W3C | TAG | Previous: 2 Dec teleconf | Next: 16 Dec 2002 teleconf

Minutes of 9 Dec 2002 TAG teleconference

Nearby: IRC log | Teleconference details issues list www-tag archive

1. Administrative

  1. Roll call: SW (Chair), TBL (Scribe), DO, DC, CL, NW (also scribed), PC, IJ. Regrets: TB, RF.
  2. Accepted 2 Dec minutes with correction that DC action was to point tag, not www-tag to doc for uriMediaType-9.
  3. Accepted this agenda
  4. Next meeting: 16 Dec 2002
  5. Following meeting: 6 Jan 2003

1.1 Completed actions

1.2 Meeting planning

2. Technical

  1. uriMediaType-9
  2. namespaceDocument-8
  3. xmlProfiles-29

2.1 uriMediaType-9

IJ notes that TBL did not minute who said what in some of what follows.

  1. uriMediaType-9:
    1. Action DC 2002/12.02: Point to this draft on www-tag: "A Registry of Assignments using Ubiquitous Technologies and Careful Policies."


Dan: I got endorsement of this from Tim Bray and Norm W.
One question is, publish in TAG's name?
A TAG finding reference is missing
Should one bring up with IETF-W#C meeting first? No, March is too long
I propose that the TAG adopt this in being co-author with Mark darft-w3c-registries*
Note that darft-w3c is a possibility since last IETF-W3C telcon.
Anyone want to discuss, object, etc?
RESOLVED: The TAG adopts A Registry of Assignments using Ubiquitous Technologies and Careful Policies as suitable for publication as draft-w3c*
Stuart: Thanks to Dan and to Mark B
Dan: One thing to discuss:
This came up under iss 9, the idea of URIs for media typed.
TBL: "Media type document"
If we follow this, we will probably end up with a URI for text/html with no #
TimBL: This is OK if we really can just treat it as a URI for a media type document, not for an abstract concept.
In principle.

Chris arrives.

2.2 namespaceDocument-8

  1. namespaceDocument-8
    1. Action NW 2002/11/18: Take a stab at indicating pros and cons for the various RDDL/RDF/Xlink designs arising from TB's RDDL challenge.
      1. RDDL Proposal from Tim Bray.
      2. RDDL Proposal from Chris Wilper


