03:12
<ferret__>
hello
03:13
<ferret__>
hello
04:55
<MikeSmith>
Hixie: I updated the dates in the boilerplate files
04:56
<MikeSmith>
so if you could please push to W3C so those get rebuilt I'd appreciate it
04:56
<MikeSmith>
so if you could please push to W3C so those get rebuilt I'd appreciate it
05:25
<zcorpan>
Hixie: maybe try to fiddle with word-spacing
05:26
<zcorpan>
Hixie: maybe try to fiddle with word-spacing
05:26
<Hixie>
MikeSmith: i'm mid-way through a ridiculously involved edit right now but should be able to do that monday - is that ok?
05:26
<Hixie>
MikeSmith: (if not i can do it sooner but it's a pain to back things out then push them back in again)
05:26
<Hixie>
MikeSmith: i'm mid-way through a ridiculously involved edit right now but should be able to do that monday - is that ok?
05:26
<Hixie>
MikeSmith: (if not i can do it sooner but it's a pain to back things out then push them back in again)
05:27
<Hixie>
zcorpan: yeah, i should try that
05:27
<Hixie>
zcorpan: yeah, i should try that
05:27
<MikeSmith>
Hixie: Monday is fine
05:28
<MikeSmith>
Hixie: Monday is fine
05:28
<MikeSmith>
we publish on Tuesday
05:28
<Hixie>
k
05:28
<MikeSmith>
we publish on Tuesday
05:28
<Hixie>
k
14:06
<annevk>
hsivonen: we can just follow http://dvcs.w3.org/hg/encoding/raw-file/tip/Overview.html no?
14:06
<annevk>
hsivonen: if we are going to support other encodings, might as well support the same set we support through other APIs
14:07
<annevk>
hsivonen: we can just follow http://dvcs.w3.org/hg/encoding/raw-file/tip/Overview.html no?
14:07
<annevk>
hsivonen: if we are going to support other encodings, might as well support the same set we support through other APIs
14:07
<annevk>
hsivonen: my plan is for them all to converge anyway
14:07
<annevk>
hsivonen: my plan is for them all to converge anyway
14:10
<hsivonen>
annevk: ok
14:10
<hsivonen>
annevk: ok
14:44
<smaug____>
hmm, where is this handling for callbacks defined...
14:45
<smaug____>
hmm, where is this handling for callbacks defined...
16:46
<Stalker[lv]>
Hello, all! I've got a problem validating my webpage with one meta tag. Does w3c validator looks in whatwg's wiki when validating html5 page?
16:46
<Stalker[lv]>
Hello, all! I've got a problem validating my webpage with one meta tag. Does w3c validator looks in whatwg's wiki when validating html5 page?
16:47
<MikeSmith>
Stalker[lv]: no the meta checking is hard-coded
16:47
<MikeSmith>
for now at least
16:47
<MikeSmith>
Stalker[lv]: no the meta checking is hard-coded
16:47
<MikeSmith>
for now at least
16:47
<MikeSmith>
when something new gets registered and there's clearly agreement about it, somebody adds it to the validator
16:48
<MikeSmith>
when something new gets registered and there's clearly agreement about it, somebody adds it to the validator
16:48
<MikeSmith>
is there are specific value you're using that it's flagging as an error?
16:48
<MikeSmith>
is there are specific value you're using that it's flagging as an error?
16:50
<Stalker[lv]>
it's metatag, which tells to skype toolbar not to add ugly "call via skype" links
16:50
<Stalker[lv]>
it's metatag, which tells to skype toolbar not to add ugly "call via skype" links
16:50
<Stalker[lv]>
<meta name="SKYPE_TOOLBAR" CONTENT="SKYPE_TOOLBAR_PARSER_COMPATIBLE">
16:51
<Stalker[lv]>
<meta name="SKYPE_TOOLBAR" CONTENT="SKYPE_TOOLBAR_PARSER_COMPATIBLE">
16:52
<Stalker[lv]>
tried to use JS+jQuery skype remover, but it doesn't work with all skype toolbar versions, so I have to use this tag. Also, there's place for "valid html" badge :)
16:52
<Stalker[lv]>
tried to use JS+jQuery skype remover, but it doesn't work with all skype toolbar versions, so I have to use this tag. Also, there's place for "valid html" badge :)
16:52
<MikeSmith>
Stalker[lv]: so if there is a spec somewhere for that this "SKYPE_TOOLBAR" thing is, then you have a start on it possibly being register-able
16:53
<MikeSmith>
Stalker[lv]: so if there is a spec somewhere for that this "SKYPE_TOOLBAR" thing is, then you have a start on it possibly being register-able
16:53
<MikeSmith>
but if there is no spec anywhere, then there's no good way for people to be able to know how to use interoperably
16:54
<MikeSmith>
but if there is no spec anywhere, then there's no good way for people to be able to know how to use interoperably
16:57
<AryehGregor>
Hmm, so a failed script load fires an error event, and it bubbles in Gecko but not WebKit/Presto.
16:57
<AryehGregor>
Spec says it shouldn't bubble.
16:57
AryehGregor
tests IE
16:58
<AryehGregor>
Hmm, so a failed script load fires an error event, and it bubbles in Gecko but not WebKit/Presto.
16:58
<AryehGregor>
Spec says it shouldn't bubble.
16:58
AryehGregor
tests IE
16:58
<AryehGregor>
Nope, doesn't bubble in IE either.
16:58
<AryehGregor>
Nope, doesn't bubble in IE either.
16:58
<AryehGregor>
Why doesn't it bubble? Bubbling seems useful here. Exceptions hit window.onerror, so a failed script load logically may as well too, no?
16:59
<AryehGregor>
Why doesn't it bubble? Bubbling seems useful here. Exceptions hit window.onerror, so a failed script load logically may as well too, no?
17:00
<AryehGregor>
Nobody seems to fire error events for failed stylesheet loads. :(
17:01
<AryehGregor>
Nobody seems to fire error events for failed stylesheet loads. :(
17:10
<Stalker[lv]>
MikeSmith, https://jira.skype.com/browse/SCW-65 - there's mention of this tag. also in wiki is link to atricle about it..
17:11
<Stalker[lv]>
MikeSmith, https://jira.skype.com/browse/SCW-65 - there's mention of this tag. also in wiki is link to atricle about it..
