09:12:08 RRSAgent has joined #owl 09:12:08 logging to http://www.w3.org/2007/12/07-owl-irc 09:12:16 scribenick: JeffP 09:12:30 rrsagent, set log public 09:12:36 Major issue seems to be whether to use xsd datatype semantics 09:12:51 chair: Ian Horrocks 09:12:53 Topic: Datatypes 09:12:55 Alanr, let's meet when you get back 09:13:02 peterhaase has joined #owl 09:13:06 scribe: jeff 09:13:19 scribe: JeffP 09:14:31 Uli is presenting 09:14:55 IanH has joined #owl 09:14:58 bijan has joined #owl 09:15:11 thomassch has joined #owl 09:15:17 are we starting with issue 25 as in the agenda? 09:15:35 OWL DL does not support user defined datatypes 09:16:00 uli: users want to represent internals 09:16:19 ... and comparisons 09:16:30 s/internals/intervals/ 09:16:34 slides available at http://www.cs.man.ac.uk/~horrocks/owl1.1-datatypes.pdf 09:16:45 thanks! 09:17:22 sandro has joined #owl 09:18:03 ... in OWL DL no inverse functional datatype properties 09:18:41 vit has joined #OWL 09:18:44 ... not to mention composite keys (not even OWL Full supports this) 09:21:43 ?? 09:22:12 it got quiet (no audio yet) - I'm dependent on scribe 09:26:01 boris: we might want to keep the unit mapping out of TBox 09:27:10 jeremy: second 09:28:02 Jeremy's triangles: http://www.hpl.hp.com/personal/jjc/ 09:29:44 sebastian: there are real world examples 09:29:47 alanr has joined #owl 09:29:59 ... that we need datatype mapping in the TBox 09:30:08 bijan:both needed 09:31:48 uli: we have examples of seeing class subsumption checking based on datatype constraints 09:32:41 casten: it is difficult to choose one standard set, e.g. covering integers, rational, +, *, ... 09:32:52 uli: as many as possible 09:33:46 jeremy: each simple example is easy 09:34:30 ... but have concern on having all of them, which makes it hard 09:35:59 Sebatian's use case: http://www.w3.org/2007/OWL/wiki/N-ary_Data_predicate_use_case 09:36:12 sebastian: combining DL and data values are important and useful, there are many tasks that you could not solve if you treat datatypes externally 09:42:41 alanr: do we want to detect the problematic cases? 09:43:31 what does alan mean by "detect"? 09:46:19 jeffp: there are existing works on datatype groups, a mechanism is already there 09:46:29 Uli has just added "[Alan] add support to check whether this mechanism (second item of '3.') has been used 'safely'". Does this help? 09:46:41 dlm has joined #owl 09:47:03 ... even in OWL DL, freely combinations are not possible, e.g. transitive properties are not allowed to used in number restrictions 09:47:13 I mean during species validation, for example. Or via declarations of what features are used and flagging incompatible combinations 09:47:43 jeremy: what happen if data in the user databases having both integers, rationals + and * ... 09:47:54 link again http://www.w3.org/2007/OWL/wiki/N-ary_Data_predicate_use_case 09:48:44 boris: we need some datatype profile 09:49:16 casten: second boris point 09:53:48 jeffp: two points: 1) profiling is a good idea, there have been work there such as datatype groups 09:54:16 ... and we could provide a list of feasible datatype groups 09:55:22 2) if users have integers, rationals + and *, we could simply have type promotion, promoting integers into rational, and it is still decidable 09:56:09 alanr: maybe we could have a stroll poll on this 09:56:53 bijan: we all agree that some sort of datatypes are needed, no matter in OWL or RIF 09:57:19 ... many of our cases cannot be addressed by RIF 10:00:19 jeremy: transitive issue is different 10:05:43 dlm, GiorgosStoilos - the minutes of your sessions yesterday are now available for cleanup on the wiki: http://www.w3.org/2007/OWL/wiki/F2F1_Minutes_Session_4 10:06:15 linear polynomial (in-)equations over the reals or cardinals with order relations, 10:06:25 MarkusK - the minutes of your scribe session yesterday are now available on the wiki for you to clean up: http://www.w3.org/2007/OWL/wiki/F2F1_Minutes_Session_3 10:06:34 nonlinear multivariate polynomial (in-)equations over complex numbers, 10:07:07 (from the racer manual: http://www.sts.tu-harburg.de/~r.f.moeller/racer/racer-manual-1-7-19.pdf ) 10:07:14 page 11 10:08:26 See page 47 and 48 10:08:57 e.g., (* real AN ) (AN of type real or complex) 10:09:01 Sandro - OK, I will check. 10:12:51 JeffP has joined #owl 10:14:05 jeffp: besides Racer, an extension of FaCT (FaCT-DG) also supports n-ary and datatype groups 10:14:26 uli: we could have some more general proposal, rather than specific ones 10:16:02 bijan: we don't have to require all our implementors to implement everything, so we should be flexible somehow 10:16:21 uli: the 4th point: easy keys 10:19:20 markusk: in foaf people use b-nodes rather than individuals, so the easy key might not solve the foat problem 10:22:20 What I meant: It's a bad idea to, in committee, to significantly and somewhat arbitrarily increase the implementation burden. But without adding a hook, implementors *can't* (compatibly) experiment 10:22:40 So, let's add the hook and be cautious about how we fill in the hook 10:25:02 stall poll 1: many 1, no -1, four 0 10:25:36 s/stall/straw 10:26:00 straw poll 2: (all) 1 10:28:49 straw poll 3 (about 2-b): many 1, two (conditional) -1, six 0 10:33:57 JeffP has joined #owl 10:34:40 straw poll 4 (n-ary datatype): twelve +1, six -1, five 0 10:39:49 straw poll 5(easy key): 22 +1, one -1 10:41:55 boris: one profile proposal: a set of default profiles and allowing users to have arbitrary profiles 10:42:31 Uli has joined #owl 10:42:51 or people would be able to define their own profiles 10:43:09 sorryJeff, I got confused here 10:44:52 Another go, boris' profile: a fixed set of profile and also allowing people to define their owl profiles 10:45:07 Evan has joined #owl 10:45:09 alanr's proposal: a fixed set of profile 10:47:45 JeffP has joined #owl 10:47:50 Note the current support for unary datatypes is already fragmenty: http://spreadsheets.google.com/pub?key=pTmcCXR-dV6RpTEPxB0O-DQ 10:49:25 straw poll on profiling on datatype: eighteen +1, four 0 10:50:03 dmitry tsarkov introduced himself ... 10:50:15 ... email: tsarkov@cs.man.ac.uk 10:50:32 GiorgosStoilos has joined #owl 10:51:56 scribers should clean up yesterday's minutes by next telecon 10:52:10 (by IanH and no objections) 10:56:58 IanH has joined #owl 10:59:29 IanH_ has joined #owl 11:03:16 please let me know when audio is available. thanks! 11:11:05 Evan has joined #owl 11:12:04 ScribeNick: Evan 11:12:39 We now have Zakim connected for those who want to dial in. 11:12:41 thomassch has joined #owl 11:13:24 ScribeNick: Evan 11:13:52 peterhaase has joined #owl 11:14:55 Topic: OWL DL and OWL Full 11:15:59 IanH has joined #owl 11:18:33 Zakim has joined #owl 11:19:02 zakim, this will be owl 11:19:02 ok, Rinke, I see SW_OWL(F2F)6:00AM already started 11:19:35 ScribeNick: Evan 11:20:14 zakim, who is on the call? 11:20:14 On the phone I see ??P0 11:20:36 Peter presenting 11:20:45 Zakim, ??P0 is Meeting_Room 11:20:45 +Meeting_Room; got it 11:20:56 GiorgosStoilos has joined #owl 11:21:26 Here's a brief description of how model theoretic semantics works 11:22:09 Uli has joined #owl 11:22:20 Peter: OWL DL has a fairly straightforward take on this 11:22:29 IanH_ has joined #owl 11:22:56 Peter: OWL Full and RDF take a slightly weird take on this 11:24:13 alanr has joined #owl 11:24:14 Peter: wherein properties and classes live in the world with real objects 11:25:10 Peter: Here are the differences between OWL DL and Full semantics 11:26:23 See "Two Model Theories" slide 11:27:33 Peter: things like rdf:type and owl:Class are not in the world in DL but are in Full 11:28:42 Alan: In OWL DL Universe what is the status of Ontologies? 