stu: I haven't seen many responses to Tim Bray's challenge
http://www.w3.org/2002/11/rddl/ex1.xml <- http://rdfig.xmlhack.com/2002/11/25/2002-11-25.html#1038249302.320672
^timbl's proposal, I think
TimBL: I had one but forgot to send it.
CL: What happens if no satisfactory replies to contest?
Paul: I have been talking to a lot of people about this, and I really wonder about whether we are right to look for just one format.
IJ: I think the goal was to suggest one approach, not the only approach. The goal of this effort was to avoid saying "do this" without providing any example of how.
TBL: The Web wasn't designed like that, or we would still have HTML 0.9. Should we have many posisbilities, and guidelines about what should be there?
Ian: My understadning was that we are not proposing *the* solution but *one* solution - to show there is one which works. See minutes of 12 Feb 2002 meeting: "Resolution is incomplete. Consensus points from 12 Feb ftf meeting are:
  1. Namespace URIs should be dereferencable (to find useful explanatory material).
  2. The TAG has not yet reached consensus on the nature of the material at the end of a namespace URI. The TAG discussed the value of human readable materials, schemas, and indirections to useful adjuncts."
Chris: Saying it is human readable is easy, and if we don't need it to be machine readable.
IJ: We did say it should be machine readable....
folks should feel free to use IRC as a parallel channel, as far as I'm concerned.
Chris: ... then you don't need much more.
Norm: I didn't think we were defining *the* format, but we were defining a really good one.
I thought our excercise was to produce a good one, but not insist.
But we did not settle on only one image format e.g. GIF or else we would have never been able to permit the usuage of JPEG and SVG.
DC: RDDL is a distraction. It suggests that XHTML, RDF, and XML Schema don't get the job done.
TimBL: I think there is something to be gained from a standard here to stop grag.
DanCon, you wanted to say that XHTML, RDF, and XML Schema already work fine; RDDL is a distraction
DanC: I think a good appraoch to stick an xml schema there and an html document.
Chris disagreed with DanC
Paul: It is a tradeoff - if we had standardized on GIF would we never had had PNG?
DavidO: I was always unhappy with saying there should be a document there. This gets confused with what it is. Saying that xml schema would be a good thing there IMHO is a bad thing. If we can't get to resolution for what the best format is, then I would prefer us to say there should NOT be a document available.
gee... what's the best format for images? JPEG or SVG? surely it depends on the image, no, daveo?
Chris; I was not suggetsing that RDDL should b the *one true* format. As I said, if human-readable is all we want, then we have no problem.
Dan, this is an argument we had almost a year ago.
As for putting a schema there, clearly that would [scribe fails to catch the logic of Chris' argument]
TBL: Are you saying that putting a scheme there is bad?
CL: Yes.
CL: I am saying that having something which sits there and points to it vastly better than content negotiating.
I disagree.
DO agrees with CL
TimMIT, you wanted to respond to PaulC re SVG
we have not decided ANYTHING, actually. why is anybody surprised that the discussion continues?
We have consensus minuted at earlier meetings. I am surprised.
"Resolved: The point about URIs should have dereferencable material at their end applies to namespaces."
-- http://www.w3.org/2002/02/12-tagmem-irc
<- http://www.w3.org/2001/tag/ilist#namespaceDocument-8
Tim: I thought we had come to a consensus that itis good to have a document , and that human readable is useful and machine readable are good.
Ian: There was never any "must". No one suggested that asingle format will meet all needs.
[The TAG attempts to establish what we had already agreed]
From xml namespaces spec, 2:
"It is not a goal that it be directly usable for retrieval of a schema (if any exists). An example of a syntax that is designed with these goals in mind is that for Uniform Resource Names [RFC2141]. However, it should be noted that ordinary URLs can be managed in such a way as to achieve these same goals."
DO: I understood: we were going to come up with a format, and RDDL was a good start, but my assumption was that we would have a standard before we recommended putting anything there.
My problem is with people saying "the format" instead of "a format" or "an example format".
Ian: David, do you need one single format, or a general receommendation of one among several?
DO: Either
My problem is any work that suggests sticking an XML schema or RDF schema there isn't OK will get an objection from me.
IJ: I just want to be sure nobody expects it to be "the format".
any work that suggests sticking an XML schema or RDF schema there isn't OK will get an objection from me.
Chris: We say that it meeets the criteria, which are not well-defined.
I will object to XML Schema being OK.
PaulC: We should call it "only an example format"
DO: Do you think that the issue about which vocabulary should use HTML should be a "best practice"?
TimBL: URIs are the only thing we insist on [for the Web]. We don't insist on any data formats.
Paul: Sounds like we are design this like everything else on the web, that you can use something else if you want to.
Stu: TimBray not being here may be a problem. He and Jonathan had been working on the document. I still find there is a problem with ambiguity - namespace vs namespace document. I can see a solution here having a Namespace Docuement being identfied, and that indirectly identifying a namespace.
SW, I share your concern.
Norm: I never thought that RDDL would be part of the namespace rec. I thought we would publish it as a separate finding.
Chris: me too.
TimBL: me too
DOl: I mean 'effectively" chaneg the namespace rec, not actually.
Ian: I hoped that if the NS rec were revised, then it would be more explicit about it being OK or good to put a document there. It is easier to incorporatea tag finding into a namespaces revision if the language is clearer about utility of putting something at the end of a namespaces URI; it's easier to read one doc instead of 2.
timbL: I agree.
SKW: Any change of TAG attitude here?
Norm: Not without TimBray

SKW: I will update TimBray on our discussion.

