W3C

- DRAFT -

XHTML2 Working Group Teleconference

08 Apr 2009

Agenda

See also: IRC log

Attendees

Present
Alessio, Gregory_IRC_only, Markus, Roland, ShaneM, Tina
Regrets
Chair
Roland
Scribe
ShaneM

Contents


 

 

<trackbot> Date: 08 April 2009

<scribe> Scribe: ShaneM

Action Items

<Roland> http://www.w3.org/MarkUp/tracker/actions/pendingreview

Actions 42 and 47 continue.

Action 58 and 59 continue

<trackbot> Sorry, couldn't find user - 58

Actions 58 and 59 continue.

Discussion about Action 65

Gregory submitted http://lists.w3.org/Archives/Public/public-xhtml2/2009Mar/0102.html

<Roland> http://lists.w3.org/Archives/Public/public-xhtml2/2009Mar/0103.html

I missed these when I was making edits.

Roland is happy with examples.

Shane will update the spec with option 2 from message 102.

<oedipus> examples also on the wg's wiki

ShaneM: note that aria is not presently part of xhtml2

<oedipus> shane, i used aria to show that without explicit binding using aria, there isn't a means of binding a P with a list, save for embedding the list in the P

Tina: asks whether we *can* include aria in XHTML 2. The ARIA spec uses something *role-like* that is not XHTML role.

<oedipus> aria example ~can be used in a best practices

close ACTION-65

<trackbot> ACTION-65 - draft example of P with prose list and structural list closed

<oedipus> tina, i have asked for clarification on role-like ARIA as opposed to ARIA as an extension of Role Module

oedipus: didn't you ask for that clarification months ago?

<oedipus> answer was that the Role module was thought to be behind the ARIA spec in process

<oedipus> shane, yes, i did

<oedipus> shane, and continue to do so - will again today at PF meeeting if i make it

<oedipus> i have an action item open on it in PF

If Role were at CR then this would not be an issue, right?

<oedipus> GJR speaking as individual not happy that ARIA spec states role-like, rather than citing Role, but the argument was can't cite a spec that isn't a recommendation as normative

<oedipus> GJR: er, um - i don't like the fudging in ARIA - steven, can you add that to your review of ARIA?

<oedipus> that ARIA is derived from Role, uses its extension mechanism, etc.

<oedipus> shanem, thanks - didn't know

<oedipus> will ping steven about comment - ARIA comments due 9 April 2009

<Roland> http://www.w3.org/MarkUp/tracker/actions/42

<oedipus> shanem, steven has the action to review ARIA, right?

<Roland> http://lists.w3.org/Archives/Public/public-xhtml2/2009Mar/0084.html

Wow - that was a long time ago.

Action 51 lightly review WAI-ARIA

<Zakim> oedipus, you wanted to ask if role="form" belongs in ARIA or Role Module's predefined roles (will cross post to public-xhtml2)

<oedipus> currently, the closest thing is role="search"

Tine read through the spec a couple of times. From the XHTML 2 perspective the only real concern is the role module.

<oedipus> but there are a lot of other types of forms that people use and need efficient access to...

As far as the spec goes Tina says it is fine - should not impact us too much.

<oedipus> applets from server ("security" measures such as captchas), ecommerce, blogging, networking, etc.

<oedipus> need to have something that binds them together as a form when form a mixture of technologies (tag soup - part declarative markup, part scripting, so that form controls are outside of the FORM element (yes, this is backwards looking)

ShaneM: should we submit a comment that says the WAI-ARIA spec needs to normatively refernece the XHTML Role spec?

<oedipus> GJR: speaking as a member of this WG, YES

Tina says maybe - depends on where we are in the pipeline?

<Zakim> oedipus, you wanted to ask if this messes with discussion of extending predefined roles at last virtual f2f?

ShaneM: I think so too.

markus was concerned about extending roles.

<oedipus> so, comments so far: 1) normatively cite Role Module; 2)

<oedipus> i understand markus' concerns

ShaneM pointed out that our role vocab includes ALL of the aria terms already.

<oedipus> yes, which is why i could try and shoehorn "form" into ARIA

http://www.w3.org/TR/wai-aria/

<oedipus> as a region

<oedipus> will cross-post my post on the topic to pfwg public comments list and xhtml2 list

<oedipus> i think lack of role="form" in ARIA is a huge hole and i can't figure out why i didn't notice that earlier

Not sure when the comment deadline is.

