W3C > WebOnt

Fifth Meeting of the W3C Web Ontology Working Group

9-10 January 2003
Manchester, UK

Agenda/Contents

also: preliminary agenda of 26Dec

• Wednesday, Jan 8
  • 1930 on - Social gathering at KRO² bar (detailslocal arrangement page)
• Thursday, JAN 9: 8:30 AM - 5:30 PM

  Morning:

  • 0830: Entry, getting machines up, etc.
  • 0900: Meeting start - announcements re Local arrangements, by chairs, etc.
  • 0930: Plenary: 30 min round of comments on current WDs
  • 1000: Subgroups for all WDs to discuss documents in detail

    Aim: propose final changes for LC (note: TEST document group will discuss next steps, implementations, etc.)

  1200: Lunch

  • 1300: Plenary: Implementations and demos

    Presentation of 15-20 min. by each (potential) implementor on implementation status/plans

    (expected presentations/demos:

    • Invited Guests: Ray Fergerson (Protege),Jan Wielemaker (SWI-Prolog), Bijan Parsia (UMCP MINDSWAP)
    • WG members: Ian Horrocks/Sean Becherof, Jo De Roo, Jeremy Carroll, Jeff Heflin (?), Dan Connolly
  • 1730: debrief from editors' break-out
  • Evening: Drinks at Lass O'Gowrie, Dinner at Yang Sing (seelocal arrangement page)
• Friday, JAN 10: 9AM - 5:00PM

  Morning

  • 0900: Plenary: Discussion led by Connolly on details of
    1. Needs for LC (and how long to keep it open)
    2. Needs to move from LC to PR (skipping CR)

       expected outcome - Resolutions to move to LC by <date> for <period> for each document

  • 1000: Implementation: Implementation Report, Conformance track (main room):
    • overview by Connolly on WG expectations for implementations

      discussion on which implementations would (potentially) cover which OWL features (i.e. construction the first implementation matrix)

    • strategy for responding to LC comments

  1200-1330: Lunch

  • 1330: Afternoon (plenary, editing of docs may continue):
    • Closure on Requirements
    • Test Document Review
    • AllDifferent
    • planning for LC and CR/PR
    • UML/XML presentation syntaxes (only if still needed)
  • 1600 - 1700: Plenary: WG Schedule (telcon)
    • topics for Tech Plenary

Meeting Start: Attendance, Announcements

from IRC notes starting 09:07:51Z and WG membership records and registration data, internal reg data

1. Jean-François Baget, INRIA
2. Sean Bechhofer, Network Inference
3. Stephen Buswell, Stilo Technology (part)
4. Jeremy Carroll, Hewlett Packard Company
5. Dan Connolly, W3C (team contact)
6. Peter Crowther, Network Inference
7. Jonathan Dale, Fujitsu Limited
8. Jos De Roo, Agfa-Gevaert N. V.
9. D.C. De Roure, University of Southampton
10. Mike Dean
11. Jeff Heflin
12. Ziv Hellman, Unicorn Solutions Inc.
13. James Hendler, Maryland Information and Network Dynamics Lab at the University of Maryland (co-chair)
14. Bernard Horan, Sun Microsystems, Inc.
15. Ian Horrocks, Network Inference
16. Deborah McGuinness, Stanford (in part)
17. Peter Patel-Schneider, Lucent Technologies
18. Guus Schreiber, Ibrow
19. John Stanton, Defense Information Systems Agency (DISA)
20. Herman ter Horst, Philips Electronic N.V.
21. Frank van Harmelen, Ibrow
22. Raphael Volz,, Forschungszentrum Informatik (FZI)
23. Christopher Welty, IBM Corporation

Invited participants:

• Ray Fergerson

  invited from Stanford, Protoge developer

• Bijan Parsia

  UMD MINDSWAP

• Jan Wielemaker

  swi-prolog

• Jeff Pan, u. Manchester
• Berkowitz (sp?)

excused: Barnette, Eshelman, Euzenat, Fensel, Finin, Hori, Kravtsova, Maedche, Marchiori, Motta, Pike, Trastour

regrets: Buswell (2nd day), Borden, Gibbins, Hayes, Klein, Lassila, McGuinness (remote only), Obrst, Sabbouh, Shimizu, Michael Smith, Stein, Thompson, Wallace

??: Brysse, Iannuzzelli, Jeckle, Olivry, Sintek, Ned Smith, Stickler, Yanosy

Announcing the First owl.owl award

Hendler presented Frank van Harmelen with the owl.owl award for brokering the semantic truce in Bristol.

photo by Bijan

Working Draft review points, around the table

Hendler: editors: pls note new copyright statement updated for 2003 and INRIA->ERCIM transition.

