[
Lists Home |
Date Index |
Thread Index
]
Yes, it was the URN conclusion. I support that but I
was surprised. To me the major point is that
a sign (a URzed is a sign) should be used clearly and
therefore, the intended use should be clear. In this
case, the authors make the Use As Management of the Name
dominant over Dereferencing Based On Location. That
surprised me. It has been held in some corners that
there is no significant difference. I hold that that
is a systemic conclusion but in error if different
systems are interoperating. (here comes permathread#1:
universality of names and locations).
Yes, if the URzed denotes the owner, your objection
holds. It is a problem of any More Meta Than Thou
naming scheme, or why Zero is the Root Of All Theories.
len
-----Original Message-----
From: Chiusano Joseph [mailto:chiusano_joseph@bah.com]
I assume that by "unexpected conclusion" you mean the recommended use of
URNs over URLs as namespace identifiers? I was surprised at this as
well. A large part of this document was extracted from an XML schema
design rules document I co-authored while at the same company as the
authors (so I recognized quite a bit of it) - but in that document, we
did not get into "URN vs. URL" to the degree shown here.
I can see the advantages to the hierarchical/structured approach of
URNs, but also some disadvantages with the recommended approach - for
example, where one may need to provide a namespace identifier that
denotes a cross-agency initiative. Sure, there would be a "sponsor
agency", but one would lose the cross-agency significance in the
prescribed approach.
|