17:11
MikeSmith
looks
17:11
MikeSmith
looks
17:12
<MikeSmith>
Stalker[lv]: in what wiki?
17:12
<MikeSmith>
Stalker[lv]: in what wiki?
17:12
<MikeSmith>
Stalker[lv]: anyway if you wanted to take the time you could write up an actual spec for it yourself
17:13
<MikeSmith>
that's what others have done for some similar cases
17:13
<MikeSmith>
you could add it to the whatwg wiki
17:13
<MikeSmith>
that's what others have done for some similar cases
17:13
<MikeSmith>
you could add it to the whatwg wiki
17:14
<MikeSmith>
like this: http://wiki.whatwg.org/wiki/Meta_referrer
17:14
<MikeSmith>
like this: http://wiki.whatwg.org/wiki/Meta_referrer
17:20
<TabAtkins>
jwheare: The spec was actually inspired by a relatively similar JS implementation, where we just built a layout in using position:absolute hooked up to a constraint solver.
17:20
<TabAtkins>
jwheare: I can ask if it's releasable code.
17:20
<TabAtkins>
jwheare: The spec was actually inspired by a relatively similar JS implementation, where we just built a layout in using position:absolute hooked up to a constraint solver.
17:20
<TabAtkins>
jwheare: I can ask if it's releasable code.
17:20
<jwheare>
TabAtkins: yeah, alex showed me it i think. he mentioned it was internal
17:21
<jwheare>
TabAtkins: yeah, alex showed me it i think. he mentioned it was internal
17:21
<TabAtkins>
Yup.
17:21
<TabAtkins>
That doesn't mean it's secret, just that I think we have to file something with the open-source people if we want to release it.
17:21
<TabAtkins>
Yup.
17:21
<TabAtkins>
That doesn't mean it's secret, just that I think we have to file something with the open-source people if we want to release it.
17:22
<jwheare>
would be good to play around with it. i'm basically gonna need to implement something like it for irccloud
17:22
<jwheare>
would be good to play around with it. i'm basically gonna need to implement something like it for irccloud
17:25
<TabAtkins>
I'll ask about it and get back to you.
17:25
<TabAtkins>
I'll ask about it and get back to you.
17:31
<AryehGregor>
Oh, wait, people do fire error events for failed stylesheet loads, good.
17:31
<AryehGregor>
Oh, wait, people do fire error events for failed stylesheet loads, good.
18:31
<aklein>
smaug____: yt?
18:31
<aklein>
smaug____: yt?
18:34
<smaug____>
aklein: yes
18:34
<smaug____>
though, just going out
18:34
<smaug____>
aklein: yes
18:34
<smaug____>
though, just going out
18:34
<smaug____>
(in 5 mins)
18:34
<aklein>
smaug____: k, I was just going to try one more time :)
18:35
<smaug____>
(in 5 mins)
18:35
<aklein>
smaug____: k, I was just going to try one more time :)
18:35
<smaug____>
to convince case-insensitive filtering?
18:35
<smaug____>
to convince case-insensitive filtering?
18:35
<smaug____>
IMO it is simpler without it
18:35
<aklein>
smaug____: yeah...consider a hypothetical future webdev who observes HREF and then calls setAttribute('HREF', someURL). I claim they'll be surprised.
18:35
<aklein>
the implementation is certainly simpler
18:36
<smaug____>
IMO it is simpler without it
18:36
<aklein>
smaug____: yeah...consider a hypothetical future webdev who observes HREF and then calls setAttribute('HREF', someURL). I claim they'll be surprised.
18:36
<aklein>
the implementation is certainly simpler
18:36
<aklein>
but the HTML dom world very much behaves as if it were case insensitive.
18:36
<aklein>
but the HTML dom world very much behaves as if it were case insensitive.
18:36
<smaug____>
aklein: well, they they are surprised even with the current APIs
18:36
<smaug____>
Attr.name is lowercase
18:36
<aklein>
heh
18:37
<smaug____>
aklein: well, they they are surprised even with the current APIs
18:37
<smaug____>
Attr.name is lowercase
18:37
<aklein>
heh
18:38
<aklein>
yet Element.tagName is uppercase
18:38
<aklein>
yet Element.tagName is uppercase
18:38
<aklein>
yet getElementsByTagName('div') works just fine
18:38
<smaug____>
this is a new API and we don't need to complicate it with special cases
18:39
<aklein>
yet getElementsByTagName('div') works just fine
18:39
<smaug____>
this is a new API and we don't need to complicate it with special cases
18:39
<smaug____>
getElementsByTagName is ancient, so that can't be changed
18:39
<smaug____>
getElementsByTagName is ancient, so that can't be changed
18:40
<smaug____>
.tagName is lowercase
18:40
<smaug____>
er
18:40
<smaug____>
it is whatever what used when creating the element
18:40
<smaug____>
.tagName is lowercase
18:40
<smaug____>
er
18:40
<smaug____>
it is whatever what used when creating the element
18:40
<smaug____>
s/what/was/
18:40
<smaug____>
so, .tagName is case-sensitive
18:40
<aklein>
smaug____: nope: http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#dom-element-tagname
18:41
<smaug____>
s/what/was/
18:41
<smaug____>
so, .tagName is case-sensitive
18:41
<aklein>
smaug____: nope: http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#dom-element-tagname
18:41
<smaug____>
ah, is the spec wrong :)
18:41
<smaug____>
anyway, that is all legacy
18:41
<smaug____>
ah, is the spec wrong :)
18:41
<smaug____>
anyway, that is all legacy
18:41
<smaug____>
we don't have to ruin the API with legacy stugg
18:41
<smaug____>
stuff
18:42
<smaug____>
we don't have to ruin the API with legacy stugg
18:42
<smaug____>
stuff
18:42
<aklein>
but MutationObservers have to live in the same world as all the legacy stuff, so it seems weird to make them behave differently
18:42
<aklein>
but MutationObservers have to live in the same world as all the legacy stuff, so it seems weird to make them behave differently
18:42
<aklein>
ah well, I'll let you go for now
18:42
<smaug____>
I want MutationObservers to behave the way rest of the world
18:43
<aklein>
ah well, I'll let you go for now
18:43
<smaug____>
I want MutationObservers to behave the way rest of the world
18:43
<smaug____>
Attr.name is lowercase in HTML docs
18:43
smaug____
runs
18:43
<smaug____>
Attr.name is lowercase in HTML docs
18:43
smaug____
runs
18:45
<smaug____>
aklein: oh, indeed .tagName is uppercase. I was accidentally testing xhtml doc :/
18:45
<smaug____>
now I really run
18:45
<smaug____>
aklein: oh, indeed .tagName is uppercase. I was accidentally testing xhtml doc :/
18:45
<smaug____>
now I really run
18:45
<smaug____>
back later
18:46
<smaug____>
back later
18:53
<AryehGregor>
Is the fact that browsers fire a bubbling error event at scripts with uncaught exceptions documented? I don't see it glancing at HTML.
