chat-transcript_20100325f_unedited.txt PeterYim: . Welcome to the Joint SIO-OOR-Ontolog Panel Discussion: "Getting the SIO Project Going" - Thu 25-Mar-2010 * Chair: Dr. JohnSowa (Vivomind Intelligence) * Panelists: o Dr. MatthewWest (Information Junction) o Dr. NicolaGuarino (LAO-CNR) o Dr. PatCassidy (Micra) o Professor MichaelGruninger (U of Toronto) o Professor JohnBateman (U of Bremen) o Mr. MikeDean (Raytheon-BBN) o Mr. PeterYim (Ontolog; CIM3) Please refer to details on the session page at: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2010_03_25 anonymous morphed into Harold Solbrig anonymous2 morphed into MattHettinger anonymous1 morphed into BartGajderowicz BartGajderowicz: @PeterYim, are the slides for 3-Guarino and 7-Dean available? anonymous1 morphed into Kevin Keck Kevin Keck morphed into KevinKeck anonymous morphed into Lyn Niemeyer anonymous morphed into PavithraKenjige CameronRoss: hello PeterYim: @Bart - no, we don't have slides from NicolaGuarino and MikeDean ... some panelists may choose to make their remarks without slides ... I have urged them (and everyone) to document the points they are making on this chat-board, though Gary Berg-Cross: I expect that someone will define "shared" and "integrated" in the SIO title so we are clear on the goal. PeterYim: @Gary - that would be a good question for JohnSowa ... he named this project "SIO" ArturoSanchez: @Michael: is there an ontology which defines "meta-data" (as you called them) terms? Phil Murray: Gotta go (2:10). Very interesting. Let me know how I might help. KevinKeck: Did I miss a discussion of why it should be named "shared and integrated" rather than "sharing and integrating"? I personally rather like the latter, as in the title of John's motivating presentation. PeterYim: @PhilMurray - thank you ... we'll probably be starting a new [sio-forum] mailing list; I will subscribe everyone who participated today ... see you there! PeterYim: @Kevin - we did not have such a conversation KevinKeck: To the extent that we are developing "scripts" and methods in addition to ontologies, I think the -ing form is preferable. PeterYim: @Kevin - let's bring that up during the Q&A/open-discussion segment ... since JohnSowa took the lead on this effort, we did say (during the Mar-16 meeting) to let him make the call on how to label this effort ArturoSanchez: Wow! What a good interoperability example!! Destiny is laughing at us now anonymous1 morphed into JohnBateman Gary Berg-Cross: Missing the Spring time change has a bit more bite when you are trying to make a plane flight back to the US as happended to me last year. JohnBateman: I think I'm doing well when I manage to remember to get the spring/fall time change right in just *one* timezone, let alone several! PeterYim: @MatthewWest, NicolaGuarino, JohnBateman, AlexGarcia - my apologies for having posted the wrong start time for our colleagues from Europe (the US has moved over to Daylight Saving Time on Mar-14, but Europe is still on "standard time" until Mar-2 Lyn Niemeyer: Thank you so much for allowing me to listen in. I have to move on, but I am enthused that this group is looking at ontology inter-operability while the ontology concept is still in the "craft stage". I can't offer much expertise in this area as my work is on a much more lay-level, however I intend to keep tabs on your work! Have a beautiful day. anonymous1 morphed into PierreGrenon PeterYim: @Lyn Niemeyer - thank you very much for participating today, Lyn ArturoSanchez: Slide? PeterYim: JohnBateman's slide#19 now PeterYim: goign to #20 PeterYim: skide#21 now KevinKeck: Slide #20 made me think of OMG's ODM. KevinKeck: (and John's comments on the mailing list of integrating with mainstream tools) PeterYim: JohnSowa: suggest that JohnBateman or other folks to work on the ability to translate to UML MikeBennett: Indeed, I thought ODM had quie a lot of traction. Harold Solbrig: ODM, however, has a risk. Many people think that converting a UML information model to OWL gives ontology... Harold Solbrig: ... but the only ontology you get is about tables (e.g. parents are optional, but all people inherently have SSN's...) MikeBennett: Not if you use ODM properly - if defines a complete metamodel, which can be implemented. Of course that doesn't guarantee that's what built is an ontology, but to the same extent that having something in OWL doesn't guarantee that it's an ontology. Harold Solbrig: The problem is that, by and large, UML models data. It is very strange to hear the claims these days about what sorts of ontology now exists, thanks to ODM conversions of implementation models. Harold Solbrig: And even more interesting to see what they assert... MikeBennett: Well I don't know about that but I know that I have been able to use ODM (in an early draft) to create a workable ontology using a UML tool. It's down to having a clear approach to meaning, but that seems clear enough in OWL as it is. I don't see what difference the toolset makes at the level of the ontology content (obviously it makes a difference in terms of URIs etc.). MikeBennett: Anyway "UML models data" is a bit like saying "paper creates charcoal sketches". Harold Solbrig: I realize I misspoke - UML tends to be used to model data. Harold Solbrig: We are just frustrated because folks are taking shortcuts - on the basis that now that they've got the ODM, there is no longer need for ontology modeling. They've already got it. CameronRoss: @JohnBateman: Does the HETS GPL-like license support its use within commercial applications? Harold Solbrig: A risk that I think this community needs to be aware of... MikeBennett: @Harold fair point. I think the theoretical framework is as important regardless of what tools or syntax are used PeterYim: @MikeDean ... missed you calling in earlier, please make your remarks right after JohnSowa finishes ... (we only have about 15 minutes left) PeterYim: folks: one more mishap on today's session ... the audio recording will only be for this last segment (the last 20~30 minutes only ... the earlier presentations didn't make it in) ... sorry! JohnBateman: Example of relabelling: see my slide 12, where we see CASL *parameterisation*. There we have the abstract generalized mereology theory GENMEREOLOGY which is reused for time (T), space (S) and perdurants (PD). This is exactly the kind of useful relabelling meant and shows how (a) it is already used in ontology design and (b) is fully supported by HETS structuring methods. MikeBennett: Certified - in light of the above notes on ODM, should this include some sanity check on there being a workable theory of meaning inthe ontology in the first place? MikeDean: @PeterYim - sounds good MikeDean: Mike Dean is one of the co-conveners of OOR, along with Peter Yim and Leo Obrst MikeDean: I'd like to echo Peter Yim's comments about the value of this effort and bridging between related communities MikeDean: OOR is pleased to support SIO MikeDean: good opportunity to attract more OOR users and developers MikeDean: We look forward to further collaboration! PeterYim: JohnSowa: re-labeling the project "Sharing and Integrating Ontologies" PeterYim: new mailing list: [sio-dev] to address the next level of granularity PeterYim: the overall conversation on "Sharing and Integrating Ontologies" will stay on the [ontolog-forum] list PeterYim: JohnSowa will start a thread on [ontolog-forum] to work up the SIO Project Charter PeterYim: thank you everyone, for the participation ... great session! PeterYim: -- session ended; 2010.03.25-12:38 --