00:00
<gsnedders>
Peter-: sent
00:00
<aho>
woo! :)
00:02
<aho>
now, if everything goes right this stuff will go straight into preprocessors like sass/less/cssmin/whatever
00:03
<gsnedders>
(for those curious, see www-arhcive)
00:03
<gsnedders>
*www-archive
00:04
<aho>
for the lazy ones: http://lists.w3.org/Archives/Public/www-archive/2011Mar/0007.html
00:04
<aho>
<:
00:05
<aho>
sorted by length... odd :)
00:05
<aho>
-xv-voice-pitch-range
00:05
<aho>
<- what's that?
00:06
<zewt>
tts?
00:06
<aho>
-o-, -wap-, and -xv- prefixes... hum :>
00:08
<gsnedders>
aho: http://dev.opera.com/articles/view/xhtml-voice-in-style/
00:08
<aho>
-apple-dashboard-region
00:08
<aho>
<- and that one
00:08
<aho>
gsnedders, ah :)
00:11
<Philip`>
(You can get that list of properties just by running "strings" on the opera binary and looking at the right section)
00:12
<AryehGregor>
I love doing that.
00:21
<aho>
i wonder if it's possible to pull that out of document.documentElement.style somehow
00:21
<AryehGregor>
Just try assigning every possible string as a style property and see which ones work.
00:21
<aho>
but it seems like those properties are hidden by non-iter magic :l
00:22
<aho>
i.e. "in" works, but for-in doesnt
00:23
<gsnedders>
aho: That's just the [[Configurable]] property of the property in ES
00:23
<gsnedders>
Configurable? Enumerable.
00:23
gsnedders
is blatantly too tired
00:23
<aho>
ye, but there is no way around that, is there?
00:24
<aho>
i mean... except for brute force
00:24
<AryehGregor>
The longest one is only 29 characters.
00:24
<aho>
gonna try them all! :v
00:24
<gsnedders>
aho: No, there isn't.
00:24
<AryehGregor>
That's counting the hyphens, so I guess for our purposes it's really 25 characters.
00:24
<aho>
i wonder how long that would take with IE6-9's JS engines :>
00:24
<AryehGregor>
There are what, only 52 possible characters then?
00:24
<AryehGregor>
So it's only 52^25, which is, um . . .
00:25
<aho>
>>> print 52**25
00:25
<aho>
7944811378381907919170379739856654861074432
00:26
<AryehGregor>
7944811378381907919170379739856654861074432
00:26
<aho>
7.9 paininthearseillions
00:26
<gsnedders>
aho: pff, you should be using some JS prompt ;P
00:26
<AryehGregor>
I did /exec -o calc 52^25
00:26
<aho>
had that one open .)
00:26
<AryehGregor>
But XChat complained that there was already a process running.
00:27
<AryehGregor>
So I had to do pstree, ps aux a couple of times, and some kills before XChat would let me run the command.
00:27
<AryehGregor>
So you beat me. :(
00:27
<aho>
> print(Math.pow(52,25));
00:27
<aho>
7.944811378381908e+42
00:27
<gsnedders>
Though admittedly the exact behaviour of Math.pow is undefined
00:27
<aho>
meep :)
00:28
<aho>
well, i think this would take pretty long... :>
00:30
<AryehGregor>
I don't think you'd need more than about a trillion universe-lifetimes, if you had a decent supercomputer.
00:30
<AryehGregor>
Hardly worth talking about if you're used to thermodynamics.
00:32
<aho>
hm
00:32
<aho>
well, there are rules to this
00:33
<aho>
one could use some kind of dictionary
00:33
<AryehGregor>
One could also run strings on the binary.
00:33
<aho>
that's cheating :v
00:52
<AryehGregor>
Does ISSUE-56 matter?
00:52
<AryehGregor>
(IRI normative references)
05:45
<Yuhong>
FYI, <XMP> was deprecated when HTML was made based on SGML.
05:46
<Yuhong>
The problem is that they didn't like the SGML CDATA feature.
05:50
<Yuhong>
But the funny thing is that then they later added the <SCRIPT> and <STYLE> tags, which had to be CDATA. (STYLE is the older one, BTW)
05:51
<Yuhong>
Now that HTML is no longer based on SGML, I wonder whether it should be undeprecated.
08:15
<annevk>
short night o_O
08:15
<annevk>
but home safe home
08:19
<foolip_>
annevk, did you see the stuff in http://code.google.com/p/html5/issues/list ?
