04:11
<Velmont>
Thought I had booked a hotel now, but I left it mid-process I guess. Hmz. Room for 4 persons for one person, -- crazy room types at quality inn and suites mountain view
07:01
<nesta_>
good morning!
08:54
<matjas>
zcorpan: re: http://twitter.com/zcorpan/status/195834345972445186 did anyone guess the answer yet? I’m really curious
09:01
<zcorpan>
dunno, haven't checked twitter yet :)
09:01
<zcorpan>
seems not
09:02
<zcorpan>
i'll give you some more time to find out (or curse me). ;)
09:08
<matjas>
FFFFFFFFFFFFFFUUUUUUUUUUUUUUUUU
09:27
<Ms3ger>
zcorpan: el.style.filter, WebKit
09:29
<matjas>
Ms3ger: but that’s a string, so it doesn’t count
09:29
<matjas>
only objects that aren’t primitives
10:01
<zcorpan>
Ms3ger: that's what i was thinking of, yeah (though i haven't looked into how it works, indeed it looks like it's a string), and some others we have introduced in opera recently (attachEvent and its friends)
10:01
<Ms3ger>
Boo, Opera ;)
10:02
<zcorpan>
attachEvent seems to be a compat balance that's not so easy
10:05
<zcorpan>
i can't get the 'filter' property to do anything in webkit (though -webkit-filter seems to be supported)
10:05
<zcorpan>
what am i doing wrong?
11:06
<zcorpan>
http://software.hixie.ch/utilities/js/live-dom-viewer/saved/1501 - false in safari and chrome. style.filter quirk dropped?
13:38
<zcorpan>
annevk: apparently flash can go fullscreen without opt-in from the embedder... http://forums.whatwg.org/bb3/viewtopic.php?f=3&t=4563
13:39
<zcorpan>
i.e. in an iframe
13:55
<zcorpan>
"JSON is not versioned, and has no mechanism for revision. JSON cannot change." http://inkdroid.org/journal/2012/04/30/lessons-of-json/
13:56
<zewt>
nice nonsense conclusion
14:27
<annevk>
zcorpan: I think it is probably better to disallow that
14:27
<annevk>
zcorpan: we disallow popups too after all
14:29
<annevk>
"Some people complained that the X in AJAX stood for XML, and therefore JSON should not be used." this article is hilarious
14:34
<annevk>
zcorpan: hmm, we only disallow popups from sandboxed iframes I guess
14:34
<annevk>
zcorpan: I think the idea was to be stricter because it's potentially annoying, but having to embed youtube with all these attributes is annoying too
14:38
<zcorpan>
there's lots of legacy that embed videos that currently use flash
14:39
<zcorpan>
which means it's hard for video sites to switch to <video> if they can't use fullscreen for the legacy embedders
14:39
<zcorpan>
anyway, gotta go
14:58
<jzaefferer>
MikeSmith: vnu.jar is working great, we've integrated it via http://github.com/jzaefferer/grunt-html - one thing that could be useful: ignoring specific errors. Would that be feasible to add/expose?
14:59
<annevk>
why would you want to do that?
15:00
<MikeSmith>
jzaefferer: yeah, I had been expecting you guys might say there were particular errors you didn't care to be emitted dozens of times
15:01
<MikeSmith>
one answer is, update your HTML documents so that they conform to the spec :)
15:01
<MikeSmith>
jzaefferer: but to be clear, you mean errors and not just warnings, right?
15:02
<MikeSmith>
it would help me to know which particular errors are the ones you'd want to ignore
15:02
<jzaefferer>
MikeSmith: yeah, errors, though so far I can't say I really noticed a difference in the output
15:02
<MikeSmith>
OK
15:03
<MikeSmith>
well, there is no good way for us to make the configurable even if we wanted to
15:03
<jzaefferer>
For now we're just not validating files like this: https://github.com/jquery/jquery-ui/blob/master/demos/tabs/ajax/content1.html
15:03
<MikeSmith>
ah ywah
15:03
<MikeSmith>
*yeah
15:03
<MikeSmith>
scott_gonzalez or you mentioned that earlier
15:03
<MikeSmith>
t
15:04
<jzaefferer>
Elsewhere we've got an invalid <object> element, used for certain focus-related tests
15:04
<MikeSmith>
the fact that it's a document fragment, doesn't have a title element or document
15:04
<jzaefferer>
Fixing the markup breaks the test
15:04
<MikeSmith>
interesting
15:04
<jzaefferer>
That would be a better candidate to skip the error/warning, as the rest of the file is valid
15:04
<MikeSmith>
that case should be reported as a possible spec bug I guess
15:04
<jzaefferer>
Maybe, but could also be some mess on our side
15:05
<scott_gonzalez>
jzaefferer: I need to look into the <object> error.
