Joint Committee Minutes 3 July 2001

This page summarizes the telecon for the Joint US/EU Committee on Agent Markup Languages held from 1300 to 1400 PDT on 3 July 2001. These minutes were prepared by the chairman, and were approved during the 10 July telecon.



No announcements this week.

Last Week's Minutes

The minutes from June 26 were approved by those present.

A Problem With The Semantics of DAML+OIL Restrictions

We discussed Richard's message and subsequent postings.

There was agreement that the current DAML+OIL reference text

When there are multiple restrictions listed as part of a single Restriction element, the property P has to satisfy all of the restrictions (i.e., multiple restrictions are read as a conjunction).

is wrong or misleading.

ACTION (Ian and Frank): draft revisions to the reference and walkthru

Thanks to Richard for identifying this error.

DAML PI Meeting plans

The current plans for the DAML PI Meeting include: The "Language" breakout session has been renamed to "Reasoning and Rules", and will start with an hour or so session suggested by Richard Fikes where groups that have been working with reasoning in DAML+OIL (Stanford KSL, Manchester, LM/VIS/Kestrel, W3C, etc.) can share their results. Richard agreed to (co-)chair the Reasoning portion of the breakout session.

Pat and others suggested that the Rules portion of the session focus on requirements for rules and experience in working with rules in DAML+OIL, rather than on a specific DAML Rules proposal.

ACTION (Mike): provide a submission script so people can submit their own ideas for this session.

There was a brief discussion of what people found most compelling about reasoning and rules. Jim noted that the SHOE project built lots of complex reasoners, but got the most benefit when they provided a simpler web interface (with an "advanced" button). They adopted the slogan "a little semantics goes a long way" and concluded that it was best to give people lightweight buy-in with powerful tools behind it.

Deb and Ian plan to meet at the PI meeting to update their "tricks of the trade" paper.

Rulues Proposal

Stefan introduced his rules proposal, which was announced shortly before the meeting. This represents work in progress with Michael Sintek (DFKI) and Pat Hayes.

This is a layered proposal. The current proposal focuses on the lower layer, essentially providing structural manipulation rules (rewrite rules) for RDF. No one believes that rewrite rules are sufficient.

Jim and others expressed concern at referring to this lower layer as DAML Rules, since it doesn't use the features that DAML+OIL adds to RDF. He'd prefer to focus energy on higher levels that directly address DAML+OIL features (computed classes, equivalentTo, etc.).

There was also a desire that the examples focus on applications rather than DAML+OIL language constructs (e.g. uncle rather than subClassof).

There was discussion that a Horn clause at the DAML+OIL level may become something more complex at an RDF level (representing a change in logical form as you change levels).

There was some discussion about inference rules and integrity rules. Frank and Stefan are working on a paper to describe integrity rules.

ACTION (Stefan): flesh-out DAML+OIL level

Next Week

The DAML-S discussion planned for this week will be deferred to next week.

ACTION (Mike): draft slides for the PI Meeting plenary language presentation.

We'll also plan to finalize the agenda for the Face-to-Face meeting in Nashua. Proposed topics include:


Kelly Barber's raw notes
last week's minutes
Joint Committee home page

$Id: 2001-07-03.html,v 1.7 2001/07/10 21:47:04 mdean Exp $