08:52
<Molly>
hey guys
08:53
<MikeSmith>
Molly: hei
08:53
<Molly>
MikeSmith how goes it?
08:53
<MikeSmith>
swimmingly
08:54
<MikeSmith>
good to see that you finally decided to join the cabal
08:54
<othermaciej>
hi Molly
08:54
<Molly>
glad to hear it. Hahaha, which cabal would that be? (I've joined so many) ;)
08:54
<othermaciej>
hello MikeSmith
08:54
<MikeSmith>
othermaciej: greetings
08:57
Molly
just comes by from time to be annoying
09:43
<bzed>
jgraham_: this is just another status ping :) I want to update html5lib in debian within the next two weeks
09:45
<jgraham_>
bzed: http://james.html5.org/temp/html5lib-0.11.zip Have a go with that and see if it works. There seems to be one test faliure on some platforms which I think is not important enough to block the release on but I'm interested to know if you see it
09:47
<bzed>
jgraham_: will upload it to experimental and see what happens :)
11:20
<bzed>
jgraham_: which architectures should I run the tests on?
11:32
<hdh>
I'm on 386, and there's one test failing
11:33
Philip`
suggests upgrading to at least a Pentium
11:34
<hdh>
x86 :P
11:36
<Philip`>
hdh: Which test is it?
11:37
<hdh>
test_core_4, u'<span title=foo=bar>' != u'<span title="foo=bar">'
11:38
<Philip`>
Oh - that was broken in SVN when I last looked, but it worked in jgraham_'s previous release candidate so I assumed he'd fixed it
11:39
<Philip`>
(The test is incorrect, I believe)
11:39
<hdh>
wait, I'm using r1170, sorry
11:40
<Philip`>
Ah, right
11:41
<hdh>
the zip file is okay here
13:15
<zcorpan>
Hixie: should HTMLSourceElement be a public interface?
13:21
<annevk>
window.HTMLSourceElement should exist, yes
13:34
<zcorpan>
because? all interfaces defined in html5 should exist? is there a rule?
13:36
<hsivonen>
(hopefully they should exist only if the functionality of the element is implemented, too)
13:36
<annevk>
my personal rule is that it should be exposed if it identifies something unique. if it's just an additional interface implemented by some existing object it might be better if it's not (such as HTMLDocument which is just Document)
13:36
<annevk>
hsivonen, :p
13:37
annevk
thinks hsivonen is saying this because Opera has builds without <source> support, not as a general principle
13:47
<hsivonen>
as a general principle, I think it's helpful if interface sniffing means feature sniffing
14:57
<hsivonen>
dolphinling_: thanks. Name on map half-fixed now.
15:04
<annevk>
"But if you are maintaining well-formedness, and you feel like being pedantic, then serve application/xhtml+xml to Mozilla and text/html to IE. Nothing significant breaks."
15:04
<annevk>
-- http://cafe.elharo.com/web/refactoring-html/why-xhtml/#comment-236665
15:04
<annevk>
lolz
15:07
<annevk>
http://css.dzone.com/news/html5-alternative-text-and-aut :/
15:07
<Dashiiva>
What is this advantage of xhtml he speaks of, especially when served as text/html?
15:07
<annevk>
"I suspect the real issue is that some members of the HTML5 working group want authoring tools to conform to HTML5 so they can demonstrate how successful HTML5 is"
15:08
<annevk>
right...
15:10
<Philip`>
Many application/xhtml+xml sites seem to be blogs, so it's true that even if they totally break, nothing significant breaks ;-)
15:10
<Dashiiva>
s/demonstrate/improve/
15:11
<Dashiiva>
Ho. Even two stars is "okay" by that ranking. Not a lot of room for criticism :)
15:12
<hsivonen>
Dashiiva: I suppose the advantage is that you can parse application/xhtml+xml with stuff with JDK libraries without an additional jar
15:12
<hsivonen>
s/with stuff /
15:12
<hsivonen>
umm. I meant:
15:13
<hsivonen>
the advantage is that if you are trying to parse text/html with JDK Xerces, it doesn't throw if you happen to be lucky enough to have gotten a stream that is actually well-formed
15:13
<hsivonen>
(unlikely)
15:13
<Dashiiva>
Does Xerces throw if there's no xmlns?