18:54
<AryehGregor>
Is the fact that browsers fire a bubbling error event at scripts with uncaught exceptions documented? I don't see it glancing at HTML.
19:07
<smaugN900>
aklein: case-sensitive is also compatible with current DOMAttrModified with which one needs to filter using attr nodes
19:07
<smaugN900>
aklein: case-sensitive is also compatible with current DOMAttrModified with which one needs to filter using attr nodes
20:02
<AryehGregor>
Given a <script>, there's no way to tell whether it's run successfully yet or not, right?
20:02
<AryehGregor>
Meaning, if I do document.querySelectorAll("script") or something, I can't figure out which (if any) haven't run yet.
20:02
<AryehGregor>
Or never ran at all, e.g., had errors.
20:02
<AryehGregor>
Given a <script>, there's no way to tell whether it's run successfully yet or not, right?
20:02
<AryehGregor>
Meaning, if I do document.querySelectorAll("script") or something, I can't figure out which (if any) haven't run yet.
20:02
<AryehGregor>
Or never ran at all, e.g., had errors.
20:02
<TabAtkins>
Besides having the script alter itself at the end of running, right?
20:03
<TabAtkins>
Besides having the script alter itself at the end of running, right?
20:03
<AryehGregor>
Right, I don't want to modify the scripts themselves.
20:03
<AryehGregor>
(since there are zillions)
20:03
<AryehGregor>
Right, I don't want to modify the scripts themselves.
20:03
<AryehGregor>
(since there are zillions)
20:03
<TabAtkins>
Then no clue.
20:04
<TabAtkins>
Then no clue.
20:04
<AryehGregor>
Hmm, afterscriptexecute bubbles.
20:04
<AryehGregor>
Maybe I could use that.
20:04
<AryehGregor>
Hmm, afterscriptexecute bubbles.
20:04
<AryehGregor>
Maybe I could use that.
20:04
<AryehGregor>
With a capturing event listener.
20:05
<AryehGregor>
With a capturing event listener.
20:05
<AryehGregor>
Won't help for scripts that have already run, of course.
20:06
<AryehGregor>
Won't help for scripts that have already run, of course.
20:06
<Hixie>
AryehGregor: change their contents and reinsert them into the dom; if they do anything then they weren't run. :-)
20:07
<Hixie>
AryehGregor: change their contents and reinsert them into the dom; if they do anything then they weren't run. :-)
20:07
<Hixie>
AryehGregor: (the key being a <script> can only run once)
20:07
<Hixie>
AryehGregor: (the key being a <script> can only run once)
20:07
<AryehGregor>
Hixie, so something like: script.textContent = "alert('bad')"; var nextSibling = script.nextSibling; script.parentNode.removeChild(script); script.insertBefore(script, nextSibling); ?
20:08
<AryehGregor>
Hixie, so something like: script.textContent = "alert('bad')"; var nextSibling = script.nextSibling; script.parentNode.removeChild(script); script.insertBefore(script, nextSibling); ?
20:08
<AryehGregor>
And unset the src.
20:08
<AryehGregor>
That's kind of evil genius.
20:08
AryehGregor
tries
20:08
<AryehGregor>
And unset the src.
20:08
<AryehGregor>
That's kind of evil genius.
20:08
AryehGregor
tries
20:09
<Hixie>
AryehGregor: something like that. or maybe even better, if you want to not screw up the original element, clone it, remove .src, set the contents to something harmless like ';', and then insert that
20:09
<Hixie>
AryehGregor: since the "has executed" state propagates to clones
20:10
<Hixie>
AryehGregor: something like that. or maybe even better, if you want to not screw up the original element, clone it, remove .src, set the contents to something harmless like ';', and then insert that
20:10
<Hixie>
AryehGregor: since the "has executed" state propagates to clones
20:10
<Hixie>
(remove async and defer as well)
20:10
<Hixie>
(remove async and defer as well)
20:18
<Hixie>
are any of the people who wanted path objects not to have transforms but instead for the methods to get matrix arguments around?
20:19
<Hixie>
are any of the people who wanted path objects not to have transforms but instead for the methods to get matrix arguments around?
20:20
<TabAtkins>
Matrix-based transforms are the devil.
20:20
<TabAtkins>
Matrix-based transforms are the devil.
20:21
<Philip`>
As in requiring users of the API to do the matrix maths themselves and pass the result into the API?
20:22
<Philip`>
As in requiring users of the API to do the matrix maths themselves and pass the result into the API?
20:22
<Hixie>
the matrix object has the regular transformation stuff on it, that's not hte issue
20:22
<Hixie>
(it's just SVGMatrix)
20:22
<Philip`>
Ah
20:22
<Hixie>
the matrix object has the regular transformation stuff on it, that's not hte issue
20:22
<Hixie>
(it's just SVGMatrix)
20:22
<Philip`>
Ah
20:22
Philip`
hasn't had time/energy/nonlaziness to try looking at any of this yet :-(
20:23
Philip`
hasn't had time/energy/nonlaziness to try looking at any of this yet :-(
20:23
<Hixie>
my question is whether those people think it's ok for Pattern to implement the transform stuff directly or whether there should be some method that takes a matrix there, too
20:23
<Hixie>
and if the latter, which method
20:23
<Hixie>
my question is whether those people think it's ok for Pattern to implement the transform stuff directly or whether there should be some method that takes a matrix there, too
20:23
<Hixie>
createPattern()?