<oedipus> 9 April 2009 i believe

markus: supports getting them to try to use the real module normatively.

<oedipus> BUT, if roland emails janina@rednote.net and wai-liaison@w3.org to say a review from the WG is on the way, we can buy some time

shane asks who will take the action to send in the formal comment?

<oedipus> http://lists.w3.org/Archives/Public/public-pfwg-comments (from memory, so caveat lector)

<oedipus> shane, i thought the action was already on steven

yes but we are trying to optimize here.

Real deadline is 17 April 2009 so that's okay. Tina will send in the comment.

<oedipus> ok

Gregory has raised the issue of whether we need a predefined value of "form" for the role attribute.

We have an agenda item for this already. So let's discuss it in that context?

Additional values for the role attribute

<Roland> ref: discussion around <address> durinf virtual f2f

Originally came up when we were discussing address and whether there was a way to indicate the role of something was an "address"

I think we can add things to the vocab document at any time really - especially additional values for role.

but when adding terms we need to carefully craft the rdf so we can get to what it really means.

<oedipus> if that is kosher, i can live with it, under whatever restrictions the WG decides are proper

XForms does not have a "form" element, so yes a form role might make a lot of sense.

We dont have a concept of separate vocabularies for role that apply to different versions of XHTML, so a value of form would be redundant in HTML / XHTML 1.

But that's probably okay.

We already have a role of search - would we have BOTH?

<oedipus> it is needed for repair of mashups, etc.

<oedipus> yes, we need BOTH!!!

<oedipus> search is search - not a captcha form, an order form, a comment form

<oedipus> ARIA has values for form controls that parallell XForms, but most forms today are not validly constructed

<oedipus> my bank uses javascripted links for all of its form submission buttons, and they are not in the form's tabindex and not bound to the form save visually, and the whole thing happens to be in a DIV

<oedipus> search is a "special case" of "form

Shane says he is in favor of adding the value of "form". Tina asks that she is not really sure what it would mean in an XHTML 1 context.

Roland suggests we confer with the XForms people first.

<oedipus> it is a meta-fieldset for a collection of form controls -

<scribe> ACTION: Roland to confer with XForms group about adding a form value to the default role vocabulary. [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action01]

<trackbot> Created ACTION-71 - Confer with XForms group about adding a form value to the default role vocabulary. [on Roland Merrick - due 2009-04-15].

Gregory - do you have suggestions for the names for the notes terms?

<oedipus> ARIA needs it not for tomorrow's pages or XHTML2, but for today's and yesterday's content and delivery mechanisms

<alessio> agree, gregory

<oedipus> shanem, yes, but i will have to dig up the post i sent to PF on the topic wherein things are explained succinctly and coherently

oedipus: I assumed that was the case... we are just going to ensure we cover our bases with XForms too.

<oedipus> shanem, ok - just making sure the request was clear

<oedipus> also, app may use role="form" to load an expert handler or a plug-in such as ubiquity-xforms

I think it is a slippery slope to start adding terms for specialized classes of general items. Like sub-types of notes...

<oedipus> in XHTML2 would only be in legacy module

what would only be in legacy module?

<oedipus> 2 diff notes because 2 different roles: one documents text the other comments upon it

<oedipus> i meant the only reason one would use role="form" is in a legacy-enabled document

Roland: we already have complementary and notes to delineate different supplemental data. We could go a bit far if we are not careful.

<oedipus> ok, but can i have role="note complementary"

oedipus: I disagree. Since XForms does not really have a container, i could see using role="form" to indicate that a whole part of a page is a form.

<oedipus> syntax may be off (on dumb terminal)

<oedipus> i am happy to be wrong, then, shane

back on note values.... having multiple roles is of course fine. Do you think it is an issue Gregory?

<oedipus> in my mind, but i'm willing to post my previous suggestion to PF to the XHTML2 list - it is not make-or-break, as long as one can qualify a role as both complementary and a note

<oedipus> and let the WG decide

Markus points out there is a typo in the vocab document - complementary describes itself as secondary

<oedipus> i think it ok to drop the 2 kinds of notes discussion - multiple values need to be used

<scribe> ACTION: ShaneM to update vocab document to fix typo about Complementary [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action02]

<trackbot> Sorry, couldn't find user - ShaneM

<oedipus> shanem and markus, the vocab doc needs a re-sync with the LC draft of ARIA

<oedipus> i have an action item to nag you, shane, about updating the vocab doc in concert with MCooper

<oedipus> missing ARIA roles such as "math"

<oedipus> some role definitions have been tweaked, as well

MCooper owes me some updates so I can do it.

