Minutes of RDF DAWG teleconference 2004-11-09 for review

Minutes of RDF DAWG teleconference 2004-11-09 for review


IRC log: (404 at this time)

Scribe: Dave Beckett

1. Convene, take roll, review agenda

  Tom Adams
  Dave Beckett
  Kendall Clark
  Dan Connolly
  Jos De Roo
  Farrukh Najmi
  Eric Prud'hommeaux (irc)
  Simon Raboczi
  Janne Saarela
  Andy Seaborne

  Steve Harris
  Yoshio Fukushige
  Howard Katz
  Alberto Reggiori
  Hiroyuki Sato

RESOLVED: to accept 2004-11-02 telcon minutes
as a true record

Next meeting 2004-11-06 14:30 UTC
Regrets: Dan Connolly (16 Nov, 23 Nov, 30 Nov)
Chair: Eric Prud'hommeaux
Scribe: SimonR

ACTION DanC: owner of issue 'yes or no questions' (pending protocol doc)

ACTION AlbertoR: owns issue 'DESCRIBE'

ACTION: SteveH is willing to adapt his testing infrastructure to  
generate input/output RDF/XML and typed nodes into manifest file

2. Use Cases

ACTION EricP to ask KendallC to put X509 policy use case into UC&R

Kendall to look at EricP's resulting email in
and consider what to add, noting not much excitement seen on the
  telcon for the use case.

3. FTF 4

19-20 January 2005 Espoo, Finland (Hosted by Profium)

Janne has given some directions
and some further advice email will follow.  Note that the weather
can be 0C to -20C so bring a good coat.  Slight preference for
hotels in Helsinki if you want to walk around and see things

ACTION DanC: Make a dawg f2f4 registration form

4. ftf 5

ACTION EricP: find logistics re F2F5 at tech plenary in Boston March
  28 Feb - 4 Mar (2 days).  Some WG preference to Mon/Tue of that week.

See some progress: 

Note WBS form on tech plenary scheduling nightmare

and noted clashes with swig/dawg (many people, chairs)
swbpdwg/dawg (TomA, JosD, PatH).  TomA happy with overlap as Tucana's
Dave Wood is chair of SWBPD.

Next to F2F5 are opportunities for promotion and demos - at the SWIG
meeting and the plenary day (Wed).

5. Protocol, Interoperability Demonstration Services

ACTION KendallC: write a protocol document draft
pointing at 

Kendall introduced this as providing multiple methods to try to make
it possible to get consensus on different methods without affecting
the others.  It reflects things that the mindlab wants to do with RDF
queries.  There is an abstract protocol (section 5) and a concrete
one (section 7) using http with examples for all the methods,  see

There were a few questions such as what was [GRAPH_ID] in http URI
for.  Kendall described it was standing in for some http parameter,
not actually expecting to use that particular word, or hardcode query

ACTION KendallC: expose our walking tour data to SPARQL querying clients
  ETA before F2F4

ACTION TomA, JanneS: read the protocol draft at
http://monkeyfist.com/kendall/sparql-protocol/ and email a review to
the WG before the next telcon 2004-11-09

6. SPARQL update, issues

Brief discussion of progress in recent SPARQL query draft
 http://www.w3.org/2001/sw/DataAccess/rq23/ v1.128 2004/11/08 15:27:14

ACTION EricP: supply definitions for SELECT (vars ordered or not?) ala

ACTION PatH review SPARQL def'ns post-publication

ACTION: AndyS to write a test cases which use ? and $ in the same query

DanC asked for somebody else to help on reading and responding to the
public-rdf-dawg-comments list at
to reduce the burden on the WD editors.  Tom Adams volunteered
and got the job to read the list, respond and pass on things to/from
the WG and the comments list.  Chairs may poke him to respond if

5. SOURCE Issue

ACTION DaveB: Update the source section 9, add more formal links,
update the examples, try to think about extra constraints as EricP
proposed (SOURCE ?s and ?s only in in SELECT). Look at various
people's source test cases.
See proposal in:

some people had read it, more readers, questions and discussion needed.

ACTION JanneS: review
and check what happens when no source support is implemented

JosD has an issue with the proposal when the source variable is
unbound.  DaveB found it hard to grok over the phone and requested
it go to email.

Suggestion to change Query Graph to Data Graph or possibly
Federated Graph, (something) Set.  DaveB prefered Data Graph.

ACTION DaveB: update the dawg test repository to record or amend
tests to correspond to the

ACTION DanC: suggest formal definitions for SOURCE

6. Disjunction issue

Simon described his progress with thinking about disjunction and
(nested or not) optionals and is finding problems with getting a
handle on the problem.  Especially with NULLS and LEFT OUTER JOIN and
distinguishing nulls generated by disjunction versus those caused by
the variable not being bound, generated by optionals.  Is there a
difference?  Or a problem?  Unsure whether optionals can be used to
implement disjunction completely.

Quoting Simon from IRC:
  The main suspicion I have is that the nulls generated by the left
  outer join sort-of-mean "there are no values for this variable",
  whereas the "unconstrained" markers generated by disjunction mean
  almost exactly the opposite, that the variable could take any value
  without invalidating the proposition.

ACTION SimonR: explain how much of disjuction can be done with
optionals, nested or otherwise.  Point to references, problems with
LEFT OUTER JOIN in the literature.

ACTION SteveH: own (i.e. propose resolution to) disjunction issue

ACTION SteveH: owns issue 'nested optionals'

7. Next SPARQL WD: timing, reviewers

Discussion of the current timetable for drafts - a new WD in
December, before F2F4 in late January 2005.  Then CR early 2005.

Farrukh/Sun said around April 2005 would be good for a CR with his
EbXML project.  Tom/Tucana also have a scheduled product release
December 2004 that would fit well with a CR in January 2005 area, a
CR in April 2005 would fit.  DaveB, AndyS, SteveH, EricP are
potentially doing a WWW2005 tutorial on SPARQL in May 2005, Japan.

ACTION DanC: notify SemWeb CG of EbXML possible sync point in April

Andy warned that items in the first WD that had warnings they might
be dropped, are more likely to be dropped now if no champion appears.
One example given was UNSAID, AndyS at this time didn't see a clear
requirement for it.  DanC will track any such potential decision by
a WG issue.


Received on Wednesday, 10 November 2004 15:13:27 UTC