11:29:04 Peter: There is a separate place for them because of annotations 11:29:51 Peter: This description is about the spec. and not practice 11:30:28 Bijan: The things in the OWL Full universe are in there with a theory 11:31:25 Peter: None of this matters in some sense 11:31:41 Peter: What matters is the behavior which results 11:31:56 Peter: ...such as entailments 11:34:16 Peter: Differences: It's possible to make assertions about the OWL vocabulary that change their interpretation 11:35:25 Jeremy to take over presenting 11:36:44 Alan: question about the intention of compatibility to be entailments of DL and Full be identical 11:37:42 Jeremy: for me the whole point is to get compatibility with RDF 11:40:39 zakim, who is on the call? 11:40:39 On the phone I see Meeting_Room 11:40:50 jluciano, we are on the phone now, I believe. 11:40:55 Jeremy: A goal is "least surprise" for users of RDF when using OWL 11:42:59 Jeremy: OWL annotations are intended to behave as RDF annotations 11:43:27 Alan: There are implications for RDF annotations that users may not be aware of 11:43:54 Alan: I want to make a distinction between usage and consequences of the semantics 11:45:33 Bijan: I don't understand what you mean by RDF triple-by-triple semantics 11:46:44 Jeremy: In the OWL 1.0 semantics there are correspondence theorems between OWL Full and DL Semantics 11:49:01 Evan has joined #owl 11:49:23 +Joanne_Luciano 11:49:29 Jeremy listed Issues related to the FULL and DL differences 11:49:55 such as 63, 76, 81, 69, 72, 55, 73 11:50:21 pfps has joined #owl 11:50:25 Jeremy: Do we want to allow semantic subsetting for fragments 11:51:19 is that Bijan speaking? 11:51:26 Yes 11:51:55 I failed to capture it however. 11:52:40 Bijan: If we are going to support OWL Full do we need to support the full RDF umbrella 11:52:46 He said something about if he wanted to reproduce what is in Jena, he'd like to have that info available to know what to reproduce 11:53:01 Bijan: described in Jeremy presentation 11:53:38 Jeremy: The semantic of RDF reification are essentially none 11:54:11 Bijan: There exists somewhere in the known universe a Statement that includes: S, P, O 11:55:10 Jeremy: There is no clear statement in the specs for how reification can work interoperably from system to system 11:56:43 Bijan: In the OWL full situation you have to interpret the reification syntax somehow 11:57:35 More discussion about how this can be done 11:58:20 Jeremy: Punning 11:59:33 Uli has joined #owl 11:59:49 In some peoples mind the web arch specifies that a URI corresponds to a single meaning 12:00:27 Punning is weaker than OWL Full because it violates this principle 12:01:48 Jeremy: this seems to cause user confusion 12:01:57 Mapping rules 12:03:26 In my view, the mapping rules were the hardest part of the OWL Rec 12:04:38 Jeremy: The drivers behind the mapping rules in OWL 1.1 are different 12:05:28 who's speaking? 12:05:30 Jeremy: ...and this will lead to considerable change and probably 12:05:43 Boris Motik is speaking 12:06:03 Jeremy: ...issues later on. 12:06:30 What's he saying is one of the biggest problems? 12:06:41 Boris: In my opinion many of these problems are the result of 12:06:52 "these problems"??? 12:07:04 Boris: ...shoe-horning everything in the same universe. 12:07:53 Point for discussion later: 12:08:04 please distribute (and reference here for later). which sldie # of URI / slide ref 12:08:16 \me Thanks Peter. 12:09:06 Boris: if we came up with an OWL Full that has a clean model theoritic framework 12:09:17 ...then we could fix this. 12:09:46 Alan:This would be a smaller OWL Full? 12:10:21 Bijan: Punning was intended to meet the goals of Full at least quarter way 12:11:21 Peter: The dogma in this case is the same syntax extension of RDF 12:12:18 Bernardo: The people who like OWL Full should really come up 12:12:34 ...with features for OWL Full that they like and use 12:12:48 ...Then we could do some research. 12:13:48 Ian: The point I wanted to make was how much of this proposed 12:14:01 ...work will be part of the work of this WG? 12:14:28 Boris: Cleaning this up would be a huge accomplishment for this group. 12:15:28 Alan: To my mind, it's not clear that cleaning up OWL Full is desirable to 12:15:39 ... the Full/RDF community. 12:16:10 Jeremy: Dropping the comprehension principles seems like the 12:16:23 ... smallest change that would be of value. 12:17:22 Alan: Is this in scope for our group? Strictly speaking I don't think so. 12:18:03 Ian: This kind of work just isn't in scope. 12:18:19 Bijan: Form an OWLED task force to look at this. 12:19:40 Alan: We need to have a discussion about what compatibility means. 12:20:59 Alan: If we allow OWL Full semantics changing that will affect backwards compatibility 12:23:15 Discussion of semantic fragments 12:25:33 Alan: we have a delta now in the sublanguage entailments 12:26:04 Bijan: finding some delta that makes sense that makes the languages 12:26:15 ...as close as possible would be a good thing. 12:26:34 Ian: If we are comfortable with this semantic subsetting then 12:26:54 ... we should be happy with the Full - DL differences 12:27:48 Ian: One slight difference in Jeremy's proposal would be allowing 12:28:16 ... more syntactic freedom but actually reducing the entailments 12:28:36 ... by removing the comprehension principles for e.g. 12:29:18 Jeremy: HP might be happy with such a result if it is consistent 12:29:37 ... with some broader framework. 12:30:56 alanr has joined #owl 12:31:26 Jeremy: There are easy bits in the OWL 1.1 language. 12:32:04 ... getting those bits working are a bounded and achievable task. 12:33:10 Bijan: My experience is that users are concerned about not 12:33:46 can't hear Bijan 12:34:05 dlm has joined #owl 12:34:06 what did he say about channelling Jim? 12:34:08 ... being able to process large numbers of RDF graphs 12:34:15 ...with DL reasoners. 12:34:51 #/me Sandro ... I'm a mac book pro user 12:35:11 ... Features like punning improves this situation. 12:36:09 Ian: I wonder how hard it would really be to extend the status quo 12:36:45 ... with some acceptable differences. 12:37:11 Ian: This is a strawman for something that we could do. 12:37:58 Jeremy: I'd need to take this proposal back to HP before commenting on it. 12:41:10 Bijan: I would like us to keep the political and the user requirements seperate 12:41:11 Rinke has left #owl 12:41:51 Action: jeremy describe how punning and cardinality play poorly with each other 12:41:51 Created ACTION-47 - Describe how punning and cardinality play poorly with each other [on Jeremy Carroll - due 2007-12-14]. 12:42:48 Jeremy: maybe we can say in the spec that punning is a concession to implementors, not a basic part of the semantics, that univocality is intended. 12:43:36 Ian: I'd like to see suggestions for concrete ways of moving forward to address these problems 12:43:55 Rinke has joined #owl 12:44:03 Jeremy: Why don't we start with Qualified Cardinality Description? 12:45:03 PFPS: Someone made comment that Qualified Cardinality Descriptions leads to non-monitonicity 12:45:25 ... and I remember finding it believable 12:46:22 trackbot-ng, list 12:46:29 trackbot-ng, help 12:46:29 See http://www.w3.org/2005/06/tracker/ for help (use the IRC bot link) 12:46:34 trackbot-ng, info 12:46:34 Jeremy has joined #owl 12:46:51 ACTION: pfps inform the WG on absurdity of QCR / OWL Full 12:46:51 Sorry, couldn't find user - pfps 12:47:26 peterhaase has joined #owl 12:47:30 action: jeremy attempt Wiki sketch of QCR semantics OWL Full 12:47:30 Created ACTION-48 - Attempt Wiki sketch of QCR semantics OWL Full [on Jeremy Carroll - due 2007-12-14]. 12:47:56 ACTION: peter inform the WG on absurdity of QCR / OWL Full 12:47:56 Sorry, amibiguous username (more than one match) - peter 12:47:56 Try using a different identifier, such as family name or username (eg. ppatelsc, phaase) 12:48:27 ACTION: peterpatel-schneider inform the WG on the absurdity of QCR / OWL Full 12:48:28 Sorry, couldn't find user - peterpatel-schneider 12:48:31 action: ppatelsc inform the WG on absurdity of QCR / OWL Full 12:48:31 Created ACTION-49 - Inform the WG on absurdity of QCR / OWL Full [on Peter Patel-Schneider - due 2007-12-14]. 