20:23
<Hixie>
fill()?
20:23
<Hixie>
pattern.setPatternTransform()?
20:23
<Hixie>
and if the latter, which method
20:23
<WeirdAl>
Hey, folks, DOM4 question: when I call Node.addEventListener("foo", function listener(evt) { /* ... */ }, true), what is the |this| object inside the listener function?
20:23
<Hixie>
createPattern()?
20:23
<Hixie>
fill()?
20:23
<Hixie>
pattern.setPatternTransform()?
20:23
<WeirdAl>
Hey, folks, DOM4 question: when I call Node.addEventListener("foo", function listener(evt) { /* ... */ }, true), what is the |this| object inside the listener function?
20:23
<Hixie>
WeirdAl: Window
20:24
<Hixie>
WeirdAl: Window
20:24
<WeirdAl>
Hixie - reading the spec, I couldn't find a clear reference to that, even though it's the usual behavior in DOM3... worth doc'ing?
20:25
<WeirdAl>
Hixie - reading the spec, I couldn't find a clear reference to that, even though it's the usual behavior in DOM3... worth doc'ing?
20:25
<Hixie>
it's specified somewhere, though i forget where
20:25
<Hixie>
webidl maybe?
20:25
<Hixie>
maybe the js spc?
20:25
<Hixie>
it's specified somewhere, though i forget where
20:25
<Hixie>
webidl maybe?
20:25
<Hixie>
maybe the js spc?
20:25
<WeirdAl>
js spec, as in ECMAScript 5th Ed?
20:25
<Hixie>
or whatever the latest version is
20:25
<Hixie>
probably webidl
20:26
<WeirdAl>
js spec, as in ECMAScript 5th Ed?
20:26
<Hixie>
or whatever the latest version is
20:26
<Hixie>
probably webidl
20:26
<Hixie>
oh actually i was wrong
20:27
<Hixie>
oh actually i was wrong
20:27
<Hixie>
seems the |this| value should be the Function itself
20:27
<WeirdAl>
ugh
20:27
<WeirdAl>
that's almost certainly a change from DOM3
20:27
<Hixie>
http://dev.w3.org/2006/webapi/WebIDL/#dfn-callback-this-value
20:27
<Hixie>
seems the |this| value should be the Function itself
20:27
<WeirdAl>
ugh
20:27
<WeirdAl>
that's almost certainly a change from DOM3
20:27
<Hixie>
http://dev.w3.org/2006/webapi/WebIDL/#dfn-callback-this-value
20:27
<Hixie>
heycam|away: "By default, the object itself is used as the callback this value" is lacking conformance criteria (it's a description, it neither defines something nor requires something)
20:28
<Hixie>
heycam|away: "By default, the object itself is used as the callback this value" is lacking conformance criteria (it's a description, it neither defines something nor requires something)
20:28
<Hixie>
oh no i'm wrong!
20:28
<Hixie>
oh no i'm wrong!
20:28
<Hixie>
the default callback this value for Functions is 'null'
20:28
<WeirdAl>
:)
20:29
<Hixie>
the default callback this value for Functions is 'null'
20:29
<WeirdAl>
:)
20:29
<WeirdAl>
imho, that's as it should be - not a function, not window, not the global
20:29
<Hixie>
heycam|away: specifically, "By default, the callback this value when invoking a callback function is null" seems to try to override the above, but in practice it just contradicts it and doesn't leave it well defined
20:29
<WeirdAl>
imho, that's as it should be - not a function, not window, not the global
20:29
<Hixie>
heycam|away: specifically, "By default, the callback this value when invoking a callback function is null" seems to try to override the above, but in practice it just contradicts it and doesn't leave it well defined
20:29
<Hixie>
heycam|away: and "callback this value" in the algorithm below that links to the description i quoted earlier, which is why i was confused
20:30
<Hixie>
heycam|away: and "callback this value" in the algorithm below that links to the description i quoted earlier, which is why i was confused
20:30
<Hixie>
WeirdAl: yeah but i expect ES says to use the global of it's null
20:30
<Hixie>
WeirdAl: yeah but i expect ES says to use the global of it's null
20:31
<Hixie>
WeirdAl: yeah, JS spec 10.4.3 ("Entering Function Code") step 2 says "Else if thisArg is null or undefined, set the ThisBinding to the global object"
20:32
<Hixie>
WeirdAl: yeah, JS spec 10.4.3 ("Entering Function Code") step 2 says "Else if thisArg is null or undefined, set the ThisBinding to the global object"
20:33
<Hixie>
WeirdAl: and HTML says for <script> objects: "The global object is the Window object of document" http://www.whatwg.org/specs/web-apps/current-work/#the-script-settings-determined-from-the-node
20:33
<Hixie>
WeirdAl: so to answer your question, it's defined. WebIDL says it's null, ES says null means global object, and HTML says global object means Window.
20:33
<Hixie>
WeirdAl: and HTML says for <script> objects: "The global object is the Window object of document" http://www.whatwg.org/specs/web-apps/current-work/#the-script-settings-determined-from-the-node
20:33
<Hixie>
WeirdAl: so to answer your question, it's defined. WebIDL says it's null, ES says null means global object, and HTML says global object means Window.
20:43
<zcorpan>
why is it "level 1"? CSS Variables Module Level 1
20:43
<zcorpan>
why is it "level 1"? CSS Variables Module Level 1
20:43
<TabAtkins>
zcorpan: Because it's not based on anything from level 2, so it resets.
20:44
<TabAtkins>
zcorpan: Because it's not based on anything from level 2, so it resets.
20:45
<zcorpan>
animations etc is also level 1?
20:45
<zcorpan>
animations etc is also level 1?
20:45
<astearns>
all of the level 3 specs are based on 2.1 - what causes the increment?
20:46
<astearns>
all of the level 3 specs are based on 2.1 - what causes the increment?
20:46
<TabAtkins>
No, Animations is level 3 because nobody was watching too hard when it was named.
20:46
<TabAtkins>
it should have been level 1
20:46
<TabAtkins>
No, Animations is level 3 because nobody was watching too hard when it was named.
