21 Feb 2006


See also: IRC log


Norm, Vincent, Ed, Noah, Henry, Dan, Dave, TV[+0:10]



Vincent: Next meeting is f2f, Monday and Friday

Possible regrets for Norm on Monday afternoon

<DanC> I'm available 7 Mar but inclined to cancel

<ht> HST can't make 7 Mar

Next telcon: tenatively 7 Mar, though we're likely to decided at the f2f that it's unnecessary and skip to 14 Mar.

Tim has sent regrets for 14 Mar

Vincent is at risk for 14 Mar

Accept this agenda?

Dan requests discussion of IRW workshop


Approve minutes of 14 Feb?


IRW workshop

<DanC> http://www.ibiblio.org/hhalpin/irw2006/

Dan: Workshop looks interesting

Dan is slated to talk about the TAG position

Dan: I don't consider myself a disinterested party; I can give my slant and a little bit of what the TAG is up to,or I can give just my slant. I can't report impartiallhy on the TAG position.

Henry: Speaking as an organizer, we're happy for you to make that call
... Speaking as a TAG member, I'd be happy for you to make the balance as you see fit.

Noah: For the TAG side, would you intend to take mainly about positions in the Arch doc or also about unresolved, ongoing discussion?

Dan: I would probably only quote from TAG approve things; but ad libbing I'd probably leak some current discussions.

Noah: Is there more than use URIs and range-14?

Dan: Yeah, there are still questions about ambiguity.

Norm: I'm happy for Dan to wing it.

Vincent: returning to administrivia for a moment: the Jan 10 minutes are member only
... Any objection to making a public copy?

<DanC> (wow... indeed. http://lists.w3.org/Archives/Member/tag/2006Jan/att-0003/Jan102005.html )


Vincent: Ok, I'll take care of that.

Face-to-face in Cannes/Mandelieu

-> http://www.w3.org/2001/tag/2006/02/27-agenda.html#Admin

Vincent: Dinner is arranged for Monday evening at 7pm at La Riou[sp?]
... we should be able to get a quite place in the restaurant
... should we add endPointRefs-47 per Mark Baker?

Henry: Although we settled the general issue, we never addressed his original question.

Vincent: Mark also reminded me that I had an action to invite him to a telcon
... Mark will be in Cannes next week.

<DanC> (discussion of epr47 raises an intersting question about whose issue it is. Seems to me the issue is what the TAG says it is, though we're obliged to seek consensus with all interested parties.)

TV: Maybe we can invite him to have lunch with us on Monday or Friday?
... Or some other day.

Noah: If we want everyone involved, it should be at a formal meeting

Vincent: I agree.

Noah: I think Mark is telling us that if we want our comments to matter, we need to get them in soon

<DanC> "16 February 2006: Web Services Addressing 1.0 - WSDL Binding - Last Call Ends 31 March 2006" -- http://www.w3.org/TR/

Vincent: I'll meet with him during the week and see if I can arrange to meet with him at a teleconference.

Norm: Ok by me.

Vincent: Do we want to talk about httpRange-14 per David Booth?

-> http://lists.w3.org/Archives/Public/www-tag/2006Feb/0067.html

DanC: I think the questions he asks are as much about HTML-fragments-some-other-issue. I don't want to reopen httpRange-14.
... Fragments in XML seems like the one that's most relevant.

<DanC> fragmentInXML-28 : Use of fragment identifiers in XML

<DanC> http://www.w3.org/2001/tag/issues.html?type=1#fragmentInXML-28

Norm proposes that we constrain the amount of time we spend talking about it if we decide to take it up

Noah: I've had this question inside IBM. But I don't care if we do it at the f2f or put it in the queue.

<DanC> issue 47 is also nearby

Vincent: I don't see a lot of enthusiasm. Let's see if we have time left on Friday afternoon.
... My inclination is not to change the agenda for the f2f at this point.
... Any other comments about the agenda?

None offered at this time.


-> http://www.w3.org/2001/tag/2006/02/namespaceState/TR/2006/WD-namespaceState-20060223/

Norm asserts that it's ready to go except for the status

Dan: It competes with a whole bunch of things

Norm: how about we work together at the f2f and we pick another date post-publication-moritorium

Vincent: Let's do that.

Noah asks about publication for least power

<DanC> (I have no plans to publish least-power stuff as a WD)

Vincent points out that this is just an internal finding so there's no issue about the moritorium

Vincent: If I remember correctly, last week we agreed to take nsState-48 to the TR page as an experiment.
... For the moment, I think it makes sense to just publish it as a usual finding.

Noah: When we take a finding to a "next stage" should I remove the prevlocs?

DanC: My preference is that you and Norm do what you like.

Draft State finding

Dave: I published a lengthy document on state today.
... I'm hoping we can take a look at it at the f2f. Roughly, it points out that designing for state is one aspect of system design.
... To meet the various requirements of your application, you may need several kinds of state
... I tried to enumerate the kinds and give some examples.
... There's a fairly lengthy discussion of state in the web context and then another using more XML-related technologies.
... I've been working on it for a while, I'm hoping that the work the TAG was doing around end point references will fit in here.

