Re: Properties of Resources that Need More than One URI to be Identified

----- Original Message -----
From: "James Cerra" <jimbobbs@hotmail.com>
To: <www-rdf-logic@w3.org>
Sent: Tuesday, November 11, 2003 5:59 PM
Subject: Re: Properties of Resources that Need More than One URI to be
Identified


>
> > URIs are not interpreted as URLs in RDF - it's an
> > abstract identifier in some perhaps unique namespace
> > schema. ...
>
> I understood that before; however, my question was probably poorly posed.
> Sorry for any confusion.
>
> Here is my current understanding.  I am under the impression that there
are
> two different types of nodes in RDF that can be a subject: a URI reference
> string or an anonymous node.  Obviously, RDF is only useful for describing
> resources that can be represented by one of those two.  My question is:
are
> there non-anonymous resources that cannot be identified by a URI?

Not that I know of.

> If so,
> how then can they used in RDF graphs?  And what are some of their
> properties?

>
> Tom Passin seems to suggest that such a resource should be assigned an
> arbitrary URI [1].  Both of you think that predicate/object pairs could be
> added to further describe the resource.  Am I interpreting your posts
> correctly?

Yes, I think properties with a specified interpretation are a useful way to
reference specific interpretations of a URI, such as physical resources.

> In the case of the unreachable file [2], could I specify the
> following?
>
> <an-arbitrary-uri-for-my-resource> <rdfs:isDefinedBy> _:b .
> _:b <rdf:type> <rdf:Seq> .
>   _:b <rdf:_1> <a-computer-scheme-uri> .
>   _:b <rdf:_2> <a-file-scheme-url> .
>

You can do this, but note there is no formal interpretation of
rdfs:isDefinedBy, so it would be unclear what you mean by this.  Your
"an-arbitrary-uri-for-my-resource" is not really so arbitrary, the values of
_:b <rdf:_1> and _:b <rdf:_2> can change, but your resource still be
considered the same thing, which I gather is not what you want - i.e. you
mention that these uris, which are the values of the sequence items, are the
defining characteristics for identifying that resource.  Perhaps you should
use a custom Class and Property construct, or even some OWL(which would give
you necessary and sufficient constructs).

> Am I correctly understanding you so far?

Mostly.  What I'm trying to get at is that RDF likes named and unamed
resources, if it is named, then a single URI gives it a unique reource
identifier(which anonymous resources also need to be interpreted as).  To
provide an unambiguous interpretation of a physical locator, then it is
useful to formally state that 1) it is a physical locator, and 2) it is
being used as one of possibly many values of a particular property that is
interpreted in a particular way.

>
> --
> Jimmy Cerra
>
> ] "A good decision is based on knowledge
> ]  and not on numbers" - Plato
>
> [1] http://lists.w3.org/Archives/Public/www-rdf-logic/2003Nov/0024.html
>
> [2] http://lists.w3.org/Archives/Public/www-rdf-logic/2003Nov/0023.html
>
>

Received on Tuesday, 11 November 2003 05:34:56 UTC