00:04
<nattokirai>
heycam: ping
00:05
<heycam>
nattokirai, pong
01:02
<crocket>
Does mozilla developer network provide proper learning materials for HTML, CSS, and javscript?
01:02
<crocket>
w3schools is said to suck.
01:03
<zewt>
w3schools was the first site i blocked on google as soon as google grew that feature
01:03
<zewt>
can't speak for learning materials but mdn is usually okay as a reference
01:04
<MikeSmith>
MDN has some how-to docs as well
01:04
<MikeSmith>
gimme a minute
01:05
<MikeSmith>
MDN JavaScript Guide: https://developer.mozilla.org/en-US/docs/JavaScript/Guide
01:05
<crocket>
MDN seems to contain firefox-specific extensions.
01:05
<zewt>
(browsing mdn tends to not be so good, feels like a bunch of isolated pages connected by huge indexes, but fine for individual things reached through goog)
01:06
<MikeSmith>
crocket: yeah it has some things like that but generally they're clearly marked
01:07
<MikeSmith>
and it now includes a lot of information about things like implementation status in all major browsers
01:07
<MikeSmith>
not just Firefox
01:08
<MikeSmith>
the other thing is, for pretty much every bug that gets created for implementing a new standards feature, a "needs docs" flag gets added to the bug
01:08
<crocket>
Does <body> have border property?
01:09
<zewt>
quickest way to answer questions like that is to try it :)
01:10
<zewt>
(randomly makes me think back on something I hit a few months ago, trying to have a width: 100%; height: 100%; border: 1px whatever; body and hitting the "100% + 2px" problem; guess I should revisit that using box-sizing)
01:10
<MikeSmith>
crocket: what zewt said
01:11
<zewt>
(technically I said one relevant thing and one rambling, obscure thing; go with the first)
01:11
<MikeSmith>
open up Firebug or Web Inspector on any page, go in and add a "body { border: 5px solid green}" CSS rule
01:11
<MikeSmith>
and see what happens
01:12
<crocket>
meh
01:13
<MikeSmith>
anyway, the Mozilla docs team goes in and adds docs to MDN for all new standards feature every time any new feature is implemented in Firefox
01:13
<zewt>
if you're not willing to try things out for yourself, you'll have difficulty in any platform :)
01:13
<crocket>
zewt, I'll try
01:13
<MikeSmith>
MDN CSS guide: https://developer.mozilla.org/en-US/docs/CSS/Getting_Started
01:14
<MikeSmith>
or https://developer.mozilla.org/en-US/docs/CSS is better
01:14
<MikeSmith>
bunch of tutorial docs linked to there
01:22
<MikeSmith>
crocket: you can also use http://jigsaw.w3.org/css-validator/#validate_by_input to test whether a particular CSS is valid or not
01:23
<crocket>
MikeSmith, ok
02:00
<crocket>
How do I use "inspect element" feature in firefox to modify CSS?
02:00
<crocket>
I tried, but it doesn't work.
02:07
<crocket>
What tag do I use to specify properties of a part of a sentence?
02:07
<crocket>
<span.?
02:08
<crocket>
<span>?
02:25
<Oliver_>
Hello everyone!
02:39
<divya>
care to explain what "properties" mean crocket ?
02:39
<crocket>
divya, font-size
02:39
<crocket>
font-weight
02:40
<crocket>
such things
02:40
<divya>
ya span would do
02:40
<crocket>
divya, But span looks dirty.
02:40
<divya>
o?
02:40
<crocket>
Do I have a neater way?
02:40
<divya>
use <u>, <i> or <b>? but they all ahve default properties set
02:48
<crocket>
divya, how about <div>?
02:48
<divya>
wat no
02:48
<crocket>
divya, why not?
02:48
<divya>
because it is for a division
02:48
<divya>
a block level alteration
02:48
<divya>
not for inline content
02:48
<crocket>
divya, I don't know what it means yet.
02:48
<divya>
then checkout MDN
02:49
<crocket>
divya, What do <h1>, <h2>, ... mean?
