[tmcl-wg] Consistency of TMCL Schemas ?

[Dmitry]

----- Original Message ----- 
From: "Bernard Vatant"
To: <tmcl-wg@isotopicmaps.org>
Sent: Tuesday, January 06, 2004 4:52 AM
Subject: RE: [tmcl-wg] Consistency of TMCL Schemas ?



> > Secondly, TMCL and OWL rely on different data models or probably,
> > better to say, they use different categories to create models.
> > I can imagine rewriting OWL to make it compatible with Topic Maps set
> > of categories (such as names, occurrences, associations, subject
> > identifiers). But, I think, it will be actually much more complicated
> > than existing OWL.
>
> I'm not sure of what you mean by "rewriting OWL".

Let's say that I developed "opera" ontology in OWL. I definded
ObjectProperty "bornIn" with proper domain and range.
And I also added cardinality restriction to class "person"  on property
"bornIn".

How this OWL ontology can help me to understand that in my topic map
"Simoni, Renato" has to have only one association of type "bornIn" with role
"person" played by "Simoni, Renato"?

I can not directly use OWL. I should somehow to provide mapping between
different categories used in OWL and TM.

The "heavy way" - "rewriting OWL" - means that we introduce equivalent
constructs in terms of TM for each OWL construct.

So if OWL has a construct for restricting cardinality on object property we
should have a cardinality restriction on associations of speicific type
where type instance can play specific role.

Alternative is to define TMCL as "own"  language and provide clean mapping
between OWL and TMCL. We do this mapping on conceptual level. We map, for
example, each OWL object property to binary association. But...we also can
define in TMCL associations of arity 3 or 4 if we need to...TMCL will not
necessarily include all OWL constructs. It also can use optimized syntax.
Features balance between TMCL-Lite and Full can be different from OWL world
also.


Dmitry