01:05
<MikeSmith>
othermaciej: I had bugmail set to also send notifications when status of a bug changes
01:05
<MikeSmith>
you really want it only for new bugs?
01:05
<MikeSmith>
not for status changes?
01:11
<othermaciej>
MikeSmith: I don't terribly mind them myself, but my original suggestion was new bugs only, so maybe we should start with that so people get what they signed up for
01:11
<othermaciej>
MikeSmith: maybe we can add more notifications later if we need to
01:14
<MikeSmith>
ok
01:14
<MikeSmith>
now set to just notifications for new bugs
01:19
<nessy>
I like that - I can always subscribe to updates for those bugs that I care about
01:19
<Hixie>
yeah new bugs only makes sense to me too
05:06
<GPHemsley>
hsivonen, Hixie: Keep in mind that what I consider noise is likely very different from what you consider noise. ;)
07:24
<annevk>
oh, extensibility went to vote
10:04
<annevk>
zcorpan, maybe you should add your suggestion for what to replace non-normative with here: http://www.w3.org/Bugs/Public/show_bug.cgi?id=10626
10:04
annevk
forgot what it was
10:22
<zcorpan>
done
13:07
<hsivonen>
cool. Opera has added support for &TRADE;. More interop!
13:25
<annevk>
hsivonen, we did that in August 2006...
13:33
<jacobolus>
do events sent to dom elements work synchronously or asynchronously?
13:33
<jacobolus>
that is, if I'm in the middle of some function foo and I fire off an event to some div or svg shape, will the object handle that event before the rest of foo finishes, or after?
13:35
<annevk>
if you dispatch it yourself, before
13:35
<annevk>
otherwise, it depends
13:35
<jgraham>
Oh, dispatchEvent is sync?
13:35
<hsivonen>
annevk: I'm slow to notice, I guess. I have a note to make validator.nu whine about &TRADE; on Opera compat grounds. I guess I won't implement that feature.
13:36
<annevk>
oh wow
13:36
<jgraham>
?
13:37
<annevk>
jgraham, yes it's sync
13:38
<hsivonen>
and IE9 supports &apos;, so that took care of itself before I got around to make V.nu warn about that one
13:39
<annevk>
I wonder if IE dropped support for entities it had in addition to those that are in HTML5
13:39
<jgraham>
Oh, I guess it must be really
13:39
<jgraham>
Since it returns a value
13:39
<annevk>
I personally thought they made sense...
13:39
jgraham
has managed to mostly avoid dispatchEvent
13:40
<annevk>
jgraham, yeah, it's also "queue a task to dispatch an event" when you want async in spec lang
13:41
<jgraham>
Yeah. Seems liek you should be able to make async events from user code though
13:41
<annevk>
just use setTimeout
13:42
<jgraham>
Yeah, I guess that has the right effect
13:43
<jacobolus>
annevk: okay, thanks
13:44
<jacobolus>
annevk: I guess it's useful to have it sync, since it's not hard to wrap in a setTimeout if you want async &c.
13:45
<jacobolus>
oh, you said that
13:45
<jacobolus>
heh
13:51
<jacobolus>
maybe someone here has some advice then. I'm trying to build up some drag/drop for SVG objects.
13:51
<jacobolus>
my intended method is to have the whole svg area have a 'mousedown' handler that figures out the object being clicked using elementFromPoint, and then registers 'mousemove', 'mouseup' (again on the whole svg area) and I can then put the logic about how far the mouse should move before it's considered a drag, etc. in this one place.
13:51
<jacobolus>
to figure out what object is underneath the drag, the mousemove handler hides the dragged object, does elementFromPoint, and then shows the dragged object again, and I write my own logic to figure out when we're hovering over new drop targets, etc.
13:53
<jacobolus>
actually, hold that thought. I'm going to change seats. bb10
14:14
<jacobolus>
well, if anyone has any thoughts, I'm back, but now a question too. is there any way to get the mouse position when it's not over the viewport?
14:39
<annevk>
it's great that bug 10720 references Microsoft for good practice on not messing up standards
14:50
<Philip`>
Hmm, is test.w3.org down?
14:50
<Philip`>
Sadly I can't use downforeveryoneorjustme.com because it seems to return 404 for URLs starting with the string "test"
14:51
<jgraham>
Down for me, but dvcs is up
15:16
<jacobolus>
hmm. okay, another question then. is there any way to get a mouseup event if someone starts dragging while in the browser window and then releases the mouse somewhere outside?
15:17
<jacobolus>
annevk: ?
15:17
<nimbupani>
you can do a dragStart event on browsers that support it.
15:17
<jacobolus>
nimbupani: which is that?
15:18
<annevk>
jacobolus, maybe with setCapture stuff? (not standardized, not interoperable)
15:18
<nimbupani>
https://developer.mozilla.org/En/DragDrop/Drag_Operations
15:18
<annevk>
jacobolus, but don't really know
15:18
<jacobolus>
nimbupani: this isn't too promising http://www.quirksmode.org/blog/archives/2009/09/the_html5_drag.html
15:18
<jacobolus>
(at least the first couple sentences)
15:22
<nimbupani>
true, but Opera still does not support drag and drop http://caniuse.com/#feat=dragndrop
15:22
<annevk>
thanks Philip`
15:23
<jacobolus>
nimbupani: so do you know if there's a drag outside the browser window, and the user releases the mouse, will I get some event?
