SW: Regrets from TV, TimBL, Ashok
... Agenda as published: http://www.w3.org/2001/tag/2008/03/13-agenda.html

<DanC_lap> +1 ok http://www.w3.org/2001/tag/2008/02/21-minutes

SW: Minutes from 21 February http://www.w3.org/2001/tag/2008/02/21-minutes approved
... Minutes for f2f still coming together, please let's get them cleaned up and out

<DanC_lap> (Norm, which day of the ftf are you minuting? likewise Dave?)

SW: Next telcon 20 March, JR to scribe
... Regrets NM for 27 March, DO and DC for 20 March, NM at risk for 20 March

F2F recap

DO, NW: Went well

NM: We used to do code reviews on a project I was on, and unless everyone, or almost everyone, had actually read the code, the review was cancelled
... Sometimes at our meetings, I find myself thinking, in reply to a comment, "I don't think you would have said that if you had read the whole document". . .
... and this feels like a bit of a downer

<DanC_lap> (yes, it's worthwhile to set clearer expectations about who is to read what)

Issue tagSoupIntegration-54 (ISSUE-54)

SW: Skipping item 3 (Issue webApplicationState-60 (ISSUE-60)) in Raman's absence
... Please set realistic due dates for actions to review: HST, DC

SW: Lengthy comment from Al Gilman on ARIA and TagSoupIntegration: http://lists.w3.org/Archives/Public/www-tag/2008Mar/0028
... [Summarizes] -- : is a problem with existing browser implementations

DC: I would like to see details backing up the factual claims that AG makes
... There should be tests in their test suite

NM: Not just a browser implementation problem, but that there's a buggy dependency -- fixing it for one browser breaks it for another

SW: We'll return to ARIA when TBL is on the call

DO: IE8 announcement includes something about namespace defaulting on unknown elements, e.g. svg elements
... and you can put ns decls on the HTML element

<Stuart> dave's blog entry: http://www.pacificspirit.com/blog/2008/03/06/namespaces_in_ie_8_beta_1

DO: But there are serious limitations
... 1) NS decls don't work for attributes
... 2) Default NS decl only works once per subtree, e.g. no MathML inside SVG
... Since svg uses xlink:href, the first is a problem
... and the embedding problem looks bad also
... don't know why they made these restrictions
... Documents ways of doing [NS decls] in IE7 using the OBJECT tag
... What do we do about this?
... What if the TAG sent a comment to the IE Team
... "Glad to see you're moving in this direction
... could you add full NS support, please?"
... We should not only ask HTML 5 WG to do what we need, but the vendors as well

DC: DO, you could attend an HTML WG telcon and ask MS themselves
... Some attention from TBL on the validator architecture issue, which is good

HST: Question of clarification: We are talking about HTML here, yes?

DO: Yes

HST: So I'm guessing that the nesting constraint has to do with the empty tag issue: no way to tell when embedding stops

DO: Don't know. Do know that you can't use the XHTML namespace at all

NM: Where does media type feed in to this story?

<Zakim> Norm, you wanted to ask if this is beta-limitations or intended design or do we know?

DO: I agree that we should check on that

<noah> NM: In particular, do we know whether if IE 8 gets application/xhtml+xml?

NW: Can't tell if this is just for beta 1, not clear if this is it or we may get more?

DO: I think this is just what's in beta 1, the feature set may evolve, but not sure

NM: At the show where this was announced, this came in the context of IE 8's plan to default to standards-compliant rendering
... that was a big deal, leading off the press conference, so I don't think they'll go back on that

NM: Please be clear that I do not speak at all for Microsoft.

NW: Interesting to find out for sure about the NS stuff

NM: Not clear

<DanC_lap> (the SXSW crowd was very happy about the IE8 default change)

<Norm> I sure hope XHTML support exists and has standard XML namespace support.

SW: Feedback to browser vendors a bad idea?

<noah> I'm just passing on my understanding of what I thought I heard at the MIX conference last week. We obviously should contact Microsoft if we want to know what they've really announced and which parts they view as commitments beyond the current IE 8 beta.

DC: I'm opposed -- we have a WG for this

HT: This is not about HTML 5, which is what the WG is aimed at. It's about the existing HTML specs.

HT:It seems very appropriate for W3C or TAG to say "this is interesting and useful" or "not".

HT:It's not entirely clear to me why we'd say that to the HTML 5 workgroup.

NM: But how are you supposed to know you're looking at HTML 5?

