00:58
tantek
actually has a couple of Swatch BEAT watches that show/use Swatch Internet Time.
00:59
<tantek>
unfortunately they're still stuck with awkward Gregorian dates. I want a watch that displays ordinal ISO dates - if anyone knows of one (especially one that displays it by default on the "home" screen - without having to push any buttons) let me know.
01:58
<zewt>
heh another one
01:58
<zewt>
i try to search for an order i made on amazon for a surge protector
01:58
<zewt>
No matches for "Title, Department, Recipient...surge".
01:58
<zewt>
@placeholder go go go
07:05
<annevk>
morning
07:17
<annevk>
so replaceContents/empty are like textContent/innerHTML for nodes?
07:17
<annevk>
hmm
07:28
<annevk>
seems like Hixie replied to a few emails last light :)
07:29
<Ms2ger>
The graph does seem to suggest that
07:30
<annevk>
my inbox too
07:43
<MikeSmith>
hsivonen: regarding the Maven absurdity, I admire your persistence
07:43
<MikeSmith>
I think I would have given up already
07:48
<MikeSmith>
annevk: góðan dag
07:48
<MikeSmith>
annevk: about "Fixing something bad by introducing a document-global switch", I'm wondering did you have any specific examples in mind
07:50
<annevk>
quirks mode is one
07:50
<MikeSmith>
OK
07:50
<annevk>
ECMAScript had something similar at some point
07:50
<annevk>
well proposal for it
07:50
<MikeSmith>
yeah, I remember that
07:50
<annevk>
twice actually, first for ES versioning, then for Unicode
07:50
<MikeSmith>
yeah
07:51
<MikeSmith>
would you consider the currently specced behavior for meta@name=generator to be a document-global switch?
07:55
<annevk>
that doesn't seem like an invasive thing
07:56
<annevk>
but it does seem similar, sure
07:58
<MikeSmith>
yeah, it's certainly less heinous than quirks mode switching on doctype
07:59
<MikeSmith>
but still seems like an anti-pattern that should be avoided
08:02
<MikeSmith>
Hixie: about http://lists.w3.org/Archives/Public/public-whatwg-archive/2011Nov/0086.html ("Drag-and-drop folders/files support with directory structure using DirectoryEntry")
08:02
<MikeSmith>
looks like Kinuko's planning to implement it
08:02
<MikeSmith>
http://lists.webkit.org/pipermail/webkit-dev/2012-June/020952.html
08:04
<MikeSmith>
annevk: have you had a chance to ping John Gregg about the Notifications spec?
08:04
<MikeSmith>
if there's something I can do to help with it lemme know
08:05
<annevk>
yeah pinged him several times
08:05
<MikeSmith>
OK
08:05
<annevk>
he doesn't seem to have the bandwidth
08:05
<annevk>
I've been looking at fixing it myself, but it's quite messy
08:05
<MikeSmith>
maybe we can get Velmont interested
08:06
<annevk>
nah I can fix it
08:06
<MikeSmith>
OK
08:06
<MikeSmith>
I will get us an extension on the charter
08:06
<MikeSmith>
next week
08:06
<annevk>
when does it expire?
08:06
<MikeSmith>
end of this month
08:07
<annevk>
k
08:07
<MikeSmith>
hey if anybody's seen Wilhelm we really need his attention for getting a FPWD for WebDriver started
08:12
<annevk>
hmm I forgot, is it fine if a queued task does a potentially long running synchronous task?
08:12
<annevk>
or should that be in a "asynchronous" section
09:09
<MikeSmith>
"MS share contracts"?
09:10
<MikeSmith>
https://twitter.com/rem/status/210297546134929408
09:11
<MikeSmith>
oh
09:11
<MikeSmith>
http://msdn.microsoft.com/en-us/library/windows/apps/hh464906.aspx I guess
09:16
<tantek>
MikeSmith - the real question is - what is the concept called in BlackBerryOS - which had such device/app messaging abstractions before Android did.
09:19
<MikeSmith>
tantek: we seem to be on the road here that leads to Web developers needing to write code for several different browser-proprietary ways for addressing the same use cases
09:20
<MikeSmith>
I guess it's possible some library could provide an abstraction on top of it all
09:21
<MikeSmith>
but it would be a large missed opportunity if we can't get agreement on some standard way that Web devs can use to do this
09:21
<MikeSmith>
cross-browser
09:27
<jgraham>
annevk: Long running synchronous things would block the event loop
09:28
<jgraham>
(even if they happen in a task that is async with respect to the original DOM method invocation)
09:28
<jgraham>
(or whtever else causes the task to be fired)
09:29
<jgraham>
MikeSmith: What use cases are you talking about?
