TAG telcon

19 Mar 2007


See also: IRC log


Dan Connolly, Rhys Lewis, Noah Mendelsohn, David Orchard, T.V. Raman, Henry S. Thompson, Norm Walsh, Stuart Williams
Tim Berners-Lee
Stuart Williams
Henry S. Thompson



F2F minutes at http://www.w3.org/2001/tag/2007/03/06-minutes

<noah> +1 to holding over until they're read

SW: We'll give another week for review of those minutes

SW: Agenda stands as published, except maybe "internet as foundation" at the end if time

Telcon planning

SW: Three items with drafts outstanding, we discussed at the f2f proceeding more like a Working Group and less by just handing off to the editors -- is it time to schedule discussion on some of these?

DC: In my experience the way WGs work is by handing off to the editors.

NM: Mostly everything is given to the editors, but when we're thrashing, gathering consensus from the group takes more work.

SW: There are some substantial documents there, are some of them at the point where they need feedback from the group?

HST:On URNs and Registries in particular, I feel I've done what I can for now. On that one, for my part, I need the WG to engage with the issue and the document in more detail. Would like 30mins of telcon time to look in detail at what feel to me like contradictory inputs. I need help to take it further.

DO: There are two parts to this document, I'm still have work I can do on the part I'm responsible for
... I have that queued up behind the versioning finding

SW: Can we take the two parts forward separately?

DO: Yes

SW: How about next week for HST's part?

HST: Fine

SW: NM, you suggested you thought schemeProtocols was off the agenda for some time

NM: Yes, I was not planning to come back to that for some time, rather I'm expecting to follow up on the input I got at the f2f on self-describingWeb
... and to try to write a new draft on that topic soon -- I don't think I can do that and schemeProtocols

SW: Focussing on just one at a time is OK

NM: I would like to try to drive self-describing forward

... DO, can we look at versioning?

DO: 2 April telcon would be good -- I can try to deliver something by 27 March. . .
... What about 9 April - no, Easter Monday - 16 April then. . .

SW: OK, we will aim for the 2nd, document for review by 27 March


SW: New draft has been published, a long time after the discussion at the AC meeting in Edinburgh last May


SW: An interesting question about value space has come up

HST: I have been tracking this

<noah> FWIW, I believe I read it a few weeks ago, but was not entirely conscious that it was a new draft, and didn't read it with that eye.

HSTTheir first draft said "after the : you have an IRI". Seemed wrong. Now it says "after the : you get what would go after the # as a fragid", which also seems not right.
I think pushing at the value space would be productive in helping them to crystalize: what are you trying to do with these?

SW: Do we need to take a position as the TAG?

HST: I think we do

<Stuart> see http://lists.w3.org/Archives/Public/www-html/2007Mar/0038.html

<Stuart> and response at http://lists.w3.org/Archives/Public/www-html/2007Mar/0039.html

TVR: I think taking a position is perhaps too confrontational -- pushing for clarification of the syntax and the semantics seems the right way to go

NM: URIs are at the heart of the Web, and so we have to be very careful about something like this

<DanC> my comment on curies and compatibility, from 27 Oct 2005

NM: Also, as DC pointed out, href is not a corner case, but it's at the heart of the Web, and changing anything in this place is very risky

TVR: I'm not clear that they are proposing to change the interpretation of href. . .

SW: How do we go about raising these issues?
... Do we want to track this separately from ultimateQuestion-42 ?

<noah> NM: Dan corrected me -- I had meant that CURIEs are a seemingly incompatible syntax for URI >references<.

DC: There is RDFinXHTML-35 which we could use

HST:Mischa Wolf's constituency for this was NewsML

SW: OK, let's use RDFinXHTML-35

<DanC> http://www.w3.org/2001/tag/issues.html#RDFinXHTML-35

NM: We did discuss a lot of NewsML use cases at the f2f in Amherst, and they should remain in scope
... which might not be obvious if it is filed under RDFinXHTML-35

TVR: I would not like this to get lost in the RDFinXHTML-35 bucket

SW: I am minded to create it as a separate issue: CompactURIReferences-5?

<Rhys> +1 to having a specific issue for the CURIE issue

SW: But I'm worried about appearing to endorse the idea, so drafting the description text is not going to easy

<Rhys> Suggests something like Impact of Compact URI References on the Web

HST: I'll try to draft something in the way of a description

DC: What about rdfmsQnameUriMapping-6 as the right issue?

<DanC> http://www.w3.org/TR/2006/WD-rdf-sparql-query-20061004/#rQName

DC: I'd prefer to look at a draft description, then decide about a new issue or not

SW: HST, you happy with waiting for a decision on a short name until the description is agreed?

HST: yes

<scribe> ACTION: HST to circulate a candidate description to tag@w3.org [recorded in http://www.w3.org/2007/03/19-tagmem-minutes.html#action01]


DC: There is a new HTML working group

<DanC> ... http://www.w3.org/html/wg/

DC: there is discussion about what media type should be used for the new WG home page (see above)

<DanC> from 2002 http://www.w3.org/TR/xhtml-media-types/

DC: and Karl Dubost identified this document as relevant

TVR: I don't agree with that document

<noah> Raman: please clarify what you would prefer (I can guess, but would appreciate confirmation)

HST: Was there a proposed change to the recommendations about XHTML content

NW: Chris Lilley brought this to the XML CG

NW: The proposal is to allow serving XHTML as text/html

[no public reference forthcoming . . .]

<noah> I think this is highlighting the inflexible nature of media types. If it had more of a mixin than a hierarchy model, I think one could do this in a way that would be less disruptive. Then again, I suspect that train has long since left the station.

TVR: This is a backwards step -- it's an attempt to accommodate bad browser behaviour

DO: IE6 doesn't know what to do with application/xml+xhtml -- does IE7 also have trouble?

TVR: Yes, IE7 also breaks on this

<DanC> (HTML_WG_test_suite += what do browsers do with application/xml+xhtml? )

TVR: Firefox does not have this problem

HST: Many of us disappointed IE7 didn't fix this

<DanC> (my investigations suggest IE currently has 3 codepaths, none of which uses a conformin XML processor)

TVR: This mean that XForms plugins also don't work with IE7, because they depend on grabbing application/xml+xhtml
... I will have a preliminary draft of something about tagSoup by CoB today

<Rhys> The media type discussion that Raman mentioned may be at http://www.w3.org/TR/xhtml1/#media


DC: I haven't had much time to track this, but it's moving ahead with a lot of energy

TVR: What is SAML and openid?

[various]: Security Assertion Markup Language

<Stuart> http://openid.net/specs.bml

DC: Upcoming events may give more background on this for TAG members

<DanC> (there's concern around the position of XRIs vs, say, email addresses, in OpenID 2)

<DanC> (which the TAG should hear more about when I make progress on my urnsAndRegistries action)

SW: Adjourned

Summary of Action Items

[NEW] ACTION: HST to circulate a candidate description to tag@w3.org [recorded in http://www.w3.org/2007/03/19-tagmem-minutes.html#action01]
Minutes formatted by David Booth's scribe.perl version 1.128 (CVS log)
$Date: 2007/03/19 17:29:14 $