Leigh Dodds wrote:

>
> Jonathan Borden wrote:
>> In response to the TAG request I've updated http://www.rddl.org/ 
>> natures to deprecate the old nature URIs and suggest new URIs.  
>> This is all of the form:
>> For XML Schema (the archetypical example):
>> (old nature) http://www.w3.org/2001/XMLSchema
>> (new nature) http://www.rddl.org/natures#XMLSchema
>
> The nice feature (IMO) that is lost here is "closure" over the  
> resource natures.
>
> E.g that (in theory at least) a software agent could discover  
> additional information about a nature by dereferencing the nature  
> URI. E.g. a browser or other agent could use this mechanism to find  
> a suitable list
> of plugins or services capable of using that resource.

Strong point.

Perhaps it would be easier to forget the idea that the rddl:nature  
directly identifies the class of the related resource. Alternatively  
the class of the related resource could be discovered by looking at  
the RDDL document e.g.

Somewhere in http://www.w3.org/2001/XMLSchema it would indicate that
	"documents which conform to XML Schema validate according to ..."

I could change rddl2rdf.xsl to replace "rdf:type" with "rddl:nature"  
making RDDL nature a regular 'ole property of the related resource.

I am going to compose a note to www-tag discussing this issue.

Based upon feedback so far (you, Elliotte) I am thinking that perhaps  
we ought keep RDDL natures the way they have been.

Jonathan