Re: URN namespace

Michael Shapiro (mshapiro@ncsa.uiuc.edu)
Wed, 16 Aug 1995 10:01:27 -0500 (CDT)


From: mshapiro@ncsa.uiuc.edu (Michael Shapiro)
Message-Id: <9508161501.AA25506@void.ncsa.uiuc.edu>
Subject: Re: URN namespace
To: J.P.Knight@lut.ac.uk (Jon Knight)
Date: Wed, 16 Aug 1995 10:01:27 -0500 (CDT)
Cc: uri@bunyip.com
In-Reply-To: <Pine.SUN.3.91.950816150701.20210w-100000@weeble.lut.ac.uk> from "Jon Knight" at Aug 16, 95 03:10:16 pm

Jon Knight wrote:
|
|On Wed, 16 Aug 1995, Michael Shapiro wrote:
|> I was trying to ask these 2 questions:
|> 
|> 1. To get long term persistence, do we have to have a new name space?
|
|I think that a new name space is a definate advantage if we want long 
|term persistence.  But we might also when to support ``legacy 
|namespaces'' as well.

I am only aware of existing DNS domains - what other legacy name spaces
should be supported?  Or are you also thinking of ISBN, ISSN?

Within DNS, as long as you don't force a new name space you get both.
In very early designs of the path scheme, we wanted to force a new top
level DNS domain - path:/x/y/z/doc would start looking for x.urn, then
y.x.urn, them z.y.x.urn.  But we have since decided that if there is a
new top level, then it should be part of the path -
path:/urn/x/y/z/doc.  This allows the service to work with new and
legacy DNS name spaces. (Credit for this decision must go to TBL).

ISBN could be done in DNS as well with some translations:
    ISBN-1-2345-34-5678 = 5678 under a server at 34.2345.1.isbn
or, wiht the path scheme
    ISBN-1-2345-34-5678 = path:/isbn/1/2345/34/5678

|
|> 2. Will a new top level DNS domain be a sufficient new name space, or
|>    do we have to do something which is not DNS based?
|
|I've been a big DNS fan for a long time; some years ago now Martin 
|Hamilton and myself played about with using the DNS to resolve lookups 
|for resources because to us it seemed an obvious bit of technology that 
|the URI WG could reuse.  Lets make things easy on ourselves, eh?
|

I have thought the same things - use DNS because it is there, it works
well, it is deployed, etc.  The objections I have heard to DNS have to
do with recalcitrant DNS admins, it will be too slow, it will be
overburdened.  None of these objections seem serious to me.   However
the presistence issue remains unanswered. Could you think about
persistence wrt to a new top level domain? How would delegation occur
that would help insure (or hinder) persistence?



-- 
Michael Shapiro                   mshapiro@ncsa.uiuc.edu
NCSA                              (217) 244-6642
605 E Springfield Ave. RM 152CAB  fax: (217) 333-5973
Champaign, IL 61820