15:05
<jzaefferer>
Just wanted to check if there's generally a way to configure reporting per error type
15:05
<scott_gonzalez>
It's possible the test is just bad. That's why I didn't really want anyone fixing the markup in that file :-P
15:05
<MikeSmith>
OK
15:06
<scott_gonzalez>
MikeSmith: Here's the specific line that we're getting an error for: https://github.com/jquery/jquery-ui/blob/master/tests/unit/core/core.html#L64
15:07
<MikeSmith>
well, as far as suppressing errors the only thing I can suggest is to run the vnu output through some simple post-processing filter
15:07
<MikeSmith>
like, grep -v
15:07
<scott_gonzalez>
The related tests are about whether the element is focusable/tabbable.
15:07
MikeSmith
takes a look
15:07
<MikeSmith>
codebase?
15:07
<scott_gonzalez>
Updating that causes the tests to fail in IE 6.
15:07
<MikeSmith>
ah
15:07
<scott_gonzalez>
But I just need to spend some time figuring out whether the updated markup was correct.
15:08
<scott_gonzalez>
There used to be a lot of invalid markup in that test fixture.
15:08
<scott_gonzalez>
Because we were testing lots of invalid cases to make sure we got the "correct" results in the tests.
15:08
<MikeSmith>
object stuff is messy
15:08
<MikeSmith>
ah
15:08
<scott_gonzalez>
I ended up just deleting all of the other invalid markup a while ago.
15:09
<scott_gonzalez>
Realizing that it didn't make sense to have tests around invalid markup since if anyone ever reported a bug about it, we would close the bug as invalid anyway.
15:09
<MikeSmith>
heh
15:09
<MikeSmith>
yeah
15:09
<scott_gonzalez>
Yeah, objects are definitely messy. I don't even remember how I ended up with the markup that we have. It's been a long time.
15:10
<scott_gonzalez>
I remember spending weeks on that code though, trying to figure out markup and tests that made sense for all browsers.
15:11
<MikeSmith>
OK, well, for that specific example, I think you could maybe make a case that codebase should be "obsolete bug conforming"
15:11
<MikeSmith>
in which case, vnu would generate a warning for it instead of an error
15:11
<MikeSmith>
but I think it would be an uphill battle to make that case
15:28
<annevk>
https://twitter.com/apple curious
15:29
<michel_v>
did they ever have a twitter account?
15:30
<michel_v>
it's not like they would have much to talk about. "Hi everybody! We're working on something great! But everyone's under NDA!"
15:30
<michel_v>
"Stay tuned!"
15:38
<JVoracek>
16:54
<dglazkov>
good morning, Whatwg!
16:55
<annevk>
morning dglazkov
16:55
<Hixie>
is there no way in JS to push an array of items onto an array?
16:55
<Hixie>
concat returns a new array
16:55
<annevk>
for once it's morning for me too :)
16:55
<Hixie>
push() seems to just take one item
16:56
<annevk>
a.push.apply(a, b) via stackoverflow
16:56
<annevk>
-- http://stackoverflow.com/questions/1374126/how-to-append-an-array-to-an-existing-javascript-array
16:58
<Ms2ger>
Morning
17:03
<annevk>
Ms2ger: no longer in Europe?
17:16
<dglazkov>
WAT! Ms2ger is coming to F2f!!
17:36
<gsnedders>
Does this mean we have to work out how to pronounce "Ms2ger"?
17:42
<Philip`>
"Em ess tooger" ('tooge' as in 'stooge'), surely?
17:48
<dglazkov>
"Mastuger!"
17:53
<hober>
"miss toogah"
17:56
<dglazkov>
oooh, miss. I miss-ed that.
