News, views and what I choose to dos

Finishing up the rules for new Internet extensions

Category : Domain names, ICANN, Internet, Internet governance · by Sep 24th, 2010

ICANN logoTwo very interesting things are happening today that may have an enormous impact on the Internet for many years to come.

First, the ICANN Board is meeting at a special two-day retreat in Trondheim, Norway, in an effort to finalise the rules for new Internet extensions. This process have been going on for more than five years – two years longer than was planned – and it appears that everyone is now tired of the back-and-forth and wants results. The hope is that the “Applicant Guidebook” will be formally approved at the ICANN meeting in Cartagena in December.

The second interesting thing is a letter [pdf] from the Governmental Advisory Committee (GAC) of ICANN to the Board giving very precise and useful comments about what it feels about some of the outstanding issues in the new rules. These letters are usually carefully timed and this is no exception – it appears the day before the Board retreat. It is also a fair assumption that the ICANN chairman and CEO and senior management already know much of what was in the letter and may have helped in its production.

So, here’s the big question: will the Board be able to make the decisions necessary for a fifth and final version of the Applicant Guidebook to be produced in time for the Cartagena meeting?

I’m an optimist, so I’m going to say Yes; I think they will be able to do it. Well, almost all of it, with the rest thrashed out in Cartagena. Here’s why and where I think the axe will come down.


It is no mistake that the GAC letter covers every controversial aspect remaining, save one.

They are:

  • Size of the program
  • Economic impact of new extensions
  • Vertical integration
  • Trademark protection
  • Geographic names
  • ICANN’s efforts to make itself legally unaccountable
  • Community TLDs and applications from developing countries

Size of the program

The previous two additions of new Internet extensions in 2000 and 2004 saw just seven new extensions each time. ICANN as an organisation is still worrying about what will happen if hundreds of new top-level domains appear at the same time.

There is some legitimacy in this concern, although much of it has been nannying and disingenuous. The GAC letter actually lays out a reasoned, albeit conservative case. It wants a “fast track” for non-controversial additions, as well as a limited release of new extensions. It also argues for a study of a limited number before others are released.

There has been a fair amount of work done on this issue of adding alot of new extensions to the Internet by the engineers that actually have to run the systems. Broadly their conclusion was: it can be done, but it’s going to be a struggle. Basically, the systems in place are not prepared for a massive influx of new extensions.

Now, they could be prepared – but it would require money and a huge amount of effort and, frankly, there isn’t that much willingness on the part of the technical community to put in huge resources just because ICANN has decided that’s what it wants to do. Especially when, with the stroke of a pen, ICANN could spread the introduction of new TLDs over a longer timeframe.

ICANN chairman Peter Dengate Thrush made some very interesting comments last week at the IGF in Vilnius that demonstrated a clear plan to keep the number of TLDs released over a given time. He said: “The total number that can be processed in any year is 200 or 300… a number that’s currently equal to or less than the current total number of TLDs. So we’re not talking about orders of magnitude changes we’re talking about in the first year possibly doubling the current TLD numbers. I don’t think that’s mass and I think people will be able to cope with that rate.”

Peter doesn’t make these comments lightly. And now the GAC is suggesting the same thing. So this is my first prediction:

The number of extensions added to the Internet will be limited to a certain number per year. The least controversial will be first in line.


Economic impact of new extensions

This has been a frequent complaint of those who have sought to delay the program for their own reasons. Again though, it comes with a fair amount of truth.

ICANN’s staff failed miserably by not requesting an economic analysis of the impact of new extensions on the market before the whole process kicked off. Instead, it was subsequently forced to keep ordering new economic studies, which were fudged, and so was then forced into ordering another. And so on.

Why the failure to think ahead?

  • Because ICANN staff are very rarely allowed to think for themselves, and are in a constant state of reaction
  • Because ICANN staff are policy wonks and managers. Five years ago there were zero business and market analysts, for whom an early economic study would have been an obvious step. I think there’s only two now (it should be 10)
  • Because ICANN as a whole absolutely hates being told by anyone what the situation is. The Board, staff and community frequently (and rather arrogantly) dismiss the conclusions of experts. So there was no desire at all – until forced to – to hire someone to tell ICANN about the market it oversees. Even now, people that really have no idea what they are talking about attack the economic analyses that have been produced

The GAC wants the Board to use a limited fast-track subset to carry out an economic study. This is a pretty good idea. However the GAC also wants there to be no new extensions while that is going on, which is a less good idea.

Who knows where the Board – which is notoriously risk-averse – will fall on this. There would certainly be value in having a real, live economic analysis of new extensions. But any suggestions that this then delay the already heavily delayed process further is going to be met very unfavourably. So prediction number two:

An economic study will be commissioned into a subset of approved new extensions, although this will be used as feedback to the next round of new gTLDs, rather than as a stop-block in this round.


