Joint Committee Minutes 28 August 2001
This page summarizes the telecon for the Joint US/EU
Committee on Agent Markup Languages
held from 1300 to 1400 PDT on 28 August 2001.
These minutes were prepared by the chairman,
and were approved during the 4 September telecon.
Participants
- Harold Boley (invited)
- Murray Burke (invited)
- Mike Dean
- Stefan Decker
- Richard Fikes (invited)
- Benjamin Grosof (invited)
- Frank van Harmelen
- Pat Hayes
- Ian Horrocks
- Ora Lassila
- Deb McGuinness
- Peter Patel-Schneider
- Lynn Andrea Stein
Announcements
A new version of the axiomatic semantics is available (see below).
Last Week's Minutes
The minutes from August 21 were approved by those present.
WebOnt Concerns
We discussed some of the issues alluded to in
Dan Connolly's message.
Frank van Harmelen summarized the results
here.
Membership Updates
In separate actions, the committee voted to:
- invite Richard Fikes, Harold Boley, and Benjamin Grosof to join the
Joint Committee
- invite Murray Burke to join the Joint Committee as U.S. Government representative
- invite Jim Hendler (current U.S. Government representative) to join the committee as a "regular member"
Welcome to all of our new members!
Lynn noted that we may need to take more specific action to get more W3C participation in the Joint Committee now that Dan Connolly has resigned.
Ora Lassila is a member of the DAML-S Coalition.
ACTION (Ora): check with the DAML-S Coalition to see if they would like further representation on the Joint Committee.
DAML+OIL Copyright
RESVOLVED: Peter will submit DAML+OIL (March 2001) to W3C via
Lucent, which is a member of W3C.
The Lucent internal review process should take a week or so.
ACTION (Peter): publish submission package, allow a couple
days for Committee review, clear Lucent's process, and submit to W3C.
Cyclic Subclasses
We discussed the recent response from the RDF Core WG
opposed to our suggestion to allow cycles in rdfs:subClassOf
and rdfs:subPropertyOf,
as relayed by Pat Hayes during our
August 14 telecon.
Frank subsequently summarized our discussion
internally
and
externally.
Frank suggested that we not change our approach now.
RDF needs sameClassAs if not cyclic subclasses.
Ian noted that we can get implicit cycles in defined classes and
inferred subclasses
even if they're prohibited in subClassOf.
We could define our own daml:subClassOf allowing cycles
(rdfs:subClassOf implies daml:subClassOf),
but this hurts layering by limiting the ability of RDF-only tools to
understand DAML+OIL hierarchies.
It was noted that forbidding subclass loops does not
necessarily require proper subclasses
(see the
continuing email discussion).
ACTION (Peter): investigate the impact of prohibiting rdfs:subClassOf cycles on the DAML+OIL formal semantics
Query Language
Richard had previously posted the KSL
strawman proposal,
but we ran out of time this week.
We'll discuss it next week.
ACTION (everyone): review the proposal and send feedback to Richard.
Some things to particularly think about: operators, aggregation (max, min, sum, etc.).
Axiomatic Semantics
Richard and Deb discussed their most recent updates to the axiomatic semantics
for DAML+OIL (March 2001).
The updated semantics and release notes are available
here.
Next Week
We'll discuss the proposed DAML Query Language (see above).
Links
IRC
Kelly Barber's raw notes
last week's minutes
Joint Committee home page
$Id: 2001-08-28.html,v 1.6 2001/09/05 20:35:16 mdean Exp $