18:06
<rafaelw_>
hsivonen: ping.
18:13
<annevk>
hober: on my way
18:13
<annevk>
hober: well sort of, taxi is on its way first
19:05
timeless
needs to go to airport to catch flight
19:27
<Ms2ger>
weinig++
21:16
<annevk>
notifications meetup outcome: http://lists.w3.org/Archives/Public/public-web-notification/2012Apr/0018.html
21:22
<roc>
it's sort of annoying that html5rocks.com is really chromerocks.com
21:25
<annevk>
HTML5 is passé anyway
21:30
<miketaylr>
roc: http://www.html5r0cks.com/ is the better version
21:35
<gsnedders>
roc: You mean using browser prefixes is unfair? Oh no!
21:36
<gsnedders>
Better not try and be compatible, that's undermining the platform!
21:36
<roc>
have they written about that? I didn't see it
21:37
<gsnedders>
Written about what?
21:38
<tantek>
greetings whatwg'rs, who's going to the Web Apps WG f2f and/or the HTML WG f2f this week? (I'll be at both)
21:38
<othermaciej>
what does <http://www.html5r0cks.com>; do that is Chrome-specific?
21:38
<othermaciej>
tantek: I'll be at HTML, probably not Web Apps
21:39
<annevk>
I'll be there, already in MV in fact
21:39
<tantek>
othermaciej - perhaps html5rocks has some -webkit- specific dependencies? kinda like apple.com/html5 ?
21:39
<othermaciej>
I don't even know what html5rocks is supposed to do
21:39
<tantek>
if it's not chromerocks.com then perhaps it's webkitrocks.com
21:41
<gsnedders>
I believe it uses -webkit-.
21:46
<annevk>
Hixie: any opinions on passing event handler attributes as members of an object initialization dictionary?
21:49
<annevk>
Hixie: see "Constructors" in http://dev.w3.org/2006/webapi/WebNotifications/publish/Notifications.html
21:50
<Hixie>
well as defined i don't think it works but in principle i guess it could work
21:51
<Hixie>
(i don't know what "must be assigned" means in this context)
21:52
<annevk>
yeah the language in that specification needs some work
21:53
<annevk>
I was mostly wondering because WebSocket could do something similar
21:53
<Hixie>
not much point with WebSocket
21:54
<Hixie>
the only things that need to be given to the constructor are things that the constructor needs to do its stuff
21:54
<Hixie>
everything else can just be set separately
21:54
<annevk>
so that is the case here too
21:54
<annevk>
so you do have an opinion :)
21:57
<Hixie>
i think it's just a design decision
21:57
<Hixie>
i mean you could go either way
21:57
<Hixie>
same as whether methods should return the object they were invoked on or not
21:57
<Hixie>
i think it's silly, but other people think it's a great api design
21:58
<annevk>
hmm yeah, I have resisted adding that so far
21:58
<annevk>
sicking keeps pushing for it
22:07
<annevk>
https://twitter.com/pornelski/status/197018755778351104 is brilliant
22:09
<miketaylr>
haha
22:10
<annevk>
although I think the guy he linked to might actually be serious, which is somewhat sad
22:25
<tantek>
Yeah, I think @Malarkey is being serious.
22:26
<tantek>
though I tend to wonder, is the "just implement -webkit- prefix support for EVERYTHING" crowd just a rebranding of the "Why doesn't Mozilla/Microsoft/Opera etc. 'just use WebKit' as their engine?" crowd.
22:27
<tantek>
WebKitist apologists as it were
22:27
<tantek>
(or insert web engine du jour/year/decade)
22:27
<tantek>
I remember when it was "why doesn't MS/Opera etc., 'just use Gecko' "
22:27
<othermaciej>
if anyone wants to use WebKit as their engine, we're happy to share :-)
22:28
<tantek>
the monoculture engineists have always been wrong from "what's good for the web" perspective. or they want to go back to the days of 90%+ Trident marketshare for some reason.
22:29
<tantek>
(maybe they're secretly Java-advocates trolling the web standards folks)
22:30
<annevk>
or Netscape 4 for that matter, which is why we have all that beautiful <table> parsing now
22:31
<cbright6062>
The thing that bothers me about this is that all it's doing is allowing the Webkit elitists to win.
