Joint Committee Minutes 30 January 2001
This page summarizes the telecon for the Joint US/EU
Committee on Agent Markup Languages
held from 1300 to 1400 PST on 30 January 2001.
These minutes were prepared by the chairman,
and were approved during the 6 February telecon.
- Tim Berners-Lee (joined late)
- Mike Dean
- Frank van Harmelen
- Pat Hayes
- Jeff Heflin
- Ian Horrocks
- Ora Lassila
- Deb McGuinness (joined late)
- Peter Patel-Schneider
- Lynn Andrea Stein
As previously agreed, a public archive of the joint-committee email
list is now available
Last Week's Minutes
Mike wants to follow-up with Peter on his
Otherwise, the minutes were accepted.
There was considerable discussion of the
on the table,
mostly focused on the issue of required disjointness of concrete and abstract types
(and the related issue of expressivity vs. implementation effort and tractability).
A few new points were raised:
- in RDF, an rdf:Property without a domain
it can take on either resource (abstract) or literal (concrete) values.
The DAML+OIL(+CONCRETE) specification should address this case:
there may be a middle ground between "disallowing" it in DAML+OIL,
specifying that all results are undefined,
and specifying a full semantics comparable to that for abstract and
- Frank: There is value in having some means of explicitly specifying
that a property is unconstrained
[e.g. AbstractProperty, ConcreteProperty, and HybridProperty
as subclasses of rdf:Property]
rather than just depending upon the absence of a constraint.
- Mike: In RDF and the current DAML+OIL,
the domain of a property is always an abstract type.
The ability to add properties to concrete types
(e.g. integers) is a further language extension that would be
best addressed (if at all) in a subsequent proposal.
The current concrete types proposal(s) just deal with the range of
Lynn asked how DAML+OIL is expected to relate to a future DAML-L.
Some things we may want to do there may not be compatible with the current
concrete types proposal.
Most folks expect that there will be (occasionally incompatible) changes,
and seemed to believe that
we're working on a continuum of language revisions
rather than a series of discrete languages.
Dan Connolly sent word (via Lynn) that he'd like more time to finish up
(now about 2/3 done).
Deb withdrew her previous motion to vote today.
Frank recommended that we not revisit this topic until additional
formal proposals have been received and reviewed by the committee.
DAML PI Meeting
Mike outlined the current plans for the
DAML PI Meeting,
Feb 13 - 15 in DC:
- DAML+OIL plenary presentation
We have 90 minutes, to allocate as we see fit.
|W3C Semantic Web Activities?
- Language breakout session
This is essentially a follow-up to the session at the DAML Kickoff meeting:
a chance for the committee to get feedback from the user community
(e.g. Dan King's
and discuss where we've been and where we're going.
This will not be a formal committee meeting.
Mike may have a simultaneous commitment as DAML Integrator --
if so, he'll ask others to moderate and take notes.
- Birds-of-a-Feather (BOF) sessions
There will be 2 sessions with ~4 BOFs each.
This is a chance for people with like interests to get together.
BOFs will be less formal than the breakout sessions, and
may or may not lead to follow-on activity.
Benjamin Grosof will lead a BOF on DAML Rules,
based on his early work with
Send suggestions for other BOFs to
The following folks indicated their plans to attend the PI Meeting:
- Tim Berners-Lee: no (Ralph Swick will be there; will try to call in)
- Dan Connolly: no (per Tim)
- Mike Dean: yes
- Pat Hayes: yes
- Jeff Heflin: yes
- Ian Horrocks: yes
- Frank van Harmelen: yes
- Ora Lassila: yes
- Deb McGuinness: yes
- Peter Patel-Schneider yes
- Lynn Andrea Stein: ? [missed response]
W3C membership and management is currently discussing a Semantic Web Activity.
Everything will be announced at once
(may or may not be before the PI meeting),
with activities starting at that point.
In lieu of a telecon on February 13,
we'll provide a speakerphone for the Language
breakout session(s) on February 14
(exact time is or will be on the
last week's minutes
$Id: 2001-01-30.html,v 1.14 2001/04/10 09:35:11 mdean Exp $