Webont 2001-11-12 teleconference (for review) from Dan Connolly on 2001-11-12 (www-webont-wg@w3.org from November 2001) (original) (raw)

Jim Hendler wrote: http://lists.w3.org/Archives/Public/www-webont-wg/2001Nov/0043.html

Date: Monday Nov 12

Attendance: 25 present

Y Jeremy Carroll, Hewlett Packard Company Y Dan Connolly, W3C, Team contact Y Jonathan Dale, Fujitsu Limited Y Mike Dean (invited expert) Y Stefan Decker, Stanford Y Tim Finin, University of Maryland MIND Laboratory Y Nicholas Gibbins, University of Southampton Y Jeff Heflin (invited expert) Y James Hendler, Maryland Information and Network Dynamics Lab at the University of Maryland (chair) Y Ian Horrocks, Network Inference Y Ruediger Klein, Daimler Chrysler Research and Technology Y Michael Kohlhase, German Research Center for Artificial Intelligence (DFKI) Gmbh Y Ora Lassila, Nokia Y Deborah McGuinness, Stanford Y Libby Miller, University of Bristol Y Leo Obrst, MITRE Y Peter Patel-Schneider, Lucent Technologies Y Marwan Sabbouh, MITRE Y Michael Smith, Electronic Data System (EDS) Y Ned Smith, Intel Corporation Y Lynn Stein, invited expert Y Herman ter Horst, Philips Electronic N.V. Y Lynne R. Thompson, Unisys Corporation Y David Trastour, Hewlett Packard Company Y Frank van Harmelen, Ibrow

Regets:

NR Oisen Hurley, Iona Technologies, Inc. NR Guus Schreiber, Ibrow NR Michael Sintek, German Research Center for Artificial Intelligence (DFKI) Gmbh NR Warner ten Kate, Philips Electronic N.V.

AGENDA The Agenda for this meeting is as follows. If you have additions to the agenda, please email them to the webont group list before the start of the teleconference.

  1. ACTION ITEM REVIEW - Jim H (2 min)

Completed actions: ACTION DanC: announcement email to Rdf-logic, rdf-ig, rdf-rules ACTION JimH: introduction email (and suggest people send a "who I am") ACTION: DanC arrange bridge for: first phone call - Nov 12, 1500 EST ACTION DanC - add Jim to WG chair list ACTION - Jim H mail to invited experts with how to join ACTION Jim - getting started email to WOW-G ACTION DanC - Working Group list doesn't yet include webont (in the T&S/SW area)

so noted.

Open actions: ACTION JimH - "roundtrip" email or phone with all working group members

a few outstanding.

  1. Administrivia - Jim H. (~10 min) Choice of scribe for this phone call. Discussion of best time for future telecons (constraints and process, not discussion of particular times - that will happen through email)

week of 22 Nov is not likely.

stay tuned for news of next telcon sometime in the week of 26 Nov

ACTION JimH: request telcon schedule input. NOTE WELL: everyone respond soon.

Discussion of first f2f - when, where, expectation issues.

NOTE WELL: first face-to-face meeting is scheduled for 14/15 Jan 2002 in Murray Hill, NH, near Newark airport.

ACTION PeterPS: send request regarding non-US citizens getting into Lucent. (done: http://lists.w3.org/Archives/Public/www-webont-wg/2001Nov/0061.html)

  1. Introduction and Expectations - Dan Connolly (~20 min) quick overview by Dan of what is expected of working group by when. Not a detailed process review (which will be presented on a later call). quick overview by Dan of expectations on WG members - what you need to read, what you need to write, what you need to program Group questions re: process/expectations

[no decisions/actions. see log for details]

  1. Ontology - vision/goals; charter review - Jim Hendler (~20 min) Discussion of ontologies on web, what WG is expected to do (and NOT do). Review of charter [1] [1] http://www.w3.org/2001/sw/WebOnt/charter

[no decisions/actions. see log for details]

Group questions re: goals, charter.

  1. Open discussion: issues for group, needs of members, items for getting started - Jim H. moderate (20 min)

[no decisions/actions; see log for details.]