08:20
<annevk>
oh hey cool
08:29
<annevk>
foolip_, so no, but I will patch
08:30
<foolip_>
ok
08:30
<foolip_>
I'm playing with my "log by section" thing, so I poked around a bit in your tool
08:31
<foolip_>
it's been twice now that there have been changes in the video section that I didn't notice until much later and I consider to be regressions, so it'd be good to keep track of sections
08:32
<annevk>
hmm
08:33
<annevk>
I guess I should leave my server details with someone
08:33
<annevk>
In case it gets hacked again and for updating html5.org related services
08:33
<foolip_>
it got hacked?
08:33
<annevk>
I suspect that might be tricky from an internet cafe in Bolivia
08:34
<foolip_>
in any case you shouldn't work during vacation
08:34
<annevk>
there were a couple of suspicious files
08:34
<annevk>
so far I have not managed avoiding reading a bunch of stuff while on vacation
08:34
<annevk>
including mailing lists :/
08:34
<annevk>
maybe this time it will be better, since it is so long
08:36
<foolip_>
well, if you trust me, I could take a hack at updating the icon and shorturl thing and get an account in the process
08:37
<annevk>
I do
08:38
<annevk>
the subversion stuff is somewhat disconnected from the actual code btw
08:38
<annevk>
the actual code lives somewhere on my server together with my own site and a dozen others
08:38
<foolip_>
is it in sync at least?
08:39
<annevk>
apart from the change I made the other day
08:39
<foolip_>
that it doesn't autoupdate doesn't surprise me, that'd be enterprisy
08:39
<annevk>
I can check that in and then give you the ssh password somehow
08:53
<foolip_>
gotta have breakfast, brb
09:10
<annevk>
Hixie, you online?
09:11
<annevk>
Hixie, can you add foolip to http://code.google.com/p/html5/ and maybe make someone co-owner or admin so we don't have to rely on just you?
09:48
<Peter->
gsnedders: got it, thanks!
09:59
<smaug____>
annevk: do you happen to know why <progress> is a form-associated element?
10:01
<annevk>
I remember it being changed to that at some point on request
10:01
annevk
looks in logs
10:03
<annevk>
hmm, at least over 1000 revisions ago
10:04
<annevk>
smaug____, it was for <label> I think: http://html5.org/tools/web-apps-tracker?from=4530&to=4531
10:05
<annevk>
which has http://www.w3.org/Bugs/Public/show_bug.cgi?id=8554 associated with it
10:05
smaug____
looks
10:06
<annevk>
yay for web-apps-tracker
10:06
<annevk>
answering questions for at least a few years now
10:07
<smaug____>
why does having label require form association ?
10:09
<smaug____>
or what useful does having progress.form bring in?
10:09
<annevk>
the way <label> is defined it requires a form-associated element at the moment
10:09
<annevk>
I am not sure really, maybe we should change it somewhat
10:10
<annevk>
maybe it is for form.elements to make sense
10:11
<smaug____>
if progress was resetable, this all would make some more sense
10:11
<annevk>
can you have labeled progress bar outside a form?
10:11
<smaug____>
why not?
10:12
<annevk>
just asking
10:12
<smaug____>
actually, should <progress> be even submittabled
10:12
<annevk>
don't think so
10:12
<smaug____>
submittable
10:12
<annevk>
can you file a bug?
10:12
<smaug____>
yeah
10:13
<annevk>
maybe hixie didn't consider making <label> work for non-form controls
10:18
<smaug____>
annevk: btw, progress isn't even "Listed element", so it doesn't show up in form.elements
10:19
<annevk>
oh ok
10:19
<annevk>
I remember not knowing why Hixie did it at the time but not caring enough to question it
10:19
<annevk>
hopefully I have not that too often, but I guess others pick it up in due course
10:20
<smaug____>
I was reviewing volkmar's patch for this and wondered why there is the .form
11:05
<foolip_>
annevk, so diffs are cached in the diffs/ folder to relieve the SVN server in web-apps-tracker ?
11:42
<Dashiva>
MikeSmith: Even if you use a fake URL, you're still giving away intimate knowledge of what the place is called :P
11:50
<volkmar>
smaug____: there is a .form for progress?
11:51
<smaug____>
volkmar: http://www.whatwg.org/specs/web-apps/current-work/multipage/the-button-element.html#the-progress-element
11:52
<smaug____>
volkmar: and in your patch you do implement .form
11:53
<volkmar>
eh, meter has .form too...
11:53
<smaug____>
volkmar: the w3c bug is about meter too
11:53
<volkmar>
smaug____: i guess i should try to think a little bit more when i implement something
11:53
<volkmar>
smaug____: what's the # ?
