Nicola writes:
> I have the feeling that the (scarce) economical and human resources
>related to this "Reference Ontology" effort are now concentrating on using
>"brute force" in order to i) get a reasonable coverage, and ii) establish
>useful alignment relationships between existing ontologies. I am convinced
>that a patient and systematic theoretical effort and a serious attempt to
>reach an agreement on fundamental theoretical distinctions are both
>unavoidable in order to achieve a real useful result.
>
> I am also convinced that such theoretical effort is *feasible*
>(i.e., it has good chances of success) given the present status of
>scientific (especially philosophical) research in the field, but it is not
>clear to me whether people belonging to this "Ontostd" group really share
>this feeling.
Doug concurs, adding:
>However I think that occasional 1/2 or 1 day meetings will not give us
>the speed of progress we might be capable of. I think we need to start
>using current "groupware" technology...To expedite this, I have been
>developing a system, DocKMan, but it is
And John S says:
>I agree with Nicola G. and Doug S. that more work needs to be done
>on the theoretical side. But I don't think that it can be done
>without very close collaboration with people who are working on the
>empirical side of things.
I heartily agree with all these sentiments. My colleagues and I find it
rather frustrating to be working under a very tight deadline to produce
an ontology for the HPKB-Jumpstart effort, without having the time or
personpower to establish it(s top level) on a firm theoretical basis.
It is difficult to produce something you are proud of if you have no time
or people to make the difficult "snipping" decisions in a thoroughly
considered way.
I would *love* to participate in a working group as implicitly described
above. What the precise balance would be between working face-to-face in
meetings and working via Web tools I don't know; I imagine both being done.
I think the problem is money. Unless we find some funding to support a
theoretical effort, only those who manage to get funding through some other
channel(s) (such as Fritz and I under HPKB-Jumpstart) can really work on it.
Unfortunately, such efforts are piecemeal and tend to have asymmetric growth.
I see no short-term solution to this problem. What is likely to happen is:
1. HPKB Jumpstart work will continue, and put into place an initial single
ontology, tailored to the Jumpstart domain, that contains at the top
concepts from SENSUS/Pangloss, CYC, and MIKROKOSMOS.
2. Our ANSI Ad Hoc meetings will continue, possibly less frequently, with
increasing emphasis on:
- submeetings to discuss theoretical issues (as Doug, Nicola, and John
mention);
- efforts to secure funding (possibly Nicola in Europe and us from DARPA).
3. Possible links with other ontology projects (EuroWordNet, for example).
We need a funder!
>From N:
> If there is a *real* interest in trying to make some progress in
>agreeing on basic theoretical distinctions, I will be happy to afford this
>trip (and even more happy if somebody is willing to offer some monetary
>contribution...), and contribute to the organization of a full-day session
>on theoretical issues, hopefully not limited to a few people. My own
>contribution could be a deeply revised version of my paper "Some organizing
>principles for a unified top-level ontology", which includes an top-level
>ontology consisting of about 100 basic concepts.
>From J:
>I think that the meeting at Stanford in November will be a useful
>update on what people are doing. But I doubt that it will satisfy
>Nicola and Doug's desire to reach some firm agreements about theoretical
>issues. Ed and I reached a rapid agreement on thematic roles because
>we both started from common ground on a subset that has been actively
>analyzed, studied, and implemented by a large community of linguists
>and computational linguists over the past 30 years. I don't think
>that some of the areas that Nicola and Doug are trying to address
>have yet reached that level of stability.
I would like to participate in such a one-day theory/detail meeting. It
is not clear to me though what this meeting would achieve. Could we set
up, before and by email, the beginnings of a longer-term plan, with a
set of issues to be worked on, and a clear goal to drive at?
>From N:
> I would also be happy to establish a link between this ANSI
>initiative and the conference on Formal Ontology in Information Systems we
>are organizing in Trento, Italy (http://mnemosyne.itc.it:1024/fois98). This
>broad-scope conference could be a good opportunity for augmenting the
>visibility of this ANSI initiative, and maybe we can discuss of the
>modality of such a link in the first day of the ontology meeting.
A very nice idea!!
E
----------------------------------------------------------------------------
Eduard Hovy
email: hovy@isi.edu USC Information Sciences Institute
tel: 310-822-1511 ext 731 4676 Admiralty Way
fax: 310-823-6714 Marina del Rey, CA 90292-6695
project homepage: http://www.isi.edu/natural-language/nlp-at-isi.html