RDFcore teleconference minutes: 15 June 2001

[Resend:  for some unknown reason, the copy I originally sent didn't make 
it to the list.  Contains some small updates from the original that Brian 
sent.]

[Note:  some of the review of actions went too fast for me to catch the 
details;  I'm trusting that there are enough clues here to pick up the 
threads.  A copy of the IRC log is attached.]


RDFcore teleconference minutes:  15 June 2001

Present:
   Ron Daniel
   Bill de'Hora
   Jos de Roo
   Jan Grant
   Martyn Horner
   Graham Klyne
   Frank Manola
   Stephen Petschulat
   Pat Hayes
   Brian McBride (chair)
   Segey Melnik

Regrets:
   Mike Dean
   Art Barstow
   Dave Beckett
   Frank B
   Dan Conolly


1. Review agenda:

http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Jun/0222.html
No AOB requested.


2. Review previous minutes:

http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Jun/0109.html

Corrections:
Ron D present, Ora not present, Guha on IRC
The WG decided to allow partial representation of containers

Minutes approved with corrections noted


3. Review status of actions:

Eric has organized a test case repository

Jan has renumbered [the original documents?]

Brian has done syntax stuff [to do with containers?]

[Another action complete: missed detail]

Brian is liaising with Guha [about what?]

Martyn: test cases too rigid for current state of work

FrankM:  write-up of reification issues is done
frank's reification test cases: 
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Jun/0228.html

Eric: xml:base action closed

Ora's action on aboutEach - still open

Brian has created set of test cases matching simplified version of 
container proposal
brian on containers. test cases: 
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Jun/0216.html
Aaron has looked through these;  Jan has taken cursory pass;
Also Jos ... problem with case 4 (not getting reified?  reason?)
DanBri: "this is a production where the rdf:ID names the resource pointed 
to, not the reified statement"
Reason:  production in which rdf:id names object resource, not statement


4. Discussion of repeated ordinal container-membership properties (rdf:_n, 
etc):

Brian to rephrase the proposal so that it doesn't put higher-level 
validation requirements on a parser
ACTION: Brian
         (action w.r.t test case 2) [?]

Jos:  raises problem with anonymous nodes  [Scribe: not quite sure what the 
problem is]
Test case includes single element -- empty <rdf:li/>.  Take this to email.
Need volunteer to review container test cases;  Jan volunteers.
ACTION: Jan


5. Aaron's comments:  skipped for this meeting


6. Jan on xml:base

Would like to have some way to attach base URI to RDF documents ... 
xml:base may not be the right way to do this (yet).  We may need to leave 
this for now, and revisit later

Proposal from Ron, response from Danbri.  Danbri was probably too 
enthuisastic for using xml:base, but is less so in light of Ron's comments

Need a clearer line on interaction with xml:base when RDF is mixed in 
non-RDF XML documents

Ron:  maybe this is defined w.r.t. RDF processing, rather than document URI

Maybe the RDF spec should be revised to not say relative URIs are w.r.t. 
document URI, but some compatibly-derived base URI?

Remember backward compatibilty is an issue
DanBri: [[At the same time, since the web is growing rapidly, it is the 
responsibility of this group to not let near-term deployment considerations 
grossly increase the future costs (to implementors, authors, users, etc.) 
of new features.]] (from our charter, 
http://www.w3.org/2001/sw/RDFCoreWGCharter)

GK:  is it enought to preserve b/w compatibility for documents that don't 
use xml:base?

AGREED:  ron's proposal to not add xml:base to current syntax, but need to 
address issue of RDF embedded in some other document that does have xml:base

Call for volunteer:  write up resolution for latter case -- Jan volunteers
ACTION: Jan - write up interpretation of RDF embedded in documents with 
xml:Base


7. Partitioning the problem space:

Good consensus for separating model and surface syntax (XML serialization) 
issues
Should focus on XML for surface syntax
Sergey: need to focus more on the "model"
Brian: ready to move onto deeper issues, but need some structure to order 
the debate;  hence partitioning the problem
AGREED:  General agreement to separate model/syntax

Frank:  think some of schema should be drawn in...
Brian: also separate out some vocabulary?

DanBri: of rdf schema: [[This specification describes how to use RDF to 
describe RDF vocabularies. The specification also defines a basic 
vocabulary for this purpose, as well as an extensibility mechanism to 
anticipate future additions to RDF. ]] (from the abstract)

Pat:  the semantics for the minimal core will not be the same as that for 
the added vocabulary

Brian to write up a description of a proposed partitioning of the problem 
space.  They we can discuss how to approach the problem based on 
partitioning of work based on the structure proposed by brian
ACTION: Brian


8. Discussion of F2F, and preparations

Eric:  also suggest that face-to-face will be an opportunity to move 
forward.  It will sooner than we expect!
ACTION: everybody -- jot down notes of what we want to get from the 
face-to-face meeting


9. Close

Date of next meeting:  same time, next week




------------
Graham Klyne
(GK@ACM.ORG)