Re: incorporating datatypes into the MT

[Patrick Stickler, Nokia/Finland, (+358 40) 801 9690, patrick.stickler@nokia.com]

> > In other words, in the document I am now working 
> > on, there will be no such thing as a non-datatyped interpretation: 
> > datatyping will be built into the very foundation of the language. 
> > RDF will *include* datatyping.
> > 
> > Recent messages from Dan C and Jeremy and Jos, however, have made me 
> > realize that some of us apparently expect the MT to be structured 
> > rather like it has been in the past, in that there would be a simple 
> > basic RDF notion of interpretation which had no such built-in stuff, 
> > and datatyping would be one of the later additions.
> > 
> > So my question is, will incorporating datatyping into the basic RDF 
> > MT cause anyone grief?

I think that it is necessary and reasonable to incorporate
datatyping into the basic RDF MT insofar as typed literals
are part of the core abstract syntax, and that there are
certain entailments involving typed literals that hold for 
the core RDF MT without specific knowledge of the datatypes.
E.g. the following "basic RDF" entailment holds:

   some:Thing some:Property "10"@en^^xsd:integer .

RDF entails

   some:Thing some:Property "10"@fi^^xsd:integer .

No knowledge of some:Datatype is required to define that the
above entailment holds in the basic RDF MT. It is part of the
generic characteristics of typed literals.

Yet still there needs to be defined the notions of RDF entailment
versus RDF+Datatype entailment such that there are additional
entailments that will hold with specific knowledge of the
datatypes that will not hold without knowledge of the datatypes.
Thus, the following RDF+Datatype entailment holds where it
wouldn't be a valid RDF entailment:

   some:Thing some:Property "10"^^xsd:integer .

RDF+Datatype entails

   some:Thing some:Property "010"^^xsd:integer .

etc.

So typed literals should be described in the basic RDF MT
insofar as their generic properties are concerned and insofar
as they participate in valid RDF entailments such as the
first example. And then RDF+Datatype entailments can be
defined as an extension of that basic MT.

Eh?

Patrick

Received on Thursday, 31 October 2002 02:51:41 UTC