The Wall between Registries and Registrars Comes Tumbling Down
Wherefore they are no more twain, but one flesh. What therefore God hath joined together, let not man put asunder. — Matthew 19:6
The ICANN Board has just voted to get rid of all co-ownership restrictions between registries and registrars. This is major news, with far-reaching consequences. ICANN’s announcement says:
“In the absence of existing policy or new bottom-up policy recommendations, the Board saw no rationale for placing restrictions on cross-ownership;” said Peter Dengate Thrush, Chairman of the Board. “Any possible abuses can be better addressed by properly targeted mechanisms. Co-ownership rules are not an optimal technique in this area.”
This is the only principled decision the ICANN Board could have come to, and they deserve a lot of credit for doing it. By “principled,” I mean taking ICANN’s stated institutional principles and following them to their logical conclusion.
The decision is a surprise (more below on that) and will fundamentally change and liberate what has become a stagnant commercial ecosystem. Restrictions put in place to break up the monopoly of Network Solutions had long since become a burdensome and pointless legacy. ICANN has restored the issue of competition to where it belongs by reserving to themselves the ability to “refer issues to relevant competition authorities.”
The decision has all kinds of consequences, most of them good:
- VeriSign or any other registry can now run a registrar (at least in theory), while GoDaddy and other registrars can run a registry.
- Brand owners that start branded TLDs won’t have to navigate a ridiculous obstacle course just to register their own names for themselves.
- Small cultural-linguistic TLDs will now be able to set up a public-facing registrar to service their customers in their own language, instead of going begging to registrars to carry their small-volume TLDs.
- Most important, it means that new TLDs will be able to market themselves. For many new gTLDs, this ruling will mean the difference between slow death and a profitable business.
This last critical point needs explanation. Any Internet business, which is most businesses these days, relies on being able to drive traffic to their site and converting a percentage of visitors into paying customers. Under the rules that separated registries from registrars, this would not have been allowed for new gTLDs, because registries would not have been allowed to sell directly to the public, and any marketing money spent on promoting their new gTLD would have had to be purely of the “awareness and branding” variety, which is very expensive, difficult to measure, and of uncertain effectiveness. The new ruling now makes it possible for new gTLDs to devise marketing plans that send customers directly to a website where they can actually buy the domain name being promoted. This change may well be the difference between life and death for many new gTLDs.
As for existing gTLDs, they will be able to play in this new reality, with some exceptions:
ICANN will permit existing registry operators to transition to the new form of registry agreement, except that additional conditions may be necessary and appropriate to address particular circumstances of established registries.
The decision was a surprise because betting men were giving odds that the Board would institute a cross-ownership restriction limiting registrars and registries to owning between 2% and 15% of each other. The two percent figure is what the Board voted for in March 2010, a vote widely understood to be a warning to the so-called ICANN community to reach an agreement on the issue. The fifteen percent figure is what had been lobbied for heavily by various powerful incumbent players, who buttressed their position with the argument that this was what had been done in the past. But the Board (rightly) cast that argument aside:
Whereas, historical contract prohibitions on registries acquiring registrars do not provide a compelling basis for principled decision-making….
The new landscape will require everyone in the domain name business to re-examine their business, their partners, their strategy. It will have consequences between those I enumerated above. It will re-invigorate the industry, and it will help establish the respect that ICANN has lacked for so long.
Those of us who have grown cynical from years of contrasting and comparing ICANN’s principles with its actions are feeling an uncomfortable sense of possibility — uncomfortable because in the past that feeling has always been a precursor to a cruel letdown. Let’s hope instead that this proper and even slightly courageous decision is actually a sign of ICANN growing up into the organization many of its founders hoped it could become. If so, this Board will be remembered as a group of people who loved the Internet and did what they could to help it prosper.
