Re: uri, urn and info

On Wed, 2003-10-15 at 09:43, Ray Denenberg, Library of Congress wrote:
> From: "Patrick Stickler" <patrick.stickler@nokia.com>
> >     Yes, there is no restriction on how many URIs might denote
> >     the same thing (I've pointed that out myself in this very
> >     thread) but there is much to be gained from an economy of
> >     identifiers.
> 
> Once upon a time there was a notion of multiple URLs (different schemes)
> providing access to a given resource, because there are multiple ways to
> access the resource. There might even be an 'ftp' URL for a bibliographic
> record corresponding to an lccn, along with an http URL.  Now are these
> locators or identifiers? I don't care.  The info URI for that lccn would (a)
> not be a locator, and (b) be the cannonical identifier. (Whether its
> "info:lccn......" or "urn:info:lccn ......")

But as I've asked before (and which is already the case):

Since you aren't LCCN how can you know with 100% certainty that any
semantics you _think_ are created by minting 'urn:info:lccn' are equal
in all cases to the semantics inherent in 'urn:lccn'? 

This is one of the things that still bugs me about 'info'. NISO is
making attempting to make an authoritative statement about things inside
namespaces that it has no authority over. At most a namespace found
inside an 'info' registry is simply NISO's opinion, nothing more.

In other words: urn:info:lccn IS NOT EQUIVALENT to urn:lccn and never
should be. If claims are being made that it is then that is an attempt
to make an end run around the agreed upon registration mechanisms. If
those claims are not being made then a statement to that effect should
be at the very beginning of the document in a clear applicability
statement that says they are not being made and that there MUST NOT be
any future attempts to do so.

-MM

Received on Wednesday, 15 October 2003 10:51:56 UTC