Re: namespaceDocument-8: possible interaction with Namespaces in XML 1.1

At 10:38 09/04/2003 -0700, Tim Bray wrote:
>Larry Masinter wrote:
>>In the long run, I think it's easier to make a URNs retrievable
>>than it is to make HTTP URLs permanent, and that the W3C should
>>stop trying to make an anti-URN policy.
>
>Why is that?  It seems to me that the "permanence" of a URI has two facets:
>
>1. a community of people agree on what it means, e.g. a namespace
>    name for XHTML
>2. a publisher commits to provide representations of it
>
>Nothing in heaven and earth can ensure the permanence of #2.  And I don't 
>see that #1 has anything to do with whether the URI is a URN.

I think neither of the above are reasons to favour URNs over some other form.

Reasons I do perceive are:

(1) A urn, by its form, declares an *intent* to be (relatively) permanent 
that is not indicated by most URLs, and

(2) A registered URN namespace carries with it an indication of some level 
of community consent regarding its assignment and purpose.

(3) For some groups, persistent management of web serving infrastructure is 
difficult to achieve.  In my experience, commercial organizations tend to 
regard their URL space as property of the marketing department, to be 
rearranged and reorganized at whim, along with the 
pantone-colour-of-the-day.  And if not them, the IT department whose main 
clout comes from serving the marketing arm of the organiozation.  Future 
product design may not get much attention here.  We here may all agree 
about "cool URIs", but it's often difficult to get management buy-in to 
such a non-bottom-line concept.  URNs provide another way.

I note that none of these are technical reasons, but I don't think that 
makes them less legitimate.

#g


-------------------
Graham Klyne
<GK@NineByNine.org>
PGP: 0FAA 69FF C083 000B A2E9  A131 01B9 1C7A DBCA CB5E

Received on Wednesday, 9 April 2003 15:28:20 UTC