W3C home > Mailing lists > Public > uri@w3.org > November 2003

non-IETF tree URI scheme draft

From: <hardie@qualcomm.com>
Date: Tue, 11 Nov 2003 17:17:18 -0600
Message-Id: <p06010200bbd71b8eb13b@[130.129.132.195]>
To: uri@w3.org

Larry Masinter and I had a chance to grab some time in the hallways
of this IETF meeting, to discuss the future of the draft on non-IETF
URI schemes.  One of the sticking points all along has been how to avoid
name collisions in these, and we discussed this around a couple
of straw proposals that he and I generated out of cookies and hot tea.
None of them were really satisfying, though, and we eventually
came to the conclusion that the collision problem in other
IANA registries were simply a not big enough problem to put
people through any syntactic hoops for this.  I'd like to suggest,
therefore that we revise the doc to say essentially:

Non-IETF tree URI schemes are registered by the IANA
by providing the IANA a template document that shows
who has change control, gives a name for that change
controller, the proposed scheme name, and a precis of
the URI scheme's syntax. 

 The IANA would then construct and assign a URI scheme of the form:  
orgname-schemename.

The IANA will allow only single orgname per organization, will
generally choose the one associated with the MIME registry
if it is available, and may refuse a suggested organization name
and propose another at its discretion (chiefly to avoid collision,
but at its discretion).  

The IANA will use the "expert reviewer" mechanism for assignment,
and that reviewer will assure that the proposed scheme is
a syntactically correct URI, but will not review its proposed
operation or semantics.

Does this make sense to folks as the right balance of achieving
uniqueness, ensuring syntactic correctness, and avoiding a
heavyweight, slow process?
			regards, 
				Ted Hardie
Received on Tuesday, 11 November 2003 18:17:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 13 January 2011 12:15:32 GMT