2.3 xmlProfiles-29

  1. xmlProfiles-29
    1. Action DO 2002/12/02: Talk to XMLP WG about this new issue.
    2. Action NW 2002/12/09: Talk to XML Core WG about this new issue


Chris: people have been discussing entities. TimBray sugegsted removing external entities but it wasn't clear MathML would be helped.
[Discussion about entities/MathML]
DanC: I know people on MathML who di dnot want to use entities ... they had a <mchar name="..."/> before but it went away. See email from DanC on this topic.
Norm: The i18n folks pushed back on mchar as yet another way
See email from David Carlisle, on why mchar went away.
TimBL: If schema allowed one to define character entities?
DanC: Why not use elements?
____? Can't use em in attributes.
Norm: Substring XML discussion was driven by XMLP group profiling out "internal subsets".
DanC: round tripping.
... is something was serialized with &foo; do you have to write it out like that?
Stuart, you wanted to ask about background from XMLP
SKW: The XMLP -- do we need to respond to them?
DO: We could say, Thanks - good rationale.
SKU: Are we on XML Core WG ground here?
DaveO: don't understand the question about being in scope for XML core. It is in scope as I understand their charter under examining possible future version.
Norm; I think we should invide Paul Grosso to a TAG meeting.
PaulC: I agree. People have been using "profile" and 'subsetting" in confusing ways.
CL: I agree with PaulC, some people have been thinking that 2.0 would be smaller and others that it would be larger. It doesn't say anything explicit in the charter about profiling or a subset. But equally, it says they could do a 2.0 if they think it's advisable. Perhaps, they're telling us they think it's inadvisable: We should try to herd this discussion in useful directions.
PC: Several people have put words in Paul Grosso's mouth along the lines of trying to decide what the boundaries are. We could do it by email, or here.
CL: I think it would be very useful.
SW: I'd be happy to invite Paul Grosso. We probably need to show some leadership in formulating what the issue is
DaveO: I think the TAG should make up it's mind about what it thinks of profiling/subsetting, etc. For example, if the Core WG says it's a sucky idea, that might be one response, then we should think about whether we would agree or disagree with that. I don't want to couple what the TAG thinks about this issue with what the Core WG thinks.
SW: Ahead or in parallel with meeting with Paul
That was brutely Norm volunteering to take the "high ground".
ACTION NW: Write up a first draft of the TAG position
ACTION SW: Invite Paul Grosso to a future meeting

2.1 Postponed

  1. Status of URIEquivalence-15, IRIEverywhere-27. Relation to Character Model of the Web (chapter 4)? See text from TimBL on URI canonicalization and email from Martin in particular. See more comments from Martin.
    1. Action MD 2002/11/18: Write up text about IRIEverywhere-27 for spec writers to include in their spec.
    2. Action CL 2002/11/18: Write up finding for IRIEverywhere-27 (from TB and TBL, a/b/c), to include MD's text.

    CL: Both actions pending.

  2. binaryXML-30
    1. Action CL 2002/12/02: Write up problem statement about binary XML; send to www-tag.
  3. fragmentInXML-28 : Use of fragment identifiers in XML.
  4. xlinkScope-23 (5 minutes)
    1. Action SW 2002/11/18: Organize a special-interest teleconf for discussion of this issue on linking. Pending; see email from SW (TAG-only).

2.2 Findings in progress, architecture document

See also: findings.

  1. Findings in progress:
    1. deepLinking-25
      1. Action TB 2002/09/09: Revise "Deep Linking" in light of 9 Sep minutes.
    2. URIEquivalence-15
      1. Completed Action IJ 2002/1202: Action IJ: Link to TB's "URI Comparison" from findings page.
  2. 6 Dec 2002 Editor's Draft of Arch Doc (new):

    IJ: Is it ok for me to publish Editor's Drafts at will? TAG agrees to publish early, often.

    CL: Be sure to highlight any controversial changes in Editor's Drafts.

    1. Action CL 2002/09/25: Redraft section 3 based on resolutions of 18 Nov 2002 ftf meeting.
    2. Action DC 2002/11/04: Review "Meaning" to see if there's any part of self-describing Web for the arch doc. Done.
    3. Complete review of TBs proposed principles CP9, CP10 and CP11

Ian Jacobs for Stuart Williams and TimBL
Last modified: $Date: 2002/12/09 23:01:25 $