12:49:28 Alan: on concrete actions... 12:49:32 Alan: How about we say: If you manage to game the system to have different meanings for a URI, you can't count on that 12:49:39 ... we have a set of options 12:50:10 Jeremy: a suggestion that Jeremy concentrate on OWL Full 12:50:43 ... Semantics and drop out of User Facing Documents 12:51:16 Alan: Any other specific proposals 12:51:42 Ian: Let's try and extend where are now and see where we end up. 12:52:34 Alan: When do we evaluate when this approach is failing 12:52:45 ... so that we can try another approach. 12:53:06 ... I want to have some ideas about where we would go if this 12:53:12 ... doesn't work. 12:53:34 Jeremy: We have had two variants proposed today. 12:54:29 ... Sacrifice backwards compatibility and work towards 1.1 12:55:05 jluciano has joined #owl 12:56:45 Jeremy_ has joined #owl 12:57:13 Bijan: We are spending a lot of time on this 12:57:59 ... I would like to know how much interest in this WG 12:58:12 If there were a task force, who would be on it -- Alan, Jeremy, Sandro 12:58:13 ... with Full compatibility 12:58:36 jeremy: variant 2 - peter - drop same syntax requirement, and allow OWL 1.1 DL to have different syntax from RDF 12:58:51 Stawpoll 12:59:09 How many people want to use OWL Full for 1.1? 12:59:32 2nd question... +1 (not listed yet) 13:00:10 Jeremy rephrase: When using 1.1 do you want to use Full semantics? 13:00:23 hard to hear on the phone 13:00:34 Q1- Are you a potential customer for OWL 1.1 Full -- you'll be using the document 13:00:43 customer or reseller 13:00:45 OK, thanks, no :-) 13:01:38 Q1 No. 13:01:54 5 in room 13:02:25 s/5 in room/5 yes in room/ 13:02:39 Q1 no 15 13:03:10 jim would also be in the positive count for that question 13:03:42 clu has joined #owl 13:04:06 q2 are you a potential customer for Bijan's description of patch-up 13:04:23 rules 13:04:28 Q2- Are you a potential customer/reseller of a specification of techniques for translation RDF graphs (in the spirit of OWL Full) to OWL 1.1 DL 13:04:40 Q2- Are you a potential customer/reseller of a specification of techniques for translating RDF graphs (in the spirit of OWL Full) to OWL 1.1 DL 13:04:59 Q2 yes 13:05:26 Q2 +1 13:05:35 many in favor 13:05:36 Q2 yes except for 2 abstainers 13:06:25 LUNCH 13:07:46 Jeremy_ has joined #owl 13:08:20 peterhaase has joined #owl 13:19:47 Jeremy_ has joined #owl 13:29:50 Jeremy_ has joined #owl 13:41:34 Jeremy_ has joined #owl 13:43:17 Zhe has joined #owl 13:43:51 Rinke has joined #owl 13:44:15 + +1.603.897.aaaa 13:47:05 please connect to http://conference.oracle.com 13:47:22 huh? 13:47:30 the conference id is: 96360063 13:47:38 Um, no. 13:47:42 :-) 13:49:24 this is for owlprime review 13:49:42 jim, we are having lunch 13:49:48 So, who are you talking to , Zhe? 13:49:51 the session starts in 10 minutes 13:49:59 with the fragment agenda item 13:50:02 I know that. I just want to send the ID out to the group 13:50:05 Zakim, aaaa is Zhe 13:50:05 +Zhe; got it 13:50:22 I doubt anyone but me is reading this. :) 13:50:48 eh, not true ;) 13:53:08 thanks Sandro - btw, are the logs being recorded somewhere? We used to always put those in the irc topic so people could find them (and to make them member readable) 13:55:29 thomassch has joined #owl 13:57:35 sandro has changed the topic to: http://www.w3.org/2007/OWL/wiki/F2F1 13:57:49 sandro has changed the topic to: See http://www.w3.org/2007/OWL/wiki/F2F1 for agenda, minutes, etc 13:58:20 Jim, the minutes are linked from the agenda, meeting page, etc. I put them there at the end of a session. 13:58:36 the raw irc can be found by: 13:58:39 RRSAgent, pointer? 13:58:39 See http://www.w3.org/2007/12/07-owl-irc#T13-58-39 13:59:46 right, thanks - forgot RRSAgent - been a while :-) 14:00:02 understandable. 14:00:44 I've been really wishing you were here. Occasionally one of us channels you, "Jim would say.....", but it's hard to find out more details and try to convince you, when you're not here. 14:01:01 (that last session was about OWL Full.) 14:03:42 peterhaase has joined #owl 14:05:00 yeah - it's frustrating - I had wanted to come (although I suspect I'd of spent a lot of time fighting) -- I see "channeling Jim" in the logs a couple of times :-) 14:05:32 Unfortunately, I have a telecon with Tim BL at 9:30, so was hoping to at least get some of the fragments talk in first... 14:07:03 IanH has joined #owl 14:07:03 msmith has joined #owl 14:07:10 ScribeNick: msmith 14:07:32 please connect to http://conference.oracle.com using IE 14:07:41 the conference id is: 96360063 14:07:45 Hi Jim.... another remoterer here - I'd gone to the airport but 2 minutes shy of 1 hr before flight and the closed checkin! 14:07:45 alanr has joined #owl 14:08:38 Topic: Fragments - OWL Prime 14:08:39 JeffP has joined #owl 14:09:07 zakim, who is here? 14:09:07 On the phone I see Meeting_Room, Joanne_Luciano, Zhe 14:09:08 On IRC I see JeffP, alanr, msmith, IanH, peterhaase, thomassch, Rinke, Zhe, Jeremy, clu, jluciano, Uli, pfps, GiorgosStoilos, Zakim, vit, sandro, bijan, RRSAgent, Battle, ivan, 14:09:11 ... MarkusK, seanb, Ratnesh, pascalhitzler, hendler, trackbot-ng 14:09:15 + +1.518.472.aabb 14:09:26 zakim, aabb is jhendler 14:09:26 +jhendler; got it 14:10:14 Hi zhe, I've turned off pop up blocker, where do I go? 14:10:23 please connect to http://conference.oracle.com using IE 14:10:36 you will see a join conference portlet 14:11:42 Can you repeat that? 14:11:55 the conference id is: 96360063 14:11:55 96360063 14:13:04 alanr: there is a proposal to have a joint OWL & RIF task force 14:13:12 ... peter is there. is there anyone else? 14:13:25 ...uli is a second. 14:13:32 IanH_ has joined #owl 14:13:33 sandro: I may sort of be on it for both 14:13:53 bijan: I am liason to RIF and will continue to be 14:15:16 Jim, I just send a ppt to your rpi email address. 14:15:16 slides would be better for archive purposes 14:15:24 thanks 14:15:46 dlm has joined #owl 14:15:53 jeremy is still working on getting the conference room connected 14:16:18 In future WG "events" it would be nice to get the infrastructure set up in advance! 14:16:36 I can't connect to the oracle conferencing either 14:17:27 Achille has joined #owl 14:17:28 joanne, ian will send you the slids 14:17:35 s/slids/slides/ 14:17:43 thanks! 14:17:43 alanr apologizes for not setting this up during lunch. 14:18:23 I hope we'll not have to have alanr give up his lunch, then. 14:18:43 bmotik has joined #owl 14:19:01 agenda slide 14:19:04 +[IBM] 14:19:29 Zakim, [IBM] is temporarily Achille 14:19:29 +Achille; got it 14:19:31 slides just sent to public-... list 14:19:49 slide: oracle 10gR2 RDF 14:20:45 some technical difficulties continue w.r.t slide presentation 14:21:59 bijan has joined #owl 14:22:04 zhe: many ways to insert data. 14:22:38 ...in 10r2 we also support some inferencing and rules. we use forward chaining approach 14:23:13 ...also query using a SPARQL-like syntax 14:23:21 ... this was all in 2005 14:23:32 slide 11gR1 14:24:00 zhe: this year new release with new features. faster loading, owl reasoning with proof generation 14:24:05 \me Zhe, can you speak louder again, please? 14:24:21 zakim, who is talking? 