[tmcl-wg] TMCL Discussion on IRC and 2nd meeting today: Mar 28!

[Robert Barta]

On Fri, Mar 28, 2003 at 12:46:56PM +0100, Bernard Vatant wrote:

> I completely agree on that principle. Not an easy way, though ...

Bernard, Steve,

My thinking is that the fact that "is-instanceOf" and "subclasses" are
handled special in TM (XTM, TMCL) does not necessarily mean that all
applications MUST use those (although as you point out this makes
sense in most cases):

If an application wants to introduce its own way of expresing a
relationship between an element and a set of things (class), then the
application might simply choose a different association type.

But even if there would be a loss of generality, I think the benefit
of hard-wiring some association outweights it.

\rho

> 
> -----Message d'origine-----
> De : tmcl-wg-admin@isotopicmaps.org
> [mailto:tmcl-wg-admin@isotopicmaps.org]De la part de Steve Pepper
> Envoye : vendredi 28 mars 2003 11:58
> A : tmcl-wg@isotopicmaps.org
> Objet : RE: [tmcl-wg] TMCL Discussion on IRC and 2nd meeting today: Mar
> 28!
> 
> 
> At 10:14 28.03.2003 +0100, Bernard Vatant wrote:
> >What I understood during yesterday conversation is that TMCL would require
> >that distinction, which means it would not support   topic maps that would
> >not make it. Somehow, it would mean that TMCL would be less
> >"ontology-neutral" that ISO 13250 is, in the sense that it would not
> >support some "exotic" types of topic maps that do not want to make classes
> >and instances "a priori" distinct types, and stand on more "fuzzy", or
> less
> >formal, ontologies.
> 
> In my opinion we should only be providing TMCL with the ability to enforce
> the distinction between types and instances for applications that need to
> do
> so. However, it should be done in such a way that TMCL can also be used
> with
> what you call "ontology-neutral" topic maps.
> 
> Steve
> 
> --
> Steve Pepper, Ontopian
> http://www.ontopia.net
> DUMUS DELENDA EST
> 
> _______________________________________________
> tmcl-wg mailing list
> tmcl-wg@isotopicmaps.org
> http://www.isotopicmaps.org/mailman/listinfo/tmcl-wg
> 
> 
> _______________________________________________
> tmcl-wg mailing list
> tmcl-wg@isotopicmaps.org
> http://www.isotopicmaps.org/mailman/listinfo/tmcl-wg
>