00:08
<Hixie>
man, i wish people would trim quotes
02:04
<zewt>
amused at hitting ^Z while editing a mail in gmail after dropping in an image, and it undoes the "finished loading" part (where it changes from a spinner to the actual image) ... going back to just showing the spinner permanently
06:14
<jamesr__>
is anyone tracking diffs between the w3c and whatwg dom parsing specs?
06:30
<Hixie>
yikes, that looks like it had a lot of differences
06:32
<Hixie>
wow, those are really different
06:38
<Hixie>
jamesr__: based on a cursory look, the changes seem to fall into two buckets: misguided editorial changes (e.g. the bits about exceptions), and misguided attempts to change the normative requirements (e.g. the changes to the XML serialization of the attributes are dubious at best)
06:38
<Hixie>
jamesr__: i recommend bitching at ms2ger to finish his work, and using the whatwg version :-)
10:37
<annevk>
Oh help. The attribute order bugs are popping up again in my inbox.
10:53
<annevk>
Hixie: you use <dfn title=syntax-elements>elements</dfn> for HTML to describe them
10:53
<annevk>
Hixie: seems to be the same as what I suggested
11:51
hsivonen
learns that such a thing as x-sun-unicode-india-0 has existed
11:52
<hsivonen>
(Devanagari character encoding from Sun)
12:06
<annevk>
Reminds me of windows-sami-2
12:08
<annevk>
http://developer.skolelinux.no/info/samisk/Tn35.pdf
13:26
<Ms2ger>
jamesr__, I recommend the same thing as Hixie :)
13:29
<jgraham>
Ms2ger: You should look at Travis's tests and reject any that are crazy with spec feedback to the effect that the requirements are bogus
13:30
<Ms2ger>
Yes, I should
13:31
<jgraham>
Ms2ger: Is that "SHOULD", "OUGHT TO", "MUST" or "MUST (BUT WE KNOW YOU WON'T)"?
13:32
<Ms2ger>
The phrase "OUGHT TO" conveys an optimistic assertion of an
13:32
<Ms2ger>
implementation behavior that is clearly morally right, and thus does
13:32
<Ms2ger>
not require substantiation.
13:32
<Ms2ger>
That sounds about right
14:20
<annevk>
So per my Twitter feed the a11y community is still debating the alt attribute… Given the number of features <audio> and <video> have, and given that the alt attribute debate is ongoing, will we see accessible video next millennium?
14:32
<hsivonen>
What's the deal with IETF JSON not allowing astral characters to be escaped?
14:33
<Ms2ger>
What's the deal with IETF JSON?
14:36
<annevk>
hsivonen: it does allow that: http://tools.ietf.org/html/draft-ietf-json-rfc4627bis-07#section-7
14:36
<annevk>
hsivonen: "To escape an extended character that is not in the Basic Multilingual…"
14:41
<hsivonen>
annevk: oh ok. somehow, I managed to miss that paragraph when reading the earlier one that seemed to scope escapes to the BMP
14:42
<hsivonen>
I wonder if anyone has tested UTF-16 and UTF-32 with all the implementations that are claimed as implementations of JSON
15:19
<SteveF>
annevk: not alt, but use of ARIA attributes in leiu of alt
15:26
<Domenic_>
Is this the latest CSSStyleDeclaration? http://www.w3.org/TR/DOM-Level-2-Style/css.html#CSS-CSSStyleDeclaration
15:26
<Ms2ger>
No
15:26
<Ms2ger>
It's in one of the CSSOM specs
15:26
Domenic_
shakes fist at MDN
15:27
<jgraham>
Hmm?
15:27
<Domenic_>
this looks more like it? http://dev.w3.org/csswg/cssom/#the-cssstyledeclaration-interface will update MDN
15:27
<annevk>
Domenic_: fix MDN
15:27
<jgraham>
MDN is sweet and fluffy and nice
15:27
<jgraham>
And when it isn't you should fix it
15:27
<Domenic_>
oh jeez persona fuck me
15:27
<Ms2ger>
No thanks
15:32
<Domenic_>
Hmm still can't figure out where properties like `opacity` live on CSSStyleDeclaration indices and what their property descriptors are.
15:35
<annevk>
There used to be a long list of properties in CSSOM
15:35
<annevk>
As an interface that was implemented on top of StyleDeclaration
15:35
<annevk>
Not sure where it went...
15:36
<annevk>
Oh, Domenic_, search for "attribute DOMString _camel-cased attribute;"
15:37
annevk
files https://www.w3.org/Bugs/Public/show_bug.cgi?id=23925
15:42
<Domenic_>
annevk: oh excellent, thanks.
15:51
<SimonSapin>
Can ignoring character decoding errors (rather than, say, emitting U+FFFD) cause security issues?
16:57
<Hixie>
annevk_: thinking about it more, i think elements are more similar to ports or host names. the better analogy would be documents vs Documents
16:57
<Hixie>
annevk_: that is, the text/html stream of bytes, vs the Document object and DOM tree
16:57
<Hixie>
annevk_: but i agree that the terminology around HTML isn't great
16:58
<annevk_>
I'm not sure I'd call the former a document
16:58
<annevk_>
and DOM defines document as Document object, fwiw
16:58
<Hixie>
annevk_: in URLs, though, there's never been a concept of a parsed URL object before, so nobody has ever used the term that way, so we should avoid making the problem
16:58
<Hixie>
imho, anyway
16:58
<Hixie>
your spec, you do as you wish :-)
16:58
<annevk>
actually, if you ask Roy that's what he says URL has always meant
16:59
<Hixie>
that's pretty much the strongest argument in my favour :-)
16:59
<annevk>
hah
17:31
<Domenic_>
So DOMTimeStamp is just Date? /cc annevk
17:31
<annevk>
Domenic_: no it's a number
17:31
Domenic_
goes and fixes MDN again
17:31
<annevk>
Domenic_: http://heycam.github.io/webidl/#common-DOMTimeStamp
17:32
<annevk>
oh wow, bad MDN
17:33
<Domenic_>
it would be cool if WebIDL had a single number type and then some annotations for clamping or similar.