Vertical integration

This is the most fought-over aspect of the process, at least from the status quo side. And since the status quo provides ICANN with its funds, and is in a position to make life very difficult for the organisation, it is a serious issue.

In a nutshell, the current situation where registries (those running an Internet extension) and registrars (those selling domains under those extensions) are kept separate is rapidly becoming an outdated model. The clearest example is if Canon wants dot-canon, why on earth should it have to pay a third party in order to have a domain under its own extension?

Of course, those who currently benefit most from the status quo – the biggest registries and registrars – don’t want to see this situation change. But it has to. The question is really: how long can the situation be kept in place? And, more crucially for the Board this week: what do you put in its place instead? (See this comment below for additional clarification on the positions of registries and registrars.)

There was a long and tedious working group within ICANN that tried to come up with a solution to this but inevitably it failed miserably. It really does come down to the Board to step up on this occasion and make a decision. It did actually hire an expert to help find answers to this one – and the report was inevitably torn apart by the community. But that report remains the best guide for the Board at this stage.

The GAC suggests relaxing the rules and puts out there the idea that “community TLDs” be granted exemption. It also suggests that there be relaxation of the rules but that there be a regulator aspect draw up (and it advises, quite rightly, that ICANN actually ask regulators around the world for their thoughts).

The Board has to make a call on this one. I would love to think it will be brave, do what needs to be done, and simply weather the storm of protest from its most powerful members. But it won’t. The Board is too big and so there is always dissent which means courage is rarely forthcoming.

So what will it do? Go for the safe option. Prediction three:

Vertical integration rules will be lifted for small-scale community TLDs, the current rules maintained and a study commissioned to look at the issue (possibly tacked onto the economic study mentioned above), with its report used to inform the next round of gTLDs.


Trademark protection

Another significant issue. The IP lobby (standing for intellectual property rather than internet protocol) has been all over this. It was absolutely right to be worried when the first Applicant Guidebook appeared in 2008 – the protections for companies and their trademarks had been horribly overlooked.

Over time ICANN has tried to fix this issue. Unfortunately it botched the effort several times, most famously with the creation of the Implementation Recommendation Team (IRT) which ended up infuriating just about everyone. However, the IP lobby has been backing off in recent months and ICANN’s chairman also said last week that “the IP lawyers… have had their chance… and the intellectual property protection brought in place is with a delicate balance with everyone in the community not just with IP lawyers.” It helps that Peter is an IP lawyer himself.

So, what does the GAC suggest? It worries about the cost of defensive registrations (to which there really is no solution beyond the fact that companies will have to get used to the fact that the rules of the game have changed with the Internet and you can’t expect to claim control over everything). But also it says the Uniform Rapid Suspension System – which would allow companies to quickly bring down sites infringing their rights – is too unwieldy. It also wants trademarks rights extended slightly – so that if your mark is contained within a longer phrase you can also question it.

The solution to all this is actually not that hard – it’s just that everyone will complain vociferously no matter what the outcome is. Basically, trademark holders have to be given a way to take down a website quickly and cheaply if they can clearly prove their rights are being infringed. The hard part is to provide a sufficiently large stick at the other end to stop big companies from abusing this process.

In terms of the Board making a decision this week, they will probably feel obliged to give a little bit back to the GAC and the IP lobby – maybe pull out the most onerous requirement from the URS – and then simply say that they will get an outside expert to review the process in 12 months’ time. Get something in place and then watch to see what happens.

The important thing will be to make sure that the axe falls down in favour of the registrant – because big companies will make themselves heard plain enough if the system needs tightening, but it won’t be the case the other way around. Prediction four:

The current systems in place for protecting trademarks will go through, although with a tweak or two to make trademark holders lives slightly easier. If the Board is in the mood for reviews, it will order a review into the system in 12 months’ time.


Geographic names

This has been a real GAC bugbear – governments have been panicking / being greedy that names in their countries will sprout up everywhere and they will face all sorts of problems. It is also the one area where government can legitimately exercise their innate desire to have control over things.

The Board has consistently pushed back against this landgrab – and quite rightly. Because city TLDs are a very promising form of Internet expansion, and because business in particular sees the value of having a top-level domain for specific geographic locations.

The GAC stamps its foot again in its letter but appears to be recognising that it isn’t going to get what it wants and so has offered a way out: don’t charge us for objections, and allows us more time to review applications. The “don’t charge” is the hard part. With no charge, governments can simply fire objections left, right and center. If money is attached, there has to be a process and an approval process internally.

In terms of the ICANN Board making a decision, again this one takes courage. The Board has stood up to the GAC (well, actually it has pretended not to hear) on this before and could do so again. It may have to or it risks wiping out a really promising area of global competition.

The decision will be based on how much else contained in the GAC letter the Board feels it can accommodate. If the Board gives on a range of other things, it can afford to say no to the no-charge request. In fact, I can’t honestly see the Board agreeing to the no-charge as it would risk destabilising the whole process. So, predicition five:

The current rules surrounding geographic names will be retained, although governments may be granted extra time to review applications.

