From: Ian Horrocks (horrocks@cs.man.ac.uk)
Date: 07/13/02
On July 12, Lassila Ora (NRC/Boston) writes: > Ian, > > Ian Horrocks <horrocks@cs.man.ac.uk> wrote: > > type(x,C) > > P(x,y) > > subClassOf(v,w) > > > > where C is an arbitrarily complex DAML+OIL class, P is a property that > > is NOT used in RDF, RDFS or DAML+OIL syntax, > > I do not understand this. What do you mean by "RDF, RDFS or DAML+OIL > _syntax_"? Do you mean any property defined in the rdf:, rdfs: or daml: > namespaces? If so, this would exclude useful queries... I meant to exclude query elements like onProperty(x,y), unionOf(x,y), which have doubtful/uninteresting semantics from a DAML+OIL point of view, and could be seen more as "syntax queries", i.e., queries regarding the syntactic content of the ontology. I agree that excluding all of the properties in daml+oil.daml would exclude useful queries (e.g., sameClassAs(v,w), sameIndividualAs(x,y)). The semantics for these seem pretty clear, e.g., sameClassAs(C,v) would bind v to all class names in the query ontology(s) that can be determined to have the same extension as C. What I was trying to suggest was that some of the properties (and maybe some of the classes) in daml+oil.daml would not lead to such a clear semantic account of query element bindings. E.g., for unionOf(C,w), where C is some class, what would w bind to? If queries are to have a well defined semantics, then we need to decide how to handle these cases. My suggestion was that we simply forbid them. > > Anther kind of query one could imagine are syntax queries, i.e., > > retrieve stuff that is written in the KB. I suppose we could use DQL > > to do this. At this level, the KB could be considered as RDF/RDFS, and > > queries could (in principal) be answered by one of the existing > > "triple engines". It doesn't seem to make much sense to mix this kind > > of query with the DAML+OIL semantically based queries discussed above. > > I agree, particularly remembering that even in RDF we can imagine a > "semantic query" (i.e., one that respects the "inferential" aspects of > RDF/RDFS). Who cares about triples anyway...? :-) There are some subtle problems here. E.g., if RDFS is used to generate a subProperty P of sameClassAs, how is P treated in DAML+OIL queries? According to the current (model theoretic) semantics, if I write P(C,D), then a DAML+OIL reasoner would NOT infer sameClassAs(C,D). On the other hand, if I used an RDFS engine to ask if the ontology syntactically implies sameClassAs(C,D), then the answer would be yes! This aspect of the layering problem is still being debated by the OWL WG, and it isn't clear what the result will be. Ian > > - Ora > > -- > Ora Lassila mailto:ora.lassila@nokia.com http://www.lassila.org/ > Research Fellow, Nokia Research Center > Chief Scientist, Nokia Venture Partners >
This archive was generated by hypermail 2.1.4 : 07/13/02 EDT