20:46
<TabAtkins>
it should have been level 1
20:47
<zcorpan>
i think calling it "css1 variables" (which is short for css variables level 1) is confusing because it implies it was present in css1
20:47
<Hixie>
is (x,y) a coordinate, or are (x,y) coordinates?
20:47
<zcorpan>
Hixie: former
20:47
Hixie
waves his "just drop all versioning" flag
20:47
<Hixie>
zcorpan: thanks
20:47
<zcorpan>
i think calling it "css1 variables" (which is short for css variables level 1) is confusing because it implies it was present in css1
20:47
<Hixie>
is (x,y) a coordinate, or are (x,y) coordinates?
20:47
<zcorpan>
Hixie: former
20:47
Hixie
waves his "just drop all versioning" flag
20:47
<Hixie>
zcorpan: thanks
20:47
<astearns>
'level living'
20:48
<astearns>
'level living'
20:48
<zcorpan>
levelless css wfm
20:48
<zcorpan>
levelless css wfm
20:48
<TabAtkins>
For all practical purposes CSS is levelless.
20:49
<TabAtkins>
For all practical purposes CSS is levelless.
20:49
<zcorpan>
TabAtkins: so on a different note, do you think the css wg would be willing to put quirks mode requirements in their specs?
20:50
<zcorpan>
TabAtkins: so on a different note, do you think the css wg would be willing to put quirks mode requirements in their specs?
20:50
<Hixie>
ok the spec live on whatwg.org (single page only) now has the path stuff updated to use DrawingStyle objects instead of on-path line styles and font styles
20:50
<Hixie>
now doing the transform changes
20:50
<TabAtkins>
zcorpan: In all the specs, or just a "Quirks Mode spec" kind of thing?
20:50
<Hixie>
ok the spec live on whatwg.org (single page only) now has the path stuff updated to use DrawingStyle objects instead of on-path line styles and font styles
20:50
<Hixie>
now doing the transform changes
20:50
<TabAtkins>
zcorpan: In all the specs, or just a "Quirks Mode spec" kind of thing?
20:50
<Hixie>
othermaciej: any opinion on whether patterns should implement the transform stuff or whether we should use SVGMatrix objects somewhere to transform patterns in fill too?
20:51
<Hixie>
othermaciej: any opinion on whether patterns should implement the transform stuff or whether we should use SVGMatrix objects somewhere to transform patterns in fill too?
20:51
<Hixie>
othermaciej: and if the latter, where should the matrix go? createPattern()? A method on the pattern object? fill()?
20:51
<zcorpan>
TabAtkins: well i think it would be best if it was defined as part of the proper spec because a quirk spec needs to override the proper spec and that can be a bit messy
20:51
<Hixie>
othermaciej: and if the latter, where should the matrix go? createPattern()? A method on the pattern object? fill()?
20:51
<zcorpan>
TabAtkins: well i think it would be best if it was defined as part of the proper spec because a quirk spec needs to override the proper spec and that can be a bit messy
20:52
<TabAtkins>
zcorpan: I suspect that requiring a "Quirks Mode" section in every spec wouldn't be well-received.
20:53
<TabAtkins>
zcorpan: I suspect that requiring a "Quirks Mode" section in every spec wouldn't be well-received.
20:53
<Hixie>
shouldn't be a section
20:53
<Hixie>
should just be the way html does it
20:53
<Hixie>
shouldn't be a section
20:53
<zcorpan>
what Hixie said
20:53
<Hixie>
should just be the way html does it
20:53
<zcorpan>
what Hixie said
20:54
<othermaciej>
Hixie: I am not up to speed on what is planned for patterns
20:54
<TabAtkins>
What's the way HTML does it? I haven't really noticed quirks stuff anywhere.
20:54
<TabAtkins>
What's the way HTML does it? I haven't really noticed quirks stuff anywhere.
20:54
<Hixie>
othermaciej: the old plan before you pointed out it was silly to put all the transform apis on Path objects, was to make CanvasPattern objects have a built-in transform in the same way, so as to allow the pattern itself to be transformed (rotated, translated, skewed, etc) when used for fills and strokes
20:54
<Hixie>
TabAtkins: exactly
20:55
<Hixie>
othermaciej: the old plan before you pointed out it was silly to put all the transform apis on Path objects, was to make CanvasPattern objects have a built-in transform in the same way, so as to allow the pattern itself to be transformed (rotated, translated, skewed, etc) when used for fills and strokes
20:55
<Hixie>
TabAtkins: exactly
20:55
<Hixie>
TabAtkins: it just says "and now do x. if in quirks mode, do y, otherwise do z. then do q...."
20:55
<zcorpan>
TabAtkins: e.g. http://simon.html5.org/specs/quirks-mode#the-blocks-ignore-line-height-quirk should be specified as part of http://www.w3.org/TR/CSS21/visudet.html#propdef-line-height
20:55
<Hixie>
TabAtkins: it just says "and now do x. if in quirks mode, do y, otherwise do z. then do q...."
20:55
<zcorpan>
TabAtkins: e.g. http://simon.html5.org/specs/quirks-mode#the-blocks-ignore-line-height-quirk should be specified as part of http://www.w3.org/TR/CSS21/visudet.html#propdef-line-height
20:56
<TabAtkins>
I personally prefer it in a big list somewhere.
20:56
<Hixie>
really? why?
20:57
<TabAtkins>
I personally prefer it in a big list somewhere.
20:57
<Hixie>
really? why?
20:57
<TabAtkins>
The only effect of quirks on new content is parsing-level stuff.
20:57
<TabAtkins>
The only effect of quirks on new content is parsing-level stuff.
20:57
<TabAtkins>
Which shouldn't be in individual specs anyway.
20:57
<TabAtkins>
Which shouldn't be in individual specs anyway.
20:57
<zcorpan>
new content shouldn't use quirks mode at all
20:57
<Hixie>
yeah
20:57
<Hixie>
what's new content got to do with what the specs say?
20:58
<zcorpan>
new content shouldn't use quirks mode at all
20:58
<Hixie>
yeah
20:58
<Hixie>
what's new content got to do with what the specs say?
20:58
<TabAtkins>
I meant new specs.
20:58
<TabAtkins>
I meant new specs.
