[topicmapmail] Making ontologies : RDF vs TM

[Dan Corwin]

Vincent Godefroy wrote:

> Is there any documents, white papers on making
> ontologies with xtm ? Google replies me with a
> lot of stuff with RDF/RDFS and OWL.

I am working on an ontology that will formalize Conceptual
Graphs by using XTM notation and PSIs.  Since CGs are based
on an upper ontology from John Sowa, CTM is very general:

   http://www.lexikos.com/psi/ctm/index.htm

Version one does a good job on expected role types, but it still
lacks association subtypes for situations - states, relationships,
events, processes, settings and boolean subgraphs.

The tricky part is in specifying role type patterns for all *their*
common subtypes, of which several hundred may be readily listed.

> How can we see TM in the web semantic effort ?

I believe that any TM built carefully under an upper ontology
such as CTM can be mapped automatically to any other metadata
notation desired - (e.g.) to RDF, as Lars outlines here:

   http://www.ontopia.net/topicmaps/materials/tmrdf.html#N1131

It is not yet proved that this works in the general case, but any
decent TM ontology and its embedded constraints *should* provide
enough extra data to resolve all of the troublesome ambiguities
that would arise if that ontology were missing.

Cheers,
Dan Corwin