Re: User defined datatypes & XML Schema WG

On Nov 4, 2007, at 10:24 AM, Bijan Parsia wrote:

> This is discharged what I thought was an action item, but may have  
> just been a request :)

This is writtened befored coffeed.

Sheesh!

Trivial issues:
===========

http://www.w3.org/2007/OWL/wiki/Syntax#Data_Ranges

* Syntax doc doesn't list the built-in datatypes (carried over from  
OWL 1.0). It does have a pointer to the semantics document which  
lists them:
	"""The list of the datatypes supported in OWL 1.1 is given in [OWL  
1.1 Semantics]; furthermore, this list can be extended by  
implementations as needed."""

	I'll survey existing support for these (at least in Pellet & FaCT+ 
+;)) Boris could speak to Kaon2's support.

* Syntax doc uses "datatypeUri" to refer to data predictes of  
arbitrary arity. I found this a bit confusing (as did Alan).

* I wonder if it's worth having dataOneOf for predicates of arbitrary  
arity or would it just be borrowing trouble :)

Bigger (datatype) issues:
===================

* URIs for externally defined datatypes in XML Schema documents
* Syntax for inline Datatype restrictions in RDF mapping and XML syntax.

Currently, we have:

DatatypeRestriction(dr facet ct):

	_:x rdf:type owl:DataRange
	_:x owl11:onDataRange T(dr)
	_:x owl11:facet ct

I.e.,
DatatypeRestriction(xsd:integer minInclusive 5):

	_:x rdf:type owl:DataRange.
	_:x owl11:onDataRange xsd:integer.
	_:x owl11:facet "5".

For a datatype (>= 5)

We could reuse XML Schema syntax:

	_:x rdf:type owl:DataRange.
	_:x owl11:restrictedBy """<restriction base='integer'>
     <minInclusive value='100'/>
   </restriction>"""^^rdf:XMLLiteral.

Pros: reuses XML schema syntax.
Cons: Bit more annoying for toolkits.

So, this probably needs review from the XML Schema WG, but it'd would  
be good to know what our WG thinks.

Cheers,
Bijan.