ADMIN: Minutes of June 20 telecon

June 20, 2002 telecon

Chair:  Guus Schreiber
Scribe: Larry Eshelman

1) ADMIN

1.1 Roll

Present:
Schreiber
Volz
Eshelman
Horrocks
Patel-Schneider
Heflin
de Roo
Wallace
Carroll
van Harmelen
Thompson
Marchiori
Borden
Finin
DeRoure
Connolly
Stein
ter Horst
Welty
Trastour
MacGuinness

Regrets:
Dale
Gibbins
Smith
Stanton
Miller
Hendler
Dean

1.2 Minutes June 13 telecon

Dan Connally: minutes should be mailed with 2 days.

ACTION:  Dan Connolly to fix last weeks minutes

1.3 Agenda amendments

NONE

1.4 Next meeting

- 27Jun, per regular schedule.
- scribe volunteer:  Jeremy Carroll

1.5 F2f admin issues

Debra McGuinness:
- request to register right away, deadline June 21st
- parking: either buy tag or bring a lot of quarters
- need another high resolution projector

Typos in agenda:
- July 1st: "Extra-logical" time should be 15:00-15:30
- July 2nd:  "slot for unfinished discussion" time should be 15:00 - 15:30

2) ACTION ITEM REVIEW

ACTION - Connolly: to arrange direct CVS access for Carroll:
Working on it

ACTION - McGuinness, van Harmelen, Hayes: draft Feature Synopsis OWL
by June 24:
On target, will hit

ACTION - Dean: provide draft OWL Reference Description by 17Jun
DONE: see 3.1 [as of June 13 telecon]

ACTION - Patel-Schneider: produce draft Formal Spec OWL
DONE: See 3.2 [as of June 13 telecon]

ACTION - Welty, Hayes, Hendler, Connolly to review updated requirements WD
Chris Welty: in progress

ACTION - Heflin, Carroll, Borden, Volz to review XML Presentation Syntax document
Review by Raphael Volz, others continue

3) UPDATE ON RDF Core FTF (RDF Core members of WOWG, 5-10 min)

Jeremy Carroll:
- made quite a bit of progress
- changed minds on data types (still not closed, progressing)
- axiomatic treatment of formal semantics: model theory doc called a formal semantics,
  normative content unchanged
- reserved vocabulary (dark triples): preference given to syntactic marker
Jos De Roo
- all issues resolved but data types
- dark triples: function of literal or function of literal plus data type
- lengthly discussion of reserved vocabulary: compute reserve vocabulary, triggered by URI

No action for WebOnt on dark triples at the moment.

4) ISSUES (15 min)

2.1 Pending updates:

Issue 5.11 ISSUE hasClass/ToClass-names
RESOLVED on Jun 13 telecon

2.2 Issue 4.6 "EquivalentTo"

Issue had been raised by Jim Hendler
So far only one comment (by Jeff Heflin)

Jeff Heflin recapped his main comment:
Issue of whether illegal in OWL to be both, e.g., property and class, or
class and instance.

Ziv Hellman proposed a level version

Peter Patel Schneider: thought level version easier to deal with

Ian Horrocks: would like to hear some good examples in order to
motivate requirement - example where expect particular behavior
to result from this kind of usage (resulting behavior of any reasoner).

Jonathan Borden: integration of XML schema types with OWL - when
are the times we need to treat classes as instances?

Dan Connolly: doubt if problem with XML schema types

VW beetle example (instance: model or car on lot) for different
ontologies.

ACTION: Dan Connolly: to produce two test cases: (1) (classes as instances)
from requirements document (2) oracle data base example

ACTION: Jeremy Carroll: will also produce a test case

5) TEST work (focus area participants)

Jeremy Carroll: need to decide scope of tests, (i.e., what the goals are,
what do we want the test cases to show?): range from testing every feature,
to more specific issues.  Would like 3 or 4 suggestions.

Dan Connolly: tests are useful when discussing issues -- i.e., having tests
following issues.

Discussion about how difficult it would to convert Ian's test
cases from LISP.

Debra McGuinness: has set of test cases, but doesn't have time to do
translation.

At f2f to look, informally, at existing test cases.

Jos De Roo: fixing of language makes it a lot easier to work with test cases

General consensus of discussion: test cases should be issue based

6) DOCUMENTS (editors)

3.1 OWL Reference description (Dean)

Dan Connolly summarized for Dean: most comments, suggestions
have come Peter.  Peter thought title should be
"Language Description" rather than "Reference description"

3.2 OWL Formal specification (Patel-Schneider)

No questions.

Important to read documents before f2f so can make resolutions at f2f.

7) A.O.B

NONE


Philips Research USA
**My opinions do not reflect those of my employer**

Received on Thursday, 20 June 2002 14:16:38 UTC