Re: Tentative contribution to the "URI issue"

> Michael Kifer wrote:
> 
> > The question is how does one differentiate URIs from strings.
> > Are they strings? Should they be? If all we (collectively) want
> > is to write "http://jos@debrujin.com/salary"("2020-11-22", "E100000")
> > where "..." are strings, then there is no problem.
> > But I don't think think this is what we collectively had in mind.
> > I believe we wanted to distinguish URIs from strings and other types of
> > constants.
> 
> Distinguishing symbolic constants identified by a URI from literal 
> constants such as strings and integers still seems like a syntax issue 
> addressable at that level in the way proposed earlier.

It is a syntactic issue in the sense of a precise syntax for the language
of a logic - not in the sense of choosing this keyword/construct or that
keyword/construct (as in programming languages).  These are two quite
different things.


> The conceptual proposal is that we *only* have symbolic constants 
> identified by a URI and literal constants (strings, integers etc) and 
> only the former are allowed to have boolean and arrow sorts.
> 
> > There is also an issue of signatures when we will start allowing or
> > disallowing certain individuals to play roles of predicates, functions,
> > etc.  We cannot assign a signature to any given constant, so sorts could be
> > one of the grouping mechanisms here. For instance, if a dialect (like, say,
> > WSML) allows only URIs to be concepts then only the constants of the sort
> > rif:uri will have Boolean signatures.
> 
> What I was asking for in the meeting and my earlier message was what's 
> the use case where we would allow something *other* than symbolic 
> constants identified by a URI to be used as concepts?  The point being 
> that I'm not convinced there is one.

RIF is an exchange language for rule sets that are not necessarily web-related.
There are tons of such sets.

Second, in a complex KB, you always have internal predicates that shouldn't
be visible outside. Why should they be given a URI?


> I can see the value in having those symbolic constants be sorted but 
> they would still be identified by URIs. [*]
> 
> In which case perhaps all we are asking is "what's the name for the top 
> sort?".
> 
> If so then in RDF terms that is rdfs:Resource - everything in your 
> domain is an rdfs:Resource including literals, people, unicorns and web 
> information resources.
> 
> If you want a sort which is disjoint from literals (strings, integers 
> etc) then from an OWL/DL point of view that would be owl:Thing.

I don't think we need a top sort. Also, there are some practical problems
with sorted literals in the presence of a top sort.


	--michael  


> Dave
> 
> 
> [*] Indeed in a sorted semantic web compatible dialect then one might 
> identify RIF sorts with RDF Classes so in that dialect you could say 
> things like:
> 
>    all symbols of rdf:type owl:DatatypeProperty have a RIF boolean
>    signature:
>       OWL:Thing * rdfs:Literal
> 
> but you could also then have application-specific sorts defined using 
> RDFS/OWL ontologies.
> 
> 
> 
> 
>

Received on Wednesday, 21 March 2007 12:13:56 UTC