14:24:23 ... overhauled performance w.r.t. load and query 14:24:31 sandro, listening for 10 seconds I heard sound from the following: Meeting_Room (52%), Zhe (32%), jhendler (33%) 14:24:40 zakim, mute jhendler 14:24:40 jhendler should now be muted 14:24:49 ... just recently added Jena / Oracle adapter 14:24:51 hendler, I muted you. 14:24:56 ... joint with HP 14:25:15 new slide 14:25:16 -jhendler 14:25:57 zhe: subset of owl is supported 14:26:10 uli: i'm curious about what scalable and efficient means 14:26:20 zhe: i will show some numbers later 14:27:24 ... re: what is supported - forward chaining rules implementation for fast query answer 14:27:31 slide "why?" 14:29:15 zhe: ... conclusion in ISWC 2006 paper was that existing reasoners had problems with large ABox data 14:29:33 q+ 14:29:37 slide 7 - owl subsets supported 14:30:13 zhe: rdfs++ added as a "minimal" extension to RDFS 14:30:50 ...owl prime, what is now proposed as rdfs 3.0 14:31:25 slide - semantics characterized by entailment rules 14:31:41 zhe: owl prime has ~50 rules 14:31:41 does somebody know the exact literature reference for OWLPrime and can send it? 14:32:13 slide - applications of partial dl semantics 14:32:37 similar for OWLSIF - literature reference ... 14:33:11 BTW The pdf with these slides can be found under http://ontolog.cim3.net/file/work/DatabaseAndOntology/2007-10-18_AlanWu/RDBMS-RDFS-OWL-InferenceEngine--AlanWu_20071018.pdf :) 14:34:15 slide - support semantics beyond owl prime 14:35:27 jeremy: question about example being supported directly in the future 14:35:32 zhe: exactly 14:35:40 q? 14:36:40 achille: question about updates to abox 14:36:48 zhe: i'll get to that later 14:36:57 ack Achille 14:38:03 slide 13 - advanced options 14:39:55 alanr: question about time, can we focus on questions now 14:40:12 zhe: ok, i'll quickly browse remaining slides, then go to questions 14:40:44 BTW, the survey paper mentioned in the talk on slide 9 is at: http://www.mindswap.org/papers/2006/survey.pdf 14:41:59 It has more fine grained analyses, including in terms of AL, ALHF, SHIF, and SHOIN, on the one hand, and RDFS(DL), DL-Lite, EL++, and "non-tractable" 14:42:39 It also discusses "repairable" OWL Full ontologies and (sketchily) how the non-repairable ones fall into OWL Full 14:43:33 It's not at all clear to me how to map the analyses in that paper to OWL Prime (in part because I don't understand OWL Prime) 14:44:00 slide - implementation in rules 14:44:12 zhe: I want to stress that we did not handle one property at a time 14:44:59 paper also took a DL approach to the world, the raw data showed the great bulk of the stuff out there, pre-change, was low expressivity RDF or RDF with a little OWL - it's where the RDF 3.0 proposal came from 14:45:46 zhe: I'll jump to query answering slide 14:46:34 ...that's all I wanted to cover, open for questions 14:46:53 Uhm...I don't knwo what you mean by "raw data" and "great bulk" 14:47:18 ian: the tractable fragments doc describes fragments with known database mapping. wondering why you didn't choose one of those 14:47:42 In fact, I don't see that anything I said had anything to do with what fell into RDFS or not 14:47:53 zhe: we started by asking existing customers what they needed. most told us they just needed simple extension into owl from rdf 14:47:54 The repair had mostly to do with the nominally owl full documents. 14:48:25 ... pretty much the approach was driven by customers and need to implement efficiently 14:49:02 In fact, if you look at table 2 and table 3, the second part of your assertion is at least questionable 14:49:09 ian: but, customers said you needed something small (rdf + a bit) which is exactly what the fragments are. instead you chose a large fragment and implemented incompletely 14:49:46 zhe: so far, for those other fragments we have not found a complete rule set (except pdstar) 14:49:58 (I find the "small fragment" vs "large fragment" language very confusing, because I don't know what the metric is. large number of terms? large number of users? large implementation effort needed? 14:50:36 Expressivity, I think 14:50:38 """Of the 307 OWL Full documents that can be patched, 63% become OWL Lite documents, and just 37% become OWL DL. Two observations can be made. First, The majority (91%) of the OWL Full documents (from Table 2) can be turned into a decideable portions of the languages by adding type triples. Secondly, the majority of RDFS documents (95%) can transition to OWL easily by adding type triples and use OWL vocabulary instead of RDFS vocabulary.""" 14:50:44 uli: I want to echo ian and point out that you don't allow intersection, but a clever user would have it 14:51:00 ...and to be complete complexity becomes a problem 14:51:11 alan: they're not trying to be complete 14:51:17 the mentioned paper by ter Horst seems to be the following: Herman J. ter Horst, Completeness, decidability and complexity of entailment for RDF Schema and a semantic extension involving the OWL vocabulary, Web Semantics: Science, Services and Agents on the World Wide WebVolume 3, Issues 2-3, , Selcted Papers from the International Semantic Web Conference, 2004 - ISWC, 2004, October 2005, Pages 79-115. 14:51:17 (http://www.sciencedirect.com/science/article/B758F-4H16P4Y-1/2/d039e4784b224e95aafca856ecfb1edb) 14:51:17 Keywords: Ontology; Semantics; Entailment; Completeness; Computational complexity 14:51:17 Uil: Complete with respect to one reasoning problem is sound with respect to another. [[ In OWL? Really?? ]] 14:52:03 boris: echo ian, observes that fragments exist which can be implemented with a set of complete rules 14:52:45 bernardo: i'm worried about soundness and worried about what "sound and complete" means here. I don't understand the semantics 14:53:07 ...b/c you haven't implemented the OWL semantics, you've chosen some of the OWL DL vocabulary 14:53:38 From a spec perspective, this fragment seems to be *implementation* defined...which is a bit worrisome 14:53:49 zhe: we do care about completeness, but don't consider it critical 14:54:15 ... completeness is evaluated w.r.t. query answering for some benchmarks, etc. 14:54:34 My test for this would be, without looking at thier rules or using your rule engine per se, can i write an implementation from a publically available description? 14:54:41 jeremy: what I hear from customers echos Zhe's comments. 14:54:58 ...I note that much of the questioning is hostile 14:55:00 I would point out that DB communities tend to do language/sublanguage without model theories very comfortably 14:55:01 alan: I agree 14:55:18 I would disagree with the assessment of tone...is it even relevant? 14:55:24 I would strengthen Boris claim and say that most (if not all) other fragments admit forward chaining, which is sound and even complete, 14:55:26 and the rules are easily derived. 14:55:36 jeremy: that may be b/c much of the questioning is coming from members with different user groups 14:55:44 I also find that Web 3.0 companies, including the folks interacting with us on the billion triple challenge, also come from the perspective Zhe represents 14:56:22 http://ebiquity.umbc.edu/blogger/2007/12/06/the-semantic-web-billion-triples-challenge-at-iswc-2008/ 14:56:42 ian: it wasn't intended to be hostile. I was trying to understand whether Oracle would be interested in more well understood and explainable fragments 14:57:05 Ian: DL Lite, not PD* 14:57:06 ...e.g., dl-lite which can be implemented in a database system, and also in a rule system 14:57:14 Same applied to EL++ 14:57:26 applied => applies 14:57:59 discussion of PD* soundness and completeness in a rule based implementation 14:58:25 fwiw, those fragments got little or no traction with the folks I consult for - they care about parallelizability and performance over the more understood stuf - their work is largely heuristic anyway 14:58:25 ian: the problem with PD* is that it doesn't implement a subset of OWL, it implements PD* 14:58:41 jeremy: it depends on what you mean by fragment of OWL 14:59:28 alan: I hear interest in co-ordinating on database fragments with Oracle 14:59:30 Zhe, I might have sounded hostile, which wasn't intended: some of us simply have a specific reading for certain words like "reasoner", and I couldn't see how this could be possible. 