webgl is mentioned in webgl specs
22:47
<dekiss>
http://www.whatwg.org/specs/web-apps/current-work/multipage/the-canvas-element.html here exactly
22:48
<dekiss>
ah lol I got confused so yeah Vendors may define experimental contexts using the syntax vendorname-context, for example, moz-3d.
22:48
<dekiss>
† For example, the "webgl" value in the case of a user agent having exhausted the graphics hardware's abilities and having no software fallback implementation.
22:48
<dekiss>
this I don't understand
22:48
<dekiss>
this is part of the html5 - canvas specification
22:48
<Hixie>
that's part of the html specification
22:48
<Hixie>
(html5 is dead, long live html)
22:50
<dekiss>
:)
22:50
<dekiss>
ye
22:50
<dekiss>
ok haha Hixie ^^
22:50
<dekiss>
YOu made that descision in 2011
22:50
<dekiss>
that was part of my courses haha
22:52
<dekiss>
Hixie, like or not people, I am pretty sure webgl is the future of the web :)
22:54
<dekiss>
peopel liek it because looks awesome, developers liek it because it is awesome and it pays 150$.hour :)
22:54
<dekiss>
150$/hour
23:07
<zewt>
webgl is definitely not the "future of the web", it's a useful feature but just that
23:14
<Hixie>
zewt: it's in the future
23:14
<Hixie>
zewt: of the web
23:14
<Hixie>
zewt: as are many things.
23:16
<zewt>
in the future, sure; the future, not so much
23:18
<Hixie>
i don't know what it would mean for something to be "the future" of what basically underpins the world's information economy
23:18
<cabanier>
Hixie: :-)