A slight modification of your dataset proposal?

Antoine,

I was just wondering... we were discussing the relationships among different graphs in your setup, referring to the issue of, eg, functional property

<g1> { :john :age 23 }
<g2> { :john :age 33 }
<g3> { :age a owl:FunctionalProperty }

in your semantics, the following setup does not really make a difference:

<g1> { :john :age 23 }
<g2> { :john :age 33 }
:age a owl:FunctionalProperty.

The last triple is in the default graph, which is just as well 'separated' from the individual named graphs. However... wouldn't it be more natural to say that the triples in the default graph are 'true' for all individual graphs? Ie, they give some sort of a universal truth, just as they are today.

I guess your semantic definition could be modified quite easily by redefining the model:

For (G, (<n1>,G1), (n2,G2), ..., (nk,Gk) ) a model is (I1, I2, ... , Ik) where Ii is a model for Gi _and_ is also a model for G

the rest is identical.

Is this crazy?

Ivan 



----
Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
FOAF: http://www.ivan-herman.net/foaf.rdf