Re: Conformance model of RDF specs?

Hi Dom,

I suspect that you are writing a document from the point of view that 
you are specifying the behaviour of software.  The RDF specs are written 
more from the point of view that they define the syntax and meaning of 
RDF documents.  This would explain why you might not find some of the 
things you expect.

Some more detailed answers below.

Brian

Dominique Hazaël-Massieux wrote:
> Hi,
> 
> I understand it is too late for making Last Call comments, so I'm
> sending these remarks just as generic feedback to the WG; feel free to
> do whatever you want with them process-wise, but I would be very
> interested in getting answers to my questions.
> 
> Basically, I started to work a smallish specification built upon RDF,

:)

> and I'm not finding the following information that I would think are
> indispensable for anybody wanting to build something on top of RDF:
> - is there a list of what SpecGL [1] calls "classes of products" for the
> various RDF specs? e.g., is the notion of RDF parser defined somewhere?

No.  We haven't defined a processing model.  That is something that 
could be done in the future, but we have deliberately stayed away from 
trying to define one as we've had enough to do already.

> RDF document? 

The RDF/XML syntax document defines what is and is not a legal RDF/XML 
document.  The concepts document defines and RDF graph, i.e. the 
abstract syntax for RDF.

> - is there a list of requirements bound to these classes of products?

No.  We have not defined classes of products nor their requirements.

> e.g., is a RDF/XML parser required to be a validating XML parser?

I'm not sure what you mean by validating.  RDF defines no concept of 
validation.  The schema document explains that schema can be used for 
inference, or can be used for detecting errors in documents, but it does 
not mandate either use.

  Is
> there a minimal list of encodings that an RDF/XML parser should support?

If you mean character encodings, No.

> - How the requirements between semantics/model & syntax/vocabulary play
> together?

I'm not sure what you mean by that one.

> 
> Sorry for the late comments, and thanks for any feedback.
> 
> Dom

Brian

Received on Thursday, 27 November 2003 09:43:20 UTC