15:00:01 bijan: to standardize a fragment, we need a well defined semantics that we can all understand 15:00:19 Correction to scribe: I didn't say *semantics* I said *specification* 15:00:27 Zhe, also, if you want to see how IntersectionOf can be simulated with someValues and AllValues, ask Carsten. 15:00:31 we need an *implementation independant* spec 15:00:31 What I am arguing for is that there are some important communities out there to whom the fragments they care about are not those tied to Uli's definition of reasoner 15:00:35 thomassch has joined #owl 15:00:51 Topic: Fragments: (Tractable) Fragments and other Fragment Proposals 15:01:16 hendler, people are not really paying attention to IRC. 15:01:17 Jim, I appreciate this -- but "reasoner" was used on Zhe's slides, and i simply wanted to know in which sense. 15:01:30 bernardo presenting from slides in person 15:02:05 Bijan - agree with needing a spec, but I'd point out most programming languages get by just fine with operational semantics - in fact, since you implement Pellet in JAva, in a certain sense you're trusting that they get it right in some sense - 15:02:24 bernardo: motivation of owl-lite was easier owl. b/c owl dl and full are rich and complex. 15:02:25 Rinke has joined #owl 15:02:46 ...problem is owl-lite is broken b/c it doesn't address interactions between constructors 15:03:02 Jim, I guess what we would like to see is a consensus of what we mean by "Tool/reasoner X supports feature Y" 15:03:16 alanr has joined #owl 15:03:30 Bernardo's talk in email and at http://www.cs.man.ac.uk/~horrocks/F2F-Fragments.pdf 15:03:55 bernardo: most features held out of owl-lite can be recovered through "back doors" 15:03:56 First, I'm not saying anything about the specification style....but at the moment we don't have a clear spec. I don't know where to start other than by looking at Oracle's implementation 15:03:56 Uli, how do we show "Java" supports "begin/end loops"? that seems to be something in the real world that we could model for some (not all) of our work 15:04:39 bernardo: existing document includes fragments which 15:04:44 this is the point!! thw Owl language features all have operational semantics that are good enough for many people in many situations - so when Oracle says we support X, why do we need more than that 15:04:48 Second, there are differences between programming languages and ontology/data modeling languages. I hear your point, but find the analogy rather unconvincing. 15:04:57 Jim, I would never dream of trying to do this - but i would like to try to say what it means for a reasoner to support feature X 15:04:58 .... are well understood, documented, etc. 15:05:01 A precursor to OWL Lite giving some of the rationale: http://www.aifb.uni-karlsruhe.de/~sst/is/WebOntologyLanguage/harmelen.htm 15:05:02 Evan has joined #owl 15:05:30 Er...precursor discussion 15:05:39 What I'm arguing is that RDFS 3.0, or OWL Prime, might be better looked at less as ontology languages (leave that to OWL DL) then as useful data analysis languages 15:06:03 bernardo: we don't expect users to go over recent literature on tractable fragments, so wanted a single document 15:06:05 But I don't see why "useful data analysis langauges" don't need a clear spec 15:06:10 this is what my nose is rubbed in when I attend the Sem Tech conference and places like that 15:06:27 bijan - the question is what is the definition of a clear spec. 15:06:43 bernardo: most of the languages I will describe are "families" of languages, we decided to keep 1 from each 15:06:49 Furthermore, model theory is pretty easy way to specify something...at the moment, no one has proposed anythign else 15:06:51 if there is a set of rules defined as those in RDFS spec, is that clear? 15:07:01 bernardo: 1st is EL family 15:07:07 Zhe, perhaps 15:07:13 note that they are informative 15:07:14 ...used in bio-medical already 15:07:29 But I would be interested in looking at such 15:07:33 they may be informative, however, that is how most people understand semantics 15:07:40 but there are no model theories for many things, and model theory is not the only way to spec other things - like these rule-based examples 15:07:44 Zhe, we would call this "operational semantics" or such like and would be split about how clear this is 15:07:45 Understand != spec 15:07:53 bernardo: stress that these fragments are not academic exercises, there are direct applications to existing ontologies 15:08:04 Again, my test is can I write an interoperable implementation without looking at your implementation 15:08:13 +1 Bijan's inequality 15:08:17 At least my first test 15:08:30 i.e. Inverse(A,B) IFF s A p -> p B s 15:08:37 bernardo: 2nd is DL-Lite family 15:08:39 seems like a fine definition of inverse 15:08:42 if we agree on a set of rules, then interoperability is not an issue 15:08:49 Not at all since I don't know what you -> means 15:08:55 ... designed for large number of instances in database technology 15:08:58 Zhe, no 15:09:04 Not clear at all 15:09:18 For example, i might no use those rules *in* my implementation 15:09:24 using Hendler's example rule, 15:09:27 I might want to use a very different technique 15:09:37 if we agree on that, then we are interoperable 15:09:56 Is that rule controposable? 15:10:04 The reading of rules, for example, differ in whether you have contraposition or not 15:10:07 Was it meant as <->? 15:10:20 Bijan, that is either easily defined, or can be left to philosophers trying to write PhDs, in the real world, lots og languages work this way - but if you want something better - okay, we'll use SCL 15:10:27 and whether you "apply" it to all named individuals or to *all* individuals 15:10:31 hendler, that's not true 15:10:50 But c'mon, that wasn't even a partial spec 15:10:57 And it was of one of the easiest bits 15:10:58 bernardo: approach is similar to what zhe described, do work in tbox, then pass to database system for query answering 15:10:58 Inverse 15:10:59 so, I agree that your defintion of inverse seems clear, but when you want to implement it, there are questions coming up 15:10:59 could someone send Bernardo's presentation to the public mailing list? 15:11:02 Consider complementOf 15:11:06 ok, KIF 15:11:11 English would do 15:11:12 I agree with Zhe 15:11:29 complementOf not in RDFS 3.0 for precisely that reason 15:11:30 -Meeting_Room 15:11:36 It was in OWL Prime 15:11:38 Achille: this is done 15:11:47 room is calling back. 15:11:48 i could not hear anything 15:11:49 Again, I find englisch often clearer than things like "->' or such like 15:11:52 Or did you mean CLIF from the ISO standard, Common Logic 15:11:58 English ok w/me 15:12:11 carsten: reiterate bernardo, but contrast with zhe's approach. dl-lite change the ontology to use database technology, not change the database technology 15:12:12 -Zhe 15:12:13 would be nice to get a literature reference to OWLPrima - the description on the slides was not clear enough 15:12:29 I could find nothing on the web defining OWLPrime 15:12:29 In any case, I'm asking for a spec. We can beat on the spec and if we find problems we find problems 15:12:30 Zhe, we are dialing back in. 15:12:35 If we don't we don't 15:12:36 alan: another difference is in oracle you can query for classes, in dl-lite only instances 15:12:37 +Zhe 15:12:37 I am no longer hearing anything on the phone 15:12:40 but anyway, the point I'm making is not to oppose model theory - but the problem is to get the model heory right we have to put restrictions on the languge that some of us cannot live with easily 15:12:56 I don't know that that's true 15:13:02 I just did. still I hear nothing 15:13:07 Pascal, pD* has been described here: http://www.springerlink.com/content/366474250nl35412/ 15:13:07 bernardo: you can do tbox reasoning, but designed for abox answering. 15:13:09 Jim, I disagree: 15:13:09 WE ARE STILL DISCONNECTED 15:13:13 this is why OWL LITE is a subset of DL - not of Full, so there is no fragments for Full - which is what i often get asked for 15:13:16 should we dial in again 15:13:24 NO ONLY THE MEETING ROOM NEEDS TO REDIAL 15:13:26 +??P11 15:13:28 Jim, model theorey doesn't restrict things? 