Let me make a few comments on the "mission and core value" paper
which I found, quite frankly, a bit disappointing: I believe that at
least those core values which talk about ICANN's "benefit" give in
to the temptation of saying something soft and useless where a hard
statement would make dissent in the community visible. I had hoped
that I would find the time to produce something more elaborate than
this, but since it doesn't look like I'll be able to find the time
for that, let me just go through a few points which I found
particularly concerning.
Core value (b), for instance, talks about "limiting ICANN's
activities to those matters within ICANN's mission requiring or
significantly benefiting from global coordination." That doesn't
make much sense: _Matters_ hardly benefit from global coordination -
market players do. Trying to add any flesh to this "core value"
immediately leads to vastly different conclusions about what ICANN
should or should not be - this "core vlaue" covers about everything
from the gTLDs' cooperative ICANN idea, to a strong regulator with
government-like powers. I could imagine that adding some more meat
to this recommendation would have helped to provoke clearer input
from readers.
Core value (f) has a similar problem, where it talks about
introducing and promoting competition in the registration of domain
names where practicable and benficial. Once again, beneficial to
whom? To existing registrars? To existing registries? To trade
mark holders and their lobby? To the public? I have to admit that
the consideration of an unspecified "benefit" in conjunction with
competition and its limitations is something which makes me
extremely nervous, in particular when I see some rather powerful and
active players which could benefit quite a bit (at least short-term)
from _less_ competition, which would, however, most likely not be
in the interest of the public, of consumers, and of possible future
competitors.
Core values (h) and (j) seem to contradict each other: (h) talks
about "open and transparent policy-making," while (j) seems to, in
effect, restrict it to input "from those most affected." That
particular wording was suggested by the gTLDs during a names council
telephone conference in the context of a policy-making
recommendation, and I can easily imagine how they'd interpret it:
Namely, as meaning that "those most affected" are _always_ the ones
immediately bound by any policy (registrars, registries), as opposed
to, for instance, registrants which would only be bound indirectly.
In other words, recommendation (j) is a blanco cheque for dismissing
input from affected parties by claiming that they are not "most
affected."
Core value (l) is also interesting when it talks about "sensitivity
to governmental concerns for the public interest so that the need
for direct governmental action is minimized."
I think I can offer two conflicting interpretations of this which
give reason for opposite kinds of concerns.
On the one hand, the sentence may be read as trying to reduce
ICANN's responsibility towards the public interest - if you want a
caricature of this, it's "just do as much as necessary in order to
keep governments out, and otherwise do what Verisign wants."
On the other hand, the word "minimized" could give reason to the
opposite interpretation: ICANN as a "world government" which wants
to take over as many competences as possible from existing
governments, without paying any attention to whether or not
something would better (for whatever definition of "better" is
appropriate) be done by local governments.
I believe that ICANN's core values should not be that vague, but
should instead try to provide ICANN itself and the community with a
clear guidance concerning the benficaries of ICANN's work - that is,
with clear guidance on what to expect from ICANN, and what not to
expect. In particular, I could imagine that it would be extremely
helpful if these core values could be consulted for guidance on
whose benefit should have the highest priority in ICANN's
operations. Core value (a) gives some such guidance, but I believe
that this guidance is far too limited.
--
Thomas Roessler <roessler@does-not-exist.org>
Please find attached some thoughts I (mostly) wrote down several
weeks ago, but did not publish so far.
--
Thomas Roessler <roessler@does-not-exist.org>
Title:
Some Remarks on
ICANN Mission and Structure
|
The present document is
a write-up of some thinking I did in late April and early May, before the
E&R Committee published any of its discussion papers, but after the
Names Council conference call during which it became painfully obvious that
the DNSO's constituencies do not agree on some fundamental principles of
ICANN structure.
Starting with the assumptions in section 1, I ask what conclusions follow
for ICANN strcuture, and for public participation in policy-making?
If you feel provoked by this, that's fine. I'm looking forward to answers
and comments.
|
1. The remarks in the present document are mainly
aimed at ICANN's mission and structure with respect to gTLD domain name
policy. Some fundamental observations which will be used throughout the
argument: (1) Registries and registrars (who are the ones in direct contractual
relationships with ICANN) are commercial entities; their behavior is motivated
by economic gain. In fact, some of the players can not even be expected
to behave ethically, or to respect any rules when breaking them seems economically
attractive. (2) Economical considerations have to take into account costs
which can be generated by the legal environment (legislation, lawsuits,
...). (3) Some groups of players are capable to change the legal environment,
or to use it otherwise in order to generate cost. The IP Constituency is
an example of such a group. Some civil liberty groups may also be capable
of this, but don't seem to use this capability. (4) Other groups do not
currently have these capabilities.
2. Policy-making. ICANN policy-making
in the domain name area is (as far as "consensus policies" as defined by
the various contracts are concerned) inherently limited to those cases in
which abiding by policy - once established - is economically attractive for
registries and registrars. The most obvious situations of course cover technical
coordination, which can help to avoid unnecessary cost, and any other kind
of "win-win" situation. It should be noted that self-regulatory efforts
of registrars and registries could be expected in these cases even without
establishing an entity like ICANN, since such efforts would be in the economical
interest of those directly bound by such policies.
A slightly less obvious situation occurs when
policy is pushed by a "type (3)" group which can make a credible threat
of changing (or using) the legal environment in order to create costs.
