ppy/chat-transcript_edited_20120426b.txt ------------- Chat transcript from room: summit_20120426 2012-04-26 GMT-07:00 [PDT] ------------- PeterYim: Welcome to the = OntologySummit2012: Postmortem Session - Thu 2012-04-26 = Summit Theme: OntologySummit2012: "Ontology for Big Systems" Session Topic: Ontology Summit 2012 (postmortem) review and follow-up action planning Session Co-chair: Dr. LeoObrst, Dr. RamSriram, Professor MichaelGruninger Program: * Summary report on OntologySummit2012 - by LeoObrst on behalf of the organizing team * Open discussion-I: Postmortem of this year's summit - Moderator: RamSriram * Ongoing Follow-up activities updates - by ToddSchneider-AliHashemi, AmandaVizedom-MikeBennett-SimonSpero, HensonGraves * Open discussion-II: Follow-up Action planning - Moderator: MichaelGruninger Session page: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2012_04_26 Mute control: *7 to un-mute ... *6 to mute Can't find Skype Dial pad? ... it's under the "Call" dropdown menu as "Show Dial pad" . == Proceedings: == . [09:28] anonymous morphed into CoryCasanave [09:30] anonymous morphed into RamSriram [09:36] HensonGraves: have to check my settings,it will take me a few minutes [09:38] SimonSpero: Salut [09:38] SteveRay: Francais? [09:39] PeterYim: -- session started 9:38am PDT -- [09:40] PeterYim: == LeoObrst presenting: summary report on OntologySummit2012 ... [09:50] JoelBender: It may also reflect confidence in the already committed advisory committee members! :-) [10:02] PeterYim: == Open Discussion-I: postmortem of OntologySummit2012 - Moderator: RamSriram [10:02] PeterYim: ALL: please capture your key points onto the chat (even if you have made them verbally) [10:02] FrankOlken: I would urge that the summit be held somewhere more convenient next year, e.g., downtown Washington, DC, Arlington, or Crystal City. [10:08] TerryLongstreth: @Frank - are you offering suitable site? [10:10] ToddSchneider: Frank, who would pay? [10:11] ToddSchneider: There's is also George Mason University [10:12] CoryCasanave: I don't mind NIST, I drive from Reston. [10:13] FrankOlken: Yes, but you have to register way ahead of time for access to the NIST site. [10:30] MikeBennett: There is quite a cachet to having the Summit at NIST. People who have not heard of the Ontology Summit (or indeed of ontologies) have heard of NIST. So it's automatic credibility. [10:13] ToddSchneider: I also don't mind driving to NIST, at least once I get past Tysons [10:14] MikeBennett: Once one is driving it's easy to get to. Perhaps some of us who rent or borrow a car can carpool with those who don't? I'd be happy to do that. [10:14] ToddSchneider: Frank, most hotels and other facilities require advance scheduling. [10:15] SteveRay: @Todd: To be fair, so does NIST. [10:15] CoryCasanave: @steve, yes - OMG and other larger meetings get free rooms that you may be able to piggyback. Another site could be MITRE. [10:15] SteveRay: @Cory: Agreed [10:17] ToddSchneider: But the MITRE facility in Fairfax is in the Tysons / McLean area. [10:18] LeoObrst: Yes, MITRE would be possible, but a reservation for a large meeting would require a fairly long lead-time. We have a very large auditorium (100+), and a slightly smaller venue (about 60-70). [10:19] LeoObrst: Also: audio-visual, etc. support would require my finding some funding internally. [10:21] FrankOlken: MITRE is a lot closer to Dulles airport. [10:21] FrankOlken: Also, the Metro will run to Tysons corner in either 2013 or 2014. [10:24] FrankOlken: Tyson's Corner Metro extension will be complete in late 2013. [10:28] MichaelGruninger: Changes in venue will lead to much more organizational overhead, and I'm not sure whether we have the resources (i.e. people and time) to do this. Organizing the technical content is hard enough. [10:29] MichaelGruninger: i.e. if you want to change the venue, be prepared to be the local organization chair. [10:24] BobbinTeegarden: About location suggestions, any chance we could alternate east coast with the west coast, and maybe have the next one in Silicon Valley or environs next year? [10:01] MichaelGruninger: Comments on Symposium: spend more time on open discussions (e.g. panel) and demonstrations on the second day [10:06] AmandaVizedom: Suggestion: Schedule a Workshop(s) day for the day *after* the summit; use this day to launch or further follow-on efforts and collaborations. Could include anything from