Talk:Mapping to RDF Graphs

From OWL
Revision as of 10:29, 7 November 2007 by Peter Patel-Schneider (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Suggestion: replace center alignment with left alignment in tables RinkeHoekstra 01:19, 7 November 2007 (EST)

Moved to Document Production RinkeHoekstra 04:30, 7 November 2007 (EST)


Transformation to Triples

Table 2 translates only the URI's of entities such as Datatype, OWLClass, ObjectProperty etc. but not the entities themselves; e.g.
owlClassURI
is mapped onto
owlClassURI rdf:type owl:Class

However, the Syntax document does introduce these elements as part of the syntax.

Is it not more appropriate to map for instance
OWLClass ( owlClassURI )
onto
owlClassURI rdf:type owl:Class

I am aware this is not a small change, as the Syntax document will need to be revised, and perhaps it is not necessary. Perhaps, even, it would be enough to include a mapping from a.o. OWLClass statements to rdf:type owl:Class statements in Table 2

Nonetheless, it is not very 'clean' if the Functional Syntax way of stating that some element is an OWLClass differs from the way in which the RDF Mapping does the same.

RESPONSE:

In many places there is no progression through entities. Instead, for example, there is

description := owlClassURI | ... 

so it is necessary to have the mappings that are there.


That said, it does appear that there is a bug in the translation, as there is no translation for, e.g., OWLCLASS(x), which does need a mapping because the mapping is used in annotations. Strangely enough, declarations are treated correctly. I believe that the translation should be extended to have all these entities map as per their enclosed URI.

Peter Patel-Schneider 05:24, 7 November 2007 (EST)