November 13, 2001

Attendees:
Ora
Harold
Ian
Peter
Frank
Lynn
Pat
Sandro
Richard
Dan C.
Tim B-L


Announcements:

	Deborah is maintaining
	http://www.ksl.stanford.edu/people/dlm/daml/modelingIssues.html
	as a repository for ``Tricks of the Trade''.  Send mail to her
	(and cc joint-committee) with any additions you want.


Agenda:

Minutes for 30 October 2001 - approved

Submission news (Peter)

	   Almost all the required information has been collected.
	   Even the Lucent lawyers have responded. Package has been
	   sent in as of 2pm this afternoon.

WebOnt news (Peter/...)

       We had an administrative telecon yesterday.  Next telecon will
       be in about two weeks.  The first face-to-face is scheduled at
       Bell Labs 14 and 15 January 2002.

       More than 40 people on the working group ... 

       With the number of people interested in this, we will be
       prompted to create more educational materials.  Perhaps a
       walkthrough of the walkthrough or a tutorial.

       Jim talked about the charter and what we're not doing :)

       homepage for this group: http://www.w3.org/2001/sw/WebOnt

possible (short) report on DL+LP rules from Benjamin and Ian

	 Ian: telephone conversation last week or so.  nothing much
	 happening so far.

	 Harold(?): RuleML has made some progress ... changing DTD's
	 to XML Schema's ... the inheritance is quite difficult.

	 To provide a firm semantic ground for all of XML Schema ... a
	 courageous thing to do...

	 Harold: in February we will organize a seminar on rule markup
	 techniques.  Have about 20 participants so far, can handle
	 more. http://www.dagstuhl.de/DATA/Seminars/02/#02061 or maybe
	 http://www.dfki.uni-kl.de/ruleml/rmt?

	 Question: it seems that rules were important to get done, but
	 it is moving slowly.  Is there something more we need?

	 Answer(?): we have lots of people who can't commit a large
	 chunk of time to this, and that makes it hard.

	 This seems related to layering in DAML+OIL.  If you have
	 varying layers of expressiveness you can add rules to those
	 layers.

Richard's query issues summary

	  No feedback to the emails of the last days. 

	  Issue: these blank answers are clearly existential variables
	  who's scope is the query itself and not the knowledge base.
	  Are querries to be seen as independant, or are they part of
	  a query process (can I use these blank answers in a new
	  query ... are these two the same?)

	  Ian: This seems like running when we're not sure we can
	  crawl yet.

	  If you can query incrementally, it keeps each query simple
	  since you can always ask more.

	  Talking about the *how many answers* question. Is this too
	  hard?  I got lost in this discussion a little...

	  Peter: Anything that will happen in the next week?

	  Richard: I'm waiting for responses on what I put out.

	  Peter: Is it possible to get a couple of use cases?

	  Richard: more examples would be useful.