IRC log follows... (taken from http://ilrt.org/discovery/chatlogs/webont/2001-11-12.txt)

20:00:59 =========== telecon starts =========== 20:01:07 Frankh has joined #webont 20:01:09 JosD has joined #webont 20:02:31 wow... 17 on the phone, per http://www.w3.org/1998/12/bridge/Zakim.html 20:02:33 jdale has joined #webont 20:02:41 hello 20:02:49 * DanC joins 20:03:12 hi jon - how's SF today? 20:03:23 12 Nov agenda: http://lists.w3.org/Archives/Public/www-webont-wg/2001Nov/0043.html 20:04:05 pretty wet - very bad rain this morning 20:04:25 and actually a little lightening and thunder - odd for sf 20:04:28 not a good day to be driving - i saw at least 8 accidents on the 280 20:04:45 seems to have calmed down a bit now 20:05:14 las has joined #webont 20:05:39 JimH: pls use the IRC channel for meeting business (including getting the chair's attention, clarifying). 20:05:43 dlm is now known as DeborahMcGuinness 20:05:54 are we supposed to be phoning in, too? 20:06:01 yes 20:06:04 ah 20:06:10 so what's the IRC session for? 20:06:26 just for scribing and for clarification 20:06:36 * jdale dials in 20:06:41 timfinin has joined #webont 20:07:26 stefanjdecker has joined #webont 20:07:28 ora has joined #webont 20:08:26 hi tim 20:09:31 $ grep ^Y nov12telcon|wc 20:09:31 24 162 1186 20:09:48 ===== Roll Call 20:09:59 24 present. (details to be provided @@) 20:10:28 ===== Agenda review 20:11:40 JimH: administrative: during discussion items, please request the floor; the chair or discussion leader will give you the floor 20:13:03 +Ned Smith, Intel Corporation 20:14:39 == Teleconference times. 20:14:52 * las wonders what telecon frequency is expected to be 20:15:12 JimH: we span something like 14 time zones; it's unlikely we'll find something convenient to everyone. 20:15:31 * ora thinks once a week is typical 20:15:32 DebM: we could perhaps alternate times 20:15:53 stefanjdecker has quit 20:16:08 ACTION JimH: request telcon schedule input. NOTE WELL: everyone respond soon. 20:16:18 == Face to face schedule 20:16:27 JimH: PeterPS has offered to host 1st ftf 20:16:53 PeterPS: murray hill near [??] airport. [... scribe missed lots...]. trying to get a wireless net.... 20:22:39 <irc.openprojects.net> Disconnected from irc.openprojects.net (ERROR :Closing Link: logger_1[tatooine.ilrt.bris.ac.uk] by niven.openprojects.net (Ping timeout for logger_1[tatooine.ilrt.bris.ac.uk])) 20:23:02 logger has joined #webont 20:23:02 <asimov.openprojects.net> topic is: W3C WebOnt WG (http://www.w3.org/2001/sw/WebOnt/) 20:23:02 <asimov.openprojects.net> Users on #webont: logger las jdale JosD kohlhase Ian DeborahMcGuinness nmg pfps @JHendler @dajobe @DanC 20:23:06 current events DanC will try to keep w/in a week or two of up to date 20:23:14 telecon: we're in, you're all here... 20:23:22 stefanjdecker has joined #webont 20:23:33 membership in the group never closes, anyone who joins late has to get self up to speed, but you can always in principle join. 20:23:42 JimH: we are not actively seeking new members. 20:23:43 ora has joined #webont 20:23:51 * las apologizes, her phone connection is going in and out. 20:24:18 schedule/milestones dates got out of synch between when they were written and now. 20:24:27 jimh will update 20:25:06 recommendation is highest formal goal 20:25:18 something i missed precedes recommendation 20:25:37 candidate recommendation precedes that, has several implementations. 20:25:56 proposed rec is what i missed, and some wgs skip cand rec and go straight to prop rec if they're really in order 20:26:07 publish working drafts at least every 3 months 20:26:34 when a significant body of work is done, danc will push us to publish as a working draft 20:26:48 we can change working drafts, but as we go further along we should try to change less. 20:27:02 ian horrocks: what constitutes an implementation wrt the kind of thing we're looking at? 20:27:27 DavidTrastour has joined #webont 20:27:32 danC: I hope we'll build test suites and tools, but we'll have to talk about that as we go along 20:28:21 note sched/milestones must be agreed upon with semantic web coordination group; others are waiting for our resources (including danc only committed for one year) 20:28:32 membership: pretty much everyone here has jumped through the hoops. 20:28:53 jimH: if you haven't send intro yet, please do so ASAP. 20:29:35 danC: intellectual property: goal is to make sure that anything we come up with is usable by entire web community. there are many ways to get there. 