Re: [wg-c] Separating the Sheep from the Goats

["Kevin J. Connolly" <CONNOLLK@rspab.com>]

>>>> Milton Mueller <mueller@syr.edu> 08/20/99 06:21PM wrote:

>Excellent start!
>
>A few comments:
>
>Kevin J. Connolly wrote:
>
>> Three-phase rollout:
>> (A) Proof of concept phase.
>> Anticipated duration: six weeks
>> Domains to be added: .per, .nom, and at least one other non-commercial
>> domain.
>>
>> Registries: Iperdome, CORE, and one other (ideally, a for-profit registry allowing
>> multiple registrars write access to the zone file).
>
>Beautiful. Who can object? The first two are the safest TLDs
>a TM owner could imagine and we agree that three new registries
>should be authorized. Can you clarify what you mean by a "non-
>commercial" domain name?
>

At the time, I was thinking of something equivalent to .per/.nom, but
on reflection over the weekend, I now think we should look for something
innocuous but for which there is an expressed interest.

Perhaps .mus, a TLD for museums.  Or perhaps Alternic will propose something.
(Query: are any Alterniks on the WG-C list?)  I specifically wish to stay away from
a CORE domain.  I expect enough problems as it is from suggesting particular
operators/registries, even for something as innocuous as a six-week proof of concept
trial.

>> (B) Controlled RollOut Phase
>> Anticipated Duration: Six Months
>> Domains to be added: Those elected by the qualifying registries (not more than
>> three each in addition to the proof of concept domains), subject to certain
>> qualifications:
>
>I like the 9 registries, and I like the registry selection of the name.
>Do the 9 have to be different from the first 3, or can any of the
>first 3 propose to add gTLDs?
>

I was thinking that the first three would be invited to qualify to continue
during the Second Phase.  I'm still thinking that just because ICANN invites
a registry to participate in the proof of concept phase should have no bearing
on that registry continuing to operate that domain name during or after Phase Two.
If they qualify, then the POC registries would get to propose additional gTLDs
just as the other registries could.


>> (1) The registry must articulate a plan for preventing it from reaping monopoly
>> profits during the controlled rollout phase.  These could include (but other
>> possibilities are invited) (a) CORE-paradigm shared registration with registry
>> operated on strict cost-recovery regime, (b) Informal tariff-filing under which
>> compensation of registry/registrar is fixed during the Controlled RollOut Phase.
>
>If we are talking about six months, and the ICANN plan to
>continue adding registries and TLDs (absent problems) is made clear,
>I am not very concerned about monopoly profits. New registrants can
>wait six months, or they can register under a shared-registry com, net or
>org, which should have competitive prices.

Okay, let me refine my concern.  First of all, I defer to economists and like-
minded participants whose belief is that a six-month window does not raise
the spectre of monopoly profits.

But what if ICANN takes a decision during the controlled rollout phase to
stop the music?  Then we will have a situation in which we still have an
artificially-small number of gTLDs, making it possible for the favored registries
to realize monopoly profits.  This is a doubly-bad outcome: the process of
adding registries comes to a screeching halt, and the registry operators (who
may have had something to do with the troubles that inspire ICANN to apply
the brakes) are in control of the lion's share of a scarce resource.  

SO . . . I still think it's important that the registries who participate in the
controlled rollout phase lay out a proposal for preventing them from 
profiting from an unreasonable restraint of trade.  Perhaps there's no need
for the antimonopoly plan to operate from Day 1, so long as it springs up if
the rollout is stopped or deferred prior to <LARGENUM> of TLDs being
added to the root.

>
>> (2) The TLD must not be one which is the same as or confusingly similar to any
>> famous trademark, unless the holder of the trademark assents irrevocably to
>> the delegation of the TLD.
>
>Of course. This would be illegal regardless of what ICANN does.

I'm not focussing on illegality.  I'm trying to make clear that ICANN policies
are and will continue to be trademark-friendly.

