On Thu, 1 May 2008, Bill Andersen wrote:
> On Apr 30, 2008, at 18:08 , Pat Hayes wrote:
>> ...
>> The term 'ontology' has no definition precise
>> enough to answer that question. Myself, I'd be
>> inclined to say yes, anything in OWL is an
>> ontology. Certainly one would not expect any OWL
>> tool or engine to start distinguishing between
>> 'real ontology OWL' and 'mere OWL'.
>
> Correct.
>
> I admit the problems with coming up with anything close to a
> definition of the term "ontology" and of course no machine is going to
> be able to tell the difference. But your position above equates (the
> referents of) "ontology" with "theory expressed in a formal
> language". Or at least it makes them coextensional if we want to put
> that fine a point on it.
>
> This is at least a very permissive version of Quine's position in "On
> what there is". I don't think Quine would have bestowed the honor of
> existence on a lot of the referents of terms introduced in most of
> what are called "ontologies", written in OWL or not -- he reserved
> that for objects revealed by best practice in science. Thus his
> epistemic motivation ought to appeal to you. But certainly the
> following would not count for Quine and hopefully not for you:
>
> <a rdfs:subclassOf b>
>
> That's a legal OWL theory (I won't call it an "ontology" -- you can do
> that). The (non-reserved) terms are arbitrarily chosen, without even
> the intent that they convey any information about a world outside the
> computer.
>
> We want to admit that people screw up and say things like (exists (x)
> (= x Phlogiston)) where they do think the terms refer to the world,
> but that's at least a better story than admitting any logical theory
> whatever.
>
> And of course real philosophers can argue the bit about realism and
> whether or not we ever will have any idea whether we have access to a
> "real" world. (01)
Bill, (02)
Your view seems to me to conflate the question of what an ontology *is*
with the issue of what makes an ontology *good* or *useful*. John Sowa
has argued pretty persuasively that you at least have to have a name or
predicate in an ontology, but I'm still inclined to think, similar to
Pat, that the best *definition* of a (formal) ontology, relative to some
formal framework, is simply "set of sentences". (I like to follow
fairly standard practice and reserve "theory" for deductively closed
sets of sentences.) (03)
-chris (04)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (05)
|