Re: New version of URI Declarations

At 10:18 PM +0000 2/27/08, Booth, David (HP Software - Boston) wrote:
>I've substantially revised and expanded my write-up on URI declarations:
>http://dbooth.org/2007/uri-decl/
>
>For anyone interested in issues of resource identity and proper use 
>of URIs, I strongly recommend reading it.  If you already read an 
>earlier version, this version has added sections on granularity, 
>ancillary assertions, why it's important to distinguish URI 
>declarations from ancillary assertions, and an explanation of how a 
>URI declaration establishes resource identity.
>
>The abstract:
>[[
>A URI declaration permits assertions about a URI's associated 
>resource to be classified into two groups: core assertions, whch are 
>provided by the URI declaration, and ancillary assertions, which are 
>all others.  This distinction enables different parties to share a 
>common understanding of the associated resource (by accepting the 
>core assertions) while making different choices about which 
>ancillary assertions to accept. This paper defines these concepts 
>and proposes some related best practices and a Web architectural 
>rule specifying how URIs for non-information resources can be 
>conveniently declared using existing hash or hashless (303-redirect) 
>URI mechanisms.
>]]
>
>As usual, comments are invited.

OK, you did ask...

There is a basic problem with this idea and what you say about it. It 
comes to the surface here:

"Definition: A URI declaration is a set of statements, or "core 
assertions", that authoritatively declare the association between a 
URI and a particular resource.

A URI declaration is a performative speech act.  (See Cowen's message 
or Wikipedia.)  Its publication by someone who has the authority to 
make the declaration -- the URI owner or delegate -- defines the 
association between a URI and a resource.  "

Wrong. Or, since a definition can't be wrong: with this definition, 
URI declarations do not exist; in fact, cannot possibly exist.

No amount of just asserting can possibly create an association 
between a URI and a non-information resource, (unless it does so via 
some other URI which is itself 'associated' to a resource 
appropriately, but that just gets us into an infinite recursion.) 
There is no way of forcing a purely assertional framework, no matter 
how complex or large it is, to refer to anything non-linguistic or 
non-symbolic, by making assertions in the framework itself. (This can 
be formally stated as Herbrand's theorem: if a set of axioms has a 
satisfying interpretation at all, then it has one entirely made of 
its own symbols.)

Its not a question of authority. You can have all the authority you 
want: but if all you can do is make assertions at me, no amount of 
authority is going to prevent me consistently understanding you as 
only talking about a symbolic world. I won't be rejecting what you 
tell me: you do have the authority to assert it, and I will accept it 
all as true. But it can all be true, and still not refer to what you 
want it to refer to. You can't possibly attach symbols to reality by 
symbolic means (like making assertions using the symbols).

Just calling this a performative speech act doesn't work, either. In 
order to get a performative to work, you need to do something - 
perform a performative - with the thing involved in the act. The 
things involved in the speech act are part of the performance. The 
performative "I now pronounce you man and wife" works, in part, 
because it is said to the man and the wife themselves, in the flesh, 
so to speak. They have to be there to get married. Baptism, 
similarly, attaches a name to a child, and the child has to be there 
to get baptised. As John Cowan points out, in some cultures, simply 
saying "my real name is X' is enough to make your name X: but it has 
to be you that says it. Performatives always use indexicals (I, now, 
you, this) because they always need to be performed as part of an act 
involving the thing or person itself (or in some cases a person or 
thing appropriately associated with the thing or person, such as 
someone acting with power of attorney.)

I could make detailed comments on the rest of the document, but this 
is enough. This idea of 'declarations' of non-symbolic, 
non-computational, non-informational entities just does not make 
sense. You can't declare the moon. Give up on the idea, its nonsense.

Pat Hayes


-- 
---------------------------------------------------------------------
IHMC		(850)434 8903 or (650)494 3973   home
40 South Alcaniz St.	(850)202 4416   office
Pensacola			(850)202 4440   fax
FL 32502			(850)291 0667    cell
http://www.ihmc.us/users/phayes      phayesAT-SIGNihmc.us
http://www.flickr.com/pathayes/collections