15:13:32 it is working now 15:13:37 Zakim, ??P11 is Meeting_Room 15:13:37 +Meeting_Room; got it 15:13:41 I'm very skeptical about it, but Id on't knwo because I don't know what the current fragment actually *is* 15:13:43 CAN YOU HEAR US NOW? 15:13:46 yes 15:13:50 thanks! 15:13:52 yes 15:14:01 Uli, so we could define the langauge fragment based on other concerns and then dfine it via model theory - that doesn't bother me at all - I'd be fine with that 15:14:06 bernardo: I picked the particular dl-lite language b/c it is between rdfs schema and owl dl 15:14:16 ... next is Horn-SHIQ 15:14:23 But I'm open to being convinced otherwise...but I'm more convincable by a proof of concept (at least) than high level discussion 15:14:29 ...can reason without disjunctions 15:14:35 is there a scalable implementation of dl-lite? commercial tool? 15:14:38 Jim, what would you define via model theory? The fragment or its semantics? 15:14:42 ...and low complexity for query answering 15:14:44 Zhe, to the first, yes 15:14:47 To the second, no 15:14:50 so Oracle has implemented OWL prime - what did I miss? 15:14:50 QuOnto 15:14:52 (Not yet) 15:14:59 Zhe, yes ther is, I think: search for Quonto 15:15:02 Implementation != specification 15:15:08 Uli - whichever you want - I'm not going to need to read that document anyway ;-) 15:15:27 And this is true for programming langauges as well 15:15:45 Uli: what kind of tool? what is the scalability? 15:15:46 Jim, I think we simply disagree what it means to *implement* a fragment 15:15:52 There are langauges defined by *specs* (including Java, Common Lisp, C, C++) and those defined by *implementation* Perl, Python 15:15:53 bernardo: other fragments dlp as a bridge to rules 15:15:57 (at least historically) 15:16:09 ... but it may be more "hacky" that horn-shiq 15:16:24 So, frankly, I don't want to port Oracles implementation. That's probably not even legal 15:16:33 Zhe, it is as scalable as it can get: 15:16:34 fine - I want a fragment of OWL that is defined by *specs* 15:16:43 I want a specification sufficient for independant implementation 15:16:58 Zhe, because it translates queries into SQL queries and leaves everything in the DB. 15:17:03 bernardo: questions for wg 15:17:07 Zhe: I understand that you would like to do forward chaining. A lot of fragments can be captured in a sound 15:17:09 (and even complete) way with this technique. I would like to learn what is your idea of tractability and 15:17:10 scalability. Is it forward chaining per se, or is it a rule set that does not produce too many new facts? 15:17:11 ....1 do we fix owl lite 15:17:11 But jim, if the specs happen to do it by model theory and capture the language you want, what do you care? 15:17:23 ....2 does that mean select one of these fragments 15:17:34 ....3 or do we present a menu of fragments? 15:17:43 If the specs are clear enough for me I don't necessarily require model theory ( though it helps so we can understand the relation to existing OWL specs) 15:17:55 bernardo: not in slides - do we want semantic subsets of owl full? 15:18:07 bijan - the poiunt is I don't care - and I said that - what I care is what is in the fragment first, how to define it second 15:18:22 ....e.g., owl full versions of these fragments? do we care about complexity of the full fragments? about compatibility? 15:18:22 Clu: the scability and performan requires are determined by the market. 15:18:26 How do you know what's *in the fragment* without a definitio of what the fragmetn is? 15:18:32 What's the difference? 15:18:49 Clu: people are asking for hundreds of millions of triples and beyond 15:18:52 zakim, who is muted? 15:18:52 I see no one muted 15:18:59 zakim, who is on the phone? 15:18:59 On the phone I see Joanne_Luciano, Achille, Zhe, Meeting_Room 15:19:10 hendler, please stop talkin on IRC. 15:19:24 Bijan - you're arguing circularly - but I can cut through it - I have created a wiki page with a description of exactly which language features I want to include - how to define it in a spec is something I'm happy to discuss 15:19:28 ivan: request to drop side conversations 15:19:36 ... and focus 15:19:54 Jim, we're talkin in the room now, and need to focus on this discussion in the voice channel, so no more chatter on IRC, please. 15:20:00 alan: little time, can we start with semantic subset of owl full? 15:20:37 ian: semantic subset means no change to syntax, but sanction smaller set of conclusions 15:21:13 q 15:21:19 q+ 15:21:20 gocha 15:21:20 jeremy: example is pd*, which specifies what semantic rules are thrown away 15:21:33 q- 15:21:35 peter: pd* throws away *parts* of rules 15:21:47 ian: this is picky 15:22:02 hendler, are you able to call on the phone? 15:22:09 sandro, no. 15:22:12 alan: how comfortable are people with this type of fragment 15:22:22 ... does anyone want to say this is a lousy idea. 15:22:30 hendler, type what you want us to speak for you 15:22:34 peter: yes, its lousy b/c you can be arbitrarily picky 15:22:34 what kind of fragment? 15:22:47 ian: its a lousy idea b/c it blows away the idea of interoperability 15:23:17 Alan is chairing this session. 15:23:39 hendler, do you have access to the slides? 15:23:47 bijan: qualm that methodological design principles are "unclear" 15:24:16 ...guidance for making decisions seem more arbitrary, a dangerous rat-hole 15:24:40 ... would rather people say they are incomplete than building incompleteness into fragments 15:25:18 jeremey: in response to ian, any semantic subsetting would need to be clear that it is a subset of spec and an explicit, agreed semantic subset 15:26:12 ...e.g., oracle and hp would agree on semantic subset and interop on at-least the semantic subset 15:26:19 to Hendler: fragments of OWL 1.1, which: are the result of years of research, have “nice” computational properties, are already supported by tools 15:26:24 s/jeremey/jeremy/ 15:26:45 +1 Jeremy -- "incompleteness" is fine When It's In A Specified Fragment, that is implementated in multiple places, etc. 15:26:48 alan: if we call this fragment or conformance level, it seems useful 15:26:57 to hendler: cover most existing ontologies 15:27:12 ...that baseline entailments are necessary, but additional entailments may be ok 15:27:52 q+ 15:27:53 bijan: if we shift from language fragments to reasoner conformance I'm more comfortable 15:27:55 But there are fragments which are not included that have all those things as well - Oracle Prime being a perfect example 15:27:56 Bijan: "Reasoner Conformance" might be a more useful notion here than "Language Fragments". 15:28:15 Jim, please call in if you want to participate 15:28:21 to hendler: Fragment Goals: suggest possible fixes to OWL Lite, inform the OWL community about recent research results, help users & tool designers 15:28:24 ...I have examples of people specifying this at a tool level. 15:28:48 jeremy: i'd be happy with such a rewording. i don't see it as notable 15:28:57 alan: does such a distinction help others 15:29:00 hendler has left #owl 15:29:10 some affiermation to alan in room 15:30:00 zhe: ? 15:30:21 alan: he said it would be useful to say we support same entailments 15:31:26 ian: more comfortable defining conformance that fragments 15:31:55 q- 15:32:02 ian: and jeremy's suggestion sounds like standardising implementations 15:32:18 Jim, you still there. Hard to follow the IRC. I can read what you write if you want to respond. 15:32:31 bernardo: users are comfortable with incomplete reasoning. swoop offering rdfs reasoner as a choice is an example of this 15:32:45 +1 to bernardo 15:32:57 ...more comfortable with that than trying to specify semantic subsets 15:33:23 jeff: i agree with bernardo and others. 15:33:38 ... implementation does not specify fragment. 