RDF Meta Model and "primitive properties"

What was the reason for instantiating the "primitive" properties
"rdfs:subClassOf, "rdf:type", "rdfs:domain" and "rdfs:range" from the
concept "rdf:property"?

These "primitive" concepts are used both as concepts to be defined
within the meta schema and as meta concepts to define these very
concepts. In the RDF Schema Specification
(http://www.w3.org/TR/2000/CR-rdf-schema-20000327/) Figure 2 (Class
Hierarchy for the RDF Schema), which includes subclass and type as
concepts/nodes and as meta concepts/arcs, or Figure 3 (Constraints in
the RDF Schema), which includes domain and range as concepts/nodes and
as meta concepts/arcs, show this duality very well.

Wouldn't it make more sense to include them as "primitive properties",
which are not instantiated from "rdf:property"? In this way, the RDF
Schema Specification would in my opinion be easier to read and avoid
this self referentiality.

We have written a paper about this issue (see
http://www.kbs.uni-hannover.de/Arbeiten/Publikationen/2000/modeling2000/wolpers.pdf),
which talks about these issues in some detail (published in a German
Workshop on Modelling and Meta Modelling Issues
(http://www.uni-koblenz.de/iwi/mod2000). 

This paper includes some other ideas, especially the idea of
introducing "annotation objects" (instances of RDFAnnotationClass in
that paper), which annotate WWW pages, instead of treating WWW pages
as instances of RDF Classes, but that might be a different issue
(possibly just a restricted use of using RDF, which is however quite
useful in most cases).

Regards,

Wolfgang Nejdl

Prof. Dr. Wolfgang Nejdl              tel. +49 511 762-19710
Institut für Technische Informatik    fax. +49 511 762-19712
Rechnergestützte Wissensverarbeitung  http://www.kbs.uni-hannover.de/
und Educational Technology Lab        http://www.etl.uni-hannover.de/
Universität Hannover, Appelstraße 4, 30167 Hannover, Deutschland

Received on Tuesday, 6 June 2000 15:05:39 UTC