Re: [wg-c] registry contracts

[Milton Mueller <mueller@syr.edu>]

Kevin J. Connolly wrote:

> I recoil in horror at the thought of making ICANN or one of its satellites into an
> explicit tariff-receiving and rate-setting or -approving agency.

So do I. So do many people, including, I think, many of ICANN's Board.

I think Kevin J. Connolly is on the right track below. Although I have important
disagreements, his observations seem to be grounded in serious thought about how the
registry-registrar relationship would actually work operationally and contractually,
and I respect that kind of contribution. Let me set out my points of disagreement but
do not let the specific criticisms overshadow the sense that we are making progress (or
could be, if we continue in this vein)...

> 1.  A true non-profit registry should, when making its tender,

[snip]

> 2.  A for-profit registry, when making its tender, should simply

[snip]

> 3.  Zones held by non-profit registries should not be re-bid,

[snip]I don't agree with the idea of applying different standards to profit and
non-profit with respect to delegations. In both cases ICANN is offering to advertise a
zone file in its root; i.e., it is offering them both the same thing. If there is a
fee, both should pay it; if there is a periodic redelegation, both should have the same
period. By applying this principle, one avoids the problems K.C. astutely recognized in
#4 and #7 below.

I understand that it is attractive, from a social policy point of view, to "encourage"
non-profits and to "tax," or extract some of the value from, for-profits. However, I do
not wish to encourage the model of ICANN as a taxing authority with a
wealth-redistribution charge. If one believes that that should be done, that's for
governments to do. Nor do I think ICANN should be in the business of determining what
kinds of organizations are worthy of encouragement. I advocate a minimalist role for
ICANN. I think that view commands a consensus among the SOs, and especially among
IETF-ers who think seriously about the issue.

However, K.C. raises the valid question, what does ICANN do if confronted with
conflicting applications for the same name by registries operating along different
models?

I think that paying attention to precedent (FCFS) has both normative and practical
value. Auctions are another method. KJC's proposal of a public benefit calculus is
worth considering, too.

But the real answer is that this is, fortunately, a problem we do not need to solve
right now. Remember, we are just dealing with a testbed of 6-10 registries. In that
initial phase ICANN can choose among applications based upon the criteria of what makes
for a better experiment. It can also base its choices upon the principle of avoiding
avoidable short-term problems.

Long term, of course, it will have to deal with that issue more systematically. In
dealing with it, we need to keep in mind that we are *not* dealing with a scarce name
space. Individual strings are exclusive, of course, but the name space is not
particularly constraining.