20:30:01 could be that we need to get some more formal papers signed for working draft/face to face. 20:30:17 we're a large group, 40+, will need to figure out how to be effective 20:30:45 everything is published, we'll keep developer groups active, anyone who doesn't need to be a member is welcome not to. 20:30:59 JimH: Please say more on what's expected in terms of participation. 20:31:06 DanC: Charter says 1 day/week. 20:31:20 We expect folks in wg to be developing test cases, writing parts of document, ... 20:31:53 If your roll is mostly to keep track of what's going on and report to someone else, you probably don't need to be a member. Members are expected to do substantial technical development. 20:32:15 History section including technical history.... 20:32:42 First item is process history, not technical background (http://www.w3.org/Consortium/Process-20010719/) 20:33:05 DanC's job is to keep us all aware of what parts we need to know, but we're welcome to read whole thing. 20:33:51 dajobe has quit 20:33:54 Quick runthrough of obligations that make us mutually effective. 20:34:02 (from process document) 20:34:46 Art of consensus is supplemental, mostly aimed at chairs, less "rules" and more guidelines (member confidential) http://www.w3.org/Guide/ 20:35:12 editors will also need art of consensus, also f2f hosts. 20:35:30 technical background docs 20:35:39 crash course was airport exercise 20:35:46 danc is compiling lessons learned 20:36:40 uris are important technologies....all should read referenced documents 20:37:01 xml is important and the soohoo document is highly recommended 20:37:15 xml namespaces; rdf; rdf schema. 20:37:25 (all of the urls are on the webont home page) 20:37:35 Missing from homepage: Link to rdf validator. 20:38:01 daml/daml+oil not a lot of pointers on the page....there are other things that could be there: daml ontology library, .... 20:38:41 jimH: I don't yet have write permission for the home page. i will try to prioritize. You need some familiarity with all, but may wind up with expertise in some. 20:38:45 thanks for scribe-help, Lynn 20:38:56 jimH: everyone will need to know daml/daml+oil 20:39:07 * las is wondering whether danc has stepped down, so she can. 20:40:11 JimH: if we recon Nov 2001 as our start date, ... 20:40:17 Apologies to all, but I have a conflict and need to duck out now.... 20:40:35 ... three months later is Jan 2002, just after our ftf. 20:40:56 las has quit 20:40:59 ... so we'll need to be ready to publish something soon after our ftf; so we'll need to do significant preparation beforehand. 20:41:44 ==== 20:41:46 http://www.w3.org/2001/sw/WebOnt/charter 20:42:06 === 4) Ontology - vision/goals; charter review - Jim Hendler (~20 min) 20:42:54 Frankh has joined #webont 20:43:16 JimH: technical scope in the charter is up to date; schedule stuff needs revision, but the technical scope is set. 20:44:28 JimH: there was a lot of discussion of what the scope should be; there's a spectrum that goes out to hard research problems; if you've seen TimBL's layer cake, we fit in at the ontology layer. 20:44:34 ... note we're not doing rules. 20:44:48 [... more summary of the charter ...] 20:45:15 libby has joined #webont 20:46:58 ... we expect that a formal semantics for this language is necessary; we'll either maintiain the DAML+OIL model-theoretic/axiomatic semantics or draft some other formal semantics. 20:47:18 TimBL's layer cake http://www.w3.org/2000/Talks/1206-xml2k-tbl/slide10-0.html 20:47:32 JimH: some background on DAML+OIL ... 20:48:55 ... SHOE had a frames/rules flavor; OIL was based on description-logic stuff... 20:49:37 ... mixing in web technologies introduced a few interesting issues... 20:50:25 ... "The products of the WebONT group should not presuppose any particular approach to 20:50:26 either ontology design or ontology use. In addition, the language must support the 20:50:26 development and linking of ontologies together, in a web-like manner. " 20:51:46 axiomatic semantics - http://www.ksl.stanford.edu/people/dlm/daml-semantics/abstract-axiomatic-semantics.html 20:52:13 model theoretic semantics - http://www.daml.org/2001/03/model-theoretic-semantics.html 20:53:32 JimH: ... "maximum compatibility with XML and RDF language conventions. " so we'll be working with RDF Core on things like model theory, datatypes, etc. 20:54:30 JimH: unlike groups like the "standard upper ontology" group, we don't intend to standardize any particular base of knowledge, but rather a language for expressing ontologies. 