Re: [urn] [apps-discuss] URNs are not URIs (another look at RFC 3986)

[John C Klensin <[e-mail]>]

--On Monday, 05 May, 2014 15:24 -0700 SM <[e-mail]> wrote:

> There seems to be different groups interested in URNs.  The
> national library group, for example, consider persistence as a
> matter of decades or more.

More like centuries or more.   See below, but I don't think that
is particularly important to category-forming, i.e., not a
useful starting point for a discussion.

>  I don't know whether the other
> groups (excluding people who usually participate in the IETF)
> share the same view.  I agree that the similarity of the
> syntax is deceptive.

I think it is obvious that different groups have different
perceived requirements.  For better or worse and despite a sense
that more education and listening would be helpful, I think the
"my perspective is right and you are wrong" tone of many of the
recent discussions are not leading us toward progress.  

Perhaps a more useful way to look at this is to try to
understand the needs of real communities [1] and then figure out
what we need to have in a URN standard to meet those needs.  My
working hypothesis is that, after trying to do with within the
constraints of 3986 and the compromises people are willing to
make is that it is not possible.  I hope we can do better than
   urn:nid:<stuff>
but maybe, if all we can do is to push NIS and tail syntax to
NID definitions and registrations, that is better than permanent
(sic) paralysis.

  best,
   john

[1] It has been 14 years since RFC 2141 was published.  There
are millions of URNs and many different NIDs out there.  I
suggest that is long enough for us to be able to tell the
difference between real communities with real user populations
and needs and philosophical speculation and that it is time to
start paying more attention to the former as less to the latter.
A lot more.