W3C

Technical Architecture Group Teleconference

29 Nov 2007

Agenda

See also: IRC log

Attendees

Present
Dan Connolly, Rhys Lewis, Dave Orchard, Noah Mendelsohn, T. V. Raman, Henry S. Thompson, Norm Walsh, Stuart Williams
Regrets
Tim Berners-Lee
Chair
Stuart Williams
Scribe
Henry S. Thompson

Contents


Review of agenda

SW: Added item 6, review request

Item 5 wrt PW in the clear is to confirm a missing action on DO

TVR: My plan for my ACTION-25 is to just lightly link up Norm's blog entry on implicit namespaces and a document I wrote some years ago

Minutes of last meeting

http://www.w3.org/2001/tag/2007/11/15-minutes

SW: Approved as posted

Next meeting

<DanC> I'm at risk for 6 Dec; birthday celebration

<DanC> TVR regrets 6 Dec

SW: 6 December meeting scribe RL

NW: Regrets for 6 Dec

End of year break

SW: Propose last call for this year on 13 Dec, first call of new year on 10 Jan 2008
... Next f2f is 26--28 February 2008

RESOLUTION: Last call for this year on 13 Dec, first call of new year on 10 Jan 2008

abbreviatedURIs-56

<Stuart> http://lists.w3.org/Archives/Public/www-html-editor/2007OctDec/0033

SW: I followed up on ACTION-77

with http://lists.w3.org/Archives/Public/www-html-editor/2007OctDec/0025

WG chair replied formally with http://lists.w3.org/Archives/Public/www-html-editor/2007OctDec/0033

SW: We asked them to put some quite strong language to prevent use of CURIEs where URIs are expected
... The editor in an informal response pushed back
... In the context of HTML role
... But now there's a new publication elsewhere and asked us to comment on it there

NW: I regret the multiple publications of this document

NM: Quotes from the WG's official response

This version of the document includes a normative definition of CURIEs that is the source of your concerns. The section on CURIEs was included as a matter of convenience and will be removed from later versions of the document.

<DanC> +1 focus on particular attributes with a history of URI reference values

NM: Since they're now planning a separate document, the point about health warning is once again relevant

<DanC> (I think it's premature to factor curies out of the RDFa spec, but it's largely an editorial matter, and I'm not the editor.)

NM: because that document is no longer in the context of a particular attribute or attributes. . .

SW: So we're now invited to review http://www.w3.org/TR/2007/WD-curie-20071126/

<Noah> Well, what I said was, it looks like they are doing what we suggested, which is to not duplicate the normative definition of CURIEs (to the extent we're OK with there being any definition of CURIEs, which I'll put aside for a second).

DC: So what are the use cases?

NM: role, I think

DC: I think role is only in use much today in ARIA (accessible Javascript)

<DanC> (What I meant was: role is a technical approach, and I'm still interested to know about the use case. sounds like it's URI-based extensibility a la microformats class kludges.)

<Noah> We've said that if there is a normative definition of CURIE, that definition should be accompanied by a general admonition not to use CURIEs where URIs are expected. Since the XHTML Role specification in question will no longer carry a definition for CURIEs, I don't think we can ask them for the health warning.

<Noah> I do think we can check that none of the particular attributes (e.g. @role) that allow CURIEs also allow URI refs.

DC, TVR: [discussion of use of 'role' and 'class' which scribe did not keep up with]

<Stuart> I think they define role to expect URIorCURIE

DC: Role is URI-based?

TVR: Original spec. didn't say. WAI used it with QNames/URIs for taxonomies
... there are certainly non-URI uses out there

SW: RDFa have an interest in CURIEs, so do the OWL 1.1 folks

DC: I'm interested in use cases and/or customers, not technologies. . .
... Creative Commons people are waiting, NewsML were, but maybe they rolled their own

TVR: NewsML will proceed with or without W3C

DC: I was happy with their using their own abbreviation in their own language

DC: It's when they propose factoring it out and making it general that I get worried

<DanC> (happy, since their abbrevation mechanism is clearly grounded in real, good URIs)

<Noah> HT: I think I disagree with the implications of Noah's clarification above

HT: Since they've pulled it out, there needs to be a health warning in the place they've specced it on its own

NM: Right, not at the point where they spec. the role attribute
... But we should check that the 'role' spec. does not allow vanilla URIs

<scribe> ACTION: Henry to review http://www.w3.org/TR/2007/WD-curie-20071126/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action01]

<trackbot-ng> Created ACTION-80 - Review http://www.w3.org/TR/2007/WD-curie-20071126/ [on Henry S. Thompson - due 2007-12-06].

