UBL-Ontology Project Conference Call - Thu 2003-09-25 (7NB)
Conference Call Details (7NC)
- Subject: [UblOntology] project conference call Thu 2003-09-25 (7ND)
- Comments/Agenda: (7NE)
- Agenda Comments: ... build rapport; ... sync up; ... distribute, assign or arbitrate virtual ownership of tasks; ... and use it to deal with emergencies. ... plus, member suggested agenda items. (7NF)
- Please post your suggested agenda items if you would like to see anything put onto the agenda (7NG)
- please post or upload any material to be shared to the list, to the wiki or by ftp upload prior to the meeting (7NH)
- VNC session will be started 5 minutes before the call at: http://vnc2.cim3.net:5800/ (7NI)
- During the call, please browse the wiki page at: http://ontolog.cim3.net/cgi-bin/wiki.pl?Conference_Call_2003-09-25 (7NL)
- Date: Thursday, Sep. 25, 2003 (7NM)
- Start Time: 10:30 AM Pacific Daylight Time (7NN)
- End Time: 11:25 AM Pacific Daylight Time (7NO)
- Dial-in Number: 1-702-851-3330 (Las Vegas, Nevada) (7NP)
- Participant Access Code: "20030925#" (7NQ)
Attendees (7NR)
- Attended: (7NS)
Agenda Ideas (7O1)
Agenda & Proceedings (7O4)
1) Welcome (7O5)
- KurtConrad took the chair and welcomed everyone (7O6)
2) Appointment of secretary to take minutes (7O7)
3) Roll-call of participants (7O9)
4) Attempt to draw closure to the "KIF vs Protege " tools & process issues: (7OB)
- Going around - what is the issue?: (7OC)
- (7OF)
- PatCassidy: I understand that KIF will be the definitive representation. I consider Protege as a really fancy "word processor", how can it be worse than using another word processor and then putting the work into KIF? (7OG)
- (7OJ)
- LeoObrst: I tend to see the points on both sides. (1) KIF is more expressive, that is valuable and we want to retain that. I agree with Adam that if we use Protege, we will not be able to enumerate everything that we will be able to do so with KIF. (2) Protege is a very well developed and simple (with proper training) tool for people to learn to construct ontologies. I know Protege to KIF conversion work has not been done yet. (7OK)
- We probably want both. My position is: we can extract as much of the ontology as possible with Protege, and use the PAL laguage to extract axioms. (7OL)
- that way, more people can get involved in doing the bulk of the work and then some of us who are more skilled can come in to enumerate the axioms. (7OM)
- LeoObrst: I tend to see the points on both sides. (1) KIF is more expressive, that is valuable and we want to retain that. I agree with Adam that if we use Protege, we will not be able to enumerate everything that we will be able to do so with KIF. (2) Protege is a very well developed and simple (with proper training) tool for people to learn to construct ontologies. I know Protege to KIF conversion work has not been done yet. (7OK)
- (7ON)
- MikeDaconta: significant work has been put into an OWL plug-in for Protege. The industry momentum as well as my day-to-day work is focused on OWL. I would suggest we look into generating output in OWL as well. (7OO)
- (7OR)
- MonicaMartin: the issue seems to be a tools preference issue. We do need to move quickly to a solution though. (7OS)
- (7OV)
- Adam's response: (7OW)
- using Protege is a "looking under the lamp post" argument. Protege being is easy is irrelevant if it can't meet the objective of the group, which currently is to create a formal ontology, reusing SUMO. (7OX)
- people using Protege (or any other frame language) will have to "reinvent the wheel", see msg: (with the representation of measure function example in my post1 & post2) (7OY)
- skeleton is not capable of representing what we need (7OZ)
- Adam's response: (7OW)
- (7P5)
- Adam: (7P6)
- I have a full FOL prover (not SNARK), although one could use SNARK as well. I'm using University of Manchester's VAMPIRE <http://www.cs.man.ac.uk/~riazanoa/Vampire/> (7P7)
- Leo: check out this recent paper by Muthukkaruppan Annamalai and Leon Sterling, "Dealing with Mathematical Relations in Web-Ontologies", http://oas.otago.ac.nz/OAS2003/papers/oas03-annamalai1.pdf (ref. Leo's post (7P8)
- what they said is "KIF is more expressive, but they don't see Physicist using that to express themselves" (Adam counters that this is a bogus assertion, because Argonne Labs, a world leading center for physics research, developed Otter, which has proved new theorems in mathematics. The people that have a problem with formal logic aren't physicists and others with a solid math background). They did not talk about OWL, though. (7P9)
- OWL full still does not handle relations of arity greater than 2, which is a problem (7PA)
- Adam: (7P6)
- (7PB)
- Leo: most of the tools that are emerging are going to be dealing with OWL or RDF; I realize the expressivity argument (which is very powerful), but I don't know of any other solution except for using a "hybrid" (7PC)
- (7PD)
- Mike: I am not convinced yet that this is not a red herring. I cannot argue whether KIF can express all mathematical functions. (7PE)
- (7PH)
- PatCassidy: I am "building" something to respond to Adam's example. I do not see a technology barrier to building something that will accurately import and export KIF into and out of the Portege environment. It's not rocket science, but a lot of details. I can use help from people who are very conversant with java and the Protege environment. (7PI)
- MikeDaconta: I will be happy to assist you on the part of java. (7PJ)
- PatCassidy: I am "building" something to respond to Adam's example. I do not see a technology barrier to building something that will accurately import and export KIF into and out of the Portege environment. It's not rocket science, but a lot of details. I can use help from people who are very conversant with java and the Protege environment. (7PI)
- (7PK)
- Adam: I am disappointed if Pat's response to the issues I've pointed out is to use a tool that isn't going to be available, maybe, for another year. (7PL)
- (7PP)
- Leo: closing comment. I hope we can resolve this. I don't think we are in total discord. My recommendation would be some hybrid, but I am not sure if that could satisfy everyone. (7PQ)
- Second go around - where are we? (7PR)
- Adam: I think we still have confusion around fundamental issues. But if we push for using Protege, that would probably not be an effort I would want to spend time on, because it won't meet the objective of this group. (7PS)
- Mike: I respect Adam's knowledge, would like for him to do a tutorial on axioms, which seems to be the crux of the problem. (7PT)
- Pat: I would continue to pursue building that pluin I mentioned (which I cannot predict how long it will take). I will continue to contribute. (7PU)
- Mike: maybe you can make that another ontolog project. (7PV)
- Peter: the greatest value in the group in the gathering of people; so let stick together. Mike proposal is a great alternative. (7PW)
- Monica: I hope you guys can get pass the philosophical discussion. (7PX)
- Kurt: let's take Pat's approach as a "future proposal". (7PY)
- Adam will try to make available his suite of tools (which includes the VAMPIRE rasoner) available. (7PZ)
- it is the general consensus that we have reached closure on this matter. (7Q3)
- meeting adjourned 2003-09-25_12:06pm PDT (7Q4)
- next meeting same time next week, call commencing 2003-10-02_10:30_PDT (7Q5)
-- notes taken by PeterYim in real time on this wiki. will all those present please review and edit as appropriate. ppy / 2003-09-25_12:08_PDT (7Q6)