One example of this kind of policy is the UDRP, where registries and registrars
can reduce litigation risks and costs (and, possibly, risks of unpleasant
legislation due to lobbying efforts of the IP community) by abiding by the
policy. "Type (3) policy" is the area where the risks of mission
creep are largest, since economical incentives for further policies can be
artificially created. Since there is an economical incentive for registries
and registrars to follow the wishes of type (3) groups, other groups are
unlikely to have a chance to change the outcome of policy development.
As a corollary, explicit limits to ICANN policy-making are
currently most urgently needed with respect to all kinds of Intellectual
Property aspects of the DNS, and similarly with respect to any other aspect
which attracts (or may attract) powerful advocacy organizations.
Note: The risk with such limits is, of course, that policy-making
may happen in an uncontrolled environment when there is actually a win-win
situation. From this point of view, ICANN mission creep may, in fact, be
the lesser of two evils.
Note 2: Taking this argument another step further, the
above note is, of course, based on the assumption that using ICANN as a
policy-making framework can actually change anything. This is far from
obvious, in particular with a "truly consensus-based" ICANN.
3. The role of the "public interest." When
ICANN community members talk about the "public interest," two different
things may be meant. (1) Groups may try to push various kinds of political
agendas in ICANN policy-making, by trying to act as type (3) players [most
of these activities don't seem to have been too successful; it seems that
the really powerful players hardly showed up]. (2) Groups complaining about
the activities of existing type (3) players, which are not sufficiently
balanced.
The first case bears a considerable risk of
mission creep, and should be treated like any other type (3) player: Ignore
them as long as they aren't credible, and try to satisfy them to the extent
necessary if they can credibly threaten to create additional costs to registrars
and registries.
The second case is, in a certain sense, at the core
of ICANN's mission: Instead of leaving policy-making to registries, registrars,
and those who put political and economical pressure on them, ICANN's role
should be to ensure that the policies which evolve do not violate users'
or the public's interests.
Of course, both kinds of public interest activities
can be exercised by the same groups. Public interest groups which are strong
enough to balance other type (3) players would certainly make ICANN's mission
of balancing interests easier. However, it seems that these groups have not
been participating in actual policy-development.
4. The role of an at-large membership.
From an economical point of view, an at-large membership (or other individual
users' representatives) can be safely ignored during policy-making, unless
there are type (3) players supporting their activities. It seems that,
to some extent, certain governments are (or have been) this supporting player.
As a conclusion, an at-large membership and at-large elections to the
board will always depend on governments' (or strong advocacy groups') willingness
to put pressure onto the rest of the ICANN community, and to back at-large
activities. It seems plausible that such pressure may be exercised by governments
in favor of a certain model of board member selection. It seems much less
plausible that it would be used to back policy-making activities of an at-large
membership on a policy-by-policy basis, unless governments or other strong
advocacy groups are already directly involved with policy-making. One may
argue that governments or other strong public interest advocates may back
an at-large membership's role in policy-making in the most drastic cases
of abusive policy-making. However, in these cases, an intervention would
most likely be inevitable even without an at-large membership. As a conclusion,
an at-large membership is not suitable as a tool to represent the public
interest in policy-making: It could only achieve this goal when there are
other, strong public-interest representatives involved with the process.
5. The role of governments. Obviously,
governments (or, more precisely, nation states) are the ultimate type (3)
player. For the sake of argument, it seems reasonable to assume that their
ultimate objective is to make sure that the interests of the public at large
are respected. In the ICANN process, this means that there are various
roles governments can play; these roles are not mutually exclusive: (1)
Governments could back various type (3) players to the extent necessary
in order to balance these players. With this approach, there would at least
be a possibility to keep ICANN's mission bounded. (2) Governments could
get involved as a player themselves. The GAC has demonstrated the latter
option in the case of geographical names in .info. This kind of governmental
involvement maximizes the danger of mission creep.
6. Extreme proposals. The Lynn proposal's
idea to involve governments as directly as possible is attractive if ICANN
is supposed to act as an effective regulator which can -- if necessary --
impose policy on registries and registrars. An ICANN of this kind could
address public interest concerns such as privacy or consumer protection directly,
and may indeed lead to balanced policies. However, it would also bear the
greatest risk of mission creep.
On the other end of the spectrum, the gTLD
registry constituency's proposal that ICANN should be a cooperative effort
of registries, registrars, and registrants, minimizes some of the risks
of mission creep, but may lead to unbalanced policies in those areas where
existing type (3) groups which can't be avoided are active (think about
the IP community, if you like). Consumer protection and privacy concerns,
however, may have to be addressed elsewhere.
Note: As stated in the note in section
2 above, it is not obvious that such a consensus-based ICANN would actually
lead to any change when compared to a mostly unregulated market which is
solely governed by economical interests.
7. Current structure. The current
structure is a compromise between these somewhat extreme proposals. However,
it lacks a balance of powers between special interest groups. This lack
is, to a certain extent, caused by a lack of involvement of existing consumer
rights and advocacy groups with policy development, and the ICANN process
in general.
8. Conclusion. Without the participation of
powerful public-interest groups in ICANN process and policy-making,
any compromise between the extreme proposals described above will fail.
On the other hand, both of the extreme proposals would yield an effective
policy-making instance, for their inherent definition of "effective."