Re: UML/XMI and semantic web

In the utility domain UML has been pretty extensively used for defining
abstact schema which are then converted to RDF schema.  I maintain some web
pages with links about this [1,2].  There are various standards bodies and
discussion groups involved.

We convert these UML models to (extended) RDF schema using a tool [3].
Unfortunately, the full detail of the UML/RDFS mapping is not documented
properly yet.  But the main points are to be found in a related OMG
specification [4] (section 8.2).

I personally don't find UML as flexible as an RDF-based ontology. I would
suggest that the basic problem is that UML makes classes first class
citizens but not properties (associations/attributes). However, it is well
accepted and has a graphical representation that people like.

[1] http://www.langdale.com.au/DAF/
[2] http://www.langdale.com.au/CIMXML/
[3] http://www.langdale.com.au/styler/xpetal/
[4]
http://www.omg.org/technology/documents/formal/UMS_Data_Access_Facility.htm


--
Arnold deVos
Langdale Consultants
adv@langdale.com.au

----- Original Message -----
From: "oliver fodor" <fodor@itc.it>
To: <www-rdf-interest@w3.org>
Sent: Tuesday, November 13, 2001 10:08 PM
Subject: UML/XMI and semantic web


> Hi!
>
> I'm evaluating XML-based technologies for a tool providing
interoperability
> on the data level between different systems in the tourism domain.
> Naturally the question of building ontologies has came up. Observing the
> research in this area it is still unclear to me whether it is
> possible/feasible to use UML as modelling language for onthologies and
then
> XMI in order to provide a machine readable form? Please give me some
pros -
> contras, considering UML/XMI and let's say OIL based modelling. I'd also
> welcome some related URLs.
>
> Thanks!
>
> oliver
>

Received on Tuesday, 13 November 2001 23:52:25 UTC