>
>> (3) In the event of a conflict over delegation of TLDs (e.g., .web) the TLD will
>> not be delegated unless the ICANN Board determines that the adverse claims
>> to the TLD are not bona fide or otherwise without merit.  Note that California
>> has an expedited method for court review of the determinations of corporate
>> boards.  There is one legal trick to making this work: ICANN will need to waive,
>> by amendment to its certificate of incorporation, the right to assert that the
>> decision to add a TLD to the root falls within its exclusive business judgment.
>> This change will make it possible for adverse claimants actually to be heard in
>> court on the actual merits of their claims to the TLD in question.
>
>I don't understand the above. It seems that resolution of the conflict over .web would occur in the course of application or, as I suspect,
>be worked out among the competing applicants.
>
The crux language is: "the TLD will not be delegated."  The rest of the
provision is stuffing.  Ideally, the claimants will resolve their claims privately.  It
remains necessary for ICANN to have the reserved power to blow off
claims that are patently bogus.  E.g., INEG might assert that it has prior
rights in .info, .sex, .yada.  I do not think ICANN would be foolish enough
to try to resolve bona fide but conflicting claims to a TLD.

>> (4) All domains delegated prior to the full competition phase are subject to
>> reallocation and re-award on the basis prevailing during the full competition
>> phase.
>
>Sounds good
>
>> Number of Registries: up to nine.  Qualification of registries to be based on
>> stable, objective criteria established at the outset of the program.
>>
>> (C) First Stable Plateau
>>
>> This is the hard part.  Most of the problem will evaporate once entry into the
>> domain name business becomes unrestricted.  For the most part, annual
>> redelegation of domains based on bids will not be necessary, because a new
>> registry can simply get a new TLD.
>>
>> However, there are always going to be some uniquely hot domains. .xxx, .sex,
>> and .law come to mind most immediately.  These domains will generate
>> monopoly profits for their operators, and some of us have a lingering discomfort
>> about allocating this kind of economic power on a lottery or first-come basis.
>>
>> One idea: if a registry is operated on a strict cost recovery basis and allows
>> write access to registrars without unreasonable restraint (and I reiterate that
>> CORE does not meet this criterion because the price of the admission ticket to
>> the club is unreasonably high) then the registry can keep the domain as long as
>> it satisfies technical criteria and it remains subject to periodic audit (at the
>> expense of the registry) to keep it honest.  If the registry is operated on a for-profit basis, then the domain will be re-awarded periodically on the basis of a
>> bidding process.  What the period should be, we can work on (I'm thinking
>> annual).
>
>I think annual is a bit short

I know.  I hope that the time period between revalidations is something
on which the WG will produce, if not consensus, at least a measurable
central tendency.

>
>> How we select the winning bidder is also debatable.  I'm thinking that ICANN
>> should take into account not only the license fee proposed by the bidder for
>> being delegated the domain, but also what commitment the registry makes with
>> respect to end-user fees.  If the domain is reasonably expected to capture
>> 1,000,000 SLD registrations per annum and the market will bear a charge of
>> $50.00 per name-year, a bidder who offers $10MM up front should be treated
>> the same as one who offers $5MM up front and agrees to limit charges to
>> $45.I00 per name-year.
>
>I think ICANN needs to stay far, far away from trying to be an economic regulator.

I agree entirely. However, other participants in the discussion have floated the idea that
a "winning bid" should be selected, not only on the "price" offered for the delegation of 
the TLD but also how much the candidate proposes to charge for SLD name delegation.
So, this is not a matter of economic regulation as it is a matter of establishing criteria
for delegating TLDs that depend on something other than brute economic force.  

There is something else I'd like to clarify here: I selected the thread title "separating the
sheep from the goats" to make it clear that I do not expect these proposals to win rough
consensus support by the WG.  The very nature of the WG makes it nigh well inconceivable
that rough consensus will be achieved.  My objective is to identify what I think is a workable
policy proposal, to take comments from those who are inclined to offer constructive 
comments, and to consign the obstructors to the floor of the nethermost cyberhell :->

As usual, please disregard the silly trailer, which is an artifact of this mail client and
my IT director's judgment.

KJC 
**********************************************************************
The information contained in this electronic message is confidential
and is or may be protected by the attorney-client privilege, the work
product doctrine, joint defense privileges, trade secret protections,
and/or other applicable protections from disclosure.  If the reader of
this message is not the intended recipient, you are hereby notified
that any use, dissemination, distribution or reproduction of this com-
munication is strictly prohibited.  If you have received this communi-
cation in error, please immediately notify us by calling our Help Desk
at 212-541-2000 ext.3314, or by e-mail to helpdesk@rspab.com
**********************************************************************