Re: [xml-dev] Names and locations (WAS: Why REST?)

On 2002-02-13 17:08, "ext Bullard, Claude L (Len)" <>
wrote:

> All that is necessary is to distinguish the system
> of identification from the identity of the object
> so identified.  Then state clearly that in the discussions
> of URI, one is referring to a system identity, a
> means of representing the object.

And the semantics that expresses this should, I think,
be associated with the particular URI scheme, or with
a particular Class of URI, so that systems needing to
interpret such a URI know what it is intended to
signify and whether it should be resolved to something.

Providing (in a formal manner) the knowledge necessary for
an application to know if a given URI denotes a directly
accessible digital resource, an indirectly accessible
digital resource, a non-accessible resource, or is the
resource itself would make life alot easier.

And doing so in an economical manner, by URI scheme
and ideally by URI Class rather than on a URI by
URI basis is just common sense.

Patrick

--
               
Patrick Stickler              Phone: +358 50 483 9453
Senior Research Scientist     Fax:   +358 7180 35409
Nokia Research Center         Email: