00:00
<TabAtkins>
The styling leave something to be desired, but seems okay.
00:01
<Hixie>
yeah, too many lines
00:01
<TabAtkins>
Yeah, and I think the yellow is too much. Just bolding would probably be enough.
00:01
<TabAtkins>
Also: whoops, forgot to never ever open up the Inspector on the single-page spec.
00:02
<Hixie>
heh
00:02
<Hixie>
i make that mistake regularly
00:02
<Hixie>
in dev builds it's not too bad
00:02
<Hixie>
it just doesn't do anything useful until you've done it three or four times
00:02
<TabAtkins>
Yeah, plus my new machine helps out.
00:02
<TabAtkins>
Woo z620!
00:02
<Hixie>
i think i'll keep the yellow, for some of the other examples it's really necessary to have more than bold
00:03
<Hixie>
and it'd be weird to be inconsistent
00:03
<TabAtkins>
Could you slightly reduce the brightness, then? I don't like bright yellow against off-white backgrounds.
00:03
<Hixie>
yeah, could do that
00:03
<Hixie>
paste a hex code?
00:04
<TabAtkins>
Maybe #e6e600?
00:05
<TabAtkins>
hsl(60, 100%, 45%)
00:05
<Hixie>
reload?
00:05
<TabAtkins>
Ah, better!
00:05
<TabAtkins>
It's an ugly color, but it doesn't strain my eyes.
00:07
<Hixie>
hah
00:07
<Hixie>
yikes that is ugly
00:07
<Hixie>
i'll poke at it some more :-P
00:09
<Hixie>
how about http://www.whatwg.org/specs/web-apps/current-work/multipage/sections.html#usage-summary-0 ?
00:09
<Hixie>
too bright again?
00:09
<TabAtkins>
Yeah. It's just something about a light, bright color against gray or near-gray that causes eye-strain for me.
00:10
<TabAtkins>
I think it's because I can't clearly make out the boundary.
00:10
<Hixie>
oh i guess the examples _are_ gray, aren't they
00:10
<Hixie>
they're so light that i forget
00:10
<Hixie>
hmm
00:10
<TabAtkins>
Not a huge fan of it against white either, but I'm used to that.
00:12
<Hixie>
how about it now
00:19
<Hixie>
also, can we please make the right hand side of attribute selectors accept unquoted numbers
00:19
<TabAtkins>
Sigh, I've messed that up so often.
00:19
<Hixie>
glad it's not just me
00:20
<TabAtkins>
The colors are great, thanks.
00:20
<Hixie>
cool
00:20
<Hixie>
(every time, i spend minutes trying to work out what's wrong)
00:20
<Hixie>
reload for an attempt at cleaning up at least some of those lines
00:20
<Hixie>
wait
00:20
<Hixie>
not done yet
00:20
<TabAtkins>
Problem with accepting numbers, of course, is that it's not exact. <foo x=05> won't match foo[x=05]
00:20
Hixie
twiddles thumbs
00:20
<Hixie>
yeah, maybe if it's numbers = should do a numeric comparison
00:21
<Hixie>
i proposed <= >= < > and so on years ago
00:21
<Hixie>
[rowspan>5]
00:21
<TabAtkins>
Makes sense to me.
00:21
<Hixie>
ok reload now
00:21
<Hixie>
http://www.whatwg.org/specs/web-apps/current-work/#usage-summary-0
00:21
<TabAtkins>
Yeah, better.
00:21
<Hixie>
cool.
00:21
<Hixie>
ok
00:21
<Hixie>
well
00:22
<Hixie>
that was an entire afternoon spent on one editorial bug that asked for something slightly different than what i provided, so clearly my time was well spent here
00:22
<Hixie>
home!
00:22
<Hixie>
bbl
02:53
<Hixie>
cabanier: canvas has always used web idl, ever since, in fact since before, web idl existed...
02:53
<Hixie>
cabanier: (as has the rest of HTML, since it's just one spec really)
02:54
<cabanier>
hixie: I agree
02:54
<Hixie>
what i just says contradicts what you said in http://www.w3.org/mid/83F37C1A4497B54589EAEDC750D03A9473B4B551⊙ncac
02:55
<cabanier>
hixie: I just didn't see any bug on the recent changes
02:55
<cabanier>
hixie: oops
02:55
<Hixie>
:-)
02:55
<cabanier>
hixie: yeah. the spec always referred to webIDL
02:56
<cabanier>
hixie: I should have proofread that email
03:00
<cabanier>
hixie: If that is so, were there bugs logged on WebIDL or did you just fix the interface?
03:00
<Hixie>
how do you mean?
03:01
<cabanier>
the old interface didn't match the behavior so you fixed it so it does.
03:01
<cabanier>
was that in response to a bug report?
03:01
<cabanier>
It's not in the change log...
03:01
<Hixie>
either a bug report, an e-mail, me seeing it while going through the spec proofreading, or any number of other things
03:01
<cabanier>
:-)
03:01
<Hixie>
which revision in particular are you talking about?
03:02
<cabanier>
http://svn.whatwg.org/webapps@7129 340c8d12-0b0e-0410-8428-c7bf67bfef74
03:02
<Hixie>
oh ages ago
03:02
<cabanier>
June 11: [giow] (2) Update to take into account WebIDL's 'double' vs 'unrestricted double' change.
03:03
<Hixie>
probably just did that because i saw an e-mail go by about the changes to webidl
03:05
<cabanier>
OK. I sent out a correction.
03:05
<cabanier>
thanks for letting me know!
03:06
<Hixie>
no worries, just got confused :-)
03:14
SamB_MacG5
wonders why the CSSWG can't manage to style their ToC links like links
03:15
<Hixie>
i imagine they don't want to
03:16
<SamB_MacG5>
is it spite?
03:16
<Hixie>
unlikely
03:16
<Hixie>
probably just a different sense of aesthetics
03:17
SamB_MacG5
continues to believe it is done to confuse him
03:18
SamB_MacG5
sorta wishes there was a way to specify userstyle based on what stylesheets a document used ...
03:20
SamB_MacG5
would make one to override http://dev.w3.org/csswg/default.css
03:49
<Hixie>
jgraham: dude, you need to indent your test cases. :-P (e.g. http://software.hixie.ch/utilities/js/live-dom-viewer/?saved=1562 )
06:36
<zcorpan>
TabAtkins: apparently ed isn't on board, at least
06:51
<annevk>
oh lol, whether the ? or # is included matters for the terminology?
06:51
<hsivonen>
annevk: thanks. email sent to www-style about window.screen
07:12
<zcorpan>
annevk: why call it .isInvalid rather than simply .valid or .invalid (depending on which case is more likely to check for truthness in an if statement)?
07:13
<annevk>
But Klaasen, the Dutch national coordinator for counterterrorism and security: "Why can’t I trust people?"
07:13
<annevk>
hahahaha
07:13
<zcorpan>
(i'd think .valid since it seems more useful to do stuff with a valid url than with an invalid one)
07:13
<annevk>
-- http://arstechnica.com/tech-policy/2012/09/proposed-eu-plan-to-stop-terrorist-sites-even-more-ridiculous-than-thought/
07:14
<annevk>
mkay
07:14
<annevk>
looks like ValidityState has that too
07:14
<zcorpan>
ah yeah
07:15
<annevk>
zcorpan: you have the same nick on github?
07:15
<zcorpan>
yes
07:18
<annevk>
zewt: I think I'm going to leave relative URLs to libraries for now (in the manner you proposed on the list)
07:18
<annevk>
zewt: afaict there's no native code in browsers for those things at the moment
07:33
<annevk>
tantek: basically, I considered aligning the terminology with the API, but kept them different for now since a) names as scheme/query/fragment are quite widespread and understood (I think) and b) it makes it somewhat clearer there's an API on top of some internal concepts
07:34
<annevk>
tantek: this is in particular true for "scheme", which is the name exclusively used when defining new types of URLs
07:42
<zcorpan>
http://lists.w3.org/Archives/Public/public-webapps/2012JulSep/0835.html looks like an excellent example of how we want people to go about proposing a new feature
07:45
<zcorpan>
(asking people to come up with a solution and detailed processing model as in http://blog.whatwg.org/proposing-features has proven to be unhelpful for the forums since it makes people focus on a particular solution instead of focusing on use cases and requirements)
07:50
<annevk>
write an updated blog post ;)
07:50
<annevk>
and maybe update the FAQ
07:54
<zcorpan>
i edited http://forums.whatwg.org/bb3/viewtopic.php?f=3&t=41&p=159
07:54
<zcorpan>
but maybe the faq should say that directly and the forums post just point to the faq
07:55
<zcorpan>
for the faq, the step discussing with implementors i think puts people off and isn't really a requirement
07:55
<zcorpan>
i mean, not as a first step
07:56
<zcorpan>
i think step one should be "forget about your solution" :-P
07:58
<annevk>
yeah, I think you're right
07:58
<annevk>
the people typically directed to these things just need to know it's about "use cases"
08:51
<zcorpan>
i edited http://wiki.whatwg.org/wiki/FAQ#Is_there_a_process_for_adding_new_features_to_a_specification.3F to what i think works better in practice
09:25
<zcorpan>
Hixie: with the frameset change, don't you need to apply display:block on the frameset element?
10:05
<jgraham>
http://www.cs.washington.edu/research/constraints/web/ccss-uwtr.pdf is mildly interesting is something of a historical curio at this point
10:05
<jgraham>
*if something
10:25
<zcorpan>
hmm, document.all is undefined in standards mode in webkit now?
11:37
<zcorpan>
will "why can't i trust people" be the next meme?
14:04
<zcorpan>
[tm]: you know the status of the publishing in html wg?
14:07
<SamB_MacG5>
zcorpan: how are you checking for document.all ?
14:08
<SamB_MacG5>
you know that it's specified to pretend to be undefined, right?
