Re: User-defined Datatypes: owl:DataRange vs rdfs:Datatype

The reference you give, among other things, says: "rdfs:Datatype is  
both an instance of and a subclass of rdfs:Class. Each instance of  
rdfs:Datatype is a subclass of rdfs:Literal."

So the build in datatypes such as xsd:int, by this definition, would  
be a subclass of rdfs:Literal. However, according to [1] (OWL 1.0)

Definition: An OWL vocabulary V consists of a set of literals VL and  
seven sets of URI references, VC, VD, VI, VDP, VIP, VAP, and VO. In  
any vocabulary VC and VD are disjoint and VDP, VIP, VAP, and VOP are  
pairwise disjoint. VC, the class names of a vocabulary, contains  
owl:Thing and owl:Nothing. VD, the datatype names of a vocabulary,  
contains the URI references for the built-in OWL datatypes and  
rdfs:Literal.

If I understand this correctly,  it says that VC (the class names of  
a vocabulary) and VD(the datatype names of a vocabulary) are  
disjoint. That is, a datatype is not a class. Only classes have  
instances. So even for the built-in xsd:int, the statement from RDFS  
"xsd:int, is a subclass of rdfs:Literal" can't hold. And so user  
defined datatypes are no different than the builtins.

Does that look the appropriate answer to your question?

-Alan

[1] http://www.w3.org/TR/owl-semantics/direct.html#3.1

On Dec 6, 2006, at 7:48 PM, Holger Knublauch wrote:

>
> Thanks, Jeff.  But in RDF Schema, the pre-defined system datatypes  
> such as xsd:int are instances of rdfs:Datatype as well [1].  Are  
> user-defined datatypes different from system datatypes, or is OWL  
> 1.1 changing the RDF Schema semantics here?  Apologies if this has  
> been discussed and written down elsewhere - any pointers are  
> appreciated.
>
> Regards,
> Holger
>
> [1] http://www.w3.org/TR/rdf-schema/#ch_datatype
>
>

Received on Thursday, 7 December 2006 03:12:42 UTC