I would be remiss if I didn’t recall what was to me the spark for the spinefulness the Board has shown, and probably for the entire new gTLD process: the remarks by ex-Board member Susan Crawford during the debacle of the original vote on .XXX in March 2007, which she rightly called “weak and unprincipled” — using “unprincipled” in exactly the same sense that the current ICANN Board has acted in a principled way today. Both her stinging dissent and her later remarks set out all of the weaknesses and hypocrisy of the ICANN approach to gTLDs at that time. Her voice against ICANN’s culture of political meddling, censorship, ad-hoc policy-making, and craven concessions to intellectual property lobbyists was a lone one. But she said what many of us were thinking, and I think her cogency, reasonableness, and above all the plain-spoken rigor of her arguments was a spur to the reform of which we may finally be reaping the benefits. Thank you Susan.

[...] the pond, Minds + Machines CEO Antony Van Couvering quickly rattled off a typically eloquent blog post that focussed on what he seems to see as ICANN’s sudden spine growth: This is the only [...]
A major development to be sure.
Do you have any thoughts as to what sort of ‘properly targeted mechanisms’ might be put into place to address possible abuses? Specifically, will there be provisions to prevent monopolistic behavior? Will registries be required to allow qualified registrars to also retail their domains?
Hey Anthony, thanks for the news! Could you help me understand the practical implications of this?
For example, how exactly would this impact Verisign’s business?
I understand that now they could consider buying a registrar / “vertically-integrate” — but would they want to? And wouldn’t other registrars cry foul about the potential for Verisign to abuse their power and treat the in-house registrar preferentially?
On the other side of the coin, how could this decision hurt Verisign? What are the actual implications for the registry business– it’s not like Godaddy or some other registrar would become a “competing” registry, since the nature of the work requires you to be the exclusive control of the root name space… or am I missing something?
I see this more applicable to other gTLDs, and potentially for Verisign making an acquisition.
[...] To read the blog posting by Antony Van Couvering in full, see: http://www.mindsandmachines.com/2010/11/the-wall-between-registries-and-registrars-comes-tumbling-down/ [...]
Hi Antony,
A great article, and I totally agree with your position.
What I am currently worried about are other means of navigation than domains and the risk that poses to companies and individuals: Search, mobile apps and social all create their own monopolistic navigational structure, with so to say “Registrar” and “Registry” completely being equal. Plus, there are no checks and balances from a regulatory standpoint like ICANN.
I hope business owners and individuals alike will realize this and re-discover the true value of domain names in general.
Best,
Tim
P.S.: My full thoughts are under e.g. http://blog.ud.com/2010/09/navigation-nightmare-the-deadly-danger-of-facebook-twitter-google-and-the-iphone/
@thomas – Nothing in this resolution changes any of the equal access provisions for registrars — you can’t tell registrars they’re not allowed to sell your TLD. However, all the registrars are so focused on the “.com flow” that if you change that flow (for instance, requiring applicant check an extra box), chances are that many registrars will not want to implement that extra requirement, and so will not distribute your TLD. It all depends on how you want to distribute your TLD. Registrars are an excellent channel, especially if you are using a model that’s similar to .com. Some TLDs, though, will want to be single-sourced, and I think that that’s possible in practice, if not in theory.
@bob – Right now both GoDaddy and VeriSign have a very good business going, and I don’t see them making any big changes soon. For VeriSign, who are basically useless with anything customer-facing, registrars are a super-valuable channel and I don’t think they are going to muddy up the drinking water. For GoDaddy, the options are greater, and they are still a great reseller if you’re a TLD. We’ll have to wait and see how they want to play it.
The vote against .XXX was 9-5. Susan Crawford’s was not the only voice, and others recorded their dismay. So I should not have said that hers was a lone voice. But hers is the voice I remember.
How does “The Wall between Registries and Registrars Comes Tumbling Down” impact your business model going forward. Is this a good development for TLDH or will it make things more challenging as there will be more competition? Thx
Does this mean Image Online Design will get .web?