15:24
<nimbupani>
i saw such a demo once.
15:24
<nimbupani>
so I am looking
15:25
<nimbupani>
here you are http://londonwebstorenight.appspot.com/html5/index.html#slide20
15:26
<nimbupani>
it uses dragstart tho.
15:26
<nimbupani>
and I think it cancels events for other drag ops.
15:29
<jacobolus>
nimbupani: it looks like that's sending some info to the world that there's a thing being dragged that can be dropped outside
15:29
<jacobolus>
e.g. the interface implies I can drop on dock icons, the desktop, my chat client, etc.
15:29
<jacobolus>
which isn't quite what I want
15:30
<nimbupani>
not really it only says you can drag
15:30
<jacobolus>
I just want to be able to catch mouseup whether the user is over the browser or not
15:30
<nimbupani>
the dataTransfer is only when the drag ends
15:30
<nimbupani>
you can put whatever you want within function(e) {}
15:30
<jacobolus>
nimbupani: sure, but whether there's dataTransfer or not safari acts like I can drop stuff anywhere
15:31
<jacobolus>
nimbupani: e.g. in slide 18
15:31
<jacobolus>
or is that a safari bug?
15:31
<nimbupani>
oh i was only looking at it in chrome :)
15:32
<nimbupani>
oh i see it. you mean the + sign?
15:33
<nimbupani>
that only exists because there is a listener for it in that drop area.
15:33
<jacobolus>
nimbupani: I mean that if I drag the object into any other app on my system, it is implied that I can drop there
15:34
<nimbupani>
that is a feature of the mac itself I think.
15:34
<nimbupani>
oh no
15:34
<nimbupani>
it only works on safari
15:35
<jacobolus>
perhaps a safari bug?
15:35
<nimbupani>
or a safari feature? :P
15:35
<jacobolus>
is every draggable object supposed to be droppable on the world?
15:35
<jacobolus>
or just some of them?
15:35
<jacobolus>
clearly some should be
15:35
<jacobolus>
(so maybe everyone's buggy?)
15:36
<nimbupani>
ideally it should be droppable anywhere.
15:36
<jacobolus>
some objects should be
15:37
<jacobolus>
but I think there are some that should probably stay within the web app
15:37
<nimbupani>
i dont think you can set where it can be dropped though.
15:37
<nimbupani>
you can only add listeners to drop targets AFAIK
16:23
<Philip`>
hsivonen: Is there a known reason why htmlparser-statistics.jar might use lots more heap space than a standard htmlparser-1.2.2.jar?
16:32
<Philip`>
Ah, seems better when I give the JVM 2GB
16:42
<Philip`>
hsivonen: http://philip.html5.org/data/parser-stats-raw.xml.bz2 is the raw data
16:42
<Philip`>
(with a <processed> line for each page it started processing, and a <stats> line for each time the handler got called)
16:45
<Philip`>
http://parsetree.validator.nu/?doc=http%3A%2F%2Fwww.4clubs.ch%2Flms.php%3Fmodule%3Ddivers is quite a bit of nesting
16:51
<Philip`>
http://parsetree.validator.nu/?doc=http://www.vdofilm.com/webwinkel/contents/nl/d102_01.html does quite well on outer loop entries (around 60% down that page)
16:53
Philip`
isn't sure how to usefully analyse the data
17:07
<cardona507>
i just realized that the html-wg doesn't have digest mode for the mailing list
17:18
<annevk>
cardona507, don't you have filters and such for that?
17:18
<annevk>
digest seems so old fashioned; for the time when clients sucked
17:18
<cardona507>
i actually like diges
17:18
<cardona507>
heh
17:19
<cardona507>
*digest
17:19
<cardona507>
fine tuning my filter as we speak
18:43
<hober>
As I tweeted the other day: ISSUE-41 has gone to poll, and I know I only captured some of the relevant arguments in the zero-edit CP. I encourage everyone to respond to the poll if you feel you have a point not already covered in that CP. Thanks!
19:49
<variable>
does anyone know where the bulk of the security discussion about iframe srcdoc and sandbox attributes occured?
19:52
<variable>
I found BUG-100 which is to remove the element - but I can't find the original discussion
19:56
<Philip`>
I would assume the WHATWG list
19:57
<variable>
Philip`, I've been searching the archives but I can't find anything
19:57
<variable>
do you know the name of any threads?
20:01
<Philip`>
http://lists.w3.org/Archives/Public/public-html/2010Jan/0491.html may be relevant
20:01
<variable>
http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2008-July/015349.html is what I found
20:05
<Philip`>
http://lists.w3.org/Archives/Public/public-webapi/2008May/0326.html looks like one of the first suggestions for doc="..."
20:08
<variable>
Philip`, thanks - let me read that thread now
22:49
<Hixie>
my mac has started ignoring responses from my DHCP server
22:49
<Hixie>
wtf
22:49
<Hixie>
it connects to the wifi network, and i can see the DHCP response in a tcpdump of the network seen from the laptop
22:50
<Hixie>
but the IP remains set to one of the autoconfigured IPs