Re: RDF datatyping

On 2002-01-11 13:07, "ext Jan Grant" <Jan.Grant@bristol.ac.uk> wrote:


>> If you mean that members of the value space should have explicit
>> representation in the graph, then I disagree.
> 
> That's what I mean, yeah.
> 
>> To achieve representation
>> of actual values, you must have native support for all data types
>> for all statements, which is unreasonable and IMO unfeasible.
> 
> No. I've got no problem with my conceptual RDF graphs having any and all
> datatypes; 

Nor do I, presuming their representation is system/platform/implementation
neutral.

> I'm flexible enough to cope with that. When it comes to (RDF
> implementation X), doubtless such a conceptual ideal will be realised
> (or at least, approximated) _behind_the_scenes_ by some kind of what is,
> effectively, a lexical representation. I don't really care about that,
> though, because my conceptual view of the RDF graph (and the view I'd
> like any APIs to fake for me) is that the values are really there in the
> graph.

Hence my description of an API that both deals with the variant idioms
and provides an optimized interface to actual values, insofar as it
knows about the idioms and datatypes in question and is able to provide
an internalized representation of values for a given data type.

Sure. But we're talking about the RDF graph here, not an API.

>> There can be no native data types in RDF, only a consistent means
>> for declaring the data type of a lexical form.
> 
> Actually, the first half of this doesn't have to be true.

I don't see how it can't be true, if we are to maintain a generic,
portable, implementation-independent means of knowledge interchange.

Patrick

--
               
Patrick Stickler              Phone: +358 50 483 9453
Senior Research Scientist     Fax:   +358 7180 35409
Nokia Research Center         Email: patrick.stickler@nokia.com

Received on Friday, 11 January 2002 07:16:42 UTC