Re: URN definition -- a way out?

Mitra (mitra@path.net)
Wed, 27 Oct 1993 06:21:14 GMT

From: mitra@path.net (Mitra)
Subject: Re: URN definition -- a way out?
Message-Id: <CFJLnF.2L9@pandora.sf.ca.us>
Date: Wed, 27 Oct 1993 06:21:14 GMT

Peter,

URN->URC isnt particularly difficult.

However there are several proposals for how this could be done.
Whois++, copy of DNS, etc.

I havent seen a proposal for the URN which would define where to go looking
for the mapping tool, whois++ may be working, but I dont believe that
centroids have been proved to be scalable, and without them you dont know
which whois++ server to use to dereference the document.

Again there are proposals for how to do this, ranging from the vague "register
with IANA" to several involving using DNS to find the appropriate server.

And ...

If this is going to work, there needs to be a way for people to
register a URN/URL with a service, to tell a service that a URL has
moved and all the other good types of things described in Chris and
your documents.

I hope that all the proponents of different schemes can come to Houston and
talk them out. I dont really care which method we use, but I do hope we
come up with a scheme that is capable of handling most of the thigns
we think we want to do with the mapping.

- Mitra