Re: modelling issue?

David, all


> Furthermore,  it is perfectly legal to declare the domain of a property
> more than once, such as:
>
>    P rdfs:domain GreenItems .
>    P rdfs:domain BlueItems .
>
> Then if you write a statement like:
>
>    x P y .
>
> the domain declarations imply that x is in both GreenItems and
> BlueItems.  Note that the effect is that the domain of P is the
> *intersection* of GreenItems and BlueItems -- not the union.
>

Not exactly. We are in an open world. The above assertions only entail that
the domain of P is *a subclass* of the intersection of GreenItems and
BlueItems.

Let me take another example to illustrate. Let's define two classes "Food"
and "Perishable".
Those classes are distinct but not disjoint. Some Food are not Perishable
(salt, sugar, good-until-used-biscuits ...). Some Perishable are not eatable
(paint, batteries ...).
Now I define a property "eatBy" and I assert the following

eatBy   rdfs:domain   Food
eatBy   rdfs:domain   Perishable
eatBy   rdfs:range   xsd:date

This means that any thing bearing "eatBy" property is both Food and
Perishable. But bearing this property is not necessary to be in this
intersection. A tomato in my garden is definitely both Food and Perishable,
but has no defined "eatBy".

Bernard


>
> --
> David Booth, Ph.D.
> Cleveland Clinic (contractor)
>
> Opinions expressed herein are those of the author and do not necessarily
> reflect those of Cleveland Clinic.
>
>
>


-- 
Bernard Vatant
Senior Consultant
Vocabulary & Data Engineering
Tel:       +33 (0) 971 488 459
Mail:     bernard.vatant@mondeca.com
----------------------------------------------------
Mondeca
3, cité Nollez 75018 Paris France
Web:    http://www.mondeca.com
Blog:    http://mondeca.wordpress.com
----------------------------------------------------