00:02
<Hixie>
http://www.w3.org/mid/OF73CC06FC.50275ECF-ON852575DA.0072E82C-852575DA.00734691⊙lc
00:02
<Hixie>
given all the amazing APIs we're inventing these days, i wonder if noah has a particular reason to prefer the 2DContext one, or if he just doesn't know about the others
00:03
<Hixie>
i mean, WebSocket, File, and Geolocation would all, imho, rate as more important APIs than canvas'
00:05
<Hixie>
hsivonen: any comments on http://www.whatwg.org/specs/web-apps/current-work/#an-introduction-to-error-handling-in-the-parser ?
00:06
<Hixie>
Philip`: same to you ^
00:06
<Hixie>
or annevk5
00:06
<Hixie>
or anyone who's worked on the parser :-)
00:08
<Philip`>
Hixie: "errorneous"
00:10
<Philip`>
Hixie: The element names in the example trees should probably be lowercase, I'd imagine
00:11
<Hixie>
yeah, i guess
00:11
<Hixie>
(fixed the typo, thanks)
00:12
<Philip`>
Some of the text nodes in the example trees get thick green underlines
00:13
<Philip`>
"The interesting part is with the end tag token with the tag name "b" is parsed." - s/with/when/
00:13
<Philip`>
or 'where' or whatever
00:15
<Hixie>
ok reload, fixed all that
00:15
<Hixie>
(and thanks)
00:16
<Hixie>
bbiab, meeting
00:17
<Philip`>
I'm too lazy to check if all the details of the algorithms are correct, but I don't see anything blatantly wrong, and I think these sections seem useful as a way of giving people some intuition as to what the algorithms are trying to do
00:17
<Hixie>
cool
00:17
<Hixie>
i'll add some more then!
00:18
<Philip`>
But I could be wrong, and maybe they don't help at all for someone who's not already familiar with it
00:18
<Philip`>
I've been corrupted by years of blog posts and discussions and tests and implementations of this stuff, so I can't approach it with a fresh mind :-(
00:19
<Philip`>
(That's not to say I actually understand it, just that I'm vaguely familiar with the general concepts)
02:27
<annevk5>
Hixie, look ok, but also unfinished
03:20
<Hixie>
annevk5: yeah, wip
04:21
<Hixie>
why do i get garbage when i go to mpilgrim's site?
04:21
<Hixie>
how odd
11:13
<annevk5>
http://ejohn.org/blog/unimpressed-by-nodeiterator/ should we nuke it again (including from Acid3)...
11:16
<jgraham_>
annevk5: I guess acid3 is the only reason Mozilla implemented it. But it doesn't suck more than the other similar DOM apis, really
11:18
<jgraham_>
I'm not sure that his proposed replacement is really that valuble
11:21
<jgraham_>
Particularly I'm not sure how valuable it is to be able to select all the comment nodes or text nodes in the document and it is already rather easy to get all the element nodes
11:22
<annevk5>
I wasn't saying we should support the replacement
11:22
<annevk5>
just to nuke APIs that are not obviously helpful
23:56
<Midler>
will there be a target="_tab" in html5?
23:56
<Midler>
Forcing the browser to open a link in a tab
23:56
<Midler>
:P
23:59
<Dashiva>
No