working to set up relationships to vocamps to do rapid development of small related ontology modules. [10:08] MikeBennett: @Amanda +1. ... [10:03] CoryCasanave: Comment: The timeframe between the initial planning meetings and the start of the virtual sessions seemed short this made it difficult to plan some of the sessions and also did not leave much time to market the summit as it emerged. Particularly with cross discipline it could take quite a bit of time to engage these external communities. [10:04] JoelBender: (en-coo-see acronym spelling and decoding please) [10:05] TerryLongstreth: International Council on Systems Engineering (INCOSE) is a not-for-profit membership organization founded in 1990 [10:07] JoelBender: @Terry, thank you - for the log [10:05] AliHashemi: It would have been good to more explicitly reuse material from last year's summit. The ontology usage and metrics framework was a great resource that could have been extended to provide context to this year's discussion and allow us to provide more fine-grained recommendations. [10:04] TerryLongstreth: Topic for next year: 7 year retrospective (perhaps tied to a new website page for continuous tracking) Each subject area report have minimum of following: YEAR------THEME---Results/Accomplishments/Status TO DATE---CONTINUING/ONGOING ACTIVITIES-- [10:06] AliHashemi: Similarly, a commitment to greater systematicity in the capture of presented materials would greatly facilitate reuse and make it easier to connect the content of each year's summits to one another. [10:07] SteveRay: We could consider holding joint sessions with other organizations, either as part of the regular virtual sessions, or possibly to host the Ontology Summit Symposium. [10:08] CoryCasanave: @Steve, the OMG usually has a DC meeting in March we could schedule the week prior in collaboration with the OMG ontology SIG. [10:09] SteveRay: @Cory: I definitely like the idea of some sort of liaison with OMG. [10:09] HensonGraves: @cory OMB sounds attractive [10:14] SteveRay: Joint sessions with other organizations could possibly contain costs if we held F2F sessions elsewhere. [10:08] MikeBennett: ... I wonder if it would work to go through the material in the Communique of any given summit, identify what might be important findings or important bodies of work that merit wider exposure, and think about marketing of those. In this year's case, I would suggest that the Quality material is a useful industry resource. [10:09] ToddSchneider: It would facilitate remote participants if a video feed of the main conference room could be provided. [10:11] anonymous morphed into DilvanMoreira [10:18] FrankOlken: Peter, You may need to trademark the Ontology Summit. [10:25] PeterYim: RamSriram: because the symposium registration is free of charge, people have a tendency to register first, and decide not to show up later. I would request that people be conscientious of the fact that NIST gets charged by the number of registrants. Therefore, at the least, don't casually register (unless you are coming) and *cancel* your registration (by emailing one of the people handling the logistics) if you later find out that you can't make it to the symposium. [10:26] MikeBennett: @Peter perhaps that should be added to the registration page - I for one did not know that. [10:24] LeoObrst: I wonder what folks think about our focus this year. [10:25] SimonSpero: leo +1 [10:25] CoryCasanave: @Leo, I thought the focus was overly wide. [10:26] PeterYim: @Leo @Cory - I disagree that we lacked focus (or the focus being too broad) this year [10:26] CoryCasanave: A sharper focus as well as making it more clear what people may get out of participating would help. [10:26] ToddSchneider: Leo, no. There may have been a confusion among scope and focus. [10:27] MikeBennett: @Leo I thought at first the focus was a bit wide or woolly, but as it turned out it scared up a lot of interesting material, so I'd say no. [10:28] BobbinTeegarden: @Leo @Mike I agree, the wider perspect brought out things we would have missed imho [10:28] SteveRay: I think the overriding issue is that "ontology" is a tool looking for a problem, thus we need to apply it to various problem spaces. [10:31] BobbinTeegarden: @Steve Could I turn that over and suggest that "ontology" might be a better way to solve many (most?) problems (better understanding and online processing, cheaper integration etc), and it