14:08
<jgraham>
zcorpan: I heard that [tm] is unavaliable
14:09
<jgraham>
Maybe ask plh
14:10
<zcorpan>
SamB_MacG5: yeah i know. i got an exception when using it and removing the doctype made it work, but now when i test again it works in standards mode. dunno what happened last time
14:11
<SamB_MacG5>
typo or something?
14:11
<zcorpan>
SamB_MacG5: the test i had was http://software.hixie.ch/utilities/js/live-dom-viewer/?saved=1801
14:11
<Ms2ger>
zcorpan, throws in Gecko, as expected
14:11
<zcorpan>
(it saying 'undefined' in the log means document.all is available; it saying an exception was thrown means it wasn't)
14:12
<zcorpan>
(opera returns the param)
14:12
<SamB_MacG5>
ah
14:12
<zcorpan>
maybe i looked at firefox and thought i was looking at chrome? dunno :-)
14:13
<SamB_MacG5>
yeah, that's throwing in TenFourFox for me as well
14:13
<zcorpan>
browsers look much the same these days
14:16
SamB_MacG5
is a *bit* surprised that Gecko conditionalizes document.all on standards mode
14:17
<Ms2ger>
It's terrible, terrible code
14:17
<SamB_MacG5>
what does this "rendering mode: CSS1Compat" part mean?
14:18
<Ms2ger>
No quirks/limited quirks mode
14:18
<Ms2ger>
Also known as standards/almost standards
14:18
<zcorpan>
SamB_MacG5: http://dom.spec.whatwg.org/#dom-document-compatmode
14:19
<zcorpan>
Ms2ger: any change you're gonna make document.all available in standards mode? :-)
14:19
<Ms2ger>
No :)
14:19
SamB_MacG5
sees it in both Safari (WebKit 534.49.0) and TenFourFox 15
14:20
<Ms2ger>
zcorpan, we are planning to make doc.all emulate undefined rather than the mess we have now, though
14:20
SamB_MacG5
wonders how you're *supposed* to find out what version of WebKit Safari is using
14:20
<zcorpan>
Ms2ger: what does that mean?
14:21
<jgraham>
SamB_MacG5: Never, ever ask about how document.all works in Gecko. Trust me on this.
14:21
<SamB_MacG5>
okay
14:22
<Ms2ger>
jgraham, indeed so :)
14:22
<jgraham>
It's not so much "here be dragons" as "here be dragons guarding a minefield around a black hole"
14:24
<Ms2ger>
Try http://software.hixie.ch/utilities/js/live-dom-viewer/saved/1803 for example
14:26
<zcorpan>
jgraham: ok, i'll email plh
14:54
<jgraham>
Hixie: Browsers need to work out that I am editing mixed javascript/html and give me a keybinding that indents nicely, like tab in emacs (except that doesn't work so well in mixed documents)
15:55
<Ms2ger>
krijnh, still not updating
15:56
odinho
pats his ~/irclogs folder
16:05
<Ms2ger>
zewt++
16:48
<Hixie>
zcorpan_: thanks for updating the faq, that's way better
16:49
<zcorpan_>
Hixie: cool, thanks
16:49
<Hixie>
zcorpan_: i might change it to say that you can stop after sending the use cases to the list, since at that point if the use cases are good it'll get added regardless, and if they're bad it doesn't matter how good the solutions are
16:49
<Hixie>
zcorpan_: and some people get attached to their solutions and get upset when others are picked
16:50
<zcorpan_>
Hixie: yeah i considered the same thing
16:50
<Hixie>
see also the bottom section of http://wiki.whatwg.org/wiki/What_you_can_do
19:44
<Hixie>
abarth: do you know what https://bugs.webkit.org/show_bug.cgi?id=85778 was about and whether i resolved it?
19:44
<Hixie>
abarth: i don't see e-mail pending on the topic
20:05
<krijnh>
Ms2ger: sorry, missed a line :/ Should be fixed now and auto-update again
20:05
<abarth>
Hixie: we had a thread about it on whatwg
20:05
<Ms2ger>
Yay :)
20:05
<abarth>
Hixie: i think we decided to put the requirement in the CSP spec rather than the HTML spec
20:07
<abarth>
Hixie: May 7
20:07
<abarth>
Subject: <iframe srcdoc> and Content-Security-Policy
20:07
<abarth>
on june 22 you said "Fair enough. I think this belongs in the CSP spec, though."
20:28
SamB_MacG5
wonders wtf happened to the "media" tab in tenfourfox's "Page Info" window
20:28
SamB_MacG5
thinks maybe tenfourfox needs a restart or something ...
21:16
<Hixie>
abarth: k, so i don't have to do anything here?
21:16
<abarth>
correct
21:16
<Hixie>
excellent
22:54
<zewt>
why am I being postincremented
22:54
<zewt>
do I look like an lvalue
22:58
<TabAtkins>
...yes?
23:01
<zewt>
Hixie: you must know that nothing will stop programmers from offering solutions to problems; it's sort of in our nature :)
23:02
<Hixie>
and nothing will force me to pay attention to them :-P
23:02
<Hixie>
(at least not until the problem has been established)
23:52
<annevk>
what's the deal with using @w3c.org? my filters use @w3.org :/