W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > June 2001

Re: log:forSome (Was: Model-specific identity for anon resources, and its representation: A new issue?

From: pat hayes <phayes@ai.uwf.edu>
Date: Sat, 23 Jun 2001 23:22:15 -0500
Message-Id: <v04210100b759b79bd241@[205.160.76.176]>
To: jos.deroo.jd@belgium.agfa.com
Cc: w3c-rdfcore-wg@w3.org
> >>>>I think/hack that both anonymous *terms* can be *unified*
> >>>>which is NOT the same as equality
> >>>
> >>> Then your hack only seems to works when we can unify anonymous
> >>> terms, or they have some sort of UniqueProperty attached. How do
> >>> we solve the general case?
> >>
> >>There was some related discussion in the telecon
> >>this afternoon, and I was kind of unable to explain
> >>my point...(I'm really hopeless in that respect)
> >>
> >>The thing about anonymous nodes is that they
> >>are ***variables***
> >>if they would be constants, we would be able
> >>to identify them with ***URI constants***
> >>Now they are actually existentially quantified
> >>variables, something like: there exists an _:a
> >>or (using existing vocab): this log:forSome _:a
> >>So I think we should say that *explicitly*
> >>is the testresults (and in the model theory)
> >>(the general case?)
> >
> >Well, logically they seem to be like existentially quantified
> >variables, yes. But you CANT unify distinct existentially quantified
> >variables (validly) !  They would be distinct skolem constants, which
> >do not unify in general. (They do if they are existential variables
> >bound by the same quantifier, but then they would be the same skolem
> >constant :-)
>
>I agree that you CANT unify distinct existentially quantified
>variables in the sense that one does with universally
>quantified variables i.e. computing a MGU (Most General Unifier)
>but what we actually compute is a LCU (Least Common Unifier)

Whoa!  Ive never heard that term before. Can you specify what you 
mean? Eg give a sketch of this LCU algorithm, or some explanatory 
examples?

>for those anonymous nodes (I mean what those nodes have in
>common w.r.t. their descriptions and taking into account
>daml:UniqueProperty properties to succeed/fail)

Sorry, that isn't enough for me to reconstruct what you mean.

>, and that
>seems to work in a logical sense but I'm also wondering why?-)

I will try to explain once I know what you mean.

Pat Hayes

---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes
Received on Monday, 25 June 2001 10:42:51 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:37:15 EDT