Re: the mechanism for signatures in RIF

> Michael Kifer wrote:
> > Dave Reynolds <der@hplb.hpl.hp.com> wrote:
> 
> >> Lexically then the WD1 description says:
> >>
> >> "Symbols of this sort have the form "XYZ"^^rif:uri, where XYZ is a URI 
> >> as specified in RFC 3986."
> >>
> >> I don't think that is quite right. I think we want XYZ to be a 
> >> URI-Reference (i.e. including optional fragment identifiers).
> > 
> > But RFC 3986 doesn't preclude fragment identifiers, e.g., http://foo#bar.
> > Is it not true?
> 
> Indeed. I had meant to say "including relative references and optional 
> fragment identifiers" but brain and fingers seem to have not been 
> coordinated and missed out the important bit, sigh.
> 
> That RFC defines URI-references as "either a URI or a relative 
> reference" and gives the syntax for relative references.

So, how should we rephrase? Or should we? Right now we say that
rif:uri is a URI, not a URI ref.

> >> In at 
> >> least the XML syntax, and possibly in the linear syntax, we also want to 
> >> permit relative as well as absolute URI-References.
> > 
> > It is not clear to me whether relative URI references should be part of the
> > logic (and unduly complicate it) or whether they should be treated as 
> > syntactic sugar. 
> 
> Good point. They shouldn't really be part of the logic. We do need to 
> specify that syntactic sugar but you are right it would be better to 
> keep that out of the sort definition.
> 
> >> Turning to rif:localSymbol, if this is really needed then it would have 
> >> the same lack of domain assumptions as rif:uriSymbol (indeed presumably 
> >> it would generally have exactly the same domain).
> > 
> > In my view, the whole reason for local symbols is to make them completely
> > hidden so that they won't interfere with any other symbol. So, as I
> > mentioned before, I expect the domain of local symbols to be disjoint from
> > the rest.
> 
> Disjoint domains seems to be a stronger restriction than needed just for 
> scoping. In a RIF dialect with equality could you not imagine cases 
> where an individual denoted by a local symbol could be deduced to be 
> equal to an individual denoted by a global symbol?

I think we are thinking about different things. The reason I think local
symbols are needed is because I am thinking of how one often builds a rule
base. You have a bunch of predicates that have some good external meaning,
but other predicates are intermediate predicates that are like "subroutines".
Their names have no business of being equated with anything else, since
they are introduced for convenience. 


	--michael

Received on Tuesday, 27 March 2007 13:46:01 UTC