Re: [urn] [Gen-art] Review: draft-martin-urn-globus-02

[[e-mail] (Dale R. Worley)]

On 3/17/16 8:50 AM, Barry Leiba wrote:
> [...]
> My view is that we should not be too rigorous about this point at this stage.

"Joel M. Halpern" <[e-mail]> writes:
> If it were me, it would seem that a document using a new and 
> not-yet-approved process would require a normative reference to the new 
> process, and could not take effect until the new process was approved.

It seems that we can meet the current requirement by updating the draft
to state why Martin/Globus wants the namespace.  I.e., they must have
*some* reason.  Then per the new requirements, we can be relaxed
regarding what constitutes "sufficient" reasons for approval.

E.g., looking at the draft, I see in section 1:

   Globus creates unique identifiers which will be persisted in external
   systems, and which must be identifiable as references to Globus
   entities.

My guess is that this is why Martin/Globus are going to the effort to
register a namespace, and as such, it constitutes "the perceived need
for a new namespace (i.e., where existing namespaces fall short of the
proposer's requirements)."

Dale