Minutes of SWSA Telecon of 1 April 2003

Attending:
Bob Balzer, Teknowledge (Architectures, software composition, 
                         embedding semantic markup in COTS software)
Mark Burstein, BBN (co-chair) (Mixed-initiative control of agent teams, planning for service 
                    composition, ontology translation)
Christoph Bussler, Oracle (co-chair) (semantics for workflow, Oracle's EI offering)
Mike Dean, BBN (DAML+OIL, OWL committee member, DAML program integrator)
Tim Finin, UMBC (FIPA, DAML, mobile and ubiq. computing)
Michael Kifer, SUNY Stony Brook (co-chair SWS-L, DB, LP, KR)
David Martin, SRI (co-chair SWS-L, agent systems, DAML-S organizer)
Massimo Paolucci, CMU (Semantic service discovery, DAML-S Profile & matchmaker)
Dan Weld, U. Washington (planning and service composition, web-based agents)
Stuart Williams, HP Labs, UK (SWS, RF networks, W3C Technical Architecture Cmtee)
Chris Wroe, U. Manchester for Carole Goble (DL ontologies for medicine, 
					    Grid-based middleware)


After everyone gave brief introductions of their background and interest in
SWS (in parens above), Chris B. described briefly the initial meeting in
Innsbruck. Chris went through the slides he presented there (sent earlier
today - titled "Feedback to Case Study WP7). These slides were a response
to the document describing the case study (Work package 7 of the SWWS
project, presented in December 2003 by HP Galway and Bristol.  The case
Chris described, which he will go over in more detail next week at the F2F,
could be one, if we discuss how it could be extended to make more extensive
use of the semantic web, and how this would go beyond RosettaNet, ebXML,
EDI.  Another is the work on MyGrid at Manchester. We need to develop a
coherent set of these that stresses different aspects of SWS.


We briefly discussed the scope and objectives of the committee, which will
be fleshed out next week at the joint face-to-face meeting with the
language committee. The main goal is to develop a W3C note in about a
year's time that makes recommendations about an architecture to support
extending current web services models to capitalize on the semantic web.
To that end we need to collect use cases (current and envisioned) that we
can use as a jumping off point for discussions of the functionality of
various architectures we would want to support.

Mike Dean mentioned the development of information services based on
DAML. Mark pointed out that the capabilities of information services (both
what content can be queried for, and what query language is used)
are not well described by DAML-S or other current proposed SWS description
languages. Chris talked about the current state of the art as automatically
translating back and forth between Java interfaces and WSDL.

A key issue to resolve at the F2F is what we should/can assume about the
language of the semantic web, and associated infrastructureal supports. How
powerful a description language?  Do we assume RDF? OWL?

The F2F meeting will start next Thursday at around 3pm. 

The time for the next telecon will be determined at the F2F. We have the
general constraint that it must be several hours later in the day (whatever
day it is) in order to allow our Australian member to participate.