15:33:45 bernardo also stressed the importance of a clear semantics as reference ... 15:34:36 boris: i just looked at pd* , this seems like definition. I think it is a useful fragment if evaluated a certain way. 15:35:18 ian: i didn't say pd* was bad, that we'd be standardizing an implementation. it was a reaction to jeremey's comments on what hp and oracle might do 15:36:08 sandro: owl is unique to me b/c it doesn't specify what the tools do, people read into that. specifying the tools would be useful. as a customer I expect that and would like it 15:36:41 what's "pd*"? 15:36:48 ina: it is not just hp and oracle, owlim, allegrograph as well 15:36:53 PD* is referred to in Zhe's presentation. 15:37:10 (from Herman ter Horst) 15:37:15 bijan: justifying discomfort - seems likely that over time fragments specified in such a way are likely to move 15:37:53 bernardo: on sandro's comment - we should specify reasoning services 15:38:46 ... it's not in the spec for OWL DL. for fragments the services descriptions would be uesful 15:38:59 sandro: i don't know what the terms are, the market decides 15:39:14 ian: its difficult to imagine semantic subsets not drifting apart 15:39:30 Sandro: It should be customer driven. When they want to find on the shelf, those should be the things defined in the spec. 15:39:36 ... it has been a success for owl that interoperability is so good, considering 15:40:15 jeremy: responding to standardizing tools - yes. there is value to user if they know different tools perform the same 15:40:36 ... this wg could provide appropriate conformance levels where vendors and user community come together 15:40:50 +1 Jeremy: there is real value to the customers in knowing that a set of products will all do (at least) the same thing. It would be a service to the community for this WG to provide that. 15:41:01 ... clear that motivations from academic community are useful, but they aren't the only motivations 15:41:11 alan: no one is saying market is unimportant 15:41:31 uli: clarification on user needs? 15:41:58 jeremy: users need some sort of specification, but don't need to know behavior is exact 15:42:41 alan: I want to poll for consensus on how to procede 15:42:47 +1 15:42:48 sandro: i don't understand 15:43:09 alan: I want to know if people think these fragments are useful 15:43:18 ... defined as a minimum set of entailments 15:43:36 bijan: reasoners can conform to the language to different degrees 15:43:52 subset of language + conformance level, is something similar to the way current languages (e.g) doing, like, Deprecated apis + core language( and specialized apis) 15:43:55 subsets of entailments == conformance levels 15:44:16 alan: we should aim for something specified like pd* 15:44:22 ian: declarative... 15:44:32 alan: yes, declarative 15:45:16 dlm has joined #owl 15:45:18 Q1 - The Working Group should (formally, precisely) define conformance levels, defining groups of reasoner which can do certain kinds of reasoning (all for a given OWL Fragment). 15:45:30 Q1 - The Working Group should (formally, precisely) define conformance levels, defining groups of reasoners which can do certain kinds of reasoning (all for a given OWL Fragment). 15:45:45 uli: we would later know e.g., what it would mean for a reasoner to conform to particular level? 15:45:52 alan: yes. 15:46:07 jeff: what does conformance level mean? is it in terms of benchmark? 15:46:41 uli: provides example 15:47:06 The Working Group should (formally, precisely) define conformance levels, defining minimum levels of inference that would be found? 15:47:57 q+ 15:48:17 carsten: degrees of incompleteness? 15:48:17 The Working Group should (declaratively) define conformance levels, defining minimum levels of inference that would be found? 15:48:24 alan: degree of completeness 15:48:25 Q1 - The Working Group should (declaratively) define (one or more) conformance levels, defining minimum levels of inference which would be performed (for a given OWL Fragment). 15:48:27 yes if we are talking about a declarative way of defining minimum levels of inference 15:48:38 ...fragments are syntactic fragments 15:48:51 ...conformance levels are distinct 15:49:09 JeffP has joined #owl 15:50:30 jeff: there might be difference between alan's and uli's suggestions 15:50:41 alan: distinction is unimportant now 15:50:42 when we, as a group define confromance levels, it is very useful to look at current market 15:50:59 including HP, Oracle, AllegroGraph, OWLIM etc. 15:50:59 alan: reads Q1 as above 15:51:20 against: Jeff, Carsten, Ian 15:51:28 +1 ( for a declarative approach) 15:51:32 joanne raises hand 15:51:32 abstain: none. 15:51:38 BREAK. 15:51:48 joanne lowers hand 15:51:52 for - lots and lots of hands 15:52:03 for - lots and lots of hands 15:52:19 in favor: lots and lots of hands 15:52:30 in favor - lots and lots of hands 15:52:38 zakim, what the heck are you doing? 15:52:38 I don't understand your question, sandro. 15:52:42 q- 15:52:48 Joanne +1 h and rais e for last q. 15:52:53 lots and lots of people raise their hands in favor. 15:52:59 testing - something 15:53:03 for - testing 15:53:11 for - lots andlots of testing 15:53:15 for - lots and lots of testing 15:53:19 for - lots and lots of hands 15:53:29 for - lots and lots of raised hands 15:53:37 testing for - lots and lots of raised hands 15:53:42 testing raised hands 15:54:20 so I can talk about hands 16:00:36 thomassch has joined #owl 16:21:52 hendler has joined #owl 16:22:39 hendler has left #owl 16:31:31 dlm has joined #owl 16:33:39 Evan has joined #owl 16:36:57 good afternoon, welome back! 16:37:52 IanH has joined #owl 16:37:53 Jeremy: has resigned from UFDT, but 16:38:05 ... wants to cancel next monday? 16:38:08 scribenick: uli 16:38:23 ScribeNick: Uli 16:39:01 AlanR: will arrange next UFDT 16:39:19 ACTION on AlanR to arrange next UFDT meeting 16:39:32 ACTION: on AlanR to arrange next UFDT meeting 16:39:32 Sorry, couldn't find user - on 16:39:40 alan, contact me when you're back in town (and rested) 16:40:22 IanH_ has joined #owl 16:40:30 q? 16:40:39 Sandro: has seen 7 sessions' minutes, currently 57 pages and asks how to read to accept them 16:40:56 ... and asks the scribes, when cleaning them up, to add sub headers 16:41:10 ... syntax is "===" for sub headers 16:41:10 Action: AlanR to arrange next UFDT meeting 16:41:10 Sorry, couldn't find user - AlanR 16:41:10 ack testing 16:41:15 q? 16:41:24 ack raised 16:41:28 q? 16:42:03 msmith: asks whether to serialize shuffled subdiscussions 16:42:22 sandro: yes, please disentangle 16:42:34 please speak a little louder 16:43:06 what did alanr just say? 16:43:06 IanH: asks what to do with parallel discussions, esp. on the IRC 16:43:16 Sandro: keep them if they are relevant 16:43:40 sandro: scribes finish cleaning up this wednesday 16:44:07 Jeremy: wants to see actions & resolutions in the minutes 16:44:40 Yeah -- keep IRC threads in if they are topical. 16:45:15 Bijan: subgroups affected by discussions at F2F should update their documents with pointers to minutes 16:45:16 I'd like to see things fleshed out a little in the minutes - 16:45:32 Jeremy: suggests to minimize effort on minutes 16:45:37 add links and pointers of a few definitions 16:45:45 AlanR: asks for subjects for discussions 16:45:59 raise hand 16:46:00 Bijan: non-OWL full issues with RDF mapping 16:46:30 AlanR: agrees with Bijan, mentions reification 16:47:29 Bijan: axioms annotation asserted versus reified 16:47:38 AlanR: wants to see both 16:48:05 IanH: we already agreed that we should explore both assertions & reifications 16:48:37 Sandro: what about b-nodes and reification 16:48:53 Bijan: can we discuss now some RDF mapping issues? 16:49:04 raise hand 16:49:13 msmith: has added such an issue wrt declarations 16:50:20 bijan: impossible to determine signature in owl full 16:50:29 ... under some conditions 16:50:58 alanr has joined #owl 16:51:07 alanr, jluciano is on the queue 16:51:48 is that horridge? 