CSS Namespaces review

Steven: The reply was sent. Done

XML Base

Steven: Not done yet. Deadline is a long way away.

FtF:

Roland: I see there is a registration questionnaire, and FtF page
... please fill in everyone

http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0061.html

<Roland> http://www.w3.org/MarkUp/xhtml2/wiki/2008-06-Minneapolis-FtF

CURIE status

Roland: We worked through one set of comments last week

Steven: I sent a redraft last Friday

http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0056.html

<oedipus> http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0063.html

Steven: Today I sent a draft of part 2

http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0063.html

Steven: I would like an OK from the group on the content
... of both

Shane: Please CC the issue tracking system, so that we have the DB trail too

Steven: OK

<ShaneM> This is issue number 8035 in the tracking system.

Roland: OK with 2 mails?

Steven: I prefer it that way, general/specific

Roland: Any comments on these?
... Hearing no objections...

<oedipus> no - send 'em

<scribe> ACTION: Steven to send two replies to TAG (issue 8035) [recorded in http://www.w3.org/2008/04/23-xhtml-minutes.html#action01]

Roland: Does this mean we can go to last call?

Steven: I believe we can, I see no hurdle

Roland: Do we need to add that to the reply?

Shane: In that case I think the reply to the TAG shouldn't ask them for more input. ("What more would you like to see?")

<ShaneM> if you cc xhtml2-issues@mn.aptest.com with (PR#8032) in the subject line magic should occur.

<scribe> ACTION: Steven to announce last call of CURIEs [recorded in http://www.w3.org/2008/04/23-xhtml-minutes.html#action02]

Mime type

Roland: We should review this
... and possibly send a message to the public list

Steven: Lot of discussions internal to w3c about mime-type; TBL doesn't see problem with text/html; got impression would address at AC meeting in Beijing; haven't checked those minutes yet; both TBL and DanC agree with us on this; TBL didn't realize how successful XHTML actually is -- only hearing "a complete flop" because no one using application/xml -- pointed out real usage statistics; some misunderstanding of role of new document; unreviewed at present, still

Steven: Olivier complained that in XHTML 1.0 allowed text/html for docs that follow Appendix C - which is informative, but section normative; found difficult to decide if Appendix C informative or normative

Shane:we fixed that

Steven:doesn't satisfy Olivier; could put stake in ground by reissuing 1.0 to make it clear

Shane: proposal - let's reissue PR 1.0 removing Appendix C and instead referencing XHTML Mime

Steven: reissue 1.0 with those changes? works for me

Shane:eliminates possibility of confusion

Steven:good for 2 reasons: 1) removes ambiguity and 2) puts our stake firmly in ground

Steven:need to get some backing for XHTML Mime -- ready to review?

Shane: Roland had some concerns, but in terms of changes i've made, i think i've completed the work

<oedipus> ACTION: Everybody - review latest draft of XHTML Mime and consider RM's comments thereupon - due 2008/04/30 [recorded in http://www.w3.org/2008/04/23-xhtml-minutes.html#action03]

<Tina> Could someone remind me of the draft URI?

<Steven> Tina, http://www.w3.org/MarkUp/2008/ED-xhtmlmime-20080423/

Shane:had been planning on updating Appendix C in context of XHTML Mime, until realized, why not just replace?

M12N

Roland:anything from Steve Bratt?

Steven:most people in Beijing at moment; been very quiet week for email, meetings and IRC -- next week will follow up

<oedipus> recent email on Event from GJR: http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0064.html

script/@implements

<Roland> http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0059.html

<oedipus> quote from shane from f2f: "We can also do that with an @if that checks HASFEATURE so @implements is then a shorthand"

Roland: Was I close?

<oedipus> GJR: +1 to RM's Question 1

Steven: Yes

<oedipus> @implements tells the system that if there is an implementation of that NS, to ignore this script, or, if no script is present, to use the @implements to point to the pertinent namespace URI for the markup dialect/version/module which the user agent or an expert handler -- a piece of middleware that provides context and the language's ontology -- to invoke that namespace in order to properly parse input and output interactions with specialized knowledge doma

Roland: Is namespace too coarse a level?

<oedipus> question?

Roland: Consider Alessio's frames script

<oedipus> SP: might include several scripts for sub-parts - top-level or not?

Steven: I did intend it to be for namespaces
... but we could be more finely grained, if we could find a notation
... consider a script that implemented href everywhere, or src everywhere

Shane: And then there are problems with different versions of a spec (XForms 1.0 and 1.1, which are both in the same namespace)

Roland: and then you have @role and access

Shane: What about a list of CURIEs?
... implements="xh:role xh:access"

Gregory: If there is an implementation of a dialect or module that the UA doesn't know, it needs some middleware to help interpret the input

Steven: I think that @implements gives you exactly what you want, since your middleware then knows to ignore those bits of script

<oedipus> yes, you are correct

Gregory: What level of granularity were you thinking of Roland?

Roland: Well, maybe module

Gregory: That is what I want too!

Shane: I think I agree
... I think that since that is our granularity level of definition, we should match that

Gregory: It also gives people a reason to use M12N and to implement it correctly

Roland: I would agree with Shane that CURIEs would then be associated with modules

Shane: So we just need to develop a naming scheme - name, revision

[general agreement]

<ShaneM> in case I was not clear, I think the datatype for @implements is URIorCURIEs....

Roland: I will refine the proposal and send it for a later discussion

<ShaneM> which should probably be renamed URIorSafeCURIEs

<oedipus> yes, URI or CURIE

Access

Gregory: The issue is on the PF agenda today

<oedipus> http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0045.html

<oedipus> http://lists.w3.org/Archives/Public/public-xhtml2/2008Apr/0044.html

Gregory: This is a cross-group effort, so it is importnant we get it right in order to do just one review
... it is also a focus of the UA guidelines, and on the agenda this week
... On the issue of boolean, I may have an answer by the end of today
... on the issue of keys, maybe by Friday

Roland: Is the ARIA issue solved with HTML5?

Gregory: Not yet

[ADJOURN]