Monday, August 16, 1999, 9:09:11 PM, John Charles Broomfield
<[EMAIL PROTECTED]> wrote:
> William Walsh wrote:
>> Monday, August 16, 1999, 4:15:08 PM, Dave Crocker <[EMAIL PROTECTED]> wrote:
>> > There are basic differences between a non-profit consortium that has
>> > oversight, versus a commercial monopoly with none.
>>
>> This ignores one of the most basic economic principles.
>> Non-profit management provide no incentive for innovation and
>> improvement of services.
> Are you arguing for a commercial, money-making ICANN? I don't think that the
> USG would recognise such a beast as NewCo.
>> They use the "monopoly" or "lock-in" argument to justify it.
>>
>> This is simply faulty logic. Every single one of the possible abuses
>> they note as justification for these arguments can be answered with
>> simple, and reasonable, contractual obligations.
>>
>> This reasonable requirements can include such things as :
>>
>> 1) Restrictions on price increases, both in frequency and in amount.
>>
>> 2) Restrictions on IP assignment, requiring that should the registry
>> become insolvent or unable/unwilling to continue operating the TLD, or
>> should they default on their obligations under the contract, the
>> registry agrees the TLD shall be consigned to ICANN for reassignment,
>> along with ALL Associated IP, including the whois database.
>>
>> 3) Reasonable performance bond, to assist in funding the operation of
>> required services of the TLD while a new registry is bidded.
>> (Estimates place this at $10k to $25k US bond).
>>
>> 4) Requirement that whois information be available according to
>> uniform standards.
>>
>> They can go on and include such things as customer satisfaction
>> standards (based on a variety of factors), etc.
> If you go fully with what you are advocating above, then I think you are
> much more in agreement with Dave than you actually think. If you take the
> logic to it's full extent, and you REALLY want to provide competition, then
> every X months/years you rebid the registry operation. After all, it is a
> great method to ensure that:
This is absolutely too far, and overly restrictive. This is the point
where the line gets crossed. If the registry is meeting the
requirements, then there is no need to rebid it.
> -the model that ICANN is basing it's assumptions on is neither too lax (in
> which case there will be registries who turn up and say they can offer the
> same service at a cheaper rate), nor too constrained (in which case ICANN
> would be forewarned that the current registry would NOT want to continue
> running the system as-is, in which case ICANN would be well-advised to
> review the conditions).
> -competition CAN exist (because registry-companies will want to bid for
> running registries, more so if there are slots out there).
This plan would deprive a company of the benefit of continuing to
operate that which it has built up over time. It is ALWAYS easier and
cheaper for a company to come in AFTER the fact and run things than it
is to develop and innovate the registry.
There is no sound basis for this requirement, as there are for the
others.
> -protection exists by having them have a compatible way of transmitting the
> TLD database between themselves to protect from disaster.
>> There is no justification to limit the diversity of economic and
>> competitive models provided that the concerns regarding such issues as
>> monopoly, competition, lockin, etc, can be addressed with simple
>> contractual obligations. This will also limit the number of
>> registries, as not everyone will be able, or willing, to meet the
>> reasonable requirements, that will insure the registry is able to
>> operate a stable operation.
>>
>> Instead the spread alarmist comments about the dire consequences of
>> commercial registries. Remember, these "consequences" are easily
>> addresses and enforced. It is this task that ICANN is charged with.
>> These details come under the "Technical details" ICANN has been
>> charged with.
> I don't think anyone wants ICANN to run a registry. Rather, ICANN wants to
> bid them out under strict contractual obligations, as opposed to "giving"
> TLDs away.
I never advocated ICANN "running" a registry. Contractual
requirements should not be "strict," they should be ONLY has strict as
absolutely NECESSARY.
>> In the absence of real quantifiable harms, the broadest possible
>> spectrum of models and operations should be embraced. This has LONG
>> been shown to be to the benefit of the market and the consumers.
>>
>> Of course, the CORE supporters have a STRONG vested interest in
>> Registries NOT being able to offer services directly in a commercial
>> format, so that they can control the "registrar" industry, and thus
>> prevent the diversity of market that brings such strength to so many
>> broad areas of industries.
>>
>> Regulation must be justified, and the minimal amount of regulation
>> used as necessary to protect the market and consumers.
>>
>> This type of a proposal clearly is much less restrictive, and still
>> provides more than adequate protection, despite the alarmist rhetoric
>> engaged in by the "unbiased" CORE supporters.
> Would you mind enumerating a single reason as to why it would be a BAD idea
> to rebid the registries on a regular basis (apart from it being a bad idea
> for the company that got it initially, which -as we can see with NSI- will
> probably want to make everyone think that they got something for eternity).
You are shifting the burden of proof. Requirements must be justified.
To justify the requirement, certain obligations must be met.
Let me spell them out for you again (As I have many times before) and
then you can meet your burden of telling me how such a requirement
meets those standards.
1) Significance - The issue the requirement addresses must be a
quantifiable harm. Not adopting the restriction must have a real
quantifiable harm associated with it.
2) Inherency. The "harm" associated above would not be addresses by
other requirements, or by the other related "governance" (in other
words, the inherent system must not already be solving the harm).
3) Solvency. The proposed requirement has to be shown to actually
SOLVE the problem addressed in #1. There needs to be quantifiable
proof that such a proposal DOES indeed provide a workable and
enforceable solution to #1.
4) Ads/Disads weight - The advantages to the plan must exceed the
weight of the disadvantages to the plan to a significant degree.
5) The entire matter from #1-4 must be "Topical" and in the area of
control of the body which is charged with adopting the proposed
requirements.
In the absence of meeting such a burden for a proposed RESTRICTION,
the restriction should not be adopted. The proposed restrictions I
noted can meet that burden sufficiently.
I've yet to see anyone be able to quantify the mandatory rebid rule as
meeting that need.
Indeed I see PLENTY of problems. There is no incentive to a company
to innovate or guarantee service since they KNOW they are likely to
loose the contract in a short period of time. The only motivation
they have is to try and see how much they can milk from the registry
in the short time it is theirs, rather than working and innovating for
the long term that they have to make sure they meet their contractual
requirements, and continue to be able to compete in an open and
competitive fashion.
--
William X. Walsh - DSo Internet Services
Email: [EMAIL PROTECTED] Fax:(209) 671-7934
Editor of http://www.dnspolicy.com/
(IDNO MEMBER)
Support the Cyberspace Association, the
constituency of Individual Domain Name Owners
http://www.idno.org
