Re: Proposals for changes to current datatyping spec

On Wed, 4 Sep 2002 Patrick.Stickler@nokia.com wrote:

>
>
> I would like to propose the following changes/clarifications be made
> to the datatyping spec (the latest restructured document) based on
> recent discussions:
>
> 1. Replace the usage of rdf:type for typed literals to rdf:datatype
>    (having the pairing rdf:datatype/rdfs:Datatype mirror the current
>     pairing rdf:resource/rdfs:Resource for syntactic attributes
>     taking URIref values and which identify members of a specific
>     class).

Yep.

> 2. Move section 1.2 Desiderada to a non-normative appendix.
>
> 3. Streamline introductory material, with references to the other
>    specs as appropriate.

Yep; although it's odd to move the motivational material to the end.
Editorial stuff like this doesn't need WG approval.

> 4. Clarify the nature of literals in section 1.5 and the nature
>    of typed literals in section 2.3 so that it is clear that
>    all that is relevant to datatyping is the unicode string
>    component, no matter how many other components may exist
>    within a literal's structure.

I think this is an error; I've already put my position on this (ie, that
a string and a langstring would appear to be different types)

> 5. Clarify the relation of rdfs:Literal, rdfs:Datatype, and
>    typed literal (begs the question of whether rdfs:TypedLiteral
>    is warranted, as a subclass of rdfs:Literal...)

This does need clarification, yes. But hang on... you suggest that

     <eg:foo> <eg:age> (xsd:integer)"10".

might (with literals "at the blunt end") imply...

     (xsd:integer)"10" <rdf:type> <xsd:integer> .

But if "TypedLiteral" is the class of typed literals, then whenever

     <xsd:someType> <rdf:type> <rdf:Datatype> .

then we have that members of <xsd:sometype> are members of
rdfs:TypedLiteral, and hence members of rdfs:Literal. Which would kind
of imply (since subclassing is a relation of class extensions) that

     <xsd:integer> <rdf:type> <rdf:Datatype> .

implies

     <xsd:integer> <rdfs:subClassOf> <rdf:Literal> .

... which is why I suggested Datatype is a superfluous extension
(particularly since it seems that all literals should carry a type, but
I don't wish to descend into "dogmatic assertions" here :-) )

> 6. Clarify the verbage in section 3.1, Global Datatyping Assertions,
>    so that it is clear that RDF does not assert any constraints,
>    but only that an application is free to do so (reference to
>    other specs as appropriate which discusss this nature of
>    rdfs:range in general, Primer?)

Yes, again I think the conversation that has arisen about of the
constraining nature or otherwise of rdf:domain and range is probably
going to happen with datatypes.

> 7. Rework the presentation of Part 2 as "Suggestions for Future Work"
>    (or split it off into an entirely different document?)

Yes - again, we need a bag for "here's stuff that we'd like to do".

> 8. Expand example in 3.1 to include triple with 'age' arc, as
>    suggested.
>
> 9. Clarify agnostic position of RDF regarding datatype clashes,
>    as for any contradiction, in section 3.2

Seems fine.

> Is the above OK with the WG?

That's my tuppence :-)


-- 
jan grant, ILRT, University of Bristol. http://www.ilrt.bris.ac.uk/
Tel +44(0)117 9287088 Fax +44 (0)117 9287112 http://ioctl.org/jan/
"NOP" is a trivial implementation of an executable Z subset.

Received on Wednesday, 4 September 2002 06:18:22 UTC