02:49
<crocket>
I don't understand the meaning of numbers.
02:49
<divya>
go to MDN that MikeSmith linked to
02:49
<divya>
and read up on html page.
08:41
<annevk>
cpearce: I was indeed
08:43
<cpearce>
annevk: yeah, I was going to ask whether object-fit:contain on the :fullscreen style is supposed to only apply to <img> elements which are fullscreen?
08:45
<cpearce>
object-fit is only mentioned here: http://dev.w3.org/csswg/css3-images/#object-fit which implies it's image only, I just wanted to check,
08:46
<cpearce>
since we've had a few bugs filed against gecko recently regarding scrolling fullscreen content, and that's relevant here.
08:55
<annevk>
also for <video> and such
08:55
<annevk>
especially video
08:59
<odinho>
annevk: XHR testsuite: He updated it a bit, but yeah, nothing great. He wrote a few extra tests. And found a few bugs that he wrote in a document.
09:00
<Ms2ger>
Hey, a few is better than none
09:03
<annevk>
odinho: spec bugs?
09:03
<odinho>
annevk: testsuite bugs :P And browser ofc bugs.
09:03
<annevk>
ah good
09:47
<Ms2ger>
c = new(window.ActiveXObject || XMLHttpRequest)("Microsoft.XMLHTTP")
10:52
<annevk>
Ms2ger: so that fails in everything but IE?
10:53
<Ms2ger>
Nah, extra arguments are ignored
10:53
<Ms2ger>
Only we have a dictionary argument in that spot
10:53
<annevk>
right
10:53
<annevk>
and we wanted to maybe change the whole AnonXMLHttpRequest design to use that dictionary instead
10:54
<Ms2ger>
Yeah
10:54
<annevk>
I guess someone will keep me posted
10:54
<annevk>
is anyone else going to maintain XHR or should we WHATWG that too?
10:55
<Ms2ger>
I guess
10:56
<annevk>
that could apply to either option :p
10:57
<Ms2ger>
No idea who would maintain it at the W3C
10:57
<Ms2ger>
darobin?
10:58
<darobin>
no idea either
10:58
<darobin>
annevk: I presume you haven't heard back on your post?
10:58
<annevk>
nope
10:58
<darobin>
let me figure this one out
10:59
<annevk>
in 2006 or so timbl told rigo to make it happen, instead of being a lawyer and doing as told, rigo objected and nothing happened
10:59
<annevk>
it prolly then escalated at some point within the Team
11:00
<annevk>
and the Team being lame instead of making a decision they put it forward to the AC
11:00
<annevk>
which is where all good ideas go to die
11:00
<annevk>
400 people are not going to embrace change
11:01
<darobin>
that's not exactly what happened, but I can't say I disagree with the sentiment
11:31
<MikeSmith>
there were actually only 24 AC reps who responded to the survey about the document license
11:31
<MikeSmith>
out of 350-400 or so AC reps
11:31
<MikeSmith>
so it's hardly like there was even a resounding mandate on it from the AC
11:33
<MikeSmith>
all that the survey really indicated is that we had 300+ other AC reps who didn't feel strongly enough about it either way to take the time to respond with their opinions
11:34
<odinho>
Oh, that doesn't sound very good.