22:34
<tantek>
win what?
22:35
<Hixie>
i like how that blog post says essentially "i have no problem with any individual member of the group, but i have a problem with the decisions the group made"
22:35
<Hixie>
as if groups of humans make decisions independent of the members of the group...
22:36
<othermaciej>
which blog post, Malarkey's?
22:36
<tantek>
Hixie, vice versa - sounds like he has a problem with the decisions of particular individuals in the group
22:36
<othermaciej>
it's well known that group decision-making can induce cognitive biases and result in decisions no individual group member would have made
22:36
<tantek>
= consensus
22:37
<Hixie>
tantek: he explicitly says that he has no trouble with individuals (See hit footnote 2)
22:37
<Hixie>
othermaciej: yes, that post
22:37
<tantek>
"decisions no individual group member would have made" is awfully similar to "least objectionable option" (AKA W3C's way of going for consensus)
22:37
<cbright6062>
tantek: Winning the right to call themselves the sole "proper" rendering engine. Which is only an opinion, not a fact.
22:37
<tantek>
Hixie, individuals other than the decision maker who made the decision.
22:37
<Hixie>
othermaciej: it's true that groups make decisions that no individual may make, but that doesn't get one out of responsibility for the decision
22:38
<othermaciej>
Is there really such a category of people as "WebKit elitists"?
22:38
<annevk>
well there was not much confusion at Opera with regards to what we had to do here
22:38
<Hixie>
othermaciej: (this gets back to one of hte biggest problems i have with standards work at the w3c)
22:38
<annevk>
it was pretty much universally hated though
22:38
<othermaciej>
I suspect people writing WebKit-only content are doing it out of lack of testing/support bandwidth, not to hep WebKit's secret agenda of winning
22:38
<tantek>
annevk - I don't buy it was "necessary" for all the border-radius props
22:38
<tantek>
that was a gross error IMHO
22:38
<cbright6062>
othermaciej: Well, I would say so, from some of the people I have talked to out there.
22:39
<tantek>
as in - prove it - show me the list of sites that are abhorrently broken due to dependency on -webkit- border-radius props.
22:39
<cbright6062>
I don't consider it a "secret agenda", just an unfortunate fact of uneducated authors which are driven by the web evangilists being overly in love with webkit.
22:39
<annevk>
abhorrently broken was not our criteria
22:40
<tantek>
"doesn't quite look as pretty" ?!?
22:40
<annevk>
as it turns out users and clients care about that
22:40
tantek
waits for Opera to mimic Webkit into looking the same on http://dowebsitesneedtolookexactlythesameineverybrowser.com/
22:40
<zewt>
does breakage only matter if it's a mushroom cloud? heh
22:41
<cbright6062>
Honestly, I think the Web Browser world in general has become somewhat of a mess.
22:41
<annevk>
has become?
22:42
<cbright6062>
Well, has re-become.
22:43
<cbright6062>
I personally thought the IE/Firefox era of the majority of the 2000s was fairly stable in the long run.
22:43
<cbright6062>
Now, things are all over the place.
22:43
<Hixie>
tantek: dude that page looks the exact same to me on every browser i tested (three engines so far)
22:45
<othermaciej>
is that the joke?
22:47
<cbright6062>
Honestly, I really hope someday to see Web Browsers finally reach a point where it can be based off of preference only, not having to evangilize that one is "better" than the other due to web rendering abilities.
22:47
<karlcow>
for things like box-shadow https://github.com/miketaylr/Snort/blob/master/snort.yml#L1263
22:47
<tantek>
no when launched that site looked different in every browser
22:47
<tantek>
a bit surprised that that has changed
22:47
<tantek>
engines will always have different rendering abilities, so long as the web is evolving
22:47
<tantek>
i
22:47
<tantek>
f th
22:47
<tantek>
e
22:47
<tantek>
22:47
<zewt>
i can't remember the last time any site looked notably different in any browser than any other browser; that hasn't been a factor in which browser I use in years
22:47
<othermaciej>
cbright6062: I think that will never happen
22:48
<cbright6062>
I know. It's sad.