Vincent: Ok, Dave. I think it's too early to have a discussion of the document today, but I'll put it on the f2f.

Dave: Ok

Some discussion of whether or not there's time to read the text before the f2f

Noah: It would be helpful Vincent if you would tell us what you would like us to have read before the f2f

Vincent: I'll update the agenda tomorrow.
... It doesn't seem clear if there will be time to read it before the meeting.
... If we take it up, it will be on Friday.
... I'll put it at the end of the agenda, if enough folks have read it, then we'll discuss it.

Issue XMLVersioning-41

Vincent reviews the open actions

Vincent: Henry to make sure that what he is doing with ontology of XML infoset fits with what DanC is doing on ontology of Language etc.

Henry: that's ongoing, but in so far as I'm keeping up with what Dan's doing, I'm happy.

<DanC> msg about RDF/OWL version of versioning terms

Henry: I haven't reviewed Dan's most recent edits

<DanC> in particular, http://www.w3.org/2001/tag/2006/ext-vers/ext-vers-uml.rdf

Vincent: Dave to update extensibility finding with the result of Edinburgh F2F discussion and related diagrams.

Dave: I think I've done that.
... There have been a couple of comments since I posted my most recent stuff to the xml-versioning mailing list.
... Dan's proposed some changes and we had some discussions on the TAG call recently.
... I made some more changes taking into account some of what Dan had done. I published this to the xml-versioning list.

Vincent: Ok, so I guess that action is closed.
... Comments?

Noah: I'm struggling to put together the various pieces that different people have been contributing.

<noah> Dan is this what you were looking for: http://www.w3.org/2001/tag/2006/ext-vers/ext-vers-uml.png

Noah: I find that diagram to be pretty close to what came out of Edinburgh
... My impression is that Dave's finding still tends towards saying that a language consists of pieces of XML

<DanC> dorchard, I'm just reading your msg of 15 Feb now. I haven't gotten a round to subscribing to xml-versioning yet. sigh.

Noah: So maybe I'm misreading, but I wanted to signal that in terms of the meat of converging the content, I'm not sure we've moved as close as would be ideal.

Dan: Which differences do you see?

Noah: I've got a PNG from Dan that looks good, and then I've got [something] from Dave which has some of that diagram and some text.
... Well, probably the best thing for me to do is punt here and go back and put the pieces together.

Dan: I'd prefer that we discussed this in substance now.

<dorchard> Here's what I posted from the f2f in edinburg http://lists.w3.org/Archives/Public/public-xml-versioning/2006Feb/0000.html

Noah: I just don't have the stuff handy that I need.

<dorchard> Then there's an update.. http://lists.w3.org/Archives/Public/public-xml-versioning/2006Feb/0006.html

<noah> " By language, we just mean the set of elements and attributes, or components, used by a particular application. "

<DanC> (please don't bother with history; the TAG hasn't decided anything)

Noah: I thought in Edinburgh, we said something like "by language we mean the class of strings" or something liek that that's independent of XML.
... I'm not sure I like the title. I think the TAG could offer more architecturally if we did something about versioning of web documents (XML, N3, etc.) not just XML.

<DanC> yes, we keep having the discussion over and over. That's the cost of collaboration. it will repeat until we converge.

Dave: To the extent that we can abstract away from XML in a time-bound manner, I'm ok. But at the end of the day, the goal put into this deliverable have been about XML

Noah: Dan pressed me and I showed you an example of some text that was bothering me

Dan: Let's start suggesting text

Noah: "A language is a class of strings" or "A version of a language is a class of strings"

<DanC> http://lists.w3.org/Archives/Public/public-xml-versioning/2006Feb/0006.html

Dave: (Looking at the diagram)
... What I did here is say that a language uses a vocabulary and I extracted out the XML vocabulary
... I said that an XML vocabulary is a type of vocabulary

<DanC> (hmm... why a 0..1 on the subclass link? odd.)

Noah: That's good. I'm still trying to decide what our story is about languages and instances
... Do you have a definition for language?
... What does this diagram try to tell me?

Dan: A language is related to membership by a "string_set"

Noah: How do you read that?

Dan: An instance (a string of characters) that's in the string set of a language.
... No one was ever happy with "Membership"

Dave: Instances relate to language by being members of the language

Dan: The reason for the box is to distinguish the different languages that might have the same strings

Noah: I almost view membership as "string sets" and languages as "interpretable string sets"

<DanC> interesting question: does the UML say that every instance encodes an infoset?

General regret that several of us can't really read the UML well enough to undertand the diagram

Noah: Given that there's a star there, how do you read that in English?

<DanC> (I wonder what TV makes of this diagram-based discussion, and whether the RDF/OWL stuff helps at all.)

