DAML Language Committee Minutes 31 October 2000

Participants

• Tim Berners-Lee
• Dan Connolly
• Mike Dean
• Stefan Decker
• Frank van Harmelen
• Pat Hayes
• Jeff Heflin
• Ian Horrocks
• Ora Lassila (first half)
• Deb McGuinness
• Lynn Andrea Stein

Announcements

November 1 is Tom Hash's last day at BBN. Contact [email protected] or other specific aliases for DAML Lab requests.

Happy Halloween!

Last Week's Minutes

RDF As the Basis for DAML

Jim Hendler wasn't at this meeting, so we had only third party accounts of his early decision to base DAML on RDF. One goal was to leverage the existing Semantic Web efforts at W3C.

There was some discussion about what "based on RDF" meant:

• RDF and/or RDF Schema?
  • Ora noted that the separation of these 2 standards was primarily a timing issue in W3C, and that they should essentially be viewed as one standard.
  • Stefan views DAML at the same level as a RDF Schema, on top of RDF. He also referred people to the recent paper A Layered Approach to Information Modeling and Interoperability on the Web he and Sergey Melnik presented at the ECDL 2000 Workshop on the Semantic Web.
• RDF Model vs. Syntax
  • We are probably more closely wedded to the RDF graph model than the current XML-based syntax. Sergey Melnik and Tim Berners-Lee have experimented with simplified syntaxes for RDF.
  • Ora noted that the RDF Working Group employed an S-expression based syntax while developing the specification.
• definition and/or implementation
  • Will RDF be used to both define and implement the DAML language?
• compatible superset
  • Will we be able to use RDF tools with DAML?

Tentative conclusion: We'll continue to use RDF unless there's a compelling reason not to. There aren't currently a lot of alternatives on the table.

Technical Proposals

The chair hopes that, at equilibrium, the committee can focus about 75% of its energy on technical issues and 25% on process and administrative issues.

Committee Charter

RESOLVED: The committee should develop a charter.

The chair and others were uncomfortable with Peter's notion of not tieing the committee to any particular outside group (DARPA, W3C, Jim Hendler, etc.). DARPA is paying for most of us, and the DAML program needs language specification(s) near-term to accomplish its goals.

Some points on W3C structure:

• Working Groups have deliverables and require a time commitment; Interest Groups do not
• Interest Groups and Working Groups may or may not be open to non-members of W3C, depending on how the charter is drawn up
• there is no precedent for W3C "adopting" an outside group and turning it into an Interest Group or Working Group

There's some tension about whether our processes should be completely open. Some members are much more used to working in a public fishbowl. There are also likely tradeoffs between openness and productivity.

It was felt that final decisions on a charter would be largely up to Jim Hendler, who wasn't present. Further discussion was tabled until Nov 14 when it was expected that Jim could participate.

Links