11:36
<MikeSmith>
hmm
11:37
<darobin>
more importantly, the question was IMHO phrased completely wrong
11:38
<MikeSmith>
I guess I shouldn't say anything more
11:38
<darobin>
it was put forth as a possible exemption granted only to the HTML WG, which is silly because there's no reason any group should have an exemption, and it's a political hot button to boot
11:39
<MikeSmith>
yeah
11:39
<darobin>
IMHO, and like Mike I don't want to speak out too much on this just yet, it should be up to individual groups to select which licence to use
11:39
<darobin>
groups do the work, they have the stakeholders, they should make the call
11:39
<MikeSmith>
100% agreed about that personally
11:40
<MikeSmith>
and within groups, editors do the hands-on editing work
11:40
<MikeSmith>
anyway I'm realizing now I can't remember how much of the details are still secret
11:41
<darobin>
the number of respondents probably was :-p
11:41
<MikeSmith>
yeah
11:41
<darobin>
the rest I already knew, so it probably wasn't
11:44
<MikeSmith>
anyway I'm not sure about that number
11:44
<MikeSmith>
I could be plain wrong and thinking about something else
11:45
<MikeSmith>
and I can't document it so doesn't matter anyway
11:45
<MikeSmith>
people shouldn't listen to what I say
11:46
<Ms2ger>
Yay W3C: https://lists.w3.org/Archives/Member/w3c-css-wg/2012JulSep/0264.html
11:47
<annevk>
I wanted to click on that, but then realized I no longer can :/
11:50
<Ms2ger>
You're not missing much
11:54
<jgraham>
Ms2ger: I think what you meant to say was "Oh well then you missed out on life-changing information"
11:54
<jgraham>
But with more commas
11:58
<Ms2ger>
Well, I can't tell him what you get when you dereference that eleph^WIRI, so we'll never know if it would have been life-changing
12:47
<annevk>
MikeSmith: you around still?
12:47
<MikeSmith>
yeah
12:47
<annevk>
MikeSmith: what are you going to cover next week?
12:47
<annevk>
MikeSmith: and are we going to prepare in one way or another together?
12:48
<MikeSmith>
in part I'm going to talk about recent HTML WG developments, so we're definitely not going to have overlap there
12:49
<MikeSmith>
but we should get together to compare notes
12:49
<MikeSmith>
I'm otherwise planning to talk about new features at a pretty high level
12:49
<darobin>
maybe it would be more entertaining if you let annevk cover the HTML WG instead
12:49
<MikeSmith>
and implementation status
12:50
<MikeSmith>
heh
12:50
<MikeSmith>
yeah that would definitely be entertaining
12:50
<MikeSmith>
I'm going to work from the list of specs at http://platform.html5.org/specs/a
12:50
<MikeSmith>
http://platform.html5.org/specs/
12:50
<MikeSmith>
and http://platform.html5.org/
12:51
<annevk>
okay
12:51
<MikeSmith>
and http://www.w3.org/Mobile/mobile-web-app-state/ too probably
12:52
<MikeSmith>
maybe also talk about http://ringmark.io/
12:52
<Ms2ger>
Hah
12:52
<annevk>
sounds good
12:52
<annevk>
maybe I'll do my code point / byte talk for something a little bit more low level
12:52
darobin
isn't sure he'd cover RingMark, it's not really finalised atm IMHO
12:53
<jgraham>
I wouldn't say anything nice about RingMark at least :)
12:56
<jgraham>
I guess I should qualify that with in-principle I have no problem with the idea of a graded set of targets for browsers to implement based on developer priorities
12:56
<jgraham>
But the reality of ringmark is highly non-ideal
13:01
<tantek>
indeed, last I checked the ringmark folks were including all kinds of "-webkit-" properties as part of their "platform/standard"
13:06
darobin
is now out of CoreMob, won't comment :)
14:38
<annevk>
MikeSmith: when are you in the Netherlands?
14:39
<annevk>
MikeSmith: middle of next week?
14:49
<MikeSmith>
annevk: arriving on the 13th at around 8pm
14:49
<annevk>
ah okay, so quite late :)
14:49
<MikeSmith>
yeah
15:31
<dglazkov>
good morning, Whatwg!
15:44
<Hixie>
lol, the only thing that has happened in the last 5 days on the w3c html wg specs is that they've removed the form for sending feedback and removed links to the whatwg
15:44
<Ms2ger>
They hadn't done that yet?
15:45
<Hixie>
seems not
15:46
<annevk>
I have the feeling that's about the same that happened to the CSSOM in the last, well, I stopped counting
15:48
<annevk>
the way the CSS WG uses version control these days makes it hard to view progress for a single spec
15:48
<annevk>
or maybe there is a way that I don't know
15:48
<Ms2ger>
I love how Bert always has four commits per change
15:49
<annevk>
maybe http://dvcs.w3.org/hg/csswg/log/tip/cssom/Overview.html
15:50
<Hixie>
TabAtkins: is http://lists.w3.org/Archives/Public/www-style/2012Jul/0014.html being tracked? it doesn't look like it got a formal reply
15:51
<annevk>
doesn't look like much happened
15:51
<annevk>
just some moving things around
16:00
<Ms2ger>
Adam Sobieski? I think I've seen that name before...