09:30
<MikeSmith>
easily sharing things through whatever particular Web apps you'd like
09:31
<MikeSmith>
or sending things to particular Web apps you'd like to do something with them
09:31
<MikeSmith>
without needing to navigate a massive sheet of buttons for every known service in the universe
09:32
<MikeSmith>
and without being restricted to some set up apps somebody else has decided to restrict you to
09:32
<annevk>
jgraham: yeah that makes sense
09:32
<annevk>
jgraham: so it needs to be an asynchronous section
09:57
<hsivonen>
Does Maxthon ship a given snapshot of WebKit-Chromium earlier than Chrome?
09:58
<hsivonen>
or: Why does Maxthon score higher than Chrome but lower than upcoming Chromium on html5test.com?
09:58
Ms2ger
thought it used trident
09:59
<hsivonen>
Ms2ger: I believe it uses Trident and WebKit-Chromium and has some sort of heuristic for deciding which one to use for which site
09:59
<Ms2ger>
Fascinating
09:59
<hsivonen>
or maybe it's based on a curated list of sites rather than a heuristic
10:04
<AryehGregor>
So how does that work with, say, respecting user preferences cross-site?
10:07
<MikeSmith>
html5test.com would seem to be a lot more useful if it documented the criteria and provided the actual tests it uses
10:12
<Ms2ger>
And if it had useful tests
10:12
<MikeSmith>
yeah
10:13
<MikeSmith>
instead of arbitrary tests created by one single dude based on personal whims
10:15
<MikeSmith>
god the fascination that grown-up men in Japan have with the pseudo-drama around AKB48 members is really nauseating. This TV commercial keeps showing a clip of one the "graduated" ones crying and saying "You can hate me but please don't hate AKB48". And it's for real and I guess supposed to be emotionally moving but it just makes me want to really pummel whoever the intended audience is
10:16
<MikeSmith>
and for anybody who doesn't know what AKB48 is, count yourself lucky. seriously
10:21
Ms2ger
doesn't look it up, then
10:52
<roc>
I looked it up, but Google didn't work
10:57
<hsivonen>
What SSH client is one supposed to use on Windows these days? Putty still?
10:57
<Ms2ger>
I think so
10:57
<hsivonen>
ok. thanks
10:58
<roc>
I just use msys ssh
11:01
<hsivonen>
So msys ssh is plain OpenSSH running under cmd?
11:01
<roc>
it looks like it
11:01
<hsivonen>
does cmd support the sort of curses irssi and screen use?
11:04
<roc>
oh, I'm running under mintty.exe actually
11:06
<hsivonen>
yay. getting crypto software from a http: site
11:10
<hsivonen>
yay. 16-color icons.
11:15
<tantek>
MikeSmith - re: developers needing to write code for several different browser-proprietary ways for addressing the same use cases - there hasn't been enough experimentation to come up with even a good minimal solution - that's part of the problem.
11:16
<tantek>
As I noted in the WebApps WG f2f in MSFT SVC - Google's Web Intents seem to have an even bigger (apparently ever growing?) scope that I had thought was already far bigger than use-cases have shown a need for.
11:17
<MikeSmith>
tantek: but in the mean time as with other cases, there's a vacuum the OS+browser vendors are going to fill one way or another
11:17
<tantek>
As far as fearing developers writing code for such technologies - most of the approaches to date are overdesigned and don't take into account developer (and publisher!) motivations.
11:17
<tantek>
MikeSmith, historically browsers have experimented with numerous things like this - most of which get ignored by developers
11:18
<tantek>
and some of which is eventually abandoned by browsers as well
11:18
<tantek>
I've been quite surprised at all the experiments that IE/Windows has abandoned for example, despite having a large marketshare etc., and potentially numerous developers using such experiments
11:19
<tantek>
(all the stuff they recently - a few months ago? - noted in a blog post were being dropped)
11:19
<annevk>
hsivonen: Windows -> cygwin?
11:21
<MikeSmith>
tantek: I think Webkit experimental features that have mobile use cases as a priority target perhaps have a lot less potential to get dropped
11:21
<tantek>
MikeSmith - except those Webkit features were never shipped as "experimental" - from the day Apple announced and demo'd them - they were intended to be a solid part of the web platform
11:22
<tantek>
plus no one shipped any alternatives, and they were put on a standards track (albeit far too slowly)
11:22
<MikeSmith>
fair enough
11:22
<tantek>
also - to be fair to Apple, the design of animations, transitions, transformations (just to name a few) was quite good
11:23
<tantek>
and I'm not sure the time cost of the revisions/improvements in the CSSWG have been worth it. others may have different opinions.
11:23
<hsivonen>
The UI of PuTTY is confusing. I'm going to try mintty later in order to get a familiar ssh binary.
11:24
<tantek>
nonetheless - there's some forward progress there with standardization, especially unprefixing, so I have some hope there.
11:25
<hsivonen>
annevk: cygwin just for running ssh seems like an overkill
11:31
<annevk>
tantek: there's some forward progress indeed, there's also tremendous damage which requires vendors to waste time on copying each others prefixes
11:31
<tantek>
annevk - or at least waste time studying and arguing if we have to or not
11:32
<tantek>
no doubt the progress is too slow, no matter how fast it is at this point.