DC: No signal, you're supposed to assume 5

NM: OK, so the HTML 5 WG is addressing the general question of what browsers should do with everything that looks like HTML

DO: They are changing HTML 4

DC: They can't change the HTML 4 spec.
... It was locked in 1998

DO: So it's an add-on that is intended to work with HTML 4

<noah>Specifically, I asked Dan for a clarification as to whether HTML 5 documents are distinguished by, say, an HTML 5 doctype. Dan said "no, you are either a browser coded with knowledge of HTML 5 or not". I then said "I think Dan is right: handling namespace prefixes in such content is within the purvue of the HTML WG"

DO: I would certainly like to see the HTML 5 WG do something about namespaces in HTML
... but IE8 is going to ship first, and we should try to get them to do the right thing
... or in particular make sure that what they do doesn't hurt more than it helps

DC: What I'm offering you is a low-latency forum to achieve this

SW: I think more interaction of the sort DC suggests would be good before we do anything

DO: What about asking the TAG to come to an HTML 5 WG telcon?

HST: I don't think we're ready for that, the TAG doesn't have an opinion on this proposal yet

<DanC_lap> . http://www.w3.org/html/wg/tracker/issues/4 HTML Versioning and DOCTYPEs

<Zakim> ht, you wanted to ask the Chair to schedule discussion

SW: We'll come back to this, Henry will introduce it

<scribe> ACTION: DO to send pointers to MS and Ruby Namespace proposals [recorded in http://www.w3.org/2008/03/13-tagmem-minutes.html#action01]

<trackbot-ng> Created ACTION-123 - Send pointers to MS and Ruby Namespace proposals [on David Orchard - due 2008-03-20].

Issue UrnsAndRegistries-50 (ISSUE-50)

SW: The XRI TC have responded to our questions http://lists.w3.org/Archives/Public/www-tag/2008Feb/0104
... Also note the call for review of another XRI spec: http://lists.oasis-open.org/archives/xri/200803/msg00019.html

<scribe> ACTION: Norman to review XRI response to our questions [recorded in http://www.w3.org/2008/03/13-tagmem-minutes.html#action02]

<trackbot-ng> Created ACTION-124 - Review XRI response to our questions [on Norman Walsh - due 2008-03-20].

<scribe> ACTION: Stuart to review XRI response to our questions [recorded in http://www.w3.org/2008/03/13-tagmem-minutes.html#action03]

<trackbot-ng> Created ACTION-125 - Review XRI response to our questions [on Stuart Williams - due 2008-03-20].

Issue httpRedirections-57 (ISSUE-57)

SW: We've had a response to our input of the diagram to SWEO, expressing some confusion
... Trying to find a date when we can have Leo and Richard for a discussion about this

SW: Any replies to JR's message to www-tag about link-header?

JR: Two private replies, one positive and one suggesting a meeting to discuss further
... No reply from Mark Nottingham yet

SW: Graham Kline missed the note because of the BCC use

JR: I will follow up with him and MN directly

SW: Anyone yet read NM's input to the f2f: http://www.w3.org/2001/tag/2008/02/RepresentationResources

NM: I've been thinking a lot about HTTP redirections
... A lot of the discussion of information resources is about whether the term is well-defined
... I'm happy with it, because it captures a valuable aspect of what you know when you get a repr. of an information resource
... but it seemed to me on reflection that the problems we have with repr. of e.g. me are similar to ones we have wrt repr. of generic resources

NM: I've had some skeptical/this doesn't help feedback, but no positive feedback yet
... I wrote it to be helpful, if it isn't, we shouldn't spend time on it
... so we should only spend time on this if/when someone says "yes, that's a useful starting point"

<DanC_lap> (I see one short response from fielding; is this the one Noah referred to? http://lists.w3.org/Archives/Public/www-tag/2008Feb/0084.html )

<DanC_lap> (hmm... there are more...)

JR: I read through it, and I think there's a lot of overlap with what the Health Care and Life Sciences group have been looking at
... but we ended up in a very different place
... I think it would help to have some use cases in place

NM: I'm not pushing the solutions I offered, so much as the scenarios. . .

DC: What's the thesis statement?

<Stuart> As a quick summary: the intuition is to acknowledge that due to conneg and just general lack of consensus in the community, the current deployed use of 200 isn't sufficiently consistent and reliable for rigorous reasoning in the semantic Web.

NM: I commend in particular the section labelled "Why Information Resource is not the right abstraction"

JR: I hope AWWSW will look at this

NM: That's not the same as the TAG looking at it, given that if I'm right we might want to change something in WebArch, whereas AWWSW is just trying to formalize what we have already

JR: But I certainly expect that once we've formalized things, we'll be feeding back on problems

AOB

HST: CURIEs comments not quite ready for review, maybe next week