16:02
<Hixie>
does _anyone_ implement the dnd addElement() properly?
16:03
<Ms2ger>
I doubt it
16:03
<Ms2ger>
It's dnd
16:04
<TabAtkins>
Hixie: I don't remember seeing this! It looks like his use-cases are all addressed by :user-error now (the official name for :moz-ui-invalid).
16:09
<jwalden>
given the dnd spec is so wrong, it's not a surprise implementations are wrong</troll>
16:09
<jwalden>
:-)
16:10
<jgraham>
Hixie: Got a test?
16:21
<Hixie>
TabAtkins: k, thanks
16:21
<Hixie>
jwalden: it's not wrong :-P
16:21
<Hixie>
jgraham: live dom viewer download
16:21
<jwalden>
success troll!
16:22
<Hixie>
looks like the browsers that support addElement() just treat it like setDragImage(), and they allow any element to be passed to setDragImage()
16:23
<Hixie>
i wonder if a should just remove addElement(), or leave it as an alias to setDragImage()
16:24
<Hixie>
webkit doesn't have addElement()
16:25
<Hixie>
(which is weird because i could have sworn it webkit documentation that i got it from)
16:25
<Hixie>
firefox treats it as setDragImage() without coordinates
16:25
<Hixie>
opera doesn't do anything with it but seems to not fire an error message
16:28
<Hixie>
i'm just gonna drop it
17:24
<miketaylr>
are we going to get something like srcList (DOMTokenList) for srcset? slicing up strings by hand doesn't sound fun
17:30
<Hixie>
miketaylr: i've been asking people on whatwg@ for use cases for script manipulation
17:30
<Hixie>
miketaylr: so far nobody has described anything concrete
17:31
<Hixie>
miketaylr: when they do, i'll try to design an api that makes sense for it
17:31
<miketaylr>
fair enough
17:32
<miketaylr>
Hixie: fwiw, i don't have one either.
17:32
<Hixie>
yeah, i couldn't come up with one either, even a theoretical oe
17:32
<Hixie>
constructing it, sure
17:32
<Hixie>
but reading it?
17:32
<Hixie>
someone suggested it would be good to have a way to force a specific one to be selected
17:33
<Hixie>
but again, that seems easier to do by just constructing a srcset with only the value you want
17:33
<Hixie>
that to use an api
17:33
<miketaylr>
my use case: if any of the filenames have the string "dog" in them, change them to a cat picture.
17:33
<miketaylr>
nailed it.
17:33
<Hixie>
you can do that today without any more of an api that .srcset :-)
17:33
<miketaylr>
heh
17:33
<Hixie>
.srcset.replace() or whatever the right regexp is
17:36
<miketaylr>
guess we can wait for srcsetQuery to pave some cowpaths
17:41
<Hixie>
annevk, Ms2ger: so for seamless="" we need to fire events (e.g. mouse and key events, focus events) across the iframe boundary so that the outer page doesn't miss events
17:42
<Hixie>
annevk, Ms2ger: e.g. if the user has focus in a seamless iframe but presses an accesskey="" key or onkeypress="" key set up in the outer frame
17:42
<Hixie>
annevk, Ms2ger: or if the user has a dialog and drags it across the inner iframe, so that the page doesn't lose track of where the mouse is during the drag
17:43
<Hixie>
annevk, Ms2ger: this seems like it'd require pretty invasive changes to the event model, is it something you want to do instead of having me try to patch it in the html spec?
17:43
<Hixie>
annevk, Ms2ger: (i can provide a link to feedback on the topic if you're interested)
17:44
<annevk>
you can't just dispatch the events again?
