2 May WebOnt telcon minutes for review [was: Updated WOWG Telecon Agenda]

On Wed, 2002-05-01 at 08:53, Jim Hendler wrote:
> May 2, 2002


log: http://www.w3.org/2002/05/02-webont-irc

> Chair: Hendler
> Scribe: Tim Finin
... and Dan Connolly

> 1) Join call/attendance/admin (10 min)

roughly these folks:

PeterPS?, MDean, Guus, SDecker, LarryE?, ZivH?, DanC, TimFinin?, JimH,
Massimo, McGuinnes?, Ian.Horrocks, FrankvH?, Libby, Evan.Wallace,
Rudiger?, ??P52, HermanTerHorst?,
... JosD, ??P55, NickG, DavidT?, ChrisW, Marwan.Sabbouh,
Jonathan.Borden, ??P0

> Regrets: (please send to chair before May 2)

there were some; I don't have them handy. Jim,
maybe you could follow up?

> 2) ACTION item review (chair, 10 min)

previous telcon was 25 Apr

> ACTION (Mar 28) Chairs to clarify OWL naming problem
> DONE - (Dan C.reported to telecon)

confirmed, DONE.

> ACTION (Mar 28) Dan Connolly, Lynn Stein (prov.), Jos De Roo, to
> participate in RDF core discussions on construct for closed lists
> Committee formation - DONE

another action was added...

> Committee discussion - ONGOING  (See Agenda item 4)

> ACTION (Apr 9): Frank van Harmelen with Deborah McGuinness,
>    Mike Dean, Enrico Motta, Ziv Hellman, Raphael Volz, Ian Horrocks
> Following polls on separation of language features between
>    level 1 (OWL-lite) and level 2 (OWL-full), a group was constituted to
>    revisit the level 1/2 conformance issue: 
> http://www.w3.org/2002/04/09-webont-irc#T14-08-19


> >ACTION: (Apr 18) Dan Connolly to arrange direct CVS access for
> >appropriate
> >   members [Jeremy Carroll, Jos de Roo] of the test focus area to that
> >repository.

CONTINUES; little, if any, progress.

ACTION ChrisW: review UML/OWL stuff continues.

ACTION PatH: review UML/OWL stuff continues.

ACTION Evan: review UML/OWL stuff continues.

> 3) Compliance levels - Approach and KR report out (Frank van Harmelen, 30min)
> ISSUE Opened:
>    http://lists.w3.org/Archives/Public/www-webont-wg/2002Apr/0369.html
> Discussion:
>    http://lists.w3.org/Archives/Public/www-webont-wg/2002Apr/0380.html

Frank vH reported on discussion so far...
see discussion...

DanC noted that the Amsterdam meeting record didn't show
a decision to split into 2 levels; DebM asked to get
such a decision. A straw poll showed more
in favor than against (~11 vs !7) but no consensus.
 issue 5.2 Language Compliance Levels remains open

JimH: 0329 seems pretty close; no major objections

RESOLVED: to open 3.1 Local Restrictions, assigned to DebM

ACTION DebM: write up "these seem to be agreed;
	these have issues; these look harder"
  DebM warned that the normal "one week later" timeline
  wasn't likely, but the following weekend looked workable.
  due: 13May.

... mostly just re-iterating
but relating it to the issues list.

DebM asked for clarification about what it means to
open and own an issue; DanC/JimH replied that
opening an issue means it's in order to discuss
it on the mailing list, and owning and issue
means being prepared to report, at each telcon,
on the status of the discussion, and in particular,
being prepared to propose a resolution when it looks like
there is one that might get consensus.

> 4) Coordination with RDF Core - reports
> Closed Lists - Update (Connolly?, up to 10 min)

DanC reported on ...

 * addressing requirements around daml:collection
 Dan Connolly (Fri, Apr 19 2002)

asked if the 'longhand-OK' proposal was acceptable;
JimH, FrankvH suggested no, it's not.

> Unasserted triples - Update (Jon Borden, up to 10 min)

Borden reports: lots of traffic, much of it good stuff...
... theoretical and practical challenges in semantic layering ...
... as to specifics, PeterPS showed a class with a cycle in it; but
something's not clear about what's entailed...

discussion of entailment (DanC - explains issues and paradoxes)

Borden: the "comprehensive entailments" proposal looks perhpas
promising, but it also looks like research; not clear that we could work
out the details in a month or two.

Borden: I'd like help making test cases...

DanCon: did you see jeremy's test cases? I think PeterPS had some too...

Borden: but I don't see how those motivate dark triples

PeterPS: if we're after test cases that show the *problem*, see the
circular paradox gizmo. It does use cardinalityQ; I could get rid of
that using complementOf, but that gets hairier

In IRC, PeterPS provided:

_:1 fowl:onProperty rdf:type .
_:1 fowl:hasClass _:2 .
_:2 fowl:OneOf _:3 .
_:3 fowl:first _:4 .
_:3 fowl:rest fowl:nil .
_:4 fowl:complementOf _:1 .

  _:1 is the set of objects
  that are related to a particular complement of _:1
  via rdf:type

  if x rdf:type _:1
  then x rdf:type _:4
  but _:1 and _:4 are complements

  so not x rdf:type _:1
  if not x rdf:type _:1
  then x rdf:type _:4
  because _:1 and _:4 are complements

  but then x rdf:type _:1

  the above is another problematic situation

JimH: in sum, we're still working on SEM/dark triples stuff.

4A: WebOnt review of RDFCore working drafts

JimH: Note well, RDF Core has released drafts; as individuals,
  we're all welcome to review.
  But more weight is given to group reviews.

JimH: we should take a look at RDFS, which we're basing level
  one on, and see that it's described in a good way

Jeremy: it makes sense to align drafts with focus areas:
  Schema/LANG, GUIDE/Primer, test cases/TEST, model theory/SEM

ACTION chairs: solicit reviewers for RDFCore drafts

> 5) Issue review  (Hendler, 10 min)
> Issues list: http://www.w3.org/2001/sw/WebOnt/webont-issues.html
> Issue 3.2: Resolved on April 25 telecon - resolution recorded?
> Discussion of issue to open next.

JimH: compliance levels, dark triples, [5.3 Semantic Layering]
  remain open.

Today we opened 3.1 Local Restrictions.

> 6) A.O.B


NOTE WELL: no teleconference 9 May; next telcon is 16 May.

Dan Connolly, W3C http://www.w3.org/People/Connolly/

Received on Thursday, 2 May 2002 16:16:25 UTC