<Noah> HT: OK, we agree. A warning is needed, but with the definition of CURIEs

<scribe> ACTION: Stuart to review http://www.w3.org/TR/2007/WD-curie-20071126/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action03]

<trackbot-ng> Created ACTION-81 - Review http://www.w3.org/TR/2007/WD-curie-20071126/ [on Stuart Williams - due 2007-12-06].

<Stuart> trackbot-ng, status

<scribe> ACTION: Norman to review http://www.w3.org/TR/2007/WD-curie-20071126/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action04]

<trackbot-ng> Created ACTION-82 - Review http://www.w3.org/TR/2007/WD-curie-20071126/ [on Norman Walsh - due 2007-12-06].

Review Request "Access Control for Cross-site Requests"

SW: http://lists.w3.org/Archives/Public/www-tag/2007Nov/0050

SW: We should be involved, right?
... Volunteers?

DC: Did our concerns get captured in a test case
... Security threat models perhaps not appropriate for a test. . .

SW: I think a fresh review is called for

<DanC> (I can't do a fresh review; I've already looked at it.)

<scribe> ACTION: Stuart to review http://www.w3.org/TR/access-control/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action05]

<trackbot-ng> Created ACTION-83 - Review http://www.w3.org/TR/access-control/ [on Stuart Williams - due 2007-12-06].

<scribe> ACTION: David to review http://www.w3.org/TR/access-control/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action06]

<trackbot-ng> Created ACTION-84 - Review http://www.w3.org/TR/access-control/ [on David Orchard - due 2007-12-06].

DO: DC, did you publish comments?

<DanC> long HTTP header field name in WD-access-control http://lists.w3.org/Archives/Public/public-appformats/2007Jul/0000.html

SW: I sent our comments on the previous draft

DO: Anything else, DC?

DC: I think they pretty much covered the threat model issue
... I was also concerned that the doc. make clear what the problem was
... They added a small bit on that, I guess I'm satisfied

SW: We'll return to this when the reviews are done

Issue passwordsInTheClear-52

SW: I believe we discussed this last time and agreed that you would do another editorial pass, publish and invite comment
... but there was no official action

<DanC> 12 Nov draft was discussed...

<scribe> ACTION: David to produce another draft of Passwords in the Clear finding, based on comments from 15 November telcon, publish it and invite comment [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action07]

<trackbot-ng> Created ACTION-85 - Produce another draft of Passwords in the Clear finding, based on comments from 15 November telcon, publish it and invite comment [on David Orchard - due 2007-12-06].

DO: Will try to do this by the end of the week

DC: Which people in WSC are paying attention?

SW: They are waiting on a draft from us/a request to comment from us
... In addition to Thomas Roessler and Mez, Hal from BEA was most vocal

DC: What about the TPAC meeting?

SW: The above 3

DO: There was input from a number of folk, at least 5 or 6, plus Tyler Close on the phone -- I got a lot of input from the meeting minutes
... So do I get review from them first, or just publish

DC: Just publish

DO: Will do

<DanC> (them? us, rather, yes?)

Issue binaryXML-30

SW: HST wrote a draft response [member only] http://lists.w3.org/Archives/Member/tag/2007Nov/0071.html

<Noah> Regarding EXI before I forget. Henry's draft says: The proposed EXI format is the best technology for the job;

<Noah> I wonder if it might best be "the best choice for the job".

<Noah> or suitable choice

DO: I'm happy with HST's packaging

<dorchard> more specifically, I'm very happy with HST's packaging

HST: I will edit to take account of NM's wording suggestion

<Noah> I see some extra underbar (_) chars in the version of the note that came through my email.