<oedipus> right

Markus points out that there are many items in the vocab document that are indicated to be part of rel/rev but are useful as role values. how do we indicate it is okay to use those?

<oedipus> i don't know his time-frame, but i think that LC is a benchmark and that the vocab document should be updated to reflect what is in the LC draft of ARIA 1.0

oedipus: of course - I have been harassing MCooper about it. will get it done ASAP.

ShaneM asks if there is any problem permitting role="last" for example?

<oedipus> need to tweak introductory verbiage to explicitly state that while the following roles have traditionally been used (or limited to use) in the HEAD of a document, using the rel and rev mechanism is only one means of using these roles, which can be used inline

<oedipus> or words to better effect <grin>

Roland thinks it might be useless but it is not really a problem.

<oedipus> COA

I think I should just improve the text for section 2 and update the underlying RDFa so that they are all also legal role values.

<oedipus> should i start a role best practices wiki page on which to develop verbiage?

oedipus: Yes, good idea.

<oedipus> ACTION: Gregory - prepare Best Practice for Role Module wiki space and begin populating with examples [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action03]

<trackbot> Created ACTION-72 - - prepare Best Practice for Role Module wiki space and begin populating with examples [on Gregory Rosmaita - due 2009-04-15].

<scribe> ACTION: ShaneM to update introduction to section 2 of the vocab document AND update the RDFa so that all the terms are associated with role. [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action04]

<trackbot> Sorry, couldn't find user - ShaneM

<scribe> ACTION: Shane to update introduction to section 2 of the vocab document AND update the RDFa so that all the terms are associated with role. [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action05]

<trackbot> Created ACTION-73 - Update introduction to section 2 of the vocab document AND update the RDFa so that all the terms are associated with role. [on Shane McCarron - due 2009-04-15].

<Roland> http://lists.w3.org/Archives/Public/public-xhtml2/2009Apr/0002.html

Roland reminds the group that this arose because of the "address" element and a potential role of address. Need to ponder that more.

title and meta

Meta content model was reverted to xhtml 1 content model, so this is overcome by events.

<scribe> ACTION: Shane to reply to message 0002 [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action06]

<trackbot> Created ACTION-74 - Reply to message 0002 [on Shane McCarron - due 2009-04-15].

HTML 5 / XHTML 2 Merger?

Tina asks if this is really happening or is it just talk?

Roland indicates that Steven and Sam Ruby have actions to discuss this, but there have been no decisions.

Nothing has been done about this thus far, as far as anyone on the call knows.

There was some data in the vF2F minutes from two weeks ago.

<oedipus> should be at http://www.w3.org/2009/03/26-xhtml-minutes.html#item01 i believe

<oedipus> either item01 or item02

<oedipus> no problem

<alessio> bye all

Summary of Action Items

[NEW] ACTION: Gregory - prepare Best Practice for Role Module wiki space and begin populating with examples [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action03]
[NEW] ACTION: Roland to confer with XForms group about adding a form value to the default role vocabulary. [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action01]
[NEW] ACTION: Shane to reply to message 0002 [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action06]
[NEW] ACTION: Shane to update introduction to section 2 of the vocab document AND update the RDFa so that all the terms are associated with role. [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action05]
[NEW] ACTION: ShaneM to update introduction to section 2 of the vocab document AND update the RDFa so that all the terms are associated with role. [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action04]
[NEW] ACTION: ShaneM to update vocab document to fix typo about Complementary [recorded in http://www.w3.org/2009/04/08-xhtml-minutes.html#action02]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2009/04/08 14:53:47 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
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)

Succeeded: s/ACtion/Action/
Succeeded: s/CYA/COA/
Succeeded: s/best practices role/role best practices/
Found Scribe: ShaneM
Inferring ScribeNick: ShaneM
Default Present: Roland, ShaneM, Markus, +46.8.64.5.aaaa, Alessio
Present: Alessio Gregory_IRC_only Markus Roland ShaneM Tina
Agenda: http://lists.w3.org/Archives/Public/public-xhtml2/2009Apr/0011.html
Found Date: 08 Apr 2009
Guessing minutes URL: http://www.w3.org/2009/04/08-xhtml-minutes.html
People with action items: gregory roland shane shanem

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


[End of scribe.perl diagnostic output]