20:58
Hixie
decides to put transforms in createPattern()
20:58
<zcorpan>
having it in a bit list makes it harder for implementors to figure out how it integrates with the spec proper
20:58
Hixie
decides to put transforms in createPattern()
20:58
<zcorpan>
having it in a bit list makes it harder for implementors to figure out how it integrates with the spec proper
20:58
<zcorpan>
TabAtkins: it wouldn't need to be in new specs if the new spec doesn't have any quirks
20:58
<othermaciej>
Hixie: I'd have to do some research on what other pattern APIs do for this, if anything
20:59
<zcorpan>
TabAtkins: it wouldn't need to be in new specs if the new spec doesn't have any quirks
20:59
<othermaciej>
Hixie: I'd have to do some research on what other pattern APIs do for this, if anything
20:59
<TabAtkins>
If it's limited to CSS2.1 and Selectors, then I'm less opposed.
20:59
<TabAtkins>
If it's limited to CSS2.1 and Selectors, then I'm less opposed.
20:59
<zcorpan>
TabAtkins: some quirks are about fine details about table layout, and css 2.1 doesn't define table layout
21:00
<zcorpan>
TabAtkins: some quirks are about fine details about table layout, and css 2.1 doesn't define table layout
21:00
<WeirdAl>
Hixie - my monitor's acting up - I'll try to reread through channel logs
21:00
<Hixie>
othermaciej: k
21:00
<WeirdAl>
Hixie - my monitor's acting up - I'll try to reread through channel logs
21:00
<Hixie>
othermaciej: k
21:01
<othermaciej>
Hixie: how exactly do patterns work, are they ultimately a bitmap?
21:01
<Hixie>
TabAtkins: values and units, box model, table -- that's probably it as far as affected css specs go, right zcorpan?
21:01
<TabAtkins>
zcorpan: There's no CSS3 Table spec, so that's all right. ^_^
21:01
<WeirdAl>
Hixie - ah, your answer does make sense.
21:01
<Hixie>
othermaciej: sure
21:01
<othermaciej>
Hixie: how exactly do patterns work, are they ultimately a bitmap?
21:01
<Hixie>
TabAtkins: values and units, box model, table -- that's probably it as far as affected css specs go, right zcorpan?
21:01
<TabAtkins>
zcorpan: There's no CSS3 Table spec, so that's all right. ^_^
21:01
<WeirdAl>
Hixie - ah, your answer does make sense.
21:01
<Hixie>
othermaciej: sure
21:01
<zcorpan>
TabAtkins: and some level 3 specs that redefine 2.1 better would be better suited to cover quirks than 2.1 (like e.g. text-decoration)
21:01
<WeirdAl>
but it'd sure be nice to not have to read fifteen different specs to get the answer ;)
21:02
<zcorpan>
TabAtkins: and some level 3 specs that redefine 2.1 better would be better suited to cover quirks than 2.1 (like e.g. text-decoration)
21:02
<WeirdAl>
but it'd sure be nice to not have to read fifteen different specs to get the answer ;)
21:02
<zcorpan>
TabAtkins: i hope there will be :-P
21:02
<zcorpan>
TabAtkins: i hope there will be :-P
21:03
<Hixie>
WeirdAl: well, i'm all in favour of reducing the numebr of specs. personally i think it'd be great to merge webidl and js, maybe even webidl and dom core, but i doubt anyone's gonna go for that. and people are already arguing that html does too much, so i doubt they'd be up for merging any of those with html...
21:03
<TabAtkins>
zcorpan: When someone sufficiently masochistic volunteers, there will be.
21:03
<Hixie>
WeirdAl: well, i'm all in favour of reducing the numebr of specs. personally i think it'd be great to merge webidl and js, maybe even webidl and dom core, but i doubt anyone's gonna go for that. and people are already arguing that html does too much, so i doubt they'd be up for merging any of those with html...
21:03
<TabAtkins>
zcorpan: When someone sufficiently masochistic volunteers, there will be.
21:03
<WeirdAl>
Hixie - I'm of mixed opinion on that...
21:04
<WeirdAl>
Hixie - I'm of mixed opinion on that...
21:04
<WeirdAl>
but I do think the undo manager spec rniwa is working on should be folded in as a chapter of HTML5
21:04
<zcorpan>
Hixie: and selectors, and maybe some others -- i don't have a good overview of the various css modules
21:04
<WeirdAl>
but I do think the undo manager spec rniwa is working on should be folded in as a chapter of HTML5
21:04
<zcorpan>
Hixie: and selectors, and maybe some others -- i don't have a good overview of the various css modules
21:05
<Hixie>
WeirdAl: it will eventually, i expect. it used to be part of html, it's just easier for him to work on it separately for now
21:05
<Hixie>
zcorpan: fair enough
21:05
<Hixie>
WeirdAl: it will eventually, i expect. it used to be part of html, it's just easier for him to work on it separately for now
21:05
<rniwa>
WeirdAl: it might make more sense to merge it into AryehGregor's editing spec
21:05
Hixie
thinks most of the css modules should be folded into one anyway :-P
21:05
<Hixie>
rniwa: yeah but that will eventually make it back to html too, probably
21:05
<Hixie>
zcorpan: fair enough
21:05
<rniwa>
WeirdAl: it might make more sense to merge it into AryehGregor's editing spec
21:05
Hixie
thinks most of the css modules should be folded into one anyway :-P
21:05
<Hixie>
rniwa: yeah but that will eventually make it back to html too, probably
21:05
WeirdAl
thinks we need a metaspec: a spec saying "Here's where you find the spec for this" :p
21:05
<rniwa>
Hixie: yeah, that makes sense in the long ru
21:06
WeirdAl
thinks we need a metaspec: a spec saying "Here's where you find the spec for this" :p
21:06
<rniwa>
Hixie: yeah, that makes sense in the long ru
21:06
<Hixie>
WeirdAl: platform.html5.org
21:06
<rniwa>
Hixie: but will probably take a decade or two before that happens :(
21:06
<Hixie>
WeirdAl: platform.html5.org
21:06
<Hixie>
rniwa: prolly :-)
21:06
<WeirdAl>
yikes!
21:06
<rniwa>
Hixie: but will probably take a decade or two before that happens :(
21:06
<Hixie>
rniwa: prolly :-)
21:06
<WeirdAl>
yikes!
