Re: The meaning of a document

From: "Jeen Broekstra" <jbroeks@cs.vu.nl>

> On Sun, 18 Nov 2001, Seth Russell wrote:
>
> > Ok, I'd like read into the record here the second item too:
> >
> > 2. The meaning of an RDF statement is defined by the
> >   preciate used, and so is specified by the specification
> >   of the Property that is used as predicate.
> >
> > Hmmm .... if we take both (1) and (2) together it seems that we could
> > conclude that:
> >
> >  The logical meaning of an RDF domument is the
> >  conjunction of  the statements in the document
> >  and the statements in all the schema documents
> >  referred to by the Property terms in that document
> >  and all the statements in other documents
> >  referred  to by those schema documents.
> >
> > In other words {:ChocolateLover dmal:complementOf  :NotCholateLover} is
> > meaningless unless we AND it to:
> >
> > <rdf:Property rdf:ID="complementOf">
> >   <rdfs:label>complementOf</rdfs:label>
> >   <rdfs:comment>
> >     for complementOf(X, Y) read: X is the complement of Y; if something
is
> > in Y,
> >     then it's not in X, and vice versa.
> >     cf OIL NOT
> >   </rdfs:comment>
> >   <rdfs:domain rdf:resource="#Class"/>
> >   <rdfs:range rdf:resource="#Class"/>
> > </rdf:Property>
> >
> > and then also code the rdfs:comment in some kind of a RDF
> > document too.  Or are we to assume that our programs can
> > respond to the meaning of English sentences?
>
> I think you are overlooking the fact that these programs are
> typically "native speakers" of DAML+OIL. We don't have to
> tell them what complementOf means, they already know. These
> are the language primitives, and they are called primitive
> for this very reason.
>
> For the program, the "meaning" is in the logic, and every
> DAML+OIL primitive corresponds to a logical formula. The
> English sentence you are referring to is merely a comment,
> put there to make it more human-understandable as well
> (humans are notoriously bad at logic).

I think were talking past each other here.  For any program to be a "native
speaker of DMAL+OIL" such that it behaves differentially to a message
containing "complementOf", it must have that rule coded somewhere in its
logic statements.  Putting the rule inside the program (inside the logic of
the prorgam) seems to have some special meaning to you that I have
overlooked ... but I don't get it.

My point is: with rule ..... differential consequences , without rule .....
no differential consequences. No differential consequences, no meaning to
the program.   It doesn't matter whether the program got the rule from the
schema (my preference) or from some human coding the program.

What is the real disagreement here ?  I don't see it.

Seth Russell

Received on Monday, 19 November 2001 11:44:06 UTC