Weekly XHTML2 WG Teleconference -- 6 Jun 2007

<yamx> this is yam

<scribe> SCRIBE: ShaneM

<scribe> ACTION: Items to Alessio completed is survey of browsers and access element [recorded in http://www.w3.org/2007/06/06-xhtml-minutes.html#action01]

role attribute module

<alessio> new link for test: http://www.terrafertile.ch/w3/xhtml2/

<scribe> new draft is up at http://www.w3.org/MarkUp/2007/ED-xhtml-role-20070606/Overview.html

<scribe> ACTION: Tina to review the xhtml-role document and send in comments. [recorded in http://www.w3.org/2007/06/06-xhtml-minutes.html#action02]

<oedipus> WAI ARIA Roles: http://www.w3.org/TR/2007/WD-aria-role-20070601/

<oedipus> States & Properties Module for ARIA: http://www.w3.org/TR/2007/WD-aria-state-20070601/

Draft says: The role attribute takes as its value one or more white-space separated QNames. Any non-qualified value MUST be interpreted in the XHTML namespace, and MUST be taken from the list defined in this section.

Discussion about interpretation of non-prefixed names

Mark is suggesting we should preclude the interpretation of non-prefixed values that are NOT in our list. They should be ignored.

Schema AND DTD implementation can define only the valuies in the list

oh... no it can't. Schema can...

Mark points out that we may need to permit the collection to be extended in the context of another host language.

Mark suggests that we NOT restrict the collection of non-qualified values. That we just say this set may be extended in the future.

Rich asks if this change would require that we prefix xhtml namespace values at the source level. No.

<oedipus> Embedding Accessibility Role and State Metadata in HTML: http://www.w3.org/WAI/PF/embedding-20061212.html

Are there additional role values that SHOULD be included in this specification before we go to last call? (e.g. copyright) or should we rely upon existing taxonomies (e.g., dc:copyright)

<oedipus> Embedding Accessibility Role and State Metadata in HTML (alternate URI): http://www.w3.org/WAI/PF/adaptable/HTML4/embedding-20061212.html

If there are additional values that are needed in the context of other specifications (e.g., HTML5) then we should find a way to put them in.

<scribe> ACTION: Shane to add text to ensure it is clear that the list is extensible in the XHTML namespace if other groups need it. [recorded in http://www.w3.org/2007/06/06-xhtml-minutes.html#action03]

<oedipus> need something formal for HTML WG - i have repeatedly pointed the WG to the WAI PF document on Embedding Role and State Metadata in HTML Documents, but no one's bitten

Tina says what we have a nice small set that is easy to implement. We need to be careful to only extend it in limited, well thought out ways.

<scribe> ACTION: Shane to set up last call for xhtml-role by 15 June. [recorded in http://www.w3.org/2007/06/06-xhtml-minutes.html#action04]

CURIES

tabled for another week. Mark and Shane to work off line

Nice thing is that curies are a superset of qnames, so we can "upgrade" later.

xhtml basic

yam sent some mail about the testfest from oma.

the inputmode was NOT tested. inputmode is just a hint to the user agent. testing guidelines in OMA do not have people testing "hints" because they are not really required.

the question is: is inputmode interoperability testing required for getting basic out the door?

<scribe> ACTION: Shane to figure out if a "hint" is an interoperbility requirement. [recorded in http://www.w3.org/2007/06/06-xhtml-minutes.html#action05]

RDFa task force / module

Mark reports some discussions about the @href everywhere debate and how it is progressing

Current thinking is that we re-introduce the @resource attribute, and can use that to identify a URI. In the context of a host language, @href can duplicate / extend the functionality.

Editorially, we need to have the xhtml-rdfa document be RDFa in an XHTML context. And then a separate spec (the RDFa syntax document) needs to be updated so it is host-language neutral.

Shane asks if this syntax doc is a note or a rec?

Mark thinks that it specifies the processing rules. So to that end it must be normative. So it must be rec-track.

<scribe> ACTION: Steven to integrate the RDFa syntax document into the XHTML WG publication system. [recorded in http://www.w3.org/2007/06/06-xhtml-minutes.html#action06]

<Steven> Yes, it will be rec track

<Steven> The problem with integration is that it is in XMLSPEC at the moment

<Steven> and Ben is another editor of it

no worries - I can fix that in a heartbeat.

<Steven> (he created the initial version)

<markbirbeck> http://www.w3.org/2006/07/SWD/wiki/RDFa/ProposedStructure

<markbirbeck> Summary of the proposal I just described.

access module draft

There is a link: http://www.w3.org/MarkUp/2007/ED-xhtml-access-20070606/

Rich asked about events: Triggering an access key defined in an access element gives focus to the next element in navigation order from the current focus that has the referenced role or id value. Note that it is possible to associate additional event handlers with target which might then perform additional actions once focus is changed.

Rich is concerned that we might want to have an access element take an explicit attribute for specifying action (default is focus, but maybe they want to specify activate)

<oedipus> +1 to RichS - need to play audio triggered by assistive technology encountering cue-before, cue-after, play-during, etc. -- the purely aural canvas; the AT has to trigger the event, and the UA needs to use its default backplane audio renderer

<Tina> Thank you all.

<oedipus> i can help i know RSSAgent

<alessio> bye all :)

<oedipus> ok, the minutes are available as hypertext at: http://www.w3.org/2007/06/06-xhtml-minutes.html

<oedipus> is that sufficient, ShaneM?

- DRAFT -

Weekly XHTML2 WG Teleconference

6 Jun 2007

Attendees

Contents

role attribute module

CURIES

xhtml basic

RDFa task force / module

access module draft

Summary of Action Items

Scribe.perl diagnostic output