[SCL] The semantics of "semantics"

[Chris Menzel]

Jim, 

Thanks very much for an extremely helpful and informative (not to
mention courteous :-) reply.  Just a few comments.

On Thu, Jan 16, 2003 at 06:33:06PM -0500, Jim Hendler wrote:
>  one of the problems is that when one says "use RDF for ?x" we can 
> actually mean a lot of different things.  Here a couple of things we 
> might need different terminological terms to use amongst ourselves to 
> distinguish:
> ...
> 1) RDF is a document language that has a meaning of its own.  Thus, if
> I say http://www.cs.umd.edu/users/hendler rdf:about "james Hendler" it
> carries some sort of assertional meaning (and the model theory for
> that meaning is a document Pat has edited [1])

Yes, I know Pat's RDF MT document well -- the semantics in fact features
some elements of SCL's MT that Pat very effectively brought to RDF.  (I
think, in addition, that the document is a model of both expository
clarity and effective utilization of HTML for composing technical
papers.)   I was basing my claims about RDF's expressive power -- and
its inadequacy to serve as a foundation for "semantic translatability",
in one sense at least, for SCL languages -- on this document.

> 2) One can define langauges with their own more specific semantics 
> that sit on top of RDF.  For example, OWL, the product of my Working 
> Group, has its own model theory [2] which give special semantics to 
> certain terms such as owl:class or owl:datatypeProperty.

An SCL-conformant language could certainly be defined in this way to
"sit on top of" RDF.
 
> 3) One can use the RDF or OWL languages and their own MTs to develop 
> languages that use, rather than extend, this semantics.  
> ...
>
> I could imagine all or some of SCL being described with relation to 
> any or all of these uses of RDF.  

Yes, though in the third case it is surely only *some* of SCL that can
be described using the MT of RDF or OWL (though I am at the moment less
clear about the extent to which that is true for OWL; I will be studying
its semantics over the next couple of weeks to get a better handle on
this).  It is surely not possible, e.g., to use RDF to define an SCL
language with classical negation and full first-order quantification --
not that you didn't know that.  The exact extent of RDF's expressiveness
relative to SCL will of course be important to investigate.  The SCL
procedure for doing this would be simply to define a fully
SCL-conformant language that extends RDF, thereby establishing its
so-called "indirect" SCL conformance.  (Recall that SCL is not itself a
language, but a framework that specifies an entire class of languages;
roughly speaking, SCL is to a particular SCL-conformant language like
KIF what Boolean Algebra is to a specific algebra like the powerset
algebra over the set of natural numbers.)

> Note also that RDF is quite different from XML in all of the above -
> it has no MT and thus one first has to invent one for one's own domain
> to make it work.  This has some advantages (i.e. SCL could be defined
> from scratch without having to sit on top of another MT) but also some
> disadvantages, particularly with respect to the third use above, as
> creating an XML "langauge" only allows for syntactic verification, and
> allows no automated semantic checks with respect to a "model" being
> defined, and also requires a locality of reference that RDF doesn't.

I didn't quite follow this discussion, but will revisit it after further
study.

Regards,

-chris

--

 /\ ASCII ribbon | Chris Menzel -- http://philebus.tamu.edu/~cmenzel
 \/   campaign   | Philosophy Dept, Texas A&M University
 /\   against    | College Station, TX 77843-4237
/  \ HTML email  | voice: 979.845-5660  fax: 979.845.0458