Re: suggestions for datatyping (long)

> > [...]
> > fine Sergey, but I think we have to face a testcase like
> >   aaa bbb "xyz".
> >   rdfs:range bbb ccc.
> > It will concur somewhere presumably
> > so what sense do we/machines make out of that?
>
> Did you mean
>
> aaa bbb "xyz"
> bbb rdfs:range ccc  ?

yes I did (wondering how I could have written that)

> If ccc is a subclass of rdfs:Literal, the example looks fine. If some
> other conditions prohibit ccc from being a subclass of rdfs:Literal,
> then the example has no valid interpretation, I guess.

well, I'm wondering...
let's take another testcase
####
    :Animal a rdfs:Class;
         <http://www.daml.org/2000/12/daml+oil#restrictedBy>  [
             a <http://www.daml.org/2000/12/daml+oil#Restriction>;
             <http://www.daml.org/2000/12/daml+oil#onProperty> :parent;
             <http://www.daml.org/2000/12/daml+oil#cardinality> "2" ];
         rdfs:label "Animal" .
####
what is that "2" denoting? the number 2, isn't it?
I can't imagine that the range of
  <http://www.daml.org/2000/12/daml+oil#cardinality>
is a subclass of rdfs:Literal

--
Jos