Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: SUO: Negotiation Instead of Legislation




At 2:31 AM -0500 2/21/02, John F. Sowa wrote:
>Chris,
>
>There is another very important point, which we both
>understand, but which we are looking at from different
>points of view:  distinctions are very important.

Right.

>What I maintain:  the distinctions are even important than
>the categories and the hierarchies.  Instead of handing anyone
>a prepackaged ontology with a hierarchy of distinctions, I
>would prefer to give them a system that can recreate a new
>ontlogy on the fly from whatever selection of distinctions
>are important for their problem.

Yes, I absolutely agree

>
>>... I can't help but observe that your architecture paper
>>mentions an upper level as a component!
>
>Yes, I have to admit that I have been wavering on this issue.
>But I am coming to the conclusion that you don't need a single
>fixed upper level -- you can create whatever upper level you
>need from the choice of distinctions that are important to
>your problem.

I agree with this as well.  But I don't see a problem with selecting 
some small number of frequently used combinations and putting them 
out there for people to use.  Some people find the distinctions a bit 
tough to get around, but upper level categories are usually intuitive 
to most people.

If Hendler&Berners-Lee get their wish, there will be a lot of people 
out there building ontologies, and most of them won't know a 
distinction from a dipswitch.  But there's a chance they'll know a 
process from a living being.


-- 

Christopher A. Welty        http://www.cs.vassar.edu/faculty/welty/
Vassar College Computer Science Dept.         Voice: (845) 437-5992
Poughkeepsie, NY 12604-0462                     Fax: (845) 437-7498