might be fun to take major emerging problems and see how we would address them with Sem Web and Ontos... [10:33] SteveRay: @Bobbin: If I understand you correctly, yes, that's what I would advocate. [10:31] SteveRay: I'm still confused as to why we have the fruit fly quote on the pretty page. Funny, but I don't see the relation to our work. [10:41] DougFoxvog: @Steve: The fruit fly example is a traditional NL problem of multiple possible interpretations that can be solved by ontological encoding if modeled predicates have inter-argument type constraints. [10:42] SteveRay: Pretty subtle connection for the general technical audience. [10:42] ToddSchneider: Steve, I like to be amused while working. [10:43] SteveRay: @Todd: No problem with that, but I do think it alienates those "not in the know" [10:45] ToddSchneider: Steve, think of it (the silliness] as a gate keeper. [10:45] SteveRay: @Todd: We don't want a gatekeeper! We want to suck in the uninitiated. [10:37] FrankOlken: Amanda, if I follow the link to the survey from the Ontology Summit web page, the survey landing page asks for login before explaining what the survey is about. It is very confusing .... [10:42] MikeBennett: The simplest thing would be a small explanation on the initial page of the LimeSurvey thing. In the main Summit page it's buried a way down, and in a page which refers to past events in the future tense. So I think the LimeSurvey first page needs to stand alone. It does say that you need to register (not just asks for a login), so what's missing is only a description of what the survey itself is. [10:39] LeoObrst: @Amanda: was the survey notice sent to IAOA lists? [10:40] LeoObrst: I also wonder if notice should be sent to some Semantic Web lists. [10:44] ToddSchneider: Does anyone think asking the Semantic Web community to endorse the communique is a good idea? [10:44] SteveRay: @Todd: I think so [10:45] MichaelGruninger: Perhaps someone could write a summary article for ISWC [10:45] ToddSchneider: Michael, isn't the communique a summary? [10:47] MichaelGruninger: @Todd: An ISWC paper would be more technical than the Communique as it is currently written [10:47] LeoObrst: @Michael: I think that's a great idea, i.e., ISWC. [10:49] ToddSchneider: Michael, okay. That's doable (via spin). [10:51] AmandaVizedom: @Michael: I think an ISWC-appropriate paper focused on just the quality stuff (including survey & results) is also very doable, and hope to work on that. [10:47] EricChan: @Ali and @Todd: I have a comment about a sentence in communique: The Internet has made it far easier for different people in the different parts of the world to share and combine data, information, and knowledge. The subsequent passage implies the meanings of information and knowledge. I believe it will benefit the general audience to explain the bifurcation of information and knowledge as used in the communique. [10:54] ToddSchneider: Eric, yes it would be a useful addition, but such a change represents a substantive modification to the communique, hence can't be done at this point. [10:50] CoryCasanave: URL for marriage example: http://www.omgwiki.org/architecture-ecosystem/doku.php?id=composite_concepts [10:52] MikeBennett: To Leo's point - how much of this is language and how much is the constructs in the ontology - agree. The specific examples would be handled differently in the patterns we use in FIBO for example (with OWL constructs) than in some other ontologies e.g. we use second order "things in roles". Similarly some of the stuff in Matthew's example would be based on a 4D extensional ontology framework. So there's a cross over between ontology patterns, language syntaxes and so on. [10:52] SteveRay: I believe that sometimes we lose perspective on things we take for granted. Specifically, ontologists tend to approach a new problem by thinking about the ontology (or model) that characterizes the problem space. However, many other folks in other disciplines don't automatically think of things this way. They may think of building a system/solution in terms of functional modules with hidden internal data representations (think object oriented programming), or other paradigms that are not model-based, or at least not explicitly model-based. We need to explain to other communities these kinds of basic ways of looking at things (that we take for granted). [10:53] TerryLongstreth: @Steve +1 [11:03] CoryCasanave: The original theory of OO put OO development and domain ontologies much closer. This may not have worked out as well as expected :) [10:54] MichaelGruninger: My original idea was to propose a set of benchmark problems that could be used to compare people's approaches (including their ontology and the representation language used). By using a commonly agreed upon set of problems, we can often avoid a lot of frustration. [10:55] SteveRay: I do like the concreteness of comparing different approaches to benchmark problems. It helps us non-abstract thinkers! [10:56] MikeBennett: I have to drop off now - thanks all! [11:11] TerryLongstreth: @Cory - Federation and Integration for technical/technology purposes, or federation/integration in a social, or organizational setting? [11:13] CoryCasanave: Terry - I look at the technical federation/integration as something done in the context of the organizational integration/federation - so both. [11:17] TerryLongstreth: @Cory - I agree, in the abstract, but I think someone should tackle the integration of cognitive insights across cultures first, to provide a framework for the technologists. [11:19] TerryLongstreth: My point is that the problem itself is one of culture. [11:19] TerryLongstreth: Has anyone tried to ontologize the competing concepts of marriage across cultures ? [11:19] HensonGraves: @steve, submit a modeling challenge problem to the challenge project [11:21] SteveRay: @Henson: Fair enough. I know that a sub-team wrestled with two very specific problems in that context. I could send you the solutions that were proposed out of it. [11:22] HensonGraves: @steve, great, send the problem as well as the solution [11:22] SteveRay: @Henson: OK [11:20] TerryLongstreth: Have we ontologies for legal concepts across jurisdictional boundaries? [10:56] ToddSchneider: Abstract?? What's abstract:) [10:56] SimonSpero: Todd: It's the bit at the front of the paper that you read in order to pretend you read the whole thing [10:58] BobbinTeegarden: @Simon Nice concretion. [10:58] SimonSpero: @bobbin's meant well [10:56] AmandaVizedom: @Steve, this point {context please} is extremely important and an area in which much existing work does not get passed to ontologists-in-training. Specifically, I am thinking of work in knowledge elicitation, specifically for ontologies AND from other field such as information and library science, and also cognitive work analysis... [10:59] anonymous morphed into PavithraKenjige [11:00] SimonSpero: How about modeling common sense physics. We could start with liquids first. [11:05] ToddSchneider: Is a algorithm for heat transfer coded in Java not a model for heat transfer? [11:01] SteveRay: @Simon: Careful, I think you are about to slip into an upper ontology discussion. [11:02] SimonSpero: Have to leave now. [11:02] AmandaVizedom: I need to drop off now, folks. Thanks! [11:02] SteveRay: @Pavithra: +1 [11:03] LeoObrst: @Henson: start with the problem and what kind of notions you need to express the objects and relationships of the problem. [11:06] HensonGraves: @leo, exactly [11:05] PeterYim: == Open discussion-II: Follow-up Action planning - Moderator: MichaelGruninger - All [11:06] ToddSchneider: Michael, beer helps coordinate also. [11:07] SteveRay: How about somebody volunteering to be a DARPA program manager? [11:07] ToddSchneider: Have to drop the audio portion. Cheers. [11:09] PeterYim: == (discussion) follow-up actions for this year's summit ... [09:58] SteveRay: One additional follow-on action is the generation of new joint projects among the communities. [11:14] SteveRay: These folks just got $10M from NSF: http://amplab.cs.berkeley.edu/ [11:15] FrankOlken: @Steve, However, I do not recall that amplab is doing anything about ontologies .......... [11:16] SteveRay: @Frank: ...so maybe there is a vacuum to be filled. [11:17] FrankOlken: To the extent that the DB is interested in semantics, they are using Datalog rather than OWL or SPARQL. [11:17] SteveRay: @Leo: My impression is that the semantic web community feels unwelcome by the ontology community. [11:22] LeoObrst: My comment was that we need to bridge the gap to the Semantic Web community, and do so more strategically. We do all advocate using SW technologies for many of the everyday applications we address. Even Semantic Media Wiki uses SW technology. [11:18] TerryLongstreth: All of our examples are so ethno-centric that we lose credibility