Re: abstract model and reification

At 04:46 AM 9/20/00 -0400, Pierre-Antoine CHAMPIN wrote:
> > Brian wrote:
> > >So if we imagine that we have two resources s1 and s2
>both of which
> > >represent some statement S. Then any RDF statement
> > >that was true of s1, would in fact be a statement about
>S and so
> > >would also be true of s2.
>
> > Graham wrote:
> > I'm not sure I buy the premise here...
> > Suppose we have s1 and s2, as you say, both modelling
>(representing) S.
> > Then we may have different statements about the statement
>S attached to them:
> > s1 --assertedBy--> "Brian"
> > s2 --assertedBy--> "Graham"
>
>The point is : Brian is seeing Statements as Facts, and
>Graham is seeing them as Fact Occurences. Both have their
>utility, sounds more like a religion issue :)

I was wondering about that.  I don't mind which way it goes, as long as it 
makes a consistent story.

>The problem is that the spec is not very clear about the
>how the authors see it.

>We had a discussion about it a few months ago,
>sounded like the majority was considering statements as
>facts. I used to talk about "statings" then, when talking
>about fact occurences.

There's also that we are talking here, effectively, about quotings of 
statements.  Are two quotations of the same statement actually the same 
quotation?   I may want to talk about a quotation of a statement made by 
myself, and a quotation of the same statement made by Brian.

Example:

(1)    A statement:  pigs can fly
(2)    I assert the statement "pigs can fly"
(3)    Brian assert the statement "pigs can fly"
(4)    When I said "pigs can fly" I believed I was telling the truth
(5)    When Brian said "pigs can fly" he was lying.

So we have:
   ["pigs can fly"] (2) is a believed truth.
and in another case
   ["pigs can fly"] (3) is a lie.

Are these quotations the same resource?  If so, we seem to have a 
contradiction.  (I accept there are other ways of modelling this that 
avoids the contradiction, but I don't think that invalidates the line taken 
above.)

Lets move this to a more realistic scenario:  I create a document 
containing a statement, and send it out with a statement (signature, 
whatever) saying that I assure the content of the document.  Then I resend 
the same document with a statement saying that I repudiate its 
contents.  If the quoted statements are distinct that's fine.  But if 
they're the same it can get difficult to untangle what is going on.

--

FWIW, I think this is bound up with the issue of URI:resource being 
1:1-onto, or 1:N.  If the former, and considering that the RDF model 
clearly allows multiple reifications to be expressed with different 
identifiers, then they must be different resources.  If the latter then 
there is a choice.

#g

------------
Graham Klyne
(GK@ACM.ORG)

Received on Wednesday, 20 September 2000 05:41:37 UTC