22:48
<tantek>
if they all have the same rendering abilities, and that's stopped changing, then the web is dead
22:48
<zewt>
othermaciej: no, it happened long ago :)
22:49
<zewt>
i can't remember the last time I had a page not render reasonably in one browser and having to use another
22:49
<cbright6062>
I've never really had that issue, either.
22:49
<cbright6062>
Not since the Netscape days anyway.
22:49
<zewt>
seems like a non-issue for users, so it should be a useless thing to try to evangelize
22:50
<karlcow>
zewt: your tolerance to differences might be higher. Mine is quite high too.
22:50
<othermaciej>
unfortunately on mobile it's pretty common for sites to be broken in browsers with nontrivial share
22:50
<zewt>
well, mobile is its own mess
22:50
<karlcow>
I wish transition was not implemented at all ;) I like things which don't make me motion sick ☺
22:50
<cbright6062>
I will say what I think part of the problem with mobile is:
22:51
<cbright6062>
and part of why I think people have a webkit mindset.
22:51
<cbright6062>
if going by end users is any idea of what the world has become line:
22:51
<cbright6062>
like*
22:51
<cbright6062>
"Smart Phones have a little i at the start of it...or they're that droid thing or whatever it's called."
22:52
<bga_>
http://www.rfc-editor.org/rfc/rfc6585.txt
22:52
<cbright6062>
Yes, I have seen both end users and developers alike who either believe, or act like, iOS (and maybe Android) are the only Mobile devices in existance.
22:53
<karlcow>
cbright6062: silicon valley is sooo far from Africa and Asia ;)
22:53
<cbright6062>
karlcow: ?
22:55
<karlcow>
if some devs think that iOS/Android are the only mobile phones possible, I would say they do not travel enough. (Or more likely, I guess, they don't care about the business model for users without those)
22:55
<annevk>
they're pretty common in Asia these days too, dunno about Africa
22:55
<karlcow>
Though the last few days, I have been impressed by the number of "misconceptions" about CSS. ☺ which I guess is a kind of reality.
22:56
<karlcow>
annevk: look at the market share of Opera Mini in Africa
22:56
<zewt>
(why are you sending unicode smiley faces over IRC? heh)
22:56
<karlcow>
http://gs.statcounter.com/#mobile_browser-af-monthly-201103-201203
22:56
<zewt>
i had to paste into another app just to figure out what that was (looks like &copy; in this window)
22:57
<karlcow>
heh zewt. Why not, it is part of utf-8
22:57
<karlcow>
it reminds me 20 years ago with the controversy about accent such as é on Usenet ☺
22:57
<annevk>
part of Unicode
22:59
<karlcow>
I sent → arrows too, and ♥ hearts :)
23:00
<annevk>
💩
23:03
<karlcow>
annevk: that is my favorite!
23:10
<jwalden>
people are aware of <https://plus.google.com/109925364564856140495/posts>;, right?
23:11
<zewt>
heh, gnu grep is still hopelessly broken when it comes to character ranges and unicode
23:12
<karlcow>
jwalden: I guess I do not get the joke.
23:12
<annevk>
jwalden: yeah loving it
23:12
<zewt>
in en_US.UTF-8, [A-Z] matches "i" and "j", but not "a" (and it shouldn't match any of them) :|
23:12
<jwalden>
what's not to get? take Unicode characters, rename them in amusing ways :-)
23:13
<jwalden>
sort of a dig at Unicode for having so many characters for so much esoterica, but not all that seriously
23:13
<karlcow>
aaaah got it
23:14
<annevk>
still have to submit MAN WITH BAZOOKA
23:15
<hober>
ROTATED HEAVY MAN WITH BAZOOKA
23:15
karlcow
should submit 💩 as PILE OF POO
23:18
<annevk>
karlcow: but that's not fake
23:18
<hober>
U+1F4A9
23:18
<karlcow>
annevk: I think the irony has been lost :)
23:21
<karlcow>
23:23
<jwalden>
karlcow: https://plus.google.com/109925364564856140495/posts/WVCgkLVGUZe
23:24
<karlcow>
ah italian ice cream
23:24
<karlcow>
just need a cup and a composition character
23:42
<kennyluck>
The Web is alive as long as the amount of useful content is increasing. Driving standards (probably) helps but I don't think it's necessary.