Dave: I think instances have infosets.

Noah: There's a sense in which you can leave all the XML stuff out and the diagram would still hold up

<DanC> hmm... interesting idea... an xml-ignorant diagram, and and an xml-specialized diagram

Noah: You could talk about a lot of interesting things, like partial understanding, in the xml-ignorant diagram
... Then you could say that when you build meta-languages, you can share a lot of mechanisms across different meta-language vocabularies

<DanC> yes, please, dave, try the xml-specific and generic diagrams.

Some agreement that it would be good to try an xml-ignorant and xml-specific diagrams

Noah: And then the middle ground is to say that XML is an example of a meta-langauge. What we expect is that lots of versioning solutions that apply to this XML-language might also apply to that other XML-language. And that's true of meta-languages in general.

Noah goes on to describe how this might work out [scribe couldn't type fast enough]

Dave describes interface descriptions that might be able to express all the constraints of the various components. With an abstract notion of languages and versioning you could apply it to several parts of the interface.

Noah: My particular proposal was to scope this to roughly web documents: MIME-typed octet-streams.

<DanC> (I also find it appealing to think about protocol versioning along with language versioning, but my thoughts on that are *much* less mature.)

Dan: I think I come down on the side of doing the string thing for now and adding new issues later.

Dave: At the end of the day, I think this serves a different purpose that I have as well. Hopefully it'll hang together well anyway.

<raman> was this a two hour call?

<DanC> no, 90min

Vincent: Dave would you commit to providing two diagrams?

<scribe> ACTION: Dave to provide two diagrams: one XML-ignorant, one XML-aware. [recorded in http://www.w3.org/2006/02/21-tagmem-minutes.html#action01]

<DanC> (is it worth the meeting's time for me to talk to dorchard about CVS issues?)

Dan plans to call his action on RDF/OWL versioning done

(collaboration without CVS is a pain so, yes, from my perspective, DanC )

<dorchard> ( apparently not Dan :-)

Vincent: For the moment, let's try to finish today's agenda
... DanC, what about your action a?
... You want to consider it done, is that right?

<DanC> > http://www.w3.org/2001/tag/2006/ext-vers/

<DanC> > http://www.w3.org/2001/tag/2006/ext-vers/ext-vers-uml

<DanC> > http://www.w3.org/2001/tag/2006/ext-vers/ext-vers-uml.png

<DanC> > http://www.w3.org/2001/tag/2006/ext-vers/ext-vers-uml.rdf

<DanC> > http://www.w3.org/2001/tag/2006/ext-vers/ext-vers-uml.violet

Vincent: Ok, then let's call it done.
... What about Dave to produce a new draft of his versioning finding by the end of the year
... I'm not sure if it's really different

Dave: I'm still working on it, but I haven't finished it yet.

Vincent: anything else on versioning today?

Issue namespaceDocument-8

Norm reports failure to make progress. Too much to do after returning from vacation.

Vincent: What should we do about Tim Bray's action?
... I propose that we drop it.

<DanC> ok, then let's call it done, not dropped

Henry: I think we agreed that we got a declaration from Jonathan and that we agreed that's the best we could do.

Vincent: Right, we have that statement from Jonathan.
... Ok, then we'll call it done.
... Norm to follow up on Noah's message on namespace name?

Norm: No progress.

<DanC> (really? I thought NDW did that one. oh well.)

Vincent: Henry to track progress of #int bug 1974 in XML Schema namespace document

(Did I? I'll have to look before I do it again :-)

<ht> http://www.w3.org/XML/Group/2005/12/XMLSchema

Henry: The WG is pretty happy. There are a few changes to the draft namespace document still pending, but we're busy producing anchors for all the things we promised would have anchors.

<DanC> I see http://www.w3.org/TR/2006/WD-xmlschema11-2-20060217/ just went to last call

Henry: Not quite done, but I'm working on it.

Dan: The namespace document isn't part of the last call?

Henry: It's along side. Historically, namespace documents haven't been part of what's reviewed at transition.

<noah> I've got to drop off...see you all in France.

Henry: This draft will replace the 2001/XMLSchema namespace document.
... That should happen in a month or so.

Dan: Please let us know when that happens.

Vincent: Norm to propose to Jonathan changes the Natures

Norm: I started the discussion with Jonathan, but it hasn't reached conclusion yet.

Vincent: So it's in progress.

Norm: Yes.

Vincent: asks about Monday night's dinner at the f2f.

<DanC> count me in for dinner monday night.

<dorchard> +1 for dinner

<Ed> Count me in as well.

Norm will be there

<ht> HST will be there

<ht> Raman, will you be at the dinner Monday?

Vincent: Adjourned

Summary of Action Items

[NEW] ACTION: Dave to provide two diagrams: one XML-ignorant, one XML-aware. [recorded in http://www.w3.org/2006/02/21-tagmem-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.127 (CVS log)
$Date: 2006/02/21 20:00:42 $