20:55:06 ... e.g. if someone proposed "this is how time should be represented..." that would be out of scope. 20:56:25 ... out of scope: query rules, query language, Universal Web Logics 20:56:49 * DanC wonders who's asking 20:56:59 NedS: how do you see DAML-S fitting in? 20:57:25 JimH: I think we want to support that sort of work, but DAML-S itself is out of scope. 20:57:53 NedS: how about as an environment for testing etc? 20:58:36 JimH: not sure about "environment"... but yes, we'll be interested in their requirements w.r.t. expressing ontology. 20:59:00 OraL: I participate in the DAML-S work, in case we need a connection. 21:00:52 JimH: we'll also be keeping track of other W3C working groups as well as other groups like FIPA... 21:01:05 JonD: I've been appointed representative from FIPA to here. 21:01:20 ora has left #webont 21:01:56 JimH: note also the European Union has issued a [... scribe missed it.] 21:02:31 ... semantic web initiative. 21:02:59 [... more charter review...] 21:03:03 "To be successful, we expect the Working Group to have approximately 10 to 20 active principal 21:03:03 members for its 12-month duration. " 21:03:08 JimH: clearly we're past that. 21:03:23 JimH: note well: "Any intellectual property essential to implement specifications produced 21:03:23 by this Activity must be available on a royalty-free basis." 21:04:05 ==== end of item 4 21:04:58 FrankvH: it's been suggested to me that we collect idiomatic expressions using an ontology language... 21:05:16 http://www.ksl.stanford.edu/people/dlm/daml/modelingIssues.html 21:05:26 is what i put up to hold daml+oil modeling issues 21:05:28 ... use cases, as it were. 21:05:37 DanC: yes, let's 21:07:05 FrankvH: our charter says start at/near DAML+OIL... (a) how about a crash-course? (b) let's pore over DAML+OIL in detail and see what issues are there. 21:07:37 JimH: ideally, we'd have a ftf presentation; unfortunately, we can't wait 'till then... 21:07:48 Ian: how about the walkthru? 21:08:22 FrankvH: we could present the walkthru by phone... [thinking out loud] 21:08:59 IanH: I'm still not clear how we'd define an implementation 21:09:32 JeremyC: implementation is my main focus in participation here... at HP, we have an RDF implementation, and we're interesting in moving "up the stack"... 21:10:09 ... for example, it would have Java classes that make it straightforward to work with this language... perhaps including a DL [ description-logic ] reasoner. 21:10:13 mdean has joined #webont 21:10:19 mdean is now known as mdean_ 21:10:21 JonD: if we're [@@] 21:13:12 DanC: [use cases->testing; RDF Core WG experience: parser testing, entailment tests] 21:13:38 DebM: I'm maintaining a "modelling issues" page... http://www.ksl.stanford.edu/people/dlm/daml/modelingIssues.html 21:13:40 if there is a question on the call, can we please get an answer before going on to something else? 21:13:59 * DanC didn't realize there was a question outstanding 21:14:23 what counts as an implementation. 21:14:51 did you get cut off? 21:16:05 what counts as an implementation? supporting those test cases... 21:16:43 PeterPS: I'm still interested to know what counts as an implementation... 21:16:54 DanC: note that this group will largely decide that (with some review) 21:17:08 PeterPS: so this group will do the "this implementaiton is good enough" work? 21:17:17 DanC: yes, mostly. (see also: new QA WG) 21🔞02 JimH: it's not necessarily an integrated product... could be a mix of tools and toolkits that counts as an implementation 21🔞44 DanC: if we decide that's OK, that'll probably work. But other groups (e.g. CSS) have decided that they want one piece (or 2 pieces) of widely-deployed software that passes all their tests. 21:19:12 JeremyC: an important motivation for the implementation requirement is to test that the spec is clear. 21:19:17 [... missed other stuff...] 21:21:36 DanC/JimH: let's do start collecting use cases: real-world scenarios where "if it doesn't do this, I'm not happy" 21:22:17 ====== next meeting 21:22:24 week of 22 Nov is not likely. 21:22:45 stay tuned for news of next telcon sometime in the week of 26 Nov 21:23:00 ====== ADJOURN. 21:23:22 * JosD bye 21:23:39 Ian has quit

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

Received on Monday, 12 November 2001 17:13:27 UTC