<Noah> Ah, I guess those are for emphasis, but somehow underbar doesn't do that for me. Prefer perhaps >XXXX< to emphasize XXXX

<Noah> Also: "No aspect of the messaging should suggest" it's not clear that the use of the word messaging here is quite unambiguous

<scribe> ACTION: Henry to make editorial corrections to http://lists.w3.org/Archives/Member/tag/2007Nov/0096.html and post that to the EXI WG on behalf of the TAG [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action08]

<trackbot-ng> Created ACTION-86 - Make editorial corrections to http://lists.w3.org/Archives/Member/tag/2007Nov/0096.html and post that to the EXI WG on behalf of the TAG [on Henry S. Thompson - due 2007-12-06].

Issue namespaceDocument-8

<DanC> (I'm looking at the diffs .. http://www.w3.org/2001/tag/doc/nsDocuments-2007-11-13/diff_20071005.html bummed I didn't manage to look at them carefully before the meeting)

SW: HST has published a new draft, completing Action-65

<DanC> (ah... yes... 4.1 and 4.2 separate cases... I remember asking for that.)

SW: Action-66 has been overtaken

HST: New draft is http://www.w3.org/2001/tag/doc/nsDocuments-2007-11-13/
... Mostly added a new section to embody agreement to discuss NS vs NS-DOC by cases

<Zakim> DanC, you wanted to note { n:key range xsd:anyURI } and ask how xsd:anyURI relates to xsd:string (and wonder about RDF plain literals)

<DanC> nature:key "http://relaxng.org/ns/structure/1.0";

DC: You changed nature:key from Object Property to Datatype Property

HST: Yes, decided not to include ^^ syntax. . .Shouldn't good RDF tools infer type from property definition?

DC: Yeah, understood

<DanC> ACTION: Dan try examples from ns8 draft in OWL tools [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action09]

<trackbot-ng> Created ACTION-87 - Try examples from ns8 draft in OWL tools [on Dan Connolly - due 2007-12-06].

SW: Are we ready to go with this one?

DC: Yes, but not in a rush. . .

<scribe> ACTION: Norman to review http://www.w3.org/2001/tag/doc/nsDocuments-2007-11-13/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action10]

<trackbot-ng> Created ACTION-88 - Review http://www.w3.org/2001/tag/doc/nsDocuments-2007-11-13/ [on Norman Walsh - due 2007-12-06].

<DanC> (what diff tool did you use? that's cool.)

<Norm> Dom announced it somewhere, spec-prod? chairs?

<ht> It's not easy to find -- there's a tools section on the W3C Editors' page, which points to a 'diff' wiki page, which links to the tool.

Overdue action review

<Stuart> http://www.w3.org/2001/tag/group/track/actions/overdue

SW: Need to cull these if any overtaken or need postponed

<Zakim> Rhys, you wanted to say that I'd like to claim 50 as complete

<DanC> ACTION: DanC note the old submission about logout button under passwordsInTheClear [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action11]

<trackbot-ng> Created ACTION-89 - Note the old submission about logout button under passwordsInTheClear [on Dan Connolly - due 2007-12-06].

<Stuart> http://www.w3.org/2001/tag/group/track/actions/74 is done -- what to point to?

<DanC> this one, I think... http://lists.w3.org/Archives/Public/public-awwsw/

<DanC> so put that pointer in a note and note yours closed.

<DanC> please

HST believes http://www.w3.org/2001/tag/2007/11/15-minutes#item06 actually discharges action-69, marked as needs review

<DanC> http://www.w3.org/2001/tag/group/track/agenda

General discussion

TVR: Anyone looking at the 'ping' attribute in HTML5 draft?

<DanC> ping issue http://www.w3.org/html/wg/tracker/issues/1

<DanC> fielding 6 Nov http://lists.w3.org/Archives/Public/public-html/2007Nov/0101.html ?

DC: My first reaction was negative, but I am now not so sure

DO: Do we need an issue for this?

TVR: My primary concern was that this is buried amongst so much other stuff, it's not getting the review it deserves

DO: I agree -- I think Roy's analysis is compelling, this is bad

TVR: HTML 5 discussion has gone off onto question of whether users understand GET vs POST

DC: So this is related to Issue 7: When to use GET vs. POST
... We could reopen that for discussion of this matter

SW: Moved and seconded, I will reopen the issue
... So, someone explain?

