Re: issues to be resolved before last call

From: Lynn Andrea Stein <las@olin.edu>
Subject: Re: issues to be resolved before last call
Date: Fri, 17 Jan 2003 09:07:03 -0500

> Thank you, Peter, for clarifying something that I have long known but 
> not been able to really wrap my head around:
> 
> >
> > Issues with RDF Concepts:
> >
> > 1/ The notion of social meaning has no place in the specification of a
> >    formal system.
> >
> >    PROPOSED RESOLUTION:  Social meaning, as defined in the RDF Concepts
> >    document, has no effect whatsoever on the meaning of OWL ontologies.
> >
> >    NB:  I view this as an extremely serious issue.
> >
> >
> 
> I believe that this is precisely the distinction between a formal 
> system and a social system.  In your proposed resolution, the word 
> "meaning" is used twice with two different senses.  By "the meaning of 
> OWL ontologies", I take it you intend something like denotational 
> semantics (though I don't mean to rule out an axiomatic encoding; I 
> just intend "what the term refers to").  But by "social meaning..." I 
> take RDF to intend something I'd call "effective semantics", i.e., 
> "what work the term can do in the world".  So, for example, an ontology 
> may formally mean one thing but the courts may (in practice, perhaps 
> even incorrectly) use it as the basis for making a distinct legal 
> ruling.  The legal ruling may be at odds with the "meaning" in your 
> sense of the ontology, but it then becomes part of the "effective 
> semantics" or "social meaning" of the ontology.
> 
> If, as I suspect may happen, the WebOnt WG goes with the formal notion 
> of "meaning", I think that it is important to clarify that this is the 
> kind of meaning we're talking about and that actual (if incorrect) 
> usage in the world is outside the scope of the formal specification.  
> (In fact, the formal specification is giving meaning to the notion of 
> "correct usage".  But "correct" is sometimes different from "effective" 
> in the philosophical and pragmatic senses.)
> 
> This isn't necessarily the approach I'd take, but I think it's likely 
> to be the pragmatic solution to where we are.  I appreciate Peter's 
> phrasing it succinctly and, at least for me, really highlighting the 
> issue.  I'm fine with the proposed resolution *provided* the 
> distinction between meaning-in-Peter's-sense (and of course with a long 
> and glorious history!) and effectiveness/pragmatic utility is made.
> 
> Lynn

I would find it acceptable for the OWL documents to state that there are
social (or whatever they are called) meanings that are outside the purview
of the OWL specification, provided that it is also stated that there can be
multiple such meanings and that a single such meaning is almost invariably
impossible to achieve.  

I would prefer it if the OWL documents could be silent on this issue, and
let usage, convention, etc., determine social meaning, as they have always
done.  However, the RDF Concepts document normatively states that the
entire social meaning of an RDF document is a part of the RDF meaning of
that document.  Therefore I believe that the OWL documents must explicitly
disavow this view.

peter

Received on Friday, 17 January 2003 09:16:03 UTC