Web Resource Identity

[John Cowan]

Paul Prescod wrote:

> In the HyTime and object oriented
> worlds, I believe that the defining characteristic of things with identity
> is that you can take two references and determine if they refer to the
> same object.

This may be feasible in closed-world scenarios.  In the Real (i.e.
open) World, identity may be a very hard problem!  It took the ancient
Greeks centuries to figure out that Hesperos (the Evening Star)
and Phosphoros (the Morning Star) were the same object, what we
call "the planet Venus".  Another famous example: Tully and
Cicero are the same man, "but it is neither trivial to
say so nor absurd to doubt it".  (Saul Kripke, _Naming and Necessity_)

On the other hand, we have the equally classical problem of 
"the ship of Theseus", preserved at ancient Athens.  Eventually
every atom of it was replaced, but its *identity* remained the
same.

> How do we know, other than common sense?

Right now we don't.  But RDF can easily handle the question
by defining a property "canonicalURI": note that this might,
perhaps even should, be an URN rather than an URL.

> This is more disturbing. It makes robust, scalable hypertext linking
> essentially impossible. Consider it from an RDF point of view. If I use
> RDF to attach a hundred properties to one URL and someone else uses it to
> attach a hundred properties to another one then our property groupings
> cannot be merged.

They can, but only if RDF has a notion of the identity relationship.
Right now it doesn't, but such a thing can and should be built.
RDF, after all, expresses properties of resources, not URIs, even
though it uses URIs to refer to resources,

> The only solution, if we assume a one to one correspondence between URLs
> and objects is to have EVERY NON-CANONICAL name for the object explicitly
> do a redirect to the canonical name.

This need not, probably should not, be done at the protocol level
(it would make mirrors pointless) but rather at the RDF level.

> I believe that the Web needs a concept of a canonical URL, if it doesn't
> already have one.

Agreed, except that "canonical URI" is more like it; you implicitly
accept this below by talking of UUIDs.

> Retrieving a document or the HEAD for the document
> should describe the canonical URL.

That's plausible, especially if you store RDF metadata in the head.
(What "head" means in general for non-HTML is a question.)

-- 
John Cowan	http://www.ccil.org/~cowan		cowan at ccil.org
	You tollerday donsk?  N.  You tolkatiff scowegian?  Nn.
	You spigotty anglease?  Nnn.  You phonio saxo?  Nnnn.
		Clear all so!  'Tis a Jute.... (Finnegans Wake 16.5)

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)