On Thu, May 8, 2014 at 4:17 AM, Ivan Herman <ivan@w3.org> wrote: > > > http://www.w3.org/TR/owl2-mapping-to-rdf/#Mapping_from_RDF_Graphs_to_the_Structural_Specification > > Maybe you meant: 'The mapping from RDF to OWL is explicitly only defined > for OWL-DL'? Which is correct but, in fact, mapping to OWL-Full is not > really useful because OWL-Full is defined in RDF only... > *Thanks: That is indeed what I meant to say. * OWL 2 is defined by the AS/FSS; a restricted subset of that language is defined as OWL 2 DL. The AS/FSS does not define the term OWL 2 Full to describe the whole language; (it is used in documents like the OWL 2 Conformance Tests http://www.w3.org/TR/2012/REC-owl2-conformance-20121211/) The forward mapping specifies how the abstract syntax should be translated to RDF, and as you note, is defined for the entire language, with the the OWL 2 RDF Semantics specifying the semantics for the generated RDF. The Direct Semantics is only defined for the DL part of the language. The RDF Semantics generally correspond to the Direct Semantics for DL, and are used The problem, as you note, is that it is not possible to recreate a non-DL OWL 2 Ontology that has been stored in RDF, which is useful for editing. It *is* possible to load such an ontology that is saved using the Functional Style Syntax, and, if entity types are declared, it is possible to load many non-DL ontologies. There are some places where it is not possible to round-trip, even *with* entity type declarations. Simon
Received on Thursday, 8 May 2014 19:12:10 UTC