15:14
<hsivonen>
no, IIRC
15:14
<Dashiiva>
So nice & pretty HTML could work just as well?
15:14
<hsivonen>
then the elements just aren't in a namespace an code that expects XHTML doesn't treat the stuff as XML
15:14
<hsivonen>
as XHTML
15:14
<hsivonen>
typo++
15:15
<hsivonen>
due to Namespaces, nice and pretty HTML with Xerces would be much worse than nice & pretty HTML with the V.nu parser or TagSoup
16:11
<Philip`>
"We are busy updating the store for you and will be back shortly." - is Apple really so incompetent at web development that they can't update their web site without locking out all their potential customers for half a day? Even if the whole site was static files being FTPed up over a dialup link, it shouldn't be as bad as it is
16:15
<smedero>
http://cafe.elharo.com/web/refactoring-html/why-xhtml/#comment-236724
16:16
<smedero>
"And I repeat: that this page is served as text/html does not mean it is not XHTML. It is XHTML because it is well-formed markup in the correct namespace. Except for one funky bit that’s pulled in by one of the ads, it’s even valid."
16:16
<smedero>
lol.
16:17
<hsivonen>
that kind of XHTML advocacy is at least 5 years late
16:17
<gsnedders>
hsivonen: More than five.
16:17
<smedero>
oh yeah, I agree.
16:17
<smedero>
I was looking to preach to the choir... I just couldn't get over that comment.
16:17
<smedero>
s/was/wasn't/
16:43
<zcorpan>
hmm, would this be feasable: when following a <blockquote cite> reference, the browser scrolls into view on the quoted text
16:43
<zcorpan>
perhaps replacing <ins> elements with wildcards
16:44
<zcorpan>
and ignoring <credit> elements
16:45
<zcorpan>
(s/<blockquote cite>/<blockquote><credit><a href>/
16:47
<zcorpan>
that would be nicer than having authors write #xpointer()s
17:37
zcorpan
files a meta bug to microsoft that they should fix everything in http://simon.html5.org/test/ie7b2-bugs/
17:40
<hoopy>
hello!
17:42
<annevk>
hi
19:31
<hoopy>
hello
20:32
<annevk>
http://lists.w3.org/Archives/Public/ietf-http-wg/2008AprJun/0517.html
20:42
<hsivonen>
I suppose I should try to get .iki.fi listed
21:09
<virtuelv>
hsivonen: http://my.opera.com/dragonfly/blog/2008/06/09/the-debug-menu-and-the-new-weekly <-- now links to validator.nu
21:10
<hsivonen>
virtuelv: cool.
21:10
<virtuelv>
links to = you can perform validation directly from the menu
21:10
<hsivonen>
virtuelv: does it obsolete http://bugzilla.validator.nu/show_bug.cgi?id=16 ?
21:12
<virtuelv>
hsivonen: imo, that bug has always been invalid
21:12
<virtuelv>
err, no, i'm talking out of my ass
21:12
<virtuelv>
but yes, I think so
21:12
<virtuelv>
install the latest weekly, install the menu and try
21:12
<hsivonen>
ok
21:12
<virtuelv>
it has an item for validating by url
21:13
<virtuelv>
so you can't validate by upload yet
22:17
<hendry>
virtuelv: wonder if we can fit in dragonfly under http://wiki.whatwg.org/wiki/IDE ?
22:31
<anne-olpc>
yay, works!
22:32
<anne-olpc>
seems that the system time is still in some US timezone :)
22:38
<jgraham_>
Some people get all the fun toys :)
22:39
jgraham_
wonders if anyone will step up to give Ruby html5lib some love
22:39
<jgraham_>
bzed: Any arch you have access to is good to test :)