We went round the table, giving comments on the WG documents. seeIRC notes starting 09:13:03Z

Some email comments were noted:

• Dave Beckett's review of Semantics
• Brian McBride's review of Ref
• Mike D's email re: datatypes
• OWL Reference for F2F 5

ACTION Carroll: - add suvj pred aClass test case in I5.3 (09:43:01)

ACTION Carroll: - check sameInstanceAs/sameClassAs tests for FULL clarify different semantics (09:58:38)

No decisions were made.

Document Review Break-Outs

informal notes in irc notes start at 10:52:53Z

After some discussion of informative material (esp. role/purpose of Reference and Feature Synopsis), break-out groups were formed:

1. Guide breakout notes are in the IRC log, starting at 11:31:40Z
2. reference break-out
3. synopsis break-out
4. Semantics Breakout

we resumed in plenary at 13:33:26 (per IRC notes) to debrief from the break-outs.

ACTION Welty: explain that "in OWL Lite, all names need types" in the guide (13:43:49)

ACTION Carroll: writes a section about the constraints to go from RDF to OWL-DL/LITE (13:56:05)

No decisions.

Afternoon session - implementors

Ray Ferguson - Protege (14:51:13)

@@presentation materials?

ACTION IanH: explain how to say "range of P is A union B" in owl lite (15:07:55)

Bijan from Mindswap (15:14:53)

• Slides from MIND lab presentation at WOWG f2fJim Hendler (Thu, Jan 09 2003)

Network Inference - Sean B. (15:30:27)

@@presentation materials?

Jan from SWI Prolog (16:02:22)

@@presentation materials?

Raphael Volz (16:24:02)

@@presentation materials

Jeff Heflin (16:36:18)

@@presentation materials

Jos De Roo (16:42:12)

• OWL implementation experience in Euler

in which Jos was nominated for the 2nd owl.owl award for demonstrating running code passing (all?) our tests.

Mike Dean (16:55:02)

@@presentation materials

Dan Connolly (17:21:40)

• Semantic Web Travel Tools
  • presentation given Oct in Bristol
• cwm/otter axiom tests

Editor's Break-out: debrief

also:IRC notes 17:29:33

Chris Welty reports back from break out group; see

• editor's break-out

  and

• Editors: consistent use of terms in webont docs

We adjourned for the day at 17:39:16, and headed to the pub...

W3C end-game Process

ircnotes [09:20:33]by pfps

Connolly explains W3C last call process, associated risks, and an exemplary last call review summary (aka disposition of comments), how to request CR/PR, and other collected wisdom from the Art of Consensus guidebook.

Implementation Experience, Conformance

seeirc notes starting 10:27:20

Hendler reviews technical relationships between Lite, DL, and Full, test methodology, and the difference between defining/checking conformance and demonstrating implementation experience.

The WG discussed the costs and benefits of specifying conformance of classes of software in addition to the (unanimously supported) specification of classes of documents.

12:01:19

RESOLVED: add a test document section for Syntax checker that can tell the difference between OWL sublanguages
Abstain: Connoly, Hendler
ACTION Carroll: to add a test for syntax checker
ACTION Patel-Schneider: add definition of OWL Lite document

12:09:44

RESOLVED: The test document should specify the conformance clauses for OWL Lite, DL, and FULL documents

[13:58:13]

RESOLVED: that the OWL test document shall specify conformance of consistency checkers {Lite, DL, Full} x {Incomplete, Complete}. Complete is to be understood as "Logically complete". DeRoo, Hendler, Connolly, De Roure Abstaining,
Action: De Roo

Closure on Requirements

irc notes starting 14:02:49

discussion of

• Action: Review of satisfaction of requirements Jeff Heflin (Mon, Dec 30 2002)
• F2F Version of Requirements Document Jeff Heflin (Thu, Jan 02 2003)

[14:11:31] RESOLVED: Req Doc to be included in last call (i.e. Recommendation track); abstains IanH, Patel-Schneider
ACTION Schrieber: to get it into road map

ACTION Hefflin: update terminology (e.g. object) in requirements document

RESOLVED: to publish Requirements as a WD, subject to editorial discretion.
ACTION Heflin.

