XHTML2 WG Weekly Teleconference

14 May 2008


See also: IRC log



Roland, ShaneM, Gregory_Rosmaita, Steven, Alessio, yamx, Tina_on_IRC


Outstanding Reviews

RM: only 2 people and myself at last week's meeting -- nothing happened
... review for XML Base?

Steven: no

RM: June face 2 face -- has everyone filled in form?

Steven action to review XML Base by 30 June continued

<Roland> F2F Registration form: http://www.w3.org/2002/09/wbs/32107/xhtml2ftf200806/

<Tina> The f2f has fallen out of my list. Will get to it.

<Steven> Maybe do it entirely remotely?

RM: reads face2face page -- only 3 people listed as attendees
... worthwhile? what are people's intentions?
... yam will be on IRC rather than in person

GJR: cannot attend

RM: steven, shane and roland will attend

GJR: yes by phone and IRC

Alessio: will attend remotely

Steven: do entire F2F remotely?

Tina: will attend remotely

RM: MarkB unknown status

SM: MarkB travelling most of may, disinclined to travel in june

<ShaneM> oops

<Steven> Mark says he can't attend in person, can attend remotely

RM: once get mark's answer, will consider moving dates, moving location, keeping dates and location, doing whole remotely or whatever circumstances dictate
... question: people can do remotely -- is the problem the location?

GJR: no

Alessio: financial and logistic problem

<Tina> Ditto

GJR cannot attend due to health problems and such

<Steven> Mark says the problem is availability of time

<Steven> I concur that it is silly for three to travel

RM: only 3 people attending doesn't sound wise for those 3 to travel - might as well all be equal and have a virtual f2f

<Steven> We should try and set up a remote meeting then

GJR: plus 1 to SP

RM: shall we resolve to have virtual f2f?

<Steven> +1

GJR: yes

<ShaneM> +1 I guess

<Tina> ... not *alot*, anyway ... ;)

RM: what would be a workable time-slot for all of us?

<Tina> ShaneM: that reminds me, need a quick word afterwards.

RM: shane is UTC minus 5 -- furthest west?

SM: gerrie schultz (last name spelled incorrectly)
... easy to shift schedule

GJR: same for me

<Steven> Our flexible friend

RM: will suggest logistics - probably 3 hour chuncks

<Steven> how many chunks?

<Steven> 2?

RM: will confer next week on RM's proposed logistics

CURIE Syntax

RM: in last call; received a bit of feedback

Role Module

RM: ended last call on 10 may 2008

<Steven> SVG sent comments last week

RM: dialog with WAI about Access Module and placement of ARIA role names into XHTML namespace

<Steven> http://lists.w3.org/Archives/Public/www-html-editor/2008AprJun/0024.html

SP: SVG comment from last thursday

SM: forwarded to issue tracker system -- in "incoming" bucket -- moving to "Role" bucket now

RM: vocab document dialog

SM: not critical path for spec; vocab doc will develop independently and should let it do so

<Tina> On topic of Access Module and ARIA: am doing a brief presentation at a Swedish IT professionals organisation, Monday 19th, on topic of XHTML 2 WG. If anyone have input on that topic, feel free to contact me directly.

RM: where are normative definitions -- SM response on list; should consider -- vocab document something machine-processable at run-time -- seems to me should be normative like schema or DTD

SM: not saying not normative, just not part of spec

RM: question posed by AlG -- is it a normative document

SM: will have to reread question in light of RM's comments

Steve Bratt on Transition

RM: REALLY dragging on...

SP: sent email yesterday; got unclear reply -- not clear what root of problem is, so tried again

RM: yam sent email to list -- can't refer to spec -- holding up other OMA specs

SM: holding up other specs for us, too

RM: not just us

SP: no formal reply to transition request

RM: causing problems for OMA
... biggest problem for OMA isn't Basic transition, but M12n
... don't undestand delay -- especially without explanation

RDFa Syntax

RM: different comments from TAG

SM: the core of issue is something we discussed long time ago -- XHTML+SVG media type -- request from TAG to change doc to permit RDF
... TAG commented about mediatypes and RDFa - felt need to update media doc

SP: thought we said we didn't see what problem was

SM: right -- did you convey that to TAG?
... background - TAG has document working on - NoahM updated this week, still states need media type to change
... talked with noah privately, now understand what is going on -- semantic web and announcements: how do things that are part of semantic web know that content might contain RDF -- how helping with RDFa syntax; don't think is an issue for this WG -- don't need to do anything with mediatype and want to ascertain if we have stated that to TAG and if not, we should now

<Roland> The Self-Describing Web : http://www.w3.org/2001/tag/doc/selfDescribingDocuments-2008-05-12.html

SM: TAG suggestion: update mediatype to affirm that RDF can be served
... not necessary

SP: send again to TAG

steven's existing ACTION: recorded in http://www.w3.org/2008/02/20-xhtml-minutes.html#action06

<Roland> existing ACTION: http://www.w3.org/2008/02/20-xhtml-minutes.html#action06

SM: meanwhile will get task force to respond to Noah's draft -- bigger response could be made, but want to put to bed

<Steven> I'm sorry. I think I confused that with my other action wrt the TAG

RM: is that all on RDFa side? still another couple of outstanding actions

SM: XML data set; want to leave other actions open -- need more dialog on rel attribute; have sent out email, but so far no replies

Access Module

RM: GJR emailed Access Module rewording to list today (14 may 2008)

SM: not had time to read

<Roland> GJR note: http://lists.w3.org/Archives/Public/public-xhtml2/2008May/0023.html


<ShaneM> Shane asked about the definition of "session". Gregory will try to find a reference.