16:51:49 MattH: reports on user complaints regarding declarations 16:52:16 the issue on this is ISSUE-89 http://www.w3.org/2007/OWL/tracker/issues/89 16:52:35 JLuciano: asks for summary sections of minutes\ 16:53:00 JLuciano: wants to discuss evaluation issues 16:53:02 bijan has joined #owl 16:53:05 correction for scribe: it's that you can't specify a signature for an ontology without using the elements of that signature in an axiom or a declaration (which requires owl 11 terms) 16:53:34 PeterPS: disagrees with JLucianos suggestions: scribes should never paraphrase 16:54:38 AlanR: suggests to have summaries outside minutes 16:54:51 IanH: suggests to post summaries on the mailinlist 16:55:34 Sandro: add links to presentations in minutes 16:56:02 IanH: suggests clean up/mark up other material as well 16:56:41 Sandro: mentions that chairs could, if they wanted, blog meetings 16:57:19 Jeremy: doesn't want do them 16:57:28 or anyone else could blog meetings.... (ie summarize them). 16:57:48 AlanR: hasn't seen a lot about evaluation 16:58:32 alanr, adenda+ F2F2 ? 16:59:01 Jeremy: wants to give 2 examples reg. OWL Full compatibility 16:59:55 ... first one: we have an OWL11 document with reified annotions, we safe and modify it.... 17:00:58 MattH: do we discuss punning or declaredAs? 17:01:38 SUBPROPERTYOF[op1,...,opn] expands to rdfs:subPropertyOf if OnlyOP(opi) = true for each 1 ≤ i ≤ n, and to owl11:subObjectPropertyOf otherwise; 17:01:40 Jeremy: is worried about (starts reading out from ..please provide link) 17:01:50 That's the text 17:02:23 from http://www.w3.org/2007/OWL/wiki/Mapping_to_RDF_Graphs 17:02:43 Jeremy: there are various rules like this one, and they are the wrong ones 17:03:40 Bijan: understand why: if we have r subproperty of s, and then i add a composition, then this addition would lead to different kind of serialization 17:04:07 Bijan: suggests that using different syntax for SubPropertyOf would solve this issue 17:04:26 Bijan: this is different from round tripping 17:05:46 Jeremy: have some form of switch that safes an ontology in OWL11, then we shouldn't expect it to be safed in an OWL10 format unless i require this explicitly 17:06:12 MattH: asks whether the spec shouldn't specify this behaviour 17:06:46 msmith: asks whether Jeremy wants to safe tools only in OWL10 if explicitly asked to do so 17:07:17 Jeremy: observes a subtle relationship between the 2 OWL syntaxes 17:08:05 IanH: comes back to AlanR, and points out that it would introduce nasty non-determinism wrt serialisation 17:08:38 ... and that our n-ary disjointness axioms would cause trouble 17:09:15 AlanR: regards this as a bug 17:09:35 ... (to have these 2 possibilities for reading n-ary disjointness) 17:10:27 Bijan: sees an issue with the mapping, we need to decide what to do with it: deal with it or not 17:10:52 ... and it would be nice to be clear on our decision in the spec 17:10:57 ... asks for test cases 17:11:14 AlanR: declarations fall into similar league 17:11:30 MattH: disagree - we can throw them in/out 17:11:45 msmith: points back to issue 89 17:11:57 AlanR: asks whether we like declarations 17:12:28 Bijan: mentions that we can have both, declarations and roundtripping, but with a different mapping 17:12:45 oups - the last was MattH, not Bijan! 17:13:15 MattH: mentions discussions on the mailinglist, gives to Boris 17:13:34 Boris: explains that there are 2 readings of declarations 17:14:15 ... what is the meaning of rdf:type? To be used as linting/simple syntactic check? 17:14:21 Boris: the point of declarations is to performing 'linting' 17:14:29 Bijan: there are other use cases. 17:14:41 Bijan: adds that we can also throw out some "used terms" 17:15:34 -Achille 17:15:36 Boris: wants to distinguish declaredAs from type. 17:16:11 -Carsten 17:16:33 ... this will become tricky with imports. Since there is no notion of typing of RDF, things become problematic 17:17:07 ... eg, do we need to re-declare when importing? 17:17:40 ... in the old spec, there was no difference between "class" and "declaration" 17:18:10 AlanR: there wasn't even a notion of an ontology containing an axiom 17:19:01 Bijan: there is something about documents and ontologies (how to get one from the other) 17:19:30 Jeremy: suggests to use lateral thinking to solve this: use a new way of imports, namely one where 17:20:48 ... we put import statements at the top of our ontologies and then all declarations will be there! 17:21:17 Boris: seems to agree that this will help tools - if i knew what the type of things are, i can use streaming mode 17:21:50 Bijan: if they come late, they can still be useful (eg to find typos), but they are most useful at the top 17:22:17 Jeremy: suggests that we can do this via searching & process imports first 17:22:54 Boris: asks whether typed vocabulary will be obsolote - if yes, we can re-use it 17:23:50 Boris: we can merge the notion of typing and declarations, but cleanly 17:24:09 Jeremy: wouldn't it make a difference wrt model theory 17:24:16 Boris: no, it's all syntax 17:25:12 MattH: we need orphaned entities rather than declarations 17:26:11 msmith: parphrases that we want to be clear whether rdf type is a declaration or ...? 17:26:32 Boris: can we add a class to an ontology without adding an axiom? 17:27:03 ... declarations are a way to mention an entity outside any axiom 17:27:48 AlanR: asks whether in OWL11, can we have X owl:class Class? 17:28:17 Bijan: yes, it's in OWL Full, but it disappears in the OWL DL mapping 17:29:20 Sandro: wants to add next F2F meeting to agenda 17:29:29 Bijan: and XML syntax 17:30:00 Bijan: wants to see from Boris examples explicating differences and consequences of both solutions 17:30:25 AlanR: and we need to check our claims re. what appears/disappears in mappings 17:30:58 AlanR: wants to see backwards compatibility on the agenda 17:31:08 Topic: F2F2 17:31:25 April 3-4 17:31:34 PeterPS: April 3 and 4, in the Washington DC are, venue to be determined 17:31:37 Rinke has joined #owl 17:31:53 ... OWLED might be in the area, but perhaps not 17:32:14 where is Peter you talking about 17:32:17 ... one possibility is to make use of NIST, but access is restricted 17:32:25 I might be able to host it at MITRE 17:32:37 EvanW: access is a bit tricky, but only first time 17:32:41 Helooooo :-) 17:33:09 PeterPS: downtown DC or near to NIST are possible to 17:33:48 PeterPS: possibility to move 1 day earlier to make AlanRector happier 17:34:04 Who's the NIST person? 17:34:35 hand up 17:34:35 JLuciano: has mentioned MITRE 17:34:55 PeterPS: says that access at MITRE is even more difficult than at NIST 17:35:48 ... mentions that it will be busy and that we need to book Hotels early 17:36:07 Bijan: offers to make use of C&P rooms 17:37:07 PeterPS: reinforces the need to book hotels early 17:37:43 PeterPS: will come up with proposal together with Kendall Clark 17:38:25 how many people? 17:39:07 ACTION: on Peter to tell us by 2 weeks where F2F2 will be 17:39:07 Sorry, couldn't find user - on 17:41:10 ACTION: ppatelsc to tell us by 2 weeks where F2F2 will be 17:41:11 Created ACTION-50 - Tell us by 2 weeks where F2F2 will be [on Peter Patel-Schneider - due 2007-12-14]. 17:41:56 AlanR: wants to talk about backwards compatibility 17:42:21 seanb has left #owl 17:42:25 JeffP_ has joined #owl 17:42:28 msmith has left #owl 17:42:36 IanH: closes, thanks Sean Bechhofer for hosting 17:42:39 bye 17:42:48 ADJOURN 17:43:32 -Meeting_Room 17:43:39 -Joanne_Luciano 17:43:44 -Zhe 17:43:45 SW_OWL(F2F)6:00AM has ended 17:43:47 Attendees were Meeting_Room, Joanne_Luciano, +1.603.897.aaaa, Zhe, +1.518.472.aabb, jhendler, Achille 17:47:07 IanH has joined #owl 17:53:05 thomassch has joined #owl 17:53:10 thomassch has left #owl 20:38:52 sandro has joined #owl 21:29:28 IanH has joined #owl 21:33:26 Zakim has left #owl