RE: [xdi] Re: XDI subjects (was Groups - XDI RDF v8 Comments-Barnhill)

[Giovanni Bartolomeo <giovanni.bartolomeo@uniroma2.it>]

Some comments from my side:

At 04.24 20/03/2008, Drummond Reed wrote:
>I think the same thing applies to all of the atomic relations ($a, 
>$is, $has, etc.) in the XDI RDF Model document 
>(http://wiki.oasis-open.org/xdi/XdiRdfModel), i.e., all of them ONLY 
>express relations between two XDI resources -- a subject resource 
>and an object resource -- and therefore the object of any of these 
>predicates MUST be an XRI and not a literal.
>
>(I don't know yet about whether their object can be an XDI 
>subcontext -- our equivalent of an RDF blank node -- I need to think on that).

I agree with this. This should be part of the XDI "grammar" rules 
which we talked about during a telco some weeks ago. BTW, my former 
and "mad" idea of having XDI documents as subjects  was exactly 
coming from Drummond last sentence above..  :-)

> >
> >   Sushi is better than nothing.
> >   Nothing is better than sex.
> >   Therefore, sushi is better than sex.
> >
> > The problem with the first syllogism (and it's probably already
> > obvious that "nothing" is wrong with the second one) is its
> > implication that changing the number of beers in a standard case to be
> > 16 would also require changing the number of hours in a day to be 16,
> > which probably isn't true.

I would propose to use $equals instead of $is. $equals doesn't define 
synonyms, instead it checks the actual value contained in the literal.
Again, this is XDI "grammar". I think we need to focus on this, 
collecting all issues related to the $keywords and have a section in 
the deliverable about this.

Bests,
Giovanni