11:48
<MikeSmith>
Ms2ger: how much longer before I can get today's Firefox Nightly
11:49
<Ms2ger>
What OS? The Linux one seems finished
11:54
<MikeSmith>
Ms2ger: OSX
11:55
<Ms2ger>
Looks like that finished right when you asked
11:56
<Ms2ger>
Not sure how long it takes to get to you from there, though
11:58
<AryehGregor>
tantek, my opinion is that the design of transforms was overall very good, especially 2D transforms, and not much was gained by revisions in the CSSWG (which anyway more or less didn't occur).
11:59
<AryehGregor>
I'm still a bit ambivalent about 3D transforms, namely the whole question of the transform-style/perspective/perspective-origin properties, but I wasn't able to come up with anything better.
11:59
<AryehGregor>
Gradients, however, wound up being much nicer after work by the CSSWG, AIUI.
12:00
<tantek>
AryehGregor - what you're saying sounds about right from my experience watching the revisions (or attempts) in the CSSWG
12:01
<tantek>
I suppose the challenge is not knowing when (if?) such attempts at iteration will make any improvements
12:02
<hsivonen>
Is it just a coincidence that the syntax changes at the CSS WG if TabAtkins edits? (Gradients, Flexbox)
12:08
<AryehGregor>
hsivonen, well, it's not a coincidence that the syntax doesn't change if no one edits (transforms, transitions, animations).
12:12
<roc>
heh
12:13
<roc>
gradients syntax had to change, because Webkit gradients really sucked
12:14
<roc>
it didn't have to change more than once though
12:40
<annevk>
does HTML have some hook for supported image format?
12:41
<annevk>
not really it seems
13:20
<tantek>
annevk - what do you mean by "hook for supported image format?"
13:20
<tantek>
there's object fallback
13:21
<tantek>
<object type="image/svg+xml" href="…"><object type="image/png" href="…">alt text</object></object>
13:21
<tantek>
btw annevk - I'm fine with using public-webapps list for Fullscreen. Any existing related list is better than creating yet another list.
13:59
<annevk>
tantek: the term "supported image format"
14:10
<AryehGregor>
annevk, I thought it made the list of supported image formats totally undefined?
14:22
<annevk>
sure, but that's not relevant
14:22
<annevk>
an image format is still either supported or not supported
14:31
<annevk>
MikeSmith: getting there with cleaning up notifications
14:32
<annevk>
MikeSmith: the whole permission stuff was not done yet and everything that was done was up for a rewrite of sorts...
14:32
<annevk>
still need to do bidi
14:32
<annevk>
bidi :(
14:42
<[tm]>
annevk: progress not perfection
14:42
<[tm]>
anyway, thanks
14:43
<annevk>
I doubt I'll get it perfect, but hopefully better :)
15:14
<TabAtkins>
hsivonen: I just really, *really* like changing the syntax of things late in development. It makes me happy.
15:14
<Ms2ger>
I KNEW it
18:01
<deadowl>
Concerning the time element, wouldn't it make sense if you could specify default timezone, and possibly year, in the head using a meta tag?
18:04
<Ms2ger>
jgraham, this timeout patch....
18:12
<tantek>
deadowl - info in meta tags tends to rot and get out of sync. it's fairly well accepted that putting data in meta tags is an anti-pattern.
18:13
<tantek>
especially something like timezone which changes twice a year for most folks
18:33
<deadowl>
It would be easier to use the timezone database and specify America/New_York in a meta tag or a header than to markup -04:00 and -05:00 for changes relative to GMT.
18:34
<deadowl>
Especially since people tend to publish for a single timezone.
19:08
<zcorpan>
annevk: "so it needs to be an asynchronous section" - the spec has "synchronous section" but not "asynchronous section". for stuff that should run async, it usually just says "return but continue the remaining steps async"
21:23
<matzipan>
boo
21:23
<matzipan>
oh not here
23:48
<zewt>
http://www.gotmow.com/Sign-Up-Online i wonder if I'm the first person to even question the fact that http + cc# entry
23:49
<TabAtkins>
Probably, yes.
23:49
<zewt>
the whole "only sometimes secure" model is just useless
23:49
<zewt>
(it's actually an https form, but of course it's still insecure, since you could mitm the form itself)
23:49
<zewt>
(but I'm not even going to try explaining that to someone who runs a lawn mowing business)
23:50
<Hixie>
aw, i thought "got mow" was a reference to cats
23:50
<Hixie>
now i'm sad
23:52
tantek
looks for a picture of a cat.
23:53
<tantek>
http://statigr.am/p/203021528894527247_2975244
23:54
<tantek>
and there's much more at http://statigr.am/tag/catsofinstagram/
23:55
<Hixie>
i am not lacking in pictures of cats :-P
23:59
<astearns>
http://mubi.com/films/rent-a-cat