Re: help wanted: RDF issue rdfms-assertion

Lynn Andrea Stein wrote:
> 
> Hmmmm.....I am not concerned with what lawyers think.  I am concerned with
> Tim's statement that the *meaning* of an RDF assertion that *I* make is
> necessarily controlled by the *owner*of*the*URI* for the predicate I am
> using.  (Note that this is in direct contrast to Graham's suggestion, below.)

I think bringing lawyers into the matter only muddies the waters. What
is important is as Tim said "we must define the core algorithms for
determining meaning without hesitation or ambiguity." When someone
publishes something on the Web, they should be able to determine exactly
what is implied by their statements, and others should be able to
determine this as well. The meaning of a predicate is determined by
axioms in some selected set of ontologies. Of course, how to select this
set is still a somewhat controversial matter. I have argued that
something like daml:imports is one approach to this problem (see the
"rdf inclusion" thread elsewhere on this list for disucssion).

> 
> Since Tim is proposing that this requirement should be part of the RDF Core
> spec, I think that it *is* an RDF Core issue.  I think that there *must* be
> some way for the meaning of a predicate (or an assertion) to be separated from
> the control of a single person or URI-owner.

Is your main concern that the URI-owner may change the axioms associated
with the predicate, and thus effectively change your meaning? If so,
this is definitely an important issue. Possible solutions could be
either social or technical in nature. A social solution might be to say
that web etiquette prevents you from changing an ontology file once you
publish it, instead you have to publish a revised ontology at a new URL
(this is suggested in RDF Schema). If an ontology owner doesn't follow
this policy, then the community can choose not to use their ontologies.
A possible technical solution might involve associating a date with
every ontology reference. If the ontology has changed since that date,
then applications can refuse to process it or raise an exception.

Jeff

Received on Friday, 31 May 2002 17:40:40 UTC