ACTION Connolly: figure out privacy, security risks / issues in spec

Test Document Review

irc notes starting 14:20:16

discussion of

• TEST: document issues for f2f

ACTION Caroll: teach sean bechhofer, peter crowber editing/creating tests

Action De Roo: Put examples from guide into test suite (done 14Jan)

AllDifferent

irc notes starting 15:18:10

RESOLVED: to close the AllDifferent issue using an AllDifferent class and a distinctMembers property (to be elaborated by the semantics editors). Connolly, Carroll, Bechhofer, Dale, Volz, Horan abstaining.
ACTION guide editor (Welty)
ACTION semantics editor (Patel-Schneider)
ACTION ref editor (Dean).

ACTION Patel-Schneider: relay feedback on parseType="Collection"

WG Schedule

seeirc notes starting 15:48:59

Discussion of

• Guide: New version Smith, Michael K (Tue, Jan 07 2003)

RESOLVED: to publish Guide as last call, after edits as actioned in this ftf and editorial discretion, subject to review by Guus and Jeff H. Carroll, ter Horst abstaining
ACTION Welty: provide guide last call candidate by 15Jan
ACTION Schreiber: to review by 21Jan
ACTION Heflin: to review Guide by 21Jan
ACTION Connolly: to publish Guide, after Welty/Guus/Hefflin do their part of Guide.

Discussion of taking Abstract Syntax and Semantics to last call exposed outstanding technical issues.

ACTION Carroll: investigate this issue (rdfs:Literal etc.) in RDFCore
ACTION JimH: ask SemWeb CG about RDF model theory.
ACTION Patel-Schneider: write to webont WG (only to webont WG) about the impact of this on OWL.

Ongoing reviews of various sections of the semantics document were noted:

ACTION Carroll: review AS & S Section 4

ACTION Horrocks: review AS & S Appendices

Also...

ACTION Patel-Schneider: keep Dave Beckett informed of status of his comments.

Connolly explained the critical path:

1. WG decides to go to last call (all issues are resolved)

   some epsilon later...

2. LC publication

   at least 3 weeks later...

3. WG requests PR

   approx 2 weeks later

4. PR publication

   exactly 4 weeks later

5. Membership reviewes due

   approximately 2 weeks later...

6. Rec party and press release

We discussed schedule details up thru last call publications...

10 Jan (Fri)

today. Decide to take Guide, Requirements to last call.

16 Jan (Thu)

next telcon. Guide last call candidate available (ACTION Welty above).

test last call candidate hoped/planned.

17 Jan (Fri)

RDF Core specs to last call [stay tuned to RDFCore schedule]

23 Jan (Thu)

Guide reviews due 21 Jan (ACTION Guus, Heflin above). Guide is published, contingent on successful review.

Semantics reviews due (ACTION ter Horst, Pan) Decide to take Semantics to last call.

Ref last call candidate due 20 Jan.

Overview (nee Feature Synopsis) draft hoped for by 20Jan

30 Jan (Thu)

Overview reviews due (ACTION Welty, Horan).

Ref reviews due (ACTION Volz, Carrol/HP, Hendler)

test reviews due (ACTION Connolly, Stanton)

Connolly observed that this schedule implies we continue to meet (i.e. have telcons) past 31 Mar 2003.

Connolly reviewed tentative plans for a Semantic Web Architecture Meeting 6-7 March 2003 in Cambridge, as part of the W3C technical plenary.

Adjournment

RESOLVED: to thank the hosts, with applause.

The meeting was adjourned at 17:20Z.