W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > September 2001

RDFCore WG minutes for the telecon 2001-09-07

From: Graham Klyne <GK@NineByNine.org>
Date: Mon, 10 Sep 2001 13:50:26 +0100
Message-Id: <5.1.0.14.2.20010910134448.044ccc80@joy.songbird.com>
To: RDF core WG <w3c-rdfcore-wg@w3.org>
RDFCore WG minutes for the telecon 2001-09-07

Transcript:
   http://ilrt.org/discovery/chatlogs/rdfcore/2001-09-07
Agenda:
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0079.html

1: Allocate scribe:  Graham Klyne

2: Roll call

Participants:
    - Brian McBride (chair)
    - Art Barstow
    - Jan Grant
    - Martyn Horner
    - Graham Klyne
    - Dave Beckett
    - Dan Connolly (part of time)
    - Ron Daniels
    - Bill dehOra
    - Frank Manola
    - Stephen Petschulat
    - Aaron Swartz
    - Pat Hayes
    - Jeremey Carroll  [Not noted in IRC]
    - Mike Dean

Regrets:
    - Daniel Brickley
    - Sergey Melnik
    - Pierre G Richard
    - Jos De Roo
    - Eric Miller

Absent:
    - Frank Boumphrey
    - Ora Lassila
    - Rael Dornfest
    - Yoshiyuki Kitahara
    - Michael Kopchenov
    - KWON Hyung-Jin
    - Satoshi Nakamura
    - Guha


3: Review agenda:

    No AOB

4: Review minutes of previous meeting:

    http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Aug/0255.html
    http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Aug/0256.html

    Accepted with the corrections listed.


5: Confirm Status of completed actions:

   ACTION: 2001-08-24#3 Brian McBride
     Collect a list of other groups that the WG is interested in
     meeting with.
   DONE.
   See:
     http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0092.html)


6: Progress on publishing syntax document

ACTION: 2001-08-31#1 Art Barstow
   act as team contact for the purposes of getting
   the Syntax working draft V1.23 (plus amendments) through the W3C
   publishing process.
DONE.
See:
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0082.html

There was some discussion about what steps should be taken to
publicize this. We can expect Janet Daley to pick it up and
generate the appropriate W3C internal notices.


7: Prince Arcs:

Goal: Remove this as an obstacle to publishing the Model Theory WD.

See:

   http://lists.w3.org/Archives/Public/w3c-rdfcore-
   wg/2001Sep/0032.html

Pat Hayes says this is not a show stopper for publishing the Model
Theory.  We can be agnostic about these and put them in later, if
needed.  Pat will reword the anonymous node rules to be more
precise, and not close off future options.

DanC suggests these should be called "prince arc labels".  GK
agrees.

RESOLVED: Prince arcs are not a showstopper for publishing the model
theory


8: MODEL Theory WD

Proposal:
   Publish Model Theory WD, with: reification and containers removed
   A more visible alert that cycles in subClassOf and subPropertyOf
   are being reconsidered.
See:
   http://www.w3.org/2001/08/rdf-mt/rdf-mt

Pat Hayes will have next draft ready by next Wed/Thur.
Reification and containers will be removed.
Some Lemmas will be added back as an appendix.

Some discussion of publication schedule.

RESOLVED:  will defer publication until the group have had a week to
review the full document.

ACTION: 2001-09-07#1: Pat Hayes
   respond to indidual comments on the model theory

ACTION: 2001-09-07#2: Pat Hayes
   The publication candidate draft to be clearly signalled to the WG
   mailing list.


9: TEST CASES WORKING DRAFT

See:
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Aug/0118.html

A near-publication-candidate has been offered, and can be used with
minor changes if no WG member objects.  The text has been reviewed
by DanC, and the test cases run by AaronSw.

The existing N-triples document has been reviewed by several people,
is to be merged in to the test cases, modulo \u for Unicode,
examples to be tidied per editorial discretion and introduction to
be reworded to suit place in larger document.

RESOLVED: publish the test cases draft with minor changes noted.

ACTION: 2001-09-07#3: Dave Becket
   Update the N-triples part of the doc

ACTION: 2001-09-07#4: Art Barstow
   Drive the publication process for test cases doc.

There was some discussion about changing the test filename
convention to use 3 digits instead of 4, for reasons of consistency.
Some felt this was makework.  It's not important to freeze the test
case files at this time [implying new convention can be phased in?].

RESOLVED: CVS will be tagged with test case draft publication dates.

It was noted that older CVS versions are not publicly accessible.

RESOLVED:  create ZIP archives of test cases to correspond with test
case document publications.


10: Discuss parseType=Literal Contributions

See:
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Aug/0170.html
   http://lists.w3.org/Archives/Public/www-rdf-comments/2001JulSep/0101.html
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Aug/0076.html
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Aug/thread.html#67

There have been some exchanges with a member of the XML fragment WG
(Paul Grosso).

There is a larger issue of exactly what a Literal is in RDF:  is it
XML or is it a character sequence or a resource or something else?
The allowable forms of a literal are additionally constrained by the
serialization syntax used (e.g. in RDF/XML: literal must be well-
formed XML, in N-triples: a Unicode string).

NOTE:  literals and resource URIs are already well-distinguished in
both RDF/XML and N-triples;  we presume any other serialization
syntax must do likewise.  Hence a literal may contain URI syntax.

So what *is* a literal?  Some options mentioned:
- string as in document
- XML canonicalization of string
- per XML fragment interchange
Then there are the non-XML serialization forms.
Bill think's it's a Unicode string.
GK agrees it's character data.
Aaron wants to distinguish between a string that looks like XML, and
a literal that is an XML document.

JanG raises the issue of when a literal is used to represent (say) a
number, is that distinct from a sequence of digit characters;  if
so, how distinguished?

Bill notes if a resource is a literal, the whole issue (including
parseType) can go away -- maybe that should be our advice to a
futire WG?

ACTION: 2001-09-07#5: Jeremy Caroll
   Collaborate with Bill dehOra to produce analysis of the literal
   problem, options, pros/cons for WG consideration.

11: Next telecon - 10am Boston time, 14th September 2001.

Additional discussion:
See:
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Aug/0253.html

the ensuing thread from:
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0003.html

some other messages in this thread:
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0098.html
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0087.html
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0011.html

Bill is proposing that rdf:parseType attribute values are
namespace/prefix qualified; e.g. rdf:parseType="rdf:Literal", etc,
to supersede the old rdf:parseType="Literal", etc.  This is clearly
more than just a bug fix, but it lays a cleaner path for future
evolution.

DaveB and RonD oppose this move.  Mainly on the ground that it can
break existing code, and/or create un-needed complexity.

ACTION: 2001-09-07#6: Bill dehOra
   Prepare a proposal for namespace qualification values of
   rdf:parseType attributes upon which the group can make a decision.


Meeting closed.




------------
Graham Klyne
(GK@ACM.ORG)
Received on Monday, 10 September 2001 10:36:24 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:39:41 EDT