DC: I run news.org, a newspaper website
... When DO follows a link from my site to one of my advertisers, I would like Audit Bureau of Circulation to get a notification
... Today this is handled by masses of complex Javascript
... 'ping' is a declarative way of achieving the same thing
... I need to check that this is indeed what the spec. says

HST: I'd like to review the materials before further discussion

TVR: I'd also like TAG input on my message to www-tag last week http://lists.w3.org/Archives/Public/www-tag/2007Nov/0025.html, harking back to a video example from before that

<DanC> Subject: Toward URL-Equality For Web Servers And Web Clients

<DanC> Date: Wed, 21 Nov 2007 09:31:32 -0800 (11:31 CST)

SW: An agenda item for a future meeting. . .

TAG blog

<Stuart> http://www.w3.org/blog/tag

NM: What is status? Concerns raised, where are we?

TVR: Compromise -- blog wherever we like, post a pointer to it on the TAG blog

NM: Persistence is a problem
... There is real value to having stuff in W3C space

<DanC> (I acknowledge Noah's concerns, but I'm not inspired with any specific way to improve the situation.)

HST: I have been holding off on several possible postings because of the possibility that we will move/change

DO: I am happy with working decentralized

<DanC> (I'm open to all manner of mix-and-match... TVR's write-elsewhere-and-abstract-on-w3.org is OK by me, though I acknowledge Noah's persistence concerns.)

NM: I don't have a problem with that, but I would also like to have something in W3C space for the commitment that that represents

<Zakim> Stuart, you wanted to ask raman, as one of the people with strong motivation for a TAG blog

<dorchard> I think this issue about the TAG members blogging. I've had my domain up and running for almost 12 years now, so 15-20 years more is likely.

[various]: TAG branding gets lost/overwhelmed, too many remnants of QA branding

DC: 'QA' was rebranded to 'Q&A', for 'Questions and Answers'

<dorchard> I missed the QA rebranding as well.

<dorchard> and it never occurred to me that QA was not Quality Assurance.

<dorchard> hmm.. I could add a TAG blog to pacificspirit.com if the TAG would like

<DanC> some specific things I can do: get "Q&A Weblog" spelled out as "Question and Answer Weblog"

<DanC> and look into a better link from specific items http://www.w3.org/QA/2007/11/a_story_about_namespaces_mime.html to http://www.w3.org/blog/tag/

<DanC> (re openid support... that's an important feature, to me, and it seems to have been taken back out of /QA/ ... I wonder what's up with that.)

<DanC> ("for" or "against" isn't helpful, to me; for what we have or for something else is more helpful.)

<DanC> for reference, the explicit re-branding: http://www.w3.org/News/2007#item222 Quality Assurance Activity Completes Its Work, QA Becomes the Q&A Weblog

Summary of Action Items

[NEW] ACTION: Dan try examples from ns8 draft in OWL tools [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action09]
[NEW] ACTION: DanC note the old submission about logout button under passwordsInTheClear [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action11]
[NEW] ACTION: David to produce another draft of Passwords in the Clear finding, based on comments from 15 November telcon, publish it and invite comment [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action07]
[NEW] ACTION: David to review http://www.w3.org/TR/access-control/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action06]
[NEW] ACTION: Henry to make editorial corrections to http://lists.w3.org/Archives/Member/tag/2007Nov/0096.html and post that to the EXI WG on behalf of the TAG [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action08]
[NEW] ACTION: Henry to review http://www.w3.org/TR/2007/WD-curie-20071126/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action01]
[NEW] ACTION: Norman to review http://www.w3.org/2001/tag/doc/nsDocuments-2007-11-13/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action10]
[NEW] ACTION: Norman to review http://www.w3.org/TR/2007/WD-curie-20071126/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action04]
[NEW] ACTION: Stuart to review http://www.w3.org/TR/2007/WD-curie-20071126/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action03]
[NEW] ACTION: Stuart to review http://www.w3.org/TR/access-control/ [recorded in http://www.w3.org/2007/11/29-tagmem-minutes.html#action05]

Minutes formatted by David Booth's scribe.perl version 1.128 (CVS log)
$Date: 2007/12/07 14:49:37 $