<ShaneM> Shane asked why everything is "should" instead of "must". Gregory indicated part of the UAWG did not want to use MUST because it is too strong.


User agents <ins>MUST</ins> provide mechanisms for overriding the author setting

with user-specified settings in order to ensure that the act of moving

content focus does not cause the user agent to take any further action,

as required by UAAG 1.0, Checkpoint 9.5. [1]

<ShaneM> There is an assertion missing from @key about MUST be able to override. Gregory agreed to add it back.

RESOLUTION: accept changes proposed by GJR as well as those agreed to at this meeting; GJR will repost to public-xhtml2 ASAP

RESOLUTION: Access Module will go to last call as amended

XML Events 2

RM: done updates to XML Events 2 following f2f should work through review and take to last call asap

<Steven> please add resolution to go to last call

<Steven> (only the scribe can do it)

RM: XML Events 2 draft ready to for publish review in preparation for transition to last call -- perhaps next week
... would like to go through Alessio's work on frames and iframe
... read email from 2 weeks ago -- are you connected AC?

<Steven> I would also like to report back on XTech conference

AC: only begining of possible work on object strengthening

RM: talk us through it at next week's call?

AC: yes

RM: will put specifically on agenda

SP: feedback from XTech Conference: gave an XForms presentation -- very well received; discovered 2 more implementations didn't previously know of, but biggest joyous suprise was popularity of RDFa - major sites already adopting it, quite a buzz

RM: clear up how to treat RDFa in HTML with TAG

<Steven> Also got interviewd by BBC on the HTML5/XHTML2 thing

<Steven> will be posted on blip.tv

<Steven> THat's a good solution: we can do that in no time

<Steven> (that=changing the xhtml ns doc)

<Tina> Speaking of implementations ... if not implementation per se: there's work being done to start checking for ARIA and other similar techniques by my company's quality testing tool. No details as of yet.

SM: TBL has suggested that change XHTML namespace doc so that if RDF processor is following its nose, it knows magically that any XHTML document might contain RDFa; a GRDDL processor can get RDF out of any XHTML document - very exiting from 50,000 foot level (enable semantic web); implication - any document will contain triples despite work we've done - strips announcement mechanism -- author can use doctype or version to declare using RDFa -- need to encoura

SP: all of a sudden a bunch of docs containing triples -- all they do is represent a formal semantic way or implying that this is part of XHTML and this is part of RDFa -- don't know why objection to relationship statements; being extracted in RDF is only real concern -- what is being done with it

SM: relationships there in that there is a @rel with a value
... doesn't mean that i as author not committing to architechture; semantic web about a contract -- this is where draw the line -- have to state where appropriate

RM: trying to ascertain how to say document more than one thing -- can write a bunch of valid XHTML 1.x and valid XHTML Mobile Profile and XHTML Basic -- aught to be able to claim conformance to all

<Steven> I don't see the problem of extracting the relationships as RDF; the reationships are really there; all that has changed is representing them as RDF

<Steven> it is like changing a JPG to PNG

RM: one of these AND one of these AND one of these

<Steven> the information is the same

RM: i am one of these - by implication i am not one of those

SM: if claim is can write valid Xfoo - validator should validate; if not valid XHTML Mobile, then it is not

RM: version, doctype or mimetypes -- all contain info about pieces of compound document -- too many ways of talking about content

SM: proposal

RM: addressed when discussed script ennumeration of features -- describe features and not the content

<Steven> it might be my keyboard

RM: knee-jerk capabilities need to be considered

SM: have thought about way of saying "this doc is using a collection of features X" and that is of potential interest
... might be proposed syntax from 8 years ago

RM: more and more messy as enter mashup and fragment world -- fragment should assert what it is, not parent

<Steven> <link rel="stylesheet" href="s.css"/> says "this document uses a stylesheet called "s.css"

SM: wonderful concept, but cobbling together arbitrary content is problematic

<Steven> so does:

<Steven> <> html:stylesheet <s.css>

RM: if created valid SVG can assert it is valid and can be part of embedded SVG

GJR: question out of ignorance - can XPath be leveraged?


SP: what do we do about this -- plan change to NS that XHTML docs may contain GRDDL profile and RDFa encoding?

SM: talked with TBL and danC about this -- just change doc at end of NS so has declaration, so that magica can ocurr -- all docs that reference this NS are included - that is TBL's view
... like @link

SP: gets around media types argument

SM: bad solution -- implies that entire XHTML using internet is suddently using triples -- don't think we've thought through all implicatinos of that

RM: might be a small change, but MUST document implications of change so people are aware of that

SM: fully agree -- but not as response to a comment at this point in process

RM: may decide to do it -- not endorsing -- should make clear effect of doing it

SM: not our decision

RM: yes, but we can provide guidance

SM: would like TAG to chime in on this -- big picture about web development

RM: go chew on this one rather than curies

SM: will do as soon as given action by RDFa task force
... also interested in hearing W3C's input from logistical perspective - means a whole class of things going to leave XHTML NS - interesting impact on web infrastructure of W3C

Wrap UP

<ShaneM> I wil copy this group on my comments on the tag stuff.

RM: good discussion -- any other items that need additional prorities, send to list and will get on agenda

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.133 (CVS log)
$Date: 2008/05/23 15:28:07 $