Re: the mechanism for signatures in RIF

> <snip/>
> > I was hoping that you would read my msg addressed to
> > you, Francois, and Hassan, and will respond to the technical points.
>   
> See my e-mails in that thread.

Yeah, but I didn't think I got an answer. Maybe it wasn't clear what I was
asking...

> > This is the whole point: WHAT is a URI in the RIF language? A string?
> > Probably not.  If they are a special kind of symbol then they ARE a sort.
> >   
> Let me try to illustrate my point, i.e. that the set of URIs is a set of
> symbols, and not the sort.
> According to [1], "Each /*D*/_si is the domain of interpretation of the
> primitive sort s_i".
> This would mean that there is a special domain of interpretation for
> URIs. Now, there is no meaningful domain of interpretation especially
> for URIs. This leads me to conclude that the sorted "URI" is not a
> meaningful sort.


It is not true that a sort must have a concrete domain of interpretation.
If you open any standard book, like Enderton, then you will see
that they don't. A domain of interpretation for a sort is typically abstract.

The idea behind sorts is not to force some standard domains on sorts, but
to separate their domains so that they could be treated separately. Then
you start the engineering part to tailor the idea to your needs. For some
sorts you might want concrete domains.  For some you might want domain to
be disjoint with something else. For others you want some partial order on
the domains. And so on. It is quite surprising how far this can take you.

The idea is not mine, of course -- it is quite old (although standard texts
usually don't go into much detail on this). I just did some engineering to
tailor the idea to what I think RIF's needs are.

So, for URIs to form a meaningful sort they need to be a distinct set of
symbols and nothing more. We might also think whether their domain requires
"engineering", i.e., making it disjoint from something, a subset of
something, etc. This is how I see the meaning of the rif:uri issue
http://www.w3.org/2005/rules/wg/track/issues/30

To summarize, there seems to be no disagreement -- only a misunderstanding.
Maybe the wording needs to be tightened (suggestions appreciated), but I
see currently nothing wrong with the rif:uri sort.


	--michael

Received on Wednesday, 21 March 2007 16:54:37 UTC