W3C home > Mailing lists > Public > www-rdf-interest@w3.org > November 2000

Re: A triple is not unique.

From: Jonas Liljegren <jonas@rit.se>
Date: 20 Nov 2000 15:31:10 +0100
To: "McBride, Brian" <bwm@hplb.hpl.hp.com>
Cc: Dan Brickley <danbri@w3.org>, Seth Russell <seth@robustai.net>, RDF-IG <www-rdf-interest@w3.org>
Message-ID: <87aeauogup.fsf@jonas.rit.se>
Ok. There have been a bunch of comments on this.  I will try to
explain my view more clearly.

I would like to refere to previous discussion.  My post [1] was a
reply to Dan Brickley [2] who refered to an entry in the Issue
Tracking page [3]. Brian McBride also has some references to earlier
discussion [4]. 

The original discussion, initiated by Dan Brickley [5] is from
december 1999.  I made the same point also in that discussion [6].


It seems that we have a majority for this view:

    The statement is the triple.  But the reified statement represents
    a specific stating of a statement.  Each stating can have it's own
    URI.


 [1] http://lists.w3.org/Archives/Public/www-rdf-interest/2000Nov/0238.html
 [2] http://lists.w3.org/Archives/Public/www-rdf-interest/2000Nov/0236.html
 [3] http://www.w3.org/2000/03/rdf-tracking/#rdfms-identity-of-statements
 [4] http://www-uk.hpl.hp.com/people/bwm/rdf/issues.htm
 [5] http://lists.w3.org/Archives/Public/www-rdf-interest/1999Dec/0068.html
 [6] http://lists.w3.org/Archives/Public/www-rdf-interest/1999Dec/0099.html



And now the answer to the specific questions:


"McBride, Brian" <bwm@hplb.hpl.hp.com> writes:

> > Yes. That has been suggested before.  But that means that you can't
> > use reification in the way examplified in RDF M&S.
> 
> Just to be clear, can you point out the specific examples in M&S that
> don't work with the statements as facts interpretation.

I don't know what you mean with "the statements as facts
interpretation".

The question is:  Suppose we have two reified statements, S1 and S2.
They are both of rdf:type rdf:Statement.  They both has the same
subject, object and predicate properties.  Should S1 and S2 be
considered aliases for the same resource?

I say 'no'.


Here is the example from M&S:

  <rdf:RDF
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:a="http://description.org/schema/">
    <rdf:Description>
      <rdf:subject resource="http://www.w3.org/Home/Lassila" />
      <rdf:predicate resource="http://description.org/schema/Creator" />
      <rdf:object>Ora Lassila</rdf:object>
      <rdf:type resource="http://www.w3.org/1999/02/22-rdf-syntax-ns#Statement" />
      <a:attributedTo>Ralph Swick</a:attributedTo>
    </rdf:Description>
  </rdf:RDF>

The example uses the property a:attributedTo.  You can't attribute an
abstract statement to one person.  The example view the reified
statement as a stating.

The reasoning for this is the same as before.  Let me repeat the
example using This example.

Model 1:

  <rdf:RDF
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:a="http://description.org/schema/">
    <rdf:Description>
      <rdf:subject resource="http://www.w3.org/Home/Lassila" />
      <rdf:predicate resource="http://description.org/schema/Creator" />
      <rdf:object>Ora Lassila</rdf:object>
      <rdf:type resource="http://www.w3.org/1999/02/22-rdf-syntax-ns#Statement" />
      <a:attributedTo>Ralph Swick</a:attributedTo>
      <a:attributedTime>1999-02-22</a:attributedTime>
    </rdf:Description>
  </rdf:RDF>

Model 2:

  <rdf:RDF
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:a="http://description.org/schema/">
    <rdf:Description>
      <rdf:subject resource="http://www.w3.org/Home/Lassila" />
      <rdf:predicate resource="http://description.org/schema/Creator" />
      <rdf:object>Ora Lassila</rdf:object>
      <rdf:type resource="http://www.w3.org/1999/02/22-rdf-syntax-ns#Statement" />
      <a:attributedTo>Jonas Liljegrenk</a:attributedTo>
      <a:attributedTime>2000-11-20</a:attributedTime>
    </rdf:Description>
  </rdf:RDF>

If we import both these models, accepting both as true, and then
export them:

  <rdf:RDF
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:a="http://description.org/schema/">
    <rdf:Description>
      <rdf:subject resource="http://www.w3.org/Home/Lassila" />
      <rdf:predicate resource="http://description.org/schema/Creator" />
      <rdf:object>Ora Lassila</rdf:object>
      <rdf:type resource="http://www.w3.org/1999/02/22-rdf-syntax-ns#Statement" />
      <a:attributedTo>Jonas Liljegrenk</a:attributedTo>
      <a:attributedTo>Ralph Swick</a:attributedTo>
      <a:attributedTime>1999-02-22</a:attributedTime>
      <a:attributedTime>2000-11-20</a:attributedTime>
    </rdf:Description>
  </rdf:RDF>

How do we know which property belongs to which stating?


One suggestion to fix this is to create an explicit stating resource:

  <rdf:RDF
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:a="http://description.org/schema/">
    <rdf:Description ID="S1">
      <rdf:subject resource="http://www.w3.org/Home/Lassila" />
      <rdf:predicate resource="http://description.org/schema/Creator" />
      <rdf:object>Ora Lassila</rdf:object>
      <rdf:type resource="http://www.w3.org/1999/02/22-rdf-syntax-ns#Statement" />
    </rdf:Description>
    <rdf:Description>
      <a:states>http://description.org/schema/#S1</a:states>
      <a:attributedTo>Ralph Swick</a:attributedTo>
      <a:attributedTime>1999-02-22</a:attributedTime>
    </rdf:Description>
  </rdf:RDF>

But that's not how the RDF M&S wrote the example, and that's why I
wrote "that means that you can't use reification in the way
examplified in RDF M&S".



> For the record, section 5, the formal model section of M&S states:
> 
>   There is a set called Statements, each element of which is a triple
>   of the form {pred, sub, obj}
> 
> That says that each Statement is a triple of the form (s,p,o).  A 
> triple in mathematics is uniquely determined by its three components.

Right.  But we can't say the same about the reified statement.  There
can be many reifications of the same statement.



> > My vote is on allowing identical statements with diffrent identities.
> 
> Can you clarify what that sentence means :)  How can two identical things
> have different identity?  If they  have different identity, they are not
> identical!

I was talking about the reifications of the statements.  They should
be viewed as statings.


> > And you can't avoid that with statements distributed over several
> > models over the net.
> 
> Different representations of statements.  In my world model, statements
> are abstract and don't have a location on the net.  Perhaps yours is 
> different.

You can give URI's to reified statements.  That means that there will
be many URIs that have the same predicate, subject and object
properites.

-- 
/ Jonas Liljegren

The Wraf project http://www.uxn.nu/wraf/
Sponsored by http://www.rit.se/
Received on Monday, 20 November 2000 09:33:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:46 GMT