21:07
<zcorpan>
Hixie: are there quirks i've listed under "CSS" that you think belong in html instead?
21:07
<Hixie>
WeirdAl: (to be fair, big chunks of that are just "html")
21:07
<Hixie>
zcorpan: url?
21:07
<zcorpan>
http://simon.html5.org/specs/quirks-mode#css
21:07
<zcorpan>
Hixie: are there quirks i've listed under "CSS" that you think belong in html instead?
21:07
<Hixie>
WeirdAl: (to be fair, big chunks of that are just "html")
21:07
<Hixie>
zcorpan: url?
21:07
<zcorpan>
http://simon.html5.org/specs/quirks-mode#css
21:07
<Hixie>
WeirdAl: (though of course, "css" is a bit vague in that list and should be split out into its component specs since there's not just one css spec)
21:08
<Hixie>
WeirdAl: (though of course, "css" is a bit vague in that list and should be split out into its component specs since there's not just one css spec)
21:08
<Hixie>
WeirdAl: (so it might be a bit of a wash in the end)
21:08
WeirdAl
shakes his head ruefully
21:08
<Hixie>
WeirdAl: (so it might be a bit of a wash in the end)
21:08
WeirdAl
shakes his head ruefully
21:09
<Hixie>
zcorpan: "The table cell nowrap minimum width calculation quirk" might need to have something in HTML, but it would have to refer to something in CSS, so might not be worth it
21:09
<Hixie>
zcorpan: "The table cell nowrap minimum width calculation quirk" might need to have something in HTML, but it would have to refer to something in CSS, so might not be worth it
21:09
<Hixie>
zcorpan: can "The text decoration doesn't propagate into tables quirk" be defined as a CSS rule? if so, ->html
21:10
<Hixie>
zcorpan: can "The text decoration doesn't propagate into tables quirk" be defined as a CSS rule? if so, ->html
21:10
<zcorpan>
Hixie: it can't
21:10
<Hixie>
k
21:10
<Hixie>
zcorpan: looks like i can take "The table cell width resets nowrap quirk", file a bug
21:10
<zcorpan>
Hixie: it can't
21:10
<Hixie>
k
21:10
<Hixie>
zcorpan: looks like i can take "The table cell width resets nowrap quirk", file a bug
21:10
<Hixie>
zcorpan: other than that, i think you're right and it should go into the css spec
21:10
<Hixie>
zcorpan: i'm surprised there aren't more html quirks that i've missed
21:11
<Hixie>
zcorpan: other than that, i think you're right and it should go into the css spec
21:11
<Hixie>
zcorpan: i'm surprised there aren't more html quirks that i've missed
21:11
<zcorpan>
that's already filed
21:11
<zcorpan>
that's already filed
21:11
<Hixie>
cool
21:11
<zcorpan>
there was one more that i just filed
21:11
<Hixie>
cool
21:11
<zcorpan>
there was one more that i just filed
21:11
<zcorpan>
"just" as in "didn't spec", not "right now"
21:12
<zcorpan>
"just" as in "didn't spec", not "right now"
21:12
<Hixie>
k
21:12
<Hixie>
k
21:12
<Hixie>
still, that's pretty good :-)
21:12
<zcorpan>
yeah
21:13
<Hixie>
still, that's pretty good :-)
21:13
<zcorpan>
yeah
21:16
<zcorpan>
Hixie: The font element text decoration color quirk can't be specified as a css rule, either. should it be under "css" instead?
21:17
<zcorpan>
Hixie: The font element text decoration color quirk can't be specified as a css rule, either. should it be under "css" instead?
21:18
Hixie
shrugs
21:18
Hixie
shrugs
21:18
<Hixie>
that's a mere intraspec organisation issue
21:18
<Hixie>
up to you entirely :-)
21:19
<Hixie>
that's a mere intraspec organisation issue
21:19
<Hixie>
up to you entirely :-)
21:23
<zcorpan>
i moved it
21:24
<zcorpan>
i moved it
21:24
<zcorpan>
Hixie: there's one more quirk that i hope we can drop but might not be successful, style="{ ... }"
21:24
<zcorpan>
Hixie: there's one more quirk that i hope we can drop but might not be successful, style="{ ... }"
21:28
<Hixie>
zcorpan: k
21:28
<Hixie>
zcorpan: k
21:39
<Ms2ger>
<Hixie> is (x,y) a coordinate, or are (x,y) coordinates?
21:39
<Ms2ger>
Latter
21:39
<Ms2ger>
<Hixie> is (x,y) a coordinate, or are (x,y) coordinates?
21:39
<Ms2ger>
Latter
21:40
<smaug____>
WeirdAl: you asked something about 'this'
21:40
<smaug____>
with event listeners
21:40
<smaug____>
WeirdAl: you asked something about 'this'
21:40
<smaug____>
with event listeners
21:40
<WeirdAl>
yeah, I just wanted to know what the specs say it should be when the listener is a function and not an object
21:41
<WeirdAl>
yeah, I just wanted to know what the specs say it should be when the listener is a function and not an object
21:41
<smaug____>
it is event.currentTarget
21:41
<smaug____>
DOM4
21:41
<smaug____>
it is event.currentTarget
21:41
<smaug____>
DOM4
21:41
<WeirdAl>
ok
21:42
<WeirdAl>
ok
21:43
<Hixie>
Ms2ger: zcorpan disagreed and he was first
21:43
<Hixie>
Ms2ger: zcorpan disagreed and he was first
21:43
<zcorpan>
Ms2ger: x is a coordinate for a line, y is a coordinate for a line, (x, y) is a coordinate for a point. no?
21:43
<Ms2ger>
Say what?
21:44
<zcorpan>
Ms2ger: x is a coordinate for a line, y is a coordinate for a line, (x, y) is a coordinate for a point. no?
21:44
<Ms2ger>
Say what?
21:44
<Hixie>
smaug____: it is?
21:44
<Hixie>
smaug____: it is?
