On Wednesday 22 October 2003 10:43, Norman Gray wrote:
> I'm sure it is this which causes some people (I'm thinking of Gerard
> Lemson and Pat Dowler) to gasp and, in their poster, pick out pos_eq_ra
> for special deprecation as incoherent. If you believe that principled
> generation of UCD words would be a Good Thing (and that would probably
> be my prejudice), then I suspect that paths in (say) Gerard and Pat's
> model would be a good way to do it (do Gerard and Pat claim that every
> UCD word is thus expressible?).
That particular example was chosen to show that the different parts of the POS_EQ_RA_MAIN (UCD1) are quite different types of things with different realtionships to the thing being described. In the data model, POS is the "position" phenomenon, EQ is a particular ReferenceSystem, RA is one component of a point (the type used to represent a position in that ReferenceSystem). I still don't know that MAIN would belong in a data model. It isn't "incoherent" so much as it includes some very different kinds of things.
In the concept;property style of UCD2, I don't think it is inconsistent with the data model we presented. The difference between UCD and DM is that in a model one explicitly states the relationship between the concept and the property, So, for example, in our DM we are expliclty saying that the relationship between a position and the ReferenceSystem is a different relationship that between the position and the RA (a component of the data type/structure). UCD2 leaves the relationship implicit by having only one relationship: "propertyOf". Whether that's good or bad thing is an open issue.
-- Patrick Dowler Tel/Tél: (250) 363-6914 | fax/télécopieur: (250) 363-0045 Canadian Astronomy Data Centre | Centre canadien de donnees astronomiques National Research Council Canada | Conseil national de recherches Canada Government of Canada | Gouvernement du Canada 5071 West Saanich Road | 5071, chemin West Saanich Victoria, BC | Victoria (C.-B.)Received on 2003-10-22Z19:25:42