RE: On the integration of Topic Maps and RDF

I think that we will have to ``agree to disagree'' on whether mappings
between Topic Maps and RDF should preserve meaning. 

I strongly, no, passionately, believe that such mappings have to be
model-mappins that preserve meaning, at least if one is to hold the view
that RDF is a representation formalism.  If RDF is a representation
formalism, then positive ground binary relations have to be represented as
RDF triples.  Otherwise, RDF is just some syntactic encoding, and the
entire meaning is conveyed in some outside-of-RDF (and, probably, outside of
the web) side agreement.  

Any approach that requires an outside-of-RDF approach to ascribe meaning to
the resulting RDF has, in my opinion, lost everything.  Yes, an approach
that stays within RDF has the potential of losing some things, but at least
the portion that can be naturally represented in RDF is completely
captured.

Peter F. Patel-Schneider
Bell Labs Research


From: "Martin Lacher" <lacher@db.stanford.edu>
Subject: RE: On the integration of Topic Maps and RDF
Date: Tue, 21 Aug 2001 11:25:22 -0700

> Hi Peter,
> 
> Thank you very much for your comments. Your question concerns a very
> important point in the mapping.
> 
> > One important aspect of such mappings, for me, is whether information
> > expressed naturally in the source formalism (Topic Maps) and then
> > translated into the target formalism (RDF) can be naturally
> > integrated with
> > information expressed naturally in the target formalism.  If this is not
> > the case, then I claim that there is something wrong with the translation.
> 
> You are right, there are several ways to perform such a mapping. Graham
> Moore has nicely summarized the mapping approaches in his paper on RDF/Topic
> Maps. He called the approach you propose "mapping the model" and the
> approach we took "modeling the model".
> When mapping the model, the semantics of the primitives of one model have to
> be mapped to the semantics of the primitives in the other model. It is sort
> of an all-in-one approach, which attacks the problem in one piece (i.e.
> mapping all semantics of Topic Maps in one piece to all semantics that RDF
> offers). The downside of this approach is that one is likely to incur loss
> in the mapping, since the primitive constructs won't be close enough.
> When modeling the model, one model is expressed in the syntax of another.
> The advantage of this approach is, that loss is less likely to be incurred,
> since only the syntax primitives have to be mapped.
> As an example, mapping the model one has to map what an association conveys
> in Topic Maps to what a property conveys in RDF. That is virtually
> impossible, since associations in Topic Maps are much richer. When modeling
> the model in this case on only has to map what the graph model primitives in
> the Topic Map model convey to what the graph model primitives in RDF can
> convey.
> The unifying view on both of the approaches is that modelling the model is
> just again a mapping the model on another layer (see paper of Sergey Melnik
> on layered interoperability model). That means that our graph translation is
> again a mapping of the graph semantics of Topic Maps (which is expressed in
> XTM syntax) to the graph semantics of RDF ( any syntax). However, we believe
> that by performing the mapping between the primitives, we incur less loss.
> 
> > Suppose some facts about natural resources come from topic maps, and are
> > represented in this translation to RDF, and other facts about natural
> > resources come from a natural RDF representation.  How can one query the
> > RDF to find the union of the facts?  Even if it is possible to
> > write a such
> > a query is it at all possible to write such a query without knowing that
> > some of the natural resource facts come from topic maps?
> 
> No, the query processor will have to be aware of what source it is querying.
> But that is not a bad thing, since the specifics of the source can be
> expressed in a rule language. On top of that rule base one couled possibly
> imagine some universal query laguage. However, the question is, whether that
> is desirable at all - since one would lose all the specifics of the
> respective formats. For example, one could possibly not query for a unique
> property in DAML data anymore. Or one could not query for an association
> scope of a Topic Map association anymore. That probably depends on the
> application context.
> 
> So the definite advantages of our approach are:
> - only (n-1) mappings between formats required instead of n(n-1)/2 for an
> all-capable query engine
> - mapping between the higher-level constructs can be expressed declaratively
> - existing RDF infrastructure can be re-used for Topic Maps
> 
> Cheers,
> 
> Martin

Received on Tuesday, 21 August 2001 14:49:32 UTC