21:44
<Hixie>
smaug____: oh yeah, look at that
21:44
<Hixie>
WeirdAl: smaug____ is right, dom4 overrides webidl here
21:45
<Hixie>
smaug____: oh yeah, look at that
21:45
<Hixie>
WeirdAl: smaug____ is right, dom4 overrides webidl here
21:45
<smaug____>
Hixie: yes
21:45
<WeirdAl>
ok. It wasn't clear from my cursory reading :)
21:45
<WeirdAl>
ok. It wasn't clear from my cursory reading :)
21:45
<Hixie>
search for "this value"
21:46
<Hixie>
search for "this value"
21:46
<WeirdAl>
ah, right there, ok
21:46
<WeirdAl>
ah, right there, ok
21:46
WeirdAl
is chagrined
21:47
WeirdAl
is chagrined
21:57
<zcorpan>
smaug____: i think the spec isn't wrong for tagName
21:57
<zcorpan>
smaug____: i think the spec isn't wrong for tagName
21:57
<smaug____>
zcorpan: no no, that was my mistake
21:57
<Ms2ger>
He was accidentally using XML
21:57
<Ms2ger>
How you do that, no idea
21:57
<smaug____>
I was testing accidentally xhtml
21:58
<smaug____>
zcorpan: no no, that was my mistake
21:58
<Ms2ger>
He was accidentally using XML
21:58
<Ms2ger>
How you do that, no idea
21:58
<smaug____>
I was testing accidentally xhtml
21:58
<smaug____>
Ms2ger: I guess I opened new tab
21:58
<zcorpan>
smaug____: ah. ok
21:58
<smaug____>
and about:newTab is xhtml or something
21:58
<smaug____>
Ms2ger: I guess I opened new tab
21:58
<Ms2ger>
Ah, right
21:58
<smaug____>
silly me
21:58
<zcorpan>
smaug____: ah. ok
21:58
<smaug____>
and about:newTab is xhtml or something
21:58
<Ms2ger>
Ah, right
21:58
<smaug____>
silly me
21:59
<zcorpan>
AryehGregor: uncaught exceptions in scripts don't fire a bubbling error event
22:00
<zcorpan>
AryehGregor: they invoke window.onerror() with three arguments, not involving any event
22:00
<AryehGregor>
Oh, they just fire straight at window?
22:00
<AryehGregor>
Oh, sneaky.
22:00
<zcorpan>
nope
22:00
<AryehGregor>
What defines that?
22:00
<zcorpan>
html
22:00
<zcorpan>
AryehGregor: they invoke window.onerror() with three arguments, not involving any event
22:00
<AryehGregor>
Oh, they just fire straight at window?
22:00
<AryehGregor>
Oh, sneaky.
22:00
<zcorpan>
nope
22:00
<AryehGregor>
What defines that?
22:00
<zcorpan>
html
22:03
<Ms2ger>
There are tests, even
22:03
<Ms2ger>
There are tests, even
22:04
<AryehGregor>
So window.onerror really is used for two totally different things.
22:04
<zcorpan>
the spec has two entry points
22:04
<zcorpan>
http://dev.w3.org/html5/spec/webappapis.html#create-a-script
22:04
<zcorpan>
"Otherwise, report the error ..."
22:04
<AryehGregor>
(W3C version? Really?)
22:04
<AryehGregor>
So window.onerror really is used for two totally different things.
22:04
<zcorpan>
the spec has two entry points
22:04
<zcorpan>
http://dev.w3.org/html5/spec/webappapis.html#create-a-script
22:04
<zcorpan>
"Otherwise, report the error ..."
22:04
<AryehGregor>
(W3C version? Really?)
22:04
<zcorpan>
and http://dev.w3.org/html5/spec/webappapis.html#runtime-script-errors-in-documents
22:04
<zcorpan>
whatwg didn't want to load for me :(
22:05
<zcorpan>
and http://dev.w3.org/html5/spec/webappapis.html#runtime-script-errors-in-documents
22:05
<zcorpan>
whatwg didn't want to load for me :(
22:06
<zcorpan>
and workers have onerror stuff as well
22:06
<zcorpan>
and workers have onerror stuff as well
22:39
<Hixie>
MikeSmith: ok, should be published
22:40
<Hixie>
MikeSmith: ok, should be published
22:45
<smaug____>
aklein: looks like webkit handles documentfragments inconsistently
22:46
<smaug____>
aklein: looks like webkit handles documentfragments inconsistently
22:46
<smaug____>
when inserting documentfragment to some other node, each removal from documentfragment get separate record
22:46
<smaug____>
when inserting documentfragment to some other node, each removal from documentfragment get separate record
22:46
<smaug____>
aklein: but they are all added in one record to the new parent
22:47
<smaug____>
aklein: but they are all added in one record to the new parent
22:47
<smaug____>
gecko will create only 2 records. One for removals, one for additions
22:47
<smaug____>
spec bug has been filed
22:47
<smaug____>
(it doesn't define this case)
22:47
<smaug____>
gecko will create only 2 records. One for removals, one for additions
22:47
<smaug____>
spec bug has been filed
22:47
<smaug____>
(it doesn't define this case)
22:59
<Hixie>
heycam: why are dictionary types not distinguishable?
22:59
<Hixie>
heycam: why are dictionary types not distinguishable?
23:00
<Hixie>
heycam: is there no type checking beyond "is an Object" for dictionary arguments? i.e. no checking that fields are present?
23:00
<Hixie>
heycam: is there no type checking beyond "is an Object" for dictionary arguments? i.e. no checking that fields are present?
23:08
<aklein>
smaug____: not surprising...WebKit doesn't treat document fragments nearly atomically enough, I'm hacking on that this week
23:09
<aklein>
smaug____: not surprising...WebKit doesn't treat document fragments nearly atomically enough, I'm hacking on that this week
23:09
<heycam>
Hixie, that's right
23:10
<heycam>
Hixie, that's right
23:10
<Hixie>
:-(
23:10
<aklein>
smaug____: I suspect that replaceChild may also be handled differently
23:10
<Hixie>
:-(
23:10
<aklein>
smaug____: I suspect that replaceChild may also be handled differently
23:10
<smaug____>
ok
23:11
<smaug____>
ok
23:42
<zewt>
reading the indexeddb spec makes me :|
23:43
<zewt>
reading the indexeddb spec makes me :|
23:43
<zewt>
If these steps are running asynchronously, return the created transaction and queue up the remaining steps.
23:44
<zewt>
If these steps are running asynchronously, return the created transaction and queue up the remaining steps.