11:54
<smaug____>
http://www.whatwg.org/specs/web-apps/current-work/multipage/the-button-element.html#the-progress-element
11:54
<volkmar>
smaug____: NaN :)
11:54
<smaug____>
uups
11:54
<smaug____>
http://www.w3.org/Bugs/Public/show_bug.cgi?id=12254
11:55
<volkmar>
thanks
13:04
<annevk>
foolip, right
14:01
<foolip_>
updated gecko, opera and ie icons on http://html5.org/tools/web-apps-tracker, any objections?
14:02
<annevk>
wow much better
14:03
<annevk>
although not too great on the red I guess
14:04
<annevk>
volkmar, hope my reply helped
14:05
<volkmar>
annevk: yep
14:05
<foolip_>
annevk, difficult to do much about that, the icons are the colors they are
14:05
<foolip_>
well, one could add shadows or something, but I'm not gonna
14:06
<annevk>
yeah, I was thinking maybe forcing a background color or something
14:07
<annevk>
adding a box-shadow to img is funny though :)
14:07
<annevk>
but not sure it looks better
14:08
<foolip_>
if something can be done with just CSS, that'd be nice
14:08
<foolip_>
the ie icon is especially ugly on red
14:10
<foolip_>
annevk, see http://validator.nu/?doc=http%3A%2F%2Fhtml5.org%2Ftools%2Fweb-apps-tracker
14:10
<foolip_>
was it necessary to add value="" for some browser?
14:10
<foolip_>
just removing it still leaves an empty input box for me
14:11
<foolip_>
ah, I see it's because of the template
14:11
<foolip_>
maybe value="" should just be valid?
14:12
<annevk>
img { background:#fff; box-shadow:1px 1px 2px #fff } maybe?
14:12
<annevk>
value is not valid?
14:13
<annevk>
actually
14:13
<annevk>
img { background:#fff; padding:1px }
14:13
<annevk>
is pretty good
14:13
<annevk>
in Opera anyway
14:15
<foolip_>
isn't a box going to be very boxy?
14:15
<foolip_>
if one makes it blurry enough maybe it will look good
14:16
<foolip_>
filed http://www.w3.org/Bugs/Public/show_bug.cgi?id=12256 for the value="" thing
14:18
<annevk>
foolip_, it's boxy but decent looking :)
14:18
<foolip_>
annevk, are you editing it live now so I can see?
14:18
<annevk>
live in Opera :)
14:18
<annevk>
so no
14:19
<annevk>
I just did view source and save changes
14:20
<foolip_>
img { background:#fff; padding:1px } looks very white I'll say
14:21
<foolip_>
SVG filters would be better for this I think
14:21
<foolip_>
one wants a greyscale copy of the alpha channel and then to smear that a bit, maybe tainting it a bit white/yellow
14:23
<annevk>
well yeah, you could apply rocket science
14:23
<annevk>
i was just going for visible
14:23
<annevk>
:)
14:25
<annevk>
foolip_, where does it say valid="" is invalid?
14:26
<foolip_>
not sure if it does, just assuming validator.nu is correct
14:26
<annevk>
i cannot find it
14:27
<foolip_>
can you find anything that says that the value should be a floating point number?
14:28
<annevk>
ah good point, I guess I wasn't looking in the right section
14:29
<annevk>
"The value attribute, if specified, must have a value that is a valid floating point number."
14:29
<annevk>
not allowing the empty string there does indeed seem a bug
14:31
<foolip_>
annevk, should urlshortener.py be considered part of web-apps-tracker ?
14:44
<annevk>
I guess
14:44
<annevk>
structure is all kind of adhoc
19:35
<Hixie>
foolip_: what's your e-mail address?
19:35
<Hixie>
annevk: made more of you admins
19:36
<foolip_>
Hixie, for code.google.com purposes?
19:36
<foolip_>
if so, philip⊙fo
19:37
<Hixie>
yeah
19:37
<Hixie>
k, you're added
19:38
<Hixie>
it auto-expanded you to a gmail address!
19:38
<AryehGregor>
What code is this?
19:38
<Hixie>
weird
19:40
<foolip_>
oh, I have added philip⊙fo as an alternative to my google account
19:40
<foolip_>
too bad, I don't want anyone to ever use my gmail address :)
19:45
<foolip_>
Hixie, looks like it worked, thanks!
19:45
<foolip_>
AryehGregor, just some changes to http://html5.org/tools/web-apps-tracker
19:45
<AryehGregor>
Ah.