Replacing the "Big Bang" with a self-filtering timeline

[Werner Staub <werner@xxxxxxxx>]

The main cause of opposition to, and delays in, the new gTLD
program is the currently proposed "big bang" approach: all
the changes are supposed to take place at the same time,
whether they are important or not, whether they have consensus
or not, whether they take time to prepare or not, whether they
affect everyone or just a specific sector or community.

The "big bang" can be replaced by a self-filtering and
self-prioritizing timeline, if ICANN:

 1) defines multiple application windows for each round;

 2) enables affected third parties to object to a given TLD
    application for "Excessive External Costs";

 3) stipulates that, as long as there is opposition to an
    application, additional applications for the same string
    may be submitted in the subsequent application window.

For instance, external costs arise in the management of a TLD if
it causes large-scale cybersquatting or typo-squatting or imposes
a disproportionate burden for defensive registrations. The ability
to object for "Excessive External Costs" should come in addition
to the ability to object on "Community" grounds, on "Morality and
Public Order" and on "Rights Infringement".

The objection mechanisms combined with multiple application
windows enable applicants so select the appropriate time for their
proposals. For instance, it takes more time to devise appropriate
defenses against large-scale cybersquatting for large,
unrestricted TLDs than it does for community-based TLDs. If the
process has the features described above, community-based TLDS can
apply in an early application window and will not face opposition
for "Excessive External Costs" (or any of the other grounds).
Applicants for large TLDs will go ahead as soon as they have
satisfactory system of to avoid external costs.

(The concept of external costs is well-established in economic
theory. Moreover, governmental approval processed often use
objection mechanisms based on proof of excessive external costs,
such as pollution, noise or environmental overload. In all cases
it takes judgment and expert knowledge to make a determination.
By making the analysis objection-based, expensive evaluation can
be avoided for projects that nobody expects to cause problems.
As all human activity has external costs, the question is whether
an excessive burden is imposed on affected third parties.)

It works even better if there are graceful withdrawal terms for
any application, so that even objection can made unnecessary. (The
bottom-line cost of withdrawing is still excessive in the current
Draft Applicant Guidebook.)

The self-filtering timeline has the advantage of giving the right
incentives to all the parties. It avoids pushing trademark holders
towards opposition against unproblematic TLDs. And it pushes TLD
applicants towards preparing quality applications that minimize
external costs.

I suggest that each round could have 3 or 4 application windows.
We are currently preparing Round 3. (Round 1 was in 2000 and Round
2 in 2004.) Round 3 could run from October 2009 through October
2011 and have 4 application windows, one every 6 months. After
that, Round 4 could start October 2012, again with multiple
application windows, and possibly modified terms, based on the
experience of Round 3.


Werner Staub