17:33
<Domenic_>
also since it's unsigned long long it can't represent times before 1970
17:35
<Domenic_>
i wonder who uses DOMTimeStamp as a relative time
17:37
<annevk>
You could file a bug on IDL
17:37
<Domenic_>
might be worth starting the discussion.
17:38
<Domenic_>
trying to understand what WebIDL does with unsigned long long...
17:43
<Domenic_>
https://github.com/heycam/webidl/issues/2
17:50
<zcorpan>
MikeSmith: pls implement the new script restrictions :-)
20:47
<gsnedders>
Hixie: The distinction I was making was one can build a conformance checker that does check conformance and does execute (symbolically or not) certain scripts, and it seems silly to disallow that.
20:52
<Hixie>
gsnedders: who's disallowing it?
20:56
<gsnedders>
Hixie: Nobody. The point was there's no point to doing so, which seemed to be an implicit suggestion of the email I was replying to.
21:18
<Hixie>
gsnedders: ah
21:18
<Hixie>
gsnedders: ok
21:18
<Hixie>
in other news, some of the e-mails i get (especially those i get directly, not cc'ed to whatwg) put me in mind of chapters 5 or above of "The Book" in Anathem
21:25
<Hixie>
heycam|away: ping
21:28
<Ms2ger>
Good luck
21:30
<Hixie>
good luck with what?
21:30
<Hixie>
my book penance? :-)
21:31
<Hixie>
annevk: yt?
21:32
<gsnedders>
Hixie: I'd guess the fact pinging heycam never works. :P
21:32
<Hixie>
annevk: ah, nevermind
21:32
<Hixie>
gsnedders: works for me quite often, he comes online in my midafternoon :-)
21:34
<jamesr__>
is anyone closely tracking the w3c vs whatwg dom parsing changes?
21:35
<Ms2ger>
jamesr__, I'm planning to go through them over the weekend
21:35
<Ms2ger>
But bug me again then :)
21:35
<Hixie>
jamesr__: i had some responses for you last night after you asked but after you left :-)
21:36
<jamesr__>
Hixie: ah sorry
21:36
<jamesr__>
wait i asked questions last night?
21:36
jamesr__
was somewhat heavily medicated at different parts of yesterday
21:36
<Hixie>
heh
21:37
<Hixie>
jamesr__: http://krijnhoetmer.nl/irc-logs/whatwg/20131126#l-132
21:37
<jamesr__>
ah thanks
21:37
<jamesr__>
i just totally forgot that i already asked that
21:37
<jamesr__>
yeesh
21:38
<jamesr__>
i wonder if it's worth raising objects on the LCWD thread
21:38
<jamesr__>
probably not
21:56
<Hixie>
jamesr__: would be easier to argue if Ms2ger solved the issues properly on the whatwg spec first :-)
21:56
<Ms2ger>
You don't say
22:04
<Hixie>
Ms2ger: hey, i'm just glad i'm not the only one with a backlog
22:08
<Ms2ger>
heycam!
22:08
<heycam>
hi Ms2ger
22:08
<heycam>
Hixie, pong
22:09
<Ms2ger>
There's a PR waiting for you
22:09
<heycam>
Ms2ger, I saw that, and thanks. but I haven't looked at it closely yet.
22:24
<karlcow>
marcosc: YUM! "Firefox OS lacks a way to declaratively indicate to the Web browser that the application can run standalone when lunched." — http://w3c-webmob.github.io/installable-webapps/
22:32
<Hixie>
heycam: hello
22:32
<heycam>
Hixie, hi
22:32
<Hixie>
heycam: i have a number of bugs waiting on input for you. no rush, but was wondering if they were on your radar and/or how to make sure they were on your radar.
22:33
<Hixie>
from you, rather
22:33
<heycam>
Hixie, can you tell me which ones? and then I'll look at them soon.
22:34
<Hixie>
heycam: https://www.w3.org/Bugs/Public/show_bug.cgi?id=23744 https://www.w3.org/Bugs/Public/show_bug.cgi?id=23176 and https://www.w3.org/Bugs/Public/show_bug.cgi?id=23477 are the ones i have annotated with "heycam" in the status whiteboard
22:34
<Hixie>
there might be others but if so i haven't annotated then yet
22:34
<Hixie>
(i only annotated them the second time i get to them after having added a comment asking for feedback)
22:34
<heycam>
Hixie, ok thanks. sorry for not watching bugmail closely.
22:35
<Hixie>
heycam: oh, no worries at all, like i said there's no rush. just making sure they don't fall on the floor.
22:35
<Hixie>
even if you don't get to them for years, i have plenty to keep me busy :-P
22:35
<heycam>
heh ok
22:43
<marcosc>
karlcow?
22:44
<karlcow>
marcosc: read your sentence until the end
22:44
<karlcow>
Bon appêtit
22:44
<marcosc>
oh, lunched! :)
22:44
<karlcow>
;)
22:44
<marcosc>
mmmm... lunch
22:45
<marcosc>
karlcow: that document is full of typos ... wanna help fix 'em? :)
22:45
<karlcow>
heh
22:46
<karlcow>
You want more typos?
22:46
<marcosc>
if they are in a PR, sure! :D