One thing that ICANN clearly lacks is a set of well documented and often referenced founding principles. This leaves the awkward position where everyone who has been around since the beginning has a different position on what those principles should have been and all those that have joined later know that there is something fundamental missing.

The missing principle vexing me this week is that of fair competition. Even now, long after the gTLD vote, the argument still runs on as to whether all these new domains are needed or whether we have enough already, as if competition simply did not matter. Thankfully some of those making the decision held true to this missing principle, as Steve Crocker made clear in his comments during the vote:

The case for new TLDs is a little harder to pin down. But one of the most important principles in the creation of the Internet from a very long time ago was not to stifle or prejudge what the paths for innovation are. So the default has to be that, absent a strong case that such things will cause harm, we must move forward. And I strongly support this.

The nearest ICANN has to a principle for this is in its core values where an oddly worded value relating to competition is enshrined as follows:

5. Where feasible and appropriate, depending on market mechanisms to promote and sustain a competitive environment.

A better way to do it is the way InternetNZ, the membership society that owns my company, has with the first object of the society:

To promote the competitive provision of Internet access, services and facilities in an open and uncaptureable environment.

If ICANN had a similar principle that stated that part of its role is to promote fair competition then not only could good decisions be strengthened, but bad decisions could be prevented. Here are three recent examples:

The first and scariest emerged at the public forum at ICANN Dakar where the call came for ICANN to contractually prevent accredited registrars from selling domains from alternate roots. Thankfully, sensible voices on the board declined to do this on the grounds that it would be strongly anti-competitive and out of scope for ICANN. Unfortunately I’m not sure those that raised this terrible idea will be drop it and it may even gain ground on the back of “protecting the consumer”. Whatever the outcome, the issue remains that some people think that ICANN should be eliminating competition and entrenching monopolies, rather than the opposite which the sane universe believes.

At the same time we have others misusing the word “monopoly” and aiming it to discredit a TLD, such as in the recent action against .xxx. Unfortunately this view is widespread and I’ve even heard peers describe their ccTLD as a monopoly, which often draws a strong reaction from the audience. As anyone who has thought about it knows, a TLD can only ever be a strong brand and if a consumer wants that brand then there is only supplier, but if they want a domain name then there are many brands and suppliers to choose from.

Finally, monopoly power can unfortunately be quite subtle in the ways it manifests itself and one aspect of the new gTLD process that is rarely mentioned highlights that. At the moment we have approximately 280 TLDs under approximately 250 different policy regimes as each ccTLD sets its own policy. When 500 new gTLDs are added we will then have 780 TLDs under 250 different policy regimes - can you spot the problem?

Here at .nz our policy is determined by a non-governmental regulator independent of us in the registry (a model we have jointly recommended [PDF] to the NTIA for ICANN). The .nz policy is all openly published, all aspects of it are consulted upon regularly and all the workings of the regulator are open. We could argue down in the details about whether the .nz policy or the ICANN policy is better, but overall the two are comparable. Despite this, we can’t apply for an ASCII TLD that would use the .nz policy. We can apply for a single IDN ccTLD or, if we had a closed policy, we could try the community preference route in the new gTLD process, but for any ccTLD that runs an open, first-come-first-served TLD the road ends here. Subtly, almost without notice and even with the best of intentions, a monopoly is created.