See also: IRC log
<trackbot> Date: 09 September 2009
<scribe> ACTION: smaug to check gecko's key event handling during composition events [recorded in http://www.w3.org/2009/09/09-webapps-minutes.html#action01]
<trackbot> Created ACTION-401 - Check gecko's key event handling during composition events [on Olli Pettay - due 2009-09-16].
shepazu: we have a new WD
... asking internalization people to comment about it
shepazu: did some testing on select event and made some changing to the WD
shepazu: it is only available on
... and I added .detail
smaug .detail is long!
shepazu: hard to define when
mousedown/up generates a click
... with select I'd like to know what has been selected
Travis: just use HTML5 for that
shepazu: do we want to overload
select event for mouse and keyboard selection?
... I'll have to look at HTML5 about this
... what if there is an image selected?
Travis: it is included
shepazu: what happens when pasting that data to somewhere?
Travis: UA must put the data in all possible formats to clipboard
<Travis> Selection in HTML5: http://dev.w3.org/html5/spec/Overview.html#selection
shepazu: yeah, relates to
... there are whole range of options for copy-paste and data handling
... so is it worth worth saying author can get the text somehow?
Travis: I'd just rather use other mechanism
shepazu: I'll probably add
reference to HTML5
... does the select event work everywhere, or only with form elements?
Travis: IE fires an event to body or document
shepazu: maybe I'll say that it is up to UA to decide the target
shepazu: there is event order defined for mouse, key and composition events
shepazu: ... in the latest draft
shepazu: I added a list of
events, I'm planning to add diagrams
... is this too much detail, or not enough?
Travis: certainly helpful
smaug: I agree
shepazu: will get more complicated once I start define the interaction between different event types
shepazu: I was thinking to use webidl in the spec
Travis: do it!
... make sure to use the latest editor's draft
smaug: I agree, use webidl
Travis: why are we adding namespaces to the spec?
shepazu: I'd have to go and look
back about de-centralization
... having namespace support allows people to add their own events
... JS libraries can add their own events
... Anne would like to remove namespaces
<annevk> I brought it up a few times, I'm certainly not the only one
<annevk> e.g. I'm pretty sure Maciej agrees with me
Travis: so web pages can implement their own events without interfering with others
<annevk> and Ian, etc. see the old threads
Travis: though, isn't prefix + event name the same thing.
as namespace + event name
Travis: I'll have our team reviewing the draft
shepazu: the draft has been around for a long time, and namespaces have been there
Travis: I think it is ok to keep
... vendors could continue to use prefixes, or use namespaces
<annevk> shepazu, objections to namespaces have likewise been around for a long time
<annevk> that's not a reason :)
<shepazu> annevk, people use them in production code, that is a reason
<annevk> not on web sites
<annevk> there's no browser that is shipping with namespace support
<annevk> there's no reason for us to implement them
<shepazu> annevk, if you want to join the telcon, then do so
(the world is not just about browser)
(the world is not just about browsers)
shepazu: there is the case where people are using keyboard layout
<annevk> shepazu, midnight here
shepazu: and we did decide that
we don't cover that
... but I added something about keyboard layout
<annevk> (not sure what that has to do with anything)
shepazu: I have the idea for
another spec to define keyboard layout handling in web
... the keyboard layout section is just a hint for authors
... any objections to put that
Travis: I like it
shepazu: I'll mark it as informative
smaug: DOMContentLoaded is already defined in HTML5
shepazu: SVG would benefit
... there is a good reason to have events for "everything is in place" and "page is rendered"
Travis: we were looking at
... HTML5 doesn't define whether DOMContentLoaded before or after loading CSS
... jQuery people wanted to have an event for "CSS loaded"
... maybe CSSContentLoaded?
smaug: there is the readyState
... maybe that kind of mechanism could be reused
Travis: I'll post something to mailing list
shepazu: I'd call it StyleContentLoaded
smaug: StyleContentLoaded sounds useful, but should go to HTML5
Travis: about resize event
... only IE supports resize on elements
... IE fires it for example when <div> is resized.
shepazu: could you send an email to the list
smaug: resize on elements might be implement on gecko soon
Travis: .detail on resize?
smaug: on gecko they are just normal events
Travis: should composition events bubble?
smaug: I think they do on gecko
Travis: what about cancelling them?
shepazu: maybe they can't be
... not sure if browser can control IME
Travis: you definitely don't want to cancel the compositionend
shepazu: but what if someone
... should ask masayuki about composition events
... there many different kinds of IMEs
... maybe compostionupdate should not be cancelable, others should
compositionstart should prevent IME?
... default action is to launch IME
... compostionupdate could be not cancelable for now, unless there is some usecase
... end would be cancelable
smaug: would be good to be able cancel compositionend
shepazu: I think we agreed about canceling events
Travis: add a row for all the events about default actions
<scribe> ACTION: travis to send email about StyleContentLoaded [recorded in http://www.w3.org/2009/09/09-webapps-minutes.html#action02]
<trackbot> Created ACTION-402 - Send email about StyleContentLoaded [on Travis Leithead - due 2009-09-16].
<scribe> ACTION: travis to send email about resize event [recorded in http://www.w3.org/2009/09/09-webapps-minutes.html#action03]
<trackbot> Created ACTION-403 - Send email about resize event [on Travis Leithead - due 2009-09-16].
<scribe> ACTION: travis to send email about namespaces, and vendor prefixes [recorded in http://www.w3.org/2009/09/09-webapps-minutes.html#action04]
<trackbot> Created ACTION-404 - Send email about namespaces, and vendor prefixes [on Travis Leithead - due 2009-09-16].
<shepazu> trackbot, end telcon
This is scribe.perl Revision: 1.135 of Date: 2009/03/02 03:52:20 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) No ScribeNick specified. Guessing ScribeNick: smaug Inferring Scribes: smaug Default Present: Shepazu, smaug, [Microsoft] Present: Shepazu smaug [Microsoft] Agenda: http://lists.w3.org/Archives/Public/www-dom/2009JulSep/0265.html Found Date: 09 Sep 2009 Guessing minutes URL: http://www.w3.org/2009/09/09-webapps-minutes.html People with action items: smaug travis WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]