- From: Aaron Swartz <me@aaronsw.com>
- Date: Fri, 29 Jun 2001 00:04:33 -0500
- To: pat hayes <phayes@ai.uwf.edu>
- Cc: Dan Connolly <connolly@w3.org>, w3c-rdfcore-wg@w3.org
On Thursday, June 28, 2001, at 06:22 PM, pat hayes wrote:
> If you want your order to mean "I want ANY book matching this
> description" then its not an existential: its a universal. I
> think this is what Jos DeRoos is doing with anonymous nodes as
> well, by the way. I have no beef with this, as long as we are
> clear what we want the anaymous nodes to mean.
I've tried hard to follow this conversation, and since my KIF
isn't too good and my logic worse, I've been trying to
transcribe it into the little of N3 I understand. What I follow
you to be saying is that an anonymous node has forSome
semantics, which is reasonable with Notation3's usage:
this log:forSome :x .
:DanC :wants :x .
:x a :Book ; :title "Fractals Everywhere" ... .
In other words, Dan wants a book. The book he wants has the
title "Fractals Everywhere".
However, for Dan Connolly's bookselling example requires forAll
semantics:
this log:forAll :x .
:DanC :wants :x .
:x a :Book ; :title "Fractals Everywhere" ... .
Or: Dan wants all the books with the title "Fractals Everywhere".
I don't think you can really have it both ways. The forSome
interpretation seems to be pretty standard, and not exactly
inconsistent with what we've been calling skolemization. In
other words:
{:x a :Book ; :title "Fractals Everywhere" ... } log:implies {
[ a :Book ; :title "Fractals Everywhere" ... } .
Or: If x is a book with the title "Fractals Everywhere" then
there exists a book with the title "Fractals Everywhere".
Sorry if this isn't useful at all,
--
"Aaron Swartz" | The Semantic Web
<mailto:me@aaronsw.com> | <http://logicerror.com/semanticWeb-long>
<http://www.aaronsw.com/> | i'm working to make it happen
Received on Friday, 29 June 2001 01:04:41 UTC