Re: SUO: Re: REQUEST: survey of available ontologies, taxonomies, thesauri, lexicons?
On 4/1/02 15:09, "Jim Hendler" <hendler@cs.umd.edu> wrote:
>> I agree with the following comments by Bill Andersen:
>>
>> BA> So, how well has the schema done for us? Not well. And it CAN'T do
>>> better -- its syntax and semantics don't have enough power. Of
>>> course one could try to encode all of this somehow in some arcane
>>> syntax that someone is going to have to interpret as doing what logics
>>> already do. The RDF and RDF-Schema efforts are just such encodings.
>
> Nonsense (who taught this guy Andersen's AI classes??? :->)!
Aww... I didn't think you remembered! :-D
> When
> you write classical logic with all the usual symbols it is
> meaningless scrawling on a piece of paper until we have the social
> agreement about how the symbols map to mathematical concepts.
> Similarly
> <ImplicationRule>
> <Antecedent> x </Antecedent>
> <Consequent> y </consequent>
> </ImplicationRule>
> is not something to "interpret what logics already do" but a means of
> encoding a particular bit of logic, assuming a social agreement as to
> what the meaning of this RDF is with respect to some mathematical
> model of entailment.
But, Jim, I was talking in particular about RDF-schema and XML-schema, which
include weak constraint languages in them. One can treat them as logics and
give them a model-theoretic semantics, as Pat has done.
In addition, if the above serves to "encode some particular bit of logic" it
seems reasonable to interpret it *as* an alternate syntax for some logic,
either some standard logic like FOL, or some new logic being proposed by the
proposer of such structures.
> The confusion is that the term "logic" on the layercake diagram
> doesn't mean "there exists a logic" it means "there is a formal
> logic, expressible on the web, embedded properly in web architecture
> (i.e. URIs and the like), and able to passed between web entities via
> proper protocols. These things are different than the need for a
> logic per se -- those are a dime a dozen and easily found in about
> 2000 years of literature -- heck your own books describe numerous of
> them -- just none that could be called a "web logic" at this point.
I don't know about John's point of view, but whenever you propose a language
like RDF-schema or XML-schema that includes non-logical symbols as those
languages do, those symbols ought to be *interpreted*. If the
interpretation is supposed to come later in terms of mapping the defining
axioms for these symbols to some web-expressible logic, then that seems to
be putting the cart before the horse.
Why not say up front exactly what these languages are supposed to do, either
by defining a new logic (nothing wrong with that) or by mapping it to a
well-understood existing logic like FOL?
Anyway, all of this discussion is good and healthy. One of the things I've
wanted to see for a long time is for the SUO and philosophical ontology
folks who care about such things, to start paying attention and get involved
with the Web-ontology crowd. This seems like a good start.
.bill