17:45
<Hixie>
no, because you want the capture phase on the outer frame to happen before the inner frame
17:45
<Hixie>
and stopPropagation has to work across both events
17:45
<Hixie>
and they have to be different Event objects
17:45
<Hixie>
and mouse event coordinates need to be adjusted
17:45
<Ms2ger>
Can you just mess with the propagation chain?
17:45
<Hixie>
and it has to work at multiple levels
17:46
<Hixie>
(consider multiple nested iframes)
17:46
<Ms2ger>
Oh, hmm, adjusting coordinates sounds like fun
17:46
<annevk>
hmm, sounds like something for HTML to define as DOM does not know about Window
17:46
<Hixie>
and also this might end up being a very similar system as for web components
17:46
<annevk>
alternatively, you could merge all of DOM into HTML and maintain both ;)
17:46
<Hixie>
you guys keep taking stuff out, it's too late to give it back :-P
17:47
<Hixie>
is there some way you guys could spec all that but just provide a hook for me to say what the objects are that should get the events fired on and stuff? so you don't have to know about Window/iframe/seamless?
17:48
<Hixie>
that seems like it'd be the best solution
17:48
<annevk>
1) DOM does not have mouse events
17:48
<Hixie>
well, nothing has mouse events currently
17:49
<Hixie>
not properly specced, anyway
17:49
<annevk>
yeah :/
17:49
<annevk>
definitely not signing up for that at this point though
17:51
<Ms2ger>
Hey, you'll get mouseevents.spec.whatwg.org if you do!
17:52
<annevk>
teehee, more domain names
17:52
<annevk>
Hixie: not exactly sure what you'd want DOM to define
17:52
<annevk>
Hixie: some kind of pausing in dispatch and a more global stop propagation flag?
17:53
<annevk>
Hixie: some kind of bug report with the details and the link to the use case is prolly best, can't really give an ETA
17:54
<Hixie>
basically a way to say "when an event is dispatched in this browsing context, the following tree of trees is to be used as the chain: ..." and then you deal with all the cloning events, checking dispatch cancelation and so on, and have a hook for when the event is retargetted so that event definitions can say what happens to their values and stuff
17:54
<Hixie>
k
18:07
<Hixie>
anne: https://www.w3.org/Bugs/Public/show_bug.cgi?id=18780
18:13
<smaug____>
Hixie: I thought event re-targeting was discarded
18:13
<smaug____>
(for seamless)
18:16
<smaug____>
that is at least who I interpret the email thread about it
18:16
smaug____
comments on the bug
18:27
<dsheets>
wow… ms on public-fx… wow...
18:28
<jamesr>
they're in the CSS WG and that is a joint deliverable between CSS and SVG
18:30
<dsheets>
not their membership… their approach
18:31
<dsheets>
doesn't seem very savvy
18:39
<Hixie>
smaug____: what part of the thread suggested it was discarded?
18:39
<Hixie>
smaug____: i don't see anything to suggest that, but maybe i missed something
18:39
<smaug____>
Hixie: I sent email explaining the problems
18:39
<Hixie>
url?
18:39
<smaug____>
and dglazkov said "We should probably recreate an event object at each seamless frame boundary."
18:40
<Hixie>
right
18:40
<Hixie>
retargetting always involves new objects
18:40
<Hixie>
you can't reuse the objects
18:40
<smaug____>
no
18:40
<smaug____>
retargeting is retargeting
18:41
<Hixie>
ok well call it whatever you want :-)
18:41
<smaug____>
but ok, perhaps we do agree what should be done :)
18:41
<Hixie>
i'm talking about dispatching events at multiple scopes, with new objects at each level
18:41
<Hixie>
same as xbl
18:41
<smaug____>
right
18:41
<smaug____>
well, not same as xbl1
18:42
smaug____
doesn't remember what shadow DOM does
18:42
<smaug____>
Shadow DOM: "In the cases where events cross the shadow boundaries, the event's information about the target of the event is adjusted in order to maintain upper boundary encapsulation."
