Re: [OWL-S] arguments for PAI and PAC

At 12:56 PM 8/27/2003 -0700, David Martin wrote:



>Jeff Lansing wrote:
>
>>David Martin wrote:
>>
>>>...
>>>
>>>We don't normally model a car as the set of its parts, because we've come to realize that it works better to use some other representational mechanism for the "part-of" relationship.
>>>
>>>Similarly, we don't usually represent the class of, say, dogs, as the set of "life-traces" of individual dogs, or the class of humans as the set of "life-traces" of individual humans.  (Note that I'm not saying we should never do these things; just that we don't usually find it useful to do so.)
>>>
>>>I suppose the obvious rejoinder might be: in modeling processes we are centrally concerned with behavior across time, so it *does* make sense to think of a process as a class of executions (which we most often call "execution traces" in our discussions).  And that thought is in the spirit that initially motivated us to model processes in this way.  But, at least in my view, experience has taught us that it's more natural to think of a process, and an execution of the process, as two different (though obviously very closely related) kinds of things, and model the relationships in some other way than class membership.
>>>
>>>Again, I can't give any "proof" that processes-as-classes doesn't work.  I'm just making the pragmatic argument that it's too hard to work with.  We're building an ontology, and in my view, if an ontology doesn't reflect a natural way of thinking about a domain, it's not so likely to be used.
>>>...
>>
>>As I understand it, this is a several-thousand-year-old debate, so any appeal to naturalness here can only be regarded with suspicion. 
>
>Agreed.  But it seems to me there are very few examples of folks modeling a process as the set of its executions (certainly very few examples that have widespread use).  This may just be a lack of awareness on my part - if you know of good examples (widespread or not), please let me know.
>
>>A lot of people don't find abstract Forms to be so "natural", but that's essentially what you're stuck with, when you separate processes from sets of executions of processes.
>
>I'm not sure what you mean by abstract Forms; could you elaborate a bit?

As far as examples go, the Cyc knowledge base represents concepts, including processes, events and actions as the collection of objects that instantiate that concept, i.e., for all intents and purposes as the set of it "executions".

Regarding the Platonic forms question I hope you don't mind if I throw in my two cents worth.  An obvious question concerns the relationship between a process trace, x, as you call it, and the concept of X in virtue of which x is a "trace" of X.    Presumably both of them are processes in the PAI model.   What distinguishes one from the other?   To my mind, it is entirely suggestive of Platonic forms, i.e., the prototypical process seems to be reified as some "really real" process to which all traces aspire and in virtue of which each x receives its status as a trace of X.    Under the PAC model, the relationship between a "trace" and X is simply one of set membership or class instantiation.   That's the point I was trying to make, i.e., I think this is much more intuitive/plausible than the implicit appeal to Platonic forms and the accompanying supernatural metaphysics.  Ultimately, I agree that simple appeals to "intuitiveness" or "naturalness" must be regarded with suspicion, but you did introduce them in your set of arguments for PAI.   

best,

Mike Pool

>Thanks, David
>

Received on Thursday, 28 August 2003 10:38:35 UTC