[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]


To: "Liu, Hong" <Hong.Liu@neustar.biz>
cc: ietf-provreg@cafax.se, brunner@nic-naa.net
From: Eric Brunner-Williams in Portland Maine <brunner@nic-naa.net>
Date: Tue, 26 Feb 2002 22:00:47 -0500
In-Reply-To: Your message of "Tue, 26 Feb 2002 16:49:43 CST." <23309E398D84D5119D0F00306E07513901181AEA@dc02.npac.com>
Sender: owner-ietf-provreg@cafax.se
Subject: Re: Reverse Lookup using IP Address in EPP?

>> Is "company Y" [and X] a registrar?
>
> Both X and Y are registrants.

I thought as much. Seng wanted to stretch the AAA envelope out to the edge
(registrants), cutting the registrars out of the AAA problem, and also out
of a constructive role in registry provisioning.

>>> ... how can the registry prevent IP address hijacking ...
>>
>> Let's clear up the first set of issues and identify who the actors are, then
>> see if this is an EPP issue, or a registry-on-the-brink-of-anarchy feature.
>
> I don't think the last part of the sentence is of relevance to this
> discussion.

OK. Then ".US"

It appears to me that you are trying to map EPP, which assumes some set of
boundary conditions, such as ICANN accredited, or similarly situated (read
"responsible") registrars, onto the .US booby hatch, where five fun lovin'
guys-n-gals "own" 40% of the metro areas.

You've got registrars (presumptive clue) bagging each other's registrants'
nameservers, and absent a pre-existing policy venue, you seem to be looking
for a technical mechanism to discover and/or cure bad acts. 

To stretch the authentication envelope out to the registrants, you may as
well revisit James' issues.

> If that was discussed before, I would like to know the result of the
> discussion. Any pointer would be appreciated.

   1  02/05 Patrick            Re: draft-hollenbeck-grrp-reqs-06 [Was Re: Inter
   2  02/05 "Brian W. Spolari  RE: draft-hollenbeck-grrp-reqs-06 [Was Re: Inter
   3  02/05 Patrick            Re: draft-hollenbeck-grrp-reqs-06 [Was Re: Inter
   4  02/06 Sheer El-Showk     RE: draft-hollenbeck-grrp-reqs-06 [Was Re: Inter
   5  02/07 "James Seng/Perso  Re: draft-hollenbeck-grrp-reqs-06 [Was Re: Inter
   6  02/07 Urs Eppenberger    Re: draft-hollenbeck-grrp-reqs-06 [Was Re: Inter
   7  02/07 Patrick            Re: draft-hollenbeck-grrp-reqs-06 [Was Re: Inter

All dates are 2001. I suggest a look at the London notes as well.

>>> The second question has to do with whois.
>>
>> This isn't an EPP issue.
> 
> I asked the question because whois was suggested by Scott and Patrik as
> out-of-band means to getting the job done. What I want to make sure is that
> whether ALL host objects created in the registry are available for query in
> the whois database.

It is a reasonable question to ask a registry operator -- is the data stream
sunk by a protocol A server provably consistent with and exhaustive of some
subsequent access method of some possibly other protocol B server.

EPP is protocol A. It isn't protocol B.

I know the .US roll-out has to do better -- a lot better -- than the .biz
roll-out. The $20M figure has been public for ages, hence the 1M registrant
requirement just to make debt. But this isn't this WG's problem, it isn't a
protocol problem. If the .US cyber-squatters are converted into registrars,
they win the business proposition, and that's fine. If the gTLD registrars
peel 1M off of .com, they win the churn, and that's fine too. Either way
(assumptions being true by desire alone) and a operator avoids failure.

But your registry's business model validation isn't this WG's problem. The
US, and a company, aren't the controlling interests today, just as they
weren't the controlling interests when it was another DoC problem and another
company.

Feel free to ignore the following:

	o go back to the req draft, and _liberally_ wack out anything that
	  looks and feels "normal" (.com aka "gTLD" mindset), then
	o summarize the actual situation of your ccTLD, then
	o pretend its the Falkland Islands, not the US, and try to find the
	  bits that are common to both the .US zoo and the Falkland sheep,
and then
	o conservatively extend EPP (or vary from EPP/RRP/SRS/...) to the
	  core problem you are trying to solve (which may be very close to
	  the problem James was trying to solve).
	o drop this WG a note that explains most (maybe not #2) of the above.
or
	o go forth and specify the non-EPP registry-registrar protocol for
	  .US and/or the Falkland Islands or where ever.

Cheers,
Eric

Home | Date list | Subject list