with our examples. [11:18] CoryCasanave: @Terry, so you have to do this "first", prior to solving problems? [11:18] PeterYim: == (discussion) suggestions for OntologySummit2013 - candidate themes, ... etc. [10:04] TerryLongstreth: Topic for next year: 7 year retrospective (perhaps tied to a new website page for continuous tracking) Each subject area report have minimum of following: YEAR------THEME---Results/Accomplishments/Status TO DATE---CONTINUING/ONGOING ACTIVITIES-- [11:08] CoryCasanave: Topic: I will put forward again the same one I did last time - federation and integration. Or, is that to much like this years? [11:13] LeoObrst: A previous suggestion was "ontology evaluation" for an Ontology Summit theme. [11:23] MichaelGruninger: yes, Ontology Evaluation [11:26] MichaelGruninger: Ontology Evaluation can tie together some ideas from previous summits -- relationship to standards conformance, evaluation of ontologies within the context of specific information systems, using benchmark problems (cf Henson's slides) [11:27] HensonGraves: @michael, agree that model challenge is related to ontology evaluation [11:19] BobbinTeegarden: For next year topics, could we add executable ontologies, and how we add process to an ontology model? [11:20] BobbinTeegarden: Those might be two separate topics, same bucket? [11:22] BobbinTeegarden: One liner: adding process to, executing ontologies [11:22] AliHashemi: Ontologies in Information Systems - How are they being used? How are they parts of a bigger puzzle? [11:22] EricChan: Sounds like an ontology of the modeling languages (UML, RDF, OWL, FOL, etc) clarifying the inter-relationships (A more expressive than B, C is refinement of D, etc.) among these modeling languages will be helpful. [11:25] BobbinTeegarden: @Eric Love the ontology of modeling languages idea, let's! [11:31] EricChan: @Bobbin, perhaps we can propose "ontology of modeling languages" in the modeling workshop being planned. [11:32] HensonGraves: @eric, bobbin, this is a good idea but will take a lot of ontology engineering [11:24] BobbinTeegarden: @Leo Is there a bridge to the wider community or enterprise architects, that might break down some stovepipes? Seems the architects are in the trenches trying to implement ontologies without the benefit of our input... [11:24] CoryCasanave: @Bobbin, I like breaking down stovepipe themes [11:27] CoryCasanave: Yes, and also put effort into outreach in the interim [11:28] TerryLongstreth: Suggestion: Ontologies - Building bridges across cultures [11:29] CoryCasanave: @Terry: Are ontologies a solution to stovepipe problems or another stovepipe? I don't think the mainstream users are sure. [11:30] TerryLongstreth: I'm not sure my mainstream and your mainstream are the same [11:30] TerryLongstreth: :>) [11:31] CoryCasanave: @Terry - so which is the mainstream, mainstream? [11:24] PeterYim: capturing here a suggestion that PatrickDurusau sent in along with his Communique endorsement - "The OntologySummit2012_Communique observation that systems interact with their environments, necessitating changes in an ontology for such a system is spot on! A topic for next year [2013] could be viewing ontologies as environments, enabling modeling relationships of ontologies to systems and users of both. --PatrickDurusau /2012_04_25" ... I've already captured that to: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit/Suggestions#nid3AC4 [11:25] LeoObrst: Ontology and databases [11:26] FrankOlken: I agree that Ontologies and Databases would be an interesting and useful summit topic. [11:29] RamSriram: I need to log off. [11:29] PeterYim: thank you, Ram [11:32] PavithraKenjige: Big data - Ontology discussion as a suggestion [11:31] PeterYim: === endorse the communique ... and get your friends to do it too! [11:31] PeterYim: see: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2012_Communique ... access to the document and endorsements instructions can be found near the top of that page. (Note that endorsements will close by end-of-day 12-May-2012.) ... essentially, what one needs to do is to send an email to < communique-endorsement [at] cim.us > and put your name in the body of that message. [11:18] FrankOlken: I can tweet about the communique. [11:32] PavithraKenjige: Thank you [11:32] AliHashemi: thanks, bye [11:32] PeterYim: Great session! very productive! [11:33] PeterYim: -- session ended: 11:32am PDT -- -------------