Re: UR* schemes [was: 3rd IETF/W3C coordination call...]

Howdy,

Let me leap into this stream and straighten out a few things about
URNs.

"URN" is not a notational convention, it is a URI scheme that 
defines a particular syntax and resolution specification.  That
syntax is used to allow sub-delegation of namespaces, and the
resolution specification specifically supports the further
subdelegation of namespace components -- partitioning the
space of identifiers, or providing multiple redundant services
for identifiers, or providing different end-resolution protocols
for identifiers, for instance.  

This approach was selected in an effort to maximize the ability
to support persistence in naming.  It is true this cannot be enforced.

But now that the specifications are on the table, it is time to
stop arguing about what a URN "is" philosophically, and look instead at
what the URN scheme _does_.   There are communities that have
applications for making their resources available on the net
(note:  I said net, not necessarily Web) that want a scheme that _does_ 
the above.  They need software that can recognize identifiers that 
belong to this scheme, so that the software can apply the correct 
resolution steps and achieve the expected result, in a controlled 
fashion (by selecting resolution services from those discovered, not 
by being "bounced" in a redirect from one to another).

So, if you want to argue that it could all be done in HTTP, sit down,
write up the proposal for how to use HTTP (redirects and all) to
establish and maintain a global resolution service for the URN
scheme, to do what URN does today.  If it's better than what
exists today,  let's replace the current one.

But saying URNs are "bad" is extremely passé, and some of this is
beginning to sound like "stupid user" stories.  The users are the
customers, and they'll pick what works for them.

Leslie.

-- 

------------------------------------------------------------------------

"Square peg, round hole:                         Leslie Daigle
   Social reality, or engineering problem?"      leslie@thinkingcat.com
                       -- ThinkingCat

------------------------------------------------------------------------

Received on Wednesday, 1 December 1999 23:50:56 UTC