RE: XMI reality check

["Welsh, David" <David.Welsh@nordstrom.com>]

Hi John,
> 
> Where XMI differs from RDF in transport of model content is that XMI
> conforms to the more general UML domain, while RDF can be 
> restricted by use
> of namespaces to a specific problem domain. 
Sounds like more efficiency in RDF.

> The ability to 
> express a meta
> structure in RDF and validate it means better control on the 
> applicability
> of model content.

Cool, so I can also include CAD drawings of my engine part assembly as some
of the quality control reference material with my business process model
logic expressed in my UMM/ebXML business models, for my new car assembly
plant dock receiving procedures that I'd use with parts vendors ! And
further if I embed audio track metadata to work with some of the failure
post condtions on the receiving process, I could imagine 'active documents'
in ebXML somewhere in the not to distant future. I already see the W3C has a
note on "Describing and retrieving photos using RDF and HTTP".



>  When using models in a constricted domain 
> like b2b, it is
> attractive to be able to validate model content according to a meta
> structure.
> 
Keep going. I like it !


> For what it is worth, Edifecs undertook a formal RDF 
> initiative in late 1999
> and early 2000.  This initiative resulted in GXML/RDF 
> (guidelineXML in RDF),
> which is attached.  This effort focused on the messages 
> rather than the
> process, although it does allow association to BusinessAction elements
> (placing the message in the context of a process).  Contained 
> within this
> document are formal modeling in UML of the expression of EDI, 
> Flatfile and
> BCF (which is UMM and RosettaNet compliant) structures in RDF 
> complete with
> references to complete online examples.
> 

Always head of the curve !


> The examples referenced by the document were tested with the ICS-FORTH
> Validating RDF Parser (VRP) from the FORTH Institute of 
> Computer Science.
> 
Free parsers too ! 

> If there is any interest in this work we would be happy to 
> contribute it to
> the efforts underway.
> 

Hang onto that thought.
Dave



> Regards,
> John
> 
> 
> 
> 
>