[ga] Cui bono? Possible Mission Statements.

[Thomas Roessler <roessler@does-not-exist.org>]

I will present two very high-level mission statements for ICANN. For 
each of these mission statements, I'll try to give a brief overview  
of what such an ICANN could and should do. In doing this, I'll  
ignore most of the input side (internal process, board selection,  
...), but concentrate on the output and enforcement side. In each  
case, I'll look at three reference problems in order to better  
understand what precisely the respective mission statements mean:  
(1) The introduction of new TLDs. (2) Negotiating or entering into  
new contracts (or contract changes, such as WLS) with registrars or  
registries. (3) Working out a Dispute Resolution Policy.

Of course, most of the discussion will remain on or close to the  
surface of the deeper problems which undoubtedly occur in the  
implementation of each proposal. Still, I hope that this may be  
helpful in order to better understand what's at the heart of quite a 
bit of the current debate.



The proposed mission statements: ICANN's mission is to coordinate  
technical and policy functions of the domain name system in order to 
promote a stable and secure domain name system...

	... for the benefit of the public.

	... as a cooperative effort of registries, registrars, and 
	registrants, for their mutual benefit.

Both mission statements are strongly influenced by the suggestions  
made by various constituencies' representatives during some Names  
Council telephone conferences on ICANN structure.



1. "... for the benefit of the public." This is the broader of the  
mission statements discussed. The core value is neither the benefit  
of users or registrants, nor the benefit of registrars, nor the one  
of registries: It is the public benefit, the common good (quoted  
rather frequently throughout the Evolution & Reform Committee's  
recent output). An ICANN with this mission statement is responsible 
for making sure that the DNS is operated for the benefit of the 
public at large. An ICANN of this kind has to take into account  
considerations which traditionally belong into the realm of nation  
states' authority: Anti-trust considerations, intellectual property  
interests, free speech are just examples of the scope of this ICANN. 
Some may call this mission creep; others would talk about successful 
privatization of the regulation of the Internet (and still others  
would argue that this isn't private-sector any more) - an ICANN like 
the one described by this mission would indeed be a regulator, not a 
self-regulator.

A key value of such an ICANN would be independence: The board would  
have to be able to operate by its own judgment of what the public  
good is, without having to pay attention to pressure generated by  
vested interests. As a corollary, such an ICANN would require public 
trust and accountability on a global basis - trust from suppliers,  
 from end users, and from governments, and accountability to its key  
stakeholder, the public at large. It would have to be funded in a  
way which makes it impossible for vested interests to exercise  
pressure on the board's decision-making.

In short, it would look like a government, or an association of 
governments.


Looking at the reference problems mentioned above, it is fairly  
obvious that attacking these problems would be natural and possible  
for an ICANN truly operating in the benefit of the public: It would  
have the independence needed to come up with balanced, reasonable,  
and practicable policies for the introduction of new TLDs, without  
paying attention to the interests of existing registrars and  
registries who may fear the newly added competition. It would have  
the weight needed to protect the public benefit in contract  
negotiations with registrars and registries, by, for instance,  
paying attention to anti-trust matters. Similarly, it would also be  
the appropriate body to work out a dispute resolution policy.



2. "... as a cooperative effort of registries, registrars, and  
registrants, for their mutual benefit." The ICANN described in these 
words ("ICANN.coop"?) is the servant of the three groups of  
stakeholders mentioned; it's the ICANN of Susan Crawford and David  
Johnson. It can only act in situations in which all stakeholder  
groups profit from ICANN's activities. These situations are mostly  
limited to technical coordination. A cooperative ICANN could not  
play an active role in policy-making, except as a facilitator (or  
secretariat) for discussions which could (and would) also happen  
elsewhere.  Such discussions would obviously be managed and driven  
by those most affected by the results - if that wording sounds  
familiar, you're right: It comes from a gTLD registry contribution 
to a Names Council call, just like the mission statement.


The "cooperative effort" ICANN would not be the appropriate place to 
work out policies for the addition of new TLDs: The addition of more 
competition may be against the interests of at least existing  
registries and registrars, unless there are strong incentives for  
these parties to enable such competition (like, for instance, close  
regulatory oversight from some national government interested in  
privatizing the DNS, and introducing competition into the business). 
However, the "cooperative effort" ICANN could of course administer 
policies worked out elsewhere.


The "cooperative effort" ICANN would also not be the appropriate  
partner for working out new contracts with registrars and  
registries: It would find itself in a conflict of interests,  
negotiating with some of the very stakeholders for whose benefit it  
should act.  In short, it would be in the middle of a rather bad  
exercise of self-dealing by registries and registrars.


Finally, under the assumption that a dispute resolution policy is 
beneficial for all three groups of stakeholders, this would be the 
only one of the reference problems which could appropriately be 
handled by ICANN.coop. However, it is not clear that this assumption 
is valid: Registrars and registries may have an economical interest 
in a dispute resolution policy which removes the threat of 
litigation from them; on the other hand, current registrants may 
possibly be better served by use of the traditional legal system as 
a tool for the resolution of disputes over domain names.

In any event, the first two examples already provide evidence that 
ICANN.coop could not be a tool for the full privatization of the 
domain name system. Key policies would have to be worked out with 
the public interest in mind; they would have to take, for instance, 
anti-trust and similar considerations into account. 

Key contracts would have to be negotiated between registrars and  
registries on the one side of the table, and a sufficiently strong  
regulator on the other side. This could certainly be one government, 
any collection of governments, or some international treaty  
organization. The corporation itself would not be the adequate place 
for these activities.


Making an ICANN like the one described by this second mission  
statement the contract partner of global registries is, for this 
reason, the best way to remove public interest considerations from 
the regulatory environment for much of the gTLD registries' 
activities: It means that ICANN is given the duties of an 
international regulator, but not its rights.


In short, it's a rather brilliant move if you are a gTLD registry  
and want to get rid of regulatory constraints for some time.


Regards,
-- 
Thomas Roessler                          http://log.does-not-exist.org/
--
This message was passed to you via the ga@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga" in the body of the message).
Archives at http://www.dnso.org/archives.html