RE: SUO: RE: More KIF-ified Ontology Content
Pat,
I don't think I understand your distinction. We have used Cyc as an
interlingua to translate between ontologies. It is a particular ontology
that consists of an upper ontology and "middle" and domain-specific
ontologies. I wouldn't claim that it's all things to all ontologists, but
rather that it does provide a coherent formalization of most everything in
the world you'd want to talk about (to some level of detail).
If in the SUO we create another upper level ontology then there will be
two ontologies suitable for acting as interlinguas. Neither will be
"right". Folks will fight over the differences, but groups would be able
to decide to conform to either and be just fine. Sure, there could be many
standards if groups want to take the time to develop them. I suspect
however that if there is a standard which covers enough of the world to be
useful, that practical people won't want to create their own when the IEEE
SUO will work well enough. If there is consensus on using the Cyc public
upper ontology in this study group, then that's what we'll use and just
augment it as people see fit.
Some people will think this is not a useful effort, just as people have
criticized Cyc for many years. These concerns have been raised many times
and I guess there isn't compelling evidence for either camp to change its
view. Ultimately, as Jim Schoening has pointed out, some folks will choose
to be part of the SUO effort, and some won't.
Adam
At 03:53 PM 12/7/2000 -0600, pat hayes wrote:
>>Matthew,
>> I think Pat's point is a bit simpler - that some reasonable choices in
>> ontology are incompatible, so we'll never get a useful single
>> ontology. My point is the opposite - we can make some reasonable
>> choices and get a coherent standard. I take Cyc as an existence proof
>> that reasonable choices can be made and the result is a product of
>> significant value.
>
>Hi Adam
>
>Yes, but Cyc is not an upper ontology, but an actual ontology. If the SUO
>is in fact just another O - or even the UO of an O - then say so, and I
>will stop whining about it, since we will then agree. But I would add, in
>that case do not pretend that it is somehow going to magically be all
>things to all ontologists, or solve problems of inter-ontology
>translation, because it won't. It will just be one 'standard' among many
>others.
>
>Pat
>
>---------------------------------------------------------------------
>IHMC (850)434 8903 home
>40 South Alcaniz St. (850)202 4416 office
>Pensacola, FL 32501 (850)202 4440 fax
>phayes@ai.uwf.edu http://www.coginst.uwf.edu/~phayes
-----------------
Adam Pease
Teknowledge
(650) 424-0500 x571