See also: IRC log
<ShaneM> I dont think you are late. I think we start in 15
<Tina> My timing *and* my time zone is messed up, I'm afraid
<ShaneM> why does it think it is already started I wonder
<Steven-eee> Previous: http://www.w3.org/2008/04/02-xhtml-minutes
<yamx> ??P5 is yamx
<oedipus> GJR: having very nasty computer problems -- will call in a.s.a.p. -- just want to communicate that there is no obstacle or objection from PF on moving Role forward if that is what the xHTML2 WG wants -- Role transition gets a big plus one from me (in case i miss the discussion)
<Tina> :)
<ShaneM> scribe: shane
<ShaneM> err... Scribe: ShaneM
<ShaneM> Scribe: ShaneM
<Roland> Scribe: ShaneM
<scribe> No progress yet. Seems like there is some softening of their position. Need a formal response from them.
<scribe> No progress yet.
Should have a contract resolved this week.
Public working draft is updated.
Request to publish has been made.
Values of role in the XHTML vocabulary
see http://www.w3.org/1999/xhtml/vocab
Roland: what do we need to do here?
ShaneM: Let the ARIA group do it
markbirbeck: values are going to be used quite widely. We might need to ensure that values are "correct"
Roland: perhaps the vocabulary should be rec track
ShaneM: while the document itself is not rec-track, the input to it IS.
Roland: At least there should be an appropriate process for ensuring that things only go in there after appropriate review.
oedipus: I have an action (from the PFWG) to
review Mark's comments on the vocabulary.
... there was a mail from the PF chair that was private asking about how the
vocab document evolved and whether it was tied to the role document.
Roland: need to ensure there is no collisions between the XHTML values and the ARIA values. ShaneM says that he has already worked that out with Michael Cooper.
oedipus: Any redundant values should be left to the XHTML Role document.
markbirbeck: concerned that the ARIA group can
put their values straight into the vocabulary.
... the "multiple word" approach to terms in the ARIA document sort of misses
the point of role. Should be "alert dialog", not "alertdialog".
q
<Steven-eee> as long as it's not aria:alert aria:dialog
oedipus: Some of the values in the ARIA spec
are related to how assistive technologies work in the marketplace today and
are driven my market realities.
... values like "slider" are for things we have inherited. Not for the
future. This is related to ARIA 1.0.
<Tina> We should be extra careful with inherited techniques, and only include them in a standard if they are also good ideas.
Roland: perhaps we could specify some "pure roles" in the XHTML Role superior vocabulary so that we can be more forward looking while the ARIA values are designed to help their marketplace.
ShaneM: The group agreed at the last f2f to include the values in the vocabulary space. We did not put any conditions on how those values would be included.
markbirbeck: points out that role values are not inherited becuase there was no role attribute. What we have is a need to map values to legacy implementations to support them.
Roland: There is clearly work to do here. We need a complete list.
<Tina> If they make sense, yes. We shouldn't do so just because they exist. This is important.
oedipus: If there is a disconnect about what we agreed with the PFWG?
Roland: We deferred to the ARIA guys saying
"you define what you would like in the vocabulary, and we will put them into
the space". oedipus agrees that is what the understanding was.
... The issue is that the lists are somewhat disjoint - they don't look like
they were produced by the same group of people.
oedipus: I volunteer to work with whomoever on this. I am in both groups.
Roland: A separate interesting topic is how do we manage vocabularies in general...
Roland: definition for values of @rel/@rev
... There is some IETF work in this space. And the HTML 5 group seems to be
looking at it too.
<Roland> http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0013.html
TAG work on sem web data in HTTP headers - the LINK header
There needs to be a way of managing the collection of values that go into a vocabulary so we don't end up with a mess.
ShaneM: What about CURIEs - doesn't that help?
Roland: Not really because there needs to be agreement on basic semantics so vocabularies get used.
<Roland> http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0026.html
The Math WG is trying to ensure that the entity collection is consistent among all groups that care.
We absolutely care about the collection of entities, since XHTML 1 / 1.1 / 2 all define consistent collections today.
<scribe> ACTION: Shane to ensure there are no inconsistencies between the MathML entity set and the XHTML entity set. [recorded in http://www.w3.org/2008/04/09-xhtml-minutes.html#action01]
<Steven-eee> I seem to remember we did/tried to do just that long ago
oedipus: It is not clear that the HTML 5 group has actually made any changes in that space.
<Steven-eee> I remember long discussions about values of entities being used in MathML
<Steven-eee> (where is Masaysu when you need him)
ShaneM: There should be consistency, but I am sure that we already addressed this years ago.
<Steven-eee> ok
ShaneM: we have the definitivbe collection as part of M12N, and we harmonized that with the MathML collection. But we do not defer to them 'cause our collection is the base definition for the web.
oedipus: Will check with the PFWG, but he is pretty sure that AT vendors rely upon these definitions too. I will try to get confirmation.
Roland: is there something we should be doing with @rel values too?
<markbirbeck> <link rel="license" ... />
<markbirbeck> <link rel="[foaf:knows]" ... />
Roland: What would happen if we do not harmonize?
markbirbeck: there is some risk, but RDFa has a lot of momentum so it is likely XHTML / RDFa will dominate.
<Steven-eee> shouldn' that be <link rel="foaf:knows" ...?
<Steven-eee> No need for [] in rel
Mark was pointing out that there was an alternate potential syntax historically.
markbirbeck: we should make it so important @rel values are in the vocabulary and coordinate with these other groups.
<scribe> ACTION: Shane and Mark should open up this dialog and see if we can't get the other @rel people to come play with us [recorded in http://www.w3.org/2008/04/09-xhtml-minutes.html#action02]
oedipus: Other groups that have seen the access module are looking for mappings to specific APIs.
ARIA Provides appendices about mapping ARIA stuff to APIs.
Roland: Access is not primarily about assistive technology. It is there to support the mobile community too.
oedipus: True. But from the point of view of AT people how to cascade the access effects portably among implementations.
ShaneM: there's nothing in the access module to bind to. It is only useful in conjunction with the XML Event module (and XML Handlers).
Perhaps there is some baggage associated with this in the accessibility community.
People are looking for cascade rules.
<scribe> ACTION: Gregory to ensure that we get feedback about needing cascade rules in xhtml-access [recorded in http://www.w3.org/2008/04/09-xhtml-minutes.html#action03]
<yamx> ??P5 is yamx
<RolandMerrick> Scribe: ShaneM
This is scribe.perl Revision: 1.133 of Date: 2008/01/18 18:48:51 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/non-ARIA values/values/ Succeeded: s/XHTML/ARIA/ Succeeded: s/hime/him/ Succeeded: s/l ink/link/ Found Scribe: shane Found Scribe: ShaneM Found Scribe: ShaneM Inferring ScribeNick: ShaneM Found Scribe: ShaneM Scribes: shane, ShaneM Default Present: Roland, ShaneM, markbirbeck, Gregory_Rosmaita Present: Roland Shane Mark Gregory Yam Steven(IRC) Tina_(IRC) Regrets: Alessio Agenda: http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0014.html Got date from IRC log name: 09 Apr 2008 Guessing minutes URL: http://www.w3.org/2008/04/09-xhtml-minutes.html People with action items: gregory shane WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]