Weekly XHTML2 WG Teleconference

4 Jul 2007


See also: IRC log



+0138687aaaa, Steven, Roland_, Yamx, ShaneM, +1.208.761.aabb, markbirbeck
Rich, Alessio
Roland, Steven




<markbirbeck> ah :)

<markbirbeck> Trying to find the UK one.

<markbirbeck> ta

Welcome Roland

<scribe> Scribe: Steven

Roland: I work for IBM UK
... in the past worked on XForms
... and device independence
... now back to W3C to co-chair this group, and more interesting work

New member

Steven: A new company is joining
... a Japanese company

(machine running slowly, sorry)

scribe: Mitsue-links Co Ltd


Steven: Two new questionnaires
... one for forthcoming calls
... one for FtF



Shane: I thought it was due a week from today
... I will do it this week

Steven: I will send an apology that we are running late


Steven: There are a few possibilities. I've been talking to vcarious people round W3C
... b2evolution, moinmoin. Mediawiki may be installed sometime soon

Roland: Moinmoin is OK

Shane: We could restyle it, if we don't like the style

Roland: There might be a danger in being the first to use a new system

<scribe> ACTION: Steven request a moinmoin instance [recorded in http://www.w3.org/2007/07/04-xhtml-minutes.html#action01]

Role and access

Steven: Rich sent this message: http://lists.w3.org/Archives/Public/public-xhtml2/2007Jul/0001.html

Shane: The small issue is what we want to say about an access element that doesn't specify a key attribute

Steven: The UA is allowed to override the key anyway, so binding a key when it is not specified should be fine

Roland: Some UAs will bind to softkeys

Steven: And UAs will likely bind well-knoiw roles to standard keys
... So I think we agree with Rich's comment

Shane: There are two optional attributes as well, targetrole and targetid.
... If neither is specified, what does it mean?
... It could mean that the current focus point is the place where the event is raised
... (no use case for that though)
... or say "undefined" or "nothing happens".
... Any opinions?

Roland: I wouldn't like it to be undefined.

Steven: Me neither

Shane: Or we could require one to be specified
... but we couldn't validate that

Roland: Relax NG could

Shane: But not the W3C validator

Steven: True, but that shouldn't drive our design

Shane: So what do we think?
... It could be useful to be able to raise an event without specifying target

Steven: But it would be just as easy by saying what the target is

Shane: Well, there is the focus issue
... does raising a DOMActivate also change the focus?

Mark: I think it should get focus as well

Shane: So do I

Steven: Why do you care?

Shane: Well, if the link goes to another window

Mark: It depends on how you interpret DOMActivate
... I think DOM2 Events is too vague to be sure whether the control has focus already
... Is DOMACtivate a notification or an action?

Steven: But you can cancel a DOMActivate
... which stops default processing

Roland: DOMFocusIn is not cancellable on the other hand
... DOMFocusIn is not cancellable on the other hand
... which suggests that you can't use DOMFocusIn to set focus
... it is only notification that it has happenend

Shane: I noticed that; the only alternative is 'focus', but that is only valid on a small subset of elements

"This event is valid for the following elements: LABEL, INPUT, SELECT, TEXTAREA, and BUTTON"

Mark: The only two use cases we know of are "Give focus" and "Give focus and activate"

Steven: So maybe "raise" is too general
... maybe we should just say those two things
... or say that focus is always given, and then 'raise' gives an additional optional event

Mark: Or an attribute that is default giving focus, and allows you to specify activations
... the default behaviour for role and id is surely different
... for role it is always focus
... but for id it doesn't have to

Shane: I don't want to special case it
... Shouldn't targetid be a list?

Mark: Cute

Shane: Would it be the next id in the list, or the next id in the list in document sequence?

Mark: So tab gets you onto a route now (with next focus) and this way any key could put you on a route

Shane: Shall I spec it up?

Steven: Send it to the list, I wouldn't want to add it to the spec before Rich has seen it

Shane: I think we agreed that an unsepcified key will then be assigned by the UA, and you must either specify a targetid or a targetrole

Steven: Yes
... Out of time. Let's talk longdesc next week (http://lists.w3.org/Archives/Public/www-html/2007Jul/thread.html#msg0)
... since we have a rather good solution already

Roland: 3.1.1 says something about id attribute; does that include xml:id

Steven: We ought to accomodate xml:id; use whatever XML uses to describe the id-ness of an element, and not restrict it to id attributes


Summary of Action Items

[NEW] ACTION: Steven request a moinmoin instance [recorded in http://www.w3.org/2007/07/04-xhtml-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.128 (CVS log)
$Date: 2008/03/13 16:20:38 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.128  of Date: 2007/02/23 21:38:13  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/element/focus point/
Succeeded: s/ he / the /
Succeeded: s/SH/Sh/
Succeeded: s/longdes/longdesc/
Succeeded: s/Topics:/Topic:/
FAILED: s/tweo/two/
Succeeded: s/Tweo/Two/
Found Scribe: Steven
Inferring ScribeNick: Steven
Default Present: +0138687aaaa, Steven, Roland_, Yamx, ShaneM, +1.208.761.aabb, markbirbeck
Present: +0138687aaaa Steven Roland_ Yamx ShaneM +1.208.761.aabb markbirbeck
Regrets: Rich Alessio
Agenda: http://lists.w3.org/Archives/Public/public-xhtml2/2007Jul/0002
Got date from IRC log name: 4 Jul 2007
Guessing minutes URL: http://www.w3.org/2007/07/04-xhtml-minutes.html
People with action items: steven
[End of scribe.perl diagnostic output]