18:43
<smaug____>
so it is closer to XBL1
19:46
<Hixie>
heycam|away: yt?
20:07
<deane>
Hixie: He's probably not there as it was only 07:49 here. Not sure when he would start work.
21:14
<jkew>
Mildly dumb question; is there an xml namespace for microdata? Is it considered harmful?
21:18
<gavinc>
jkew: http://www.w3.org/TR/2012/WD-microdata-rdf-20120112/ may help you, or make you more confused
21:18
<gavinc>
see http://www.w3.org/ns/md# it's not an XML namespace, but it is a well known URI
21:19
<gavinc>
err, should have been http://www.w3.org/TR/microdata-rdf/
21:19
<gavinc>
sorry
23:02
<heycam>
hi Hixie
23:07
<Hixie>
heycam: we have a weird situation with HTMLCollection and its descendants, where they introduce operations that have the same name as in HTMLCollection, with different behaviour
23:07
<Hixie>
heycam: how this works in practice probably needs to be tested
23:07
<heycam>
hmm yes I remember there was a bug on this
23:07
<Hixie>
i looked but couldn't find it
23:08
<heycam>
I remember replying on it just recently… because you asked for some IDL way to say "override"
23:08
<heycam>
https://www.w3.org/Bugs/Public/show_bug.cgi?id=17201
23:08
<Hixie>
aha, wasn't in webidl :-)
23:09
<Hixie>
ah, i see your comment
23:09
<Hixie>
the answer to all those questions is "i've no idea"
23:09
<Hixie>
:-)
23:09
<heycam>
ha
23:12
<heycam>
ignoring what happens in practice, it would be good if calling HTMLCollection's namedItem on an HTMLPropertiesCollection object does something consistent with the definition from HTMLCollection
23:13
<Hixie>
i would suspect that the answer is entirely constrained by legacy
23:13
<heycam>
might be, but HTMLPropertiesCollection isn't that new. I guess maybe for the other descendents of HTMLCollection?
23:13
<heycam>
er, isn't that old
23:14
<Hixie>
HTMLPropertiesCollection should just be consistent with the others, yeah
23:18
<heycam>
Hixie, in the green author box for HTMLOptionsCollection it says "element = collection.item(name)". is that right?
23:18
<Hixie>
yeah, there's a legacycaller in the idl
23:19
<heycam>
but "item"?
23:19
<Hixie>
oh, right. yeah, i just filed a bug on DOM Core for that.
23:19
<Hixie>
turns out all browsers support it.
23:20
<heycam>
so item takes integers and names
23:20
<Hixie>
https://www.w3.org/Bugs/Public/show_bug.cgi?id=18784
23:37
<zewt>
drawing a blank here--is there a term for a node relationship on a rooted tree, where the path from node A to B starts by going to the parent
23:37
<zewt>
i guess "nodes that aren't descendants of the current node" is closer
23:38
<say2joe>
siblings?
23:38
<zewt>
that's just other children of your parent
23:39
<say2joe>
not sure i understand what you mean.
23:39
<zewt>
parents; grandparents; siblings; aunts, uncles; anything that you'd go up to your parent to reach, rather than down to a child
23:39
<say2joe>
i can only think of ancestors or siblings as fitting your question as an answer
23:39
<zewt>
i think simply "not a descendant" is good enough
23:40
<say2joe>
ancestor
23:40
<dsheets>
and not self
23:40
<zewt>
(not a descendant nor yourself)
23:40
<say2joe>
but ancestors does not cover lateral nodes (aunt, uncles, etc).
23:40
<zewt>
siblings aren't ancestors :)
23:41
<say2joe>
i think you're right… not a descendent.
23:41
<astearns>
parents, grandparents but not great-grandparents? or great-uncles?
23:42
<heycam>
and n-th cousins m-times removed
23:58
<zewt>
laughing at my desk at someone calling { image: 'img2.jpg', density: '2x', query: '300w' } "fractal complexity"
23:59
<say2joe>
hmmm. last i checked… that's just adaptive images in light of RWD
23:59
<say2joe>
i'd be laughing too.