To:
Patrik Fältström <paf@cisco.com>
Cc:
Rick H Wesson <wessorh@ar.com>, "Hollenbeck, Scott" <shollenbeck@verisign.com>, <ietf-provreg@cafax.se>
From:
Martin Oldfield <m@mail.tc>
Date:
Sun, 11 Mar 2001 16:39:58 +0000 (GMT)
In-Reply-To:
<p05100215b6d0f371b1b4@[192.168.1.24]>
Sender:
owner-ietf-provreg@cafax.se
Subject:
RE: Unique handle generation
>>>>> "paf" == paf <Patrik> writes:
paf> At 15.48 -0800 01-03-10, Rick H Wesson wrote:
>> On Sat, 10 Mar 2001, Hollenbeck, Scott wrote:
>>
>>
>>> OK, I see your point. In an earlier message you suggested
>>> some requirements for handles; let me try to capture those
>>> thoughts (and some mentioned by others) more precisely:
>>>
>>> 1. Every object MUST have an associated handle.
>>
>> Domain Name objects DO NOT need handles, as the FQDN uniquely
>> identifies the object.
paf> We need handles on _every_ object, and maybe the handle on
paf> domain name objects (which is a very small part of all
paf> objects we talk about) have the same syntax as the FQDN. But
paf> I see talking about some object classes not having handles is
paf> not a good thing.
Even with domain names---which are unique at a given instant of
time---I think there's a case for a more abstract handle.
For example suppose foo.com is registered for a while, then it goes
dormant, but is then registered again. In some circumstances it's
better that the handles on the two instances are different.
Of course this is very much an issue for individual registries: I only
make the point because I agree with Patrick that _every_ object should
have a handle.
Cheers,
--
Martin Oldfield,
AdamsNames Ltd.