RDFS comments

Aside from the semantic and layering issues, I recently had the 
experience of seeing an ontology novice try to write two ontologies for 
a domain, on in RDFS and one in OWL (lite, it so happens). Some common 
good ontology practices (such as favoring local property restritictions 
over global ones) are impossible in RDFS due to the lack of requisite 
constructs (someValuesFrom, etc.), but one can kinda simulate them 
through a property inheritance tree (e.g., subclassing dc:title to make 
the fact that such and such a class "should" have a title). Is there 
really a point to having an ontology sublanguage that has such a weird 
expressiveness profile? As a core vocabulary/semantics?

Section 5.3 of the Primer goes further: it discusses the global nature 
of domain and range, but without acknowleging that 1) such globalness is 
dangerous and often unwise to exploit, 2) that OWL has local 
restrictions.

Indeed, in section 5.5 of the primer, local property restrictions are 
conspicuously absent.

(The RDF Schema doc replicates some of the primer material.)

(Lest I seem to be harping upon this...ok, I am. but I find this 
endorsement of a practice which is widely shunned in the ontology 
community to be worrisome.)

In general, fixing RDFS before OWL is done, particularly the 
semantically significant parts strikes me as unwise.

Relatedly, there's little mention and no explicit discussion of how 
completely alternative languages, with alternative semantics, or even 
just alternative semantics for the RDFS vocabulary, are to be dealt 
with. The Schema document reads:

""'This specification does not attempt to enumerate all the possible 
forms of vocabulary description that are useful for representing the 
meaning of RDF classes and properties. Instead, the RDF vocabulary 
description strategy is to acknowledge that there are many techniques 
through which the meaning of classes and properties can be described. 
Richer vocabulary or 'ontology' languages such as DAML+OIL, the W3C OWL 
languages, inference rule languages and other formalisms (for example 
temporal logics) will each contribute to our ability to capture 
meaningful generalizations about data in the Web. RDF vocabulary 
designers can create and deploy Semantic Web applications using the RDF 
vocabulary description language 1.0 facilities, while exploring richer 
vocabulary description languages that share this general approach."""

Does "share this general approach" mean "layered upon"? I'm unclear how 
the "contributions" of the various formalisms will be managed. In 
general, a clearer sense of how metalinguistic description is to be 
handled in RDF, especially using RDFS would be welcome. There are *many* 
traps for the unwary!

Finally, it seems that *all* of the RDF Schema document itself is 
informative:

"""This document is intended to provide a clear specification of the RDF 
vocabulary description language to those who find the formal semantics 
specification, RDF Semantics [RDF-SEMANTICS] daunting. Thus, this 
document duplicates material also specified in the RDF Semantics 
specification . Where there is disagreement between this document and 
the RDF Semantics specification, the RDF Semantics specification should 
be taken to be correct."""

If RDF Semantics always trumps, how is the "clear specification" 
normative? What if RDF Semantics is *silent* on some point? Does that 
constitute a disagreement? Is there extra information in the Schema 
document that's necessary for implementing RDFS?

Cheers,
Bijan Parsia.

Received on Saturday, 22 February 2003 00:01:15 UTC