ICANN’s efforts to make itself legally unaccountable

This will be hotly debated but only because ICANN’s General Counsel will paint 1,000 different scenarios where not giving the organisation effective legal immunity will bringing ICANN crashing down around their ears.

The Board buys this argument every time – which is why ICANN has an Ombudsman with no power, a Reconsideration Committee that always realises it was right in the first place, and a Independent Review Process that is horribly burdensome and which the Board feels it can ignore anyway.

I would absolutely love it if the ICANN Board could finally recognise that its salvation and its legitimacy will stem from it having the confidence to stop constantly grant itself complete control over everything. But I don’t think the Board is there yet. It will still try to insist in the guidebook that people will have to sign something that recognises the Board is as infallible as the Pope (or, at least, you won’t try to get the law courts to say otherwise).

The wording will change but the Applicant Guidebook will continue to give the Board the final say in all matters – which will continue to undermine the organisation in the world’s eyes. Prediction six:

The wording in parts of the Guidebook will change but ultimately it will still grant the ICANN Board, under US California law, the absolute right of final decision and reduce as far as humanely possible the right of people to go to court to dispute a Board decision.


Community TLDs and applications from developing countries

And, finally, efforts to actually get new Internet extensions in the hands of people who aren’t extremely wealthy, powerful and based in the West.

The constant refrain for ICANN to think about developing countries and community TLDs stems from the classic situation where a process is distorted by those who shout loudest and have the most money to spend on people who shout loudly. Plus, with ICANN being based in the US, and with a US mindset, the solution to problems has consistently been financial. If there is an issue here, just raise the cost of it, squeeze out the chancers.

Of course with high costs ($185,000 just for the application, don’t forget) you also squeeze out some of the really valuable uses of the Internet – allowing people to communicate.

Unfortunately, as any African leader will tell you with resignation, the powerless are always considered last. Everyone will agree that it is terrible and something needs to be done, but nothing will be. What will happen is that a significant percentage of the money set aside for communicating the gTLD program globally will be spent on telling people in developing countries about a process that they won’t be able to afford. It will achieve nothing but it will make everyone feel better. Prediction seven:

Developing countries will have more money spent on informing them about new gTLDs that they won’t be able to afford. Community TLDs will continue to have to live with a process that makes their lives harder than it needs to be because otherwise people fear that the system will be gamed by commercial interests.

And that is my summary of where we are and what I think the ICANN Board will do.

Ah yes, public morality

Of course the one area that is possibly most important that hasn’t been mentioned is the “public order and morality” discussions which are in an impossible state. Pull away all the language and basically ICANN is planning to decide what is to be banned in new Internet extensions.

The really big problem at this stage is that there are no decent models or suggestions out there. The GAC threw the wording out at the last minute in Brussels, then presented an alternative that was never going to work, and it was then left to a cross-community working group that created a dog’s breakfast of a report.

This is the one issue that may delay the new gTLD process at this stage – because there is no way the Board can agree on an approach in two days. This will be the big issue at the meeting in Cartagena and the ability to find a solution will be what pushes the process forward or stalls it further.

My feeling is that with the momentum behind getting this done and finally finished, and with the risk that careful compromises could start falling apart if there is further delay, the Board and community will come together in an indelicate fudge that everyone will hate but fortunately may never have to be used. Because, quite frankly, anyone that is willing to spend a small fortune on a purposefully offensive application is never going to get anywhere. It’s self-limiting.


I’m confident we may finally see this very long, and often tedious process finally come to an end. And the Board may come out of it with a renewed sense of confidence. I don’t expect there to be that many serious objections – even from the IP lobby. Everyone basically wants this done.

Plus of course, every one of the individuals on the Board wants to be able to say they were at ICANN when the gTLD process was finally approved. If they pull it off, it will be one hell of an emotional moment on 10 December.

By the way, if more than three of the seven predictions come true, I’ll be amazed.


(9) comments

Jeffrey Eckhaus
11 years ago ·

Would like to clarify re: Vertical Integration the statement that the largest Registries and Registrars want to maintain status quo. If you read the Initial Report and subsequent public comments Verisign , Neustar, eNom, Network Solutions and others all called for change and allowing up to 100% co-ownership in certain cases, such as not distributing your own TLD. The groups that have asked for status quo are Afilias, PIR and GoDaddy.

11 years ago ·

Up to a point, Jeffrey – the “certain cases” for example. But yes, a fair point. Will stick a comment in at the relevant point linked down here.

Michele Neylon
11 years ago ·


Thanks for the summary.

The only working group you seem to have left out was the one that was / is meant to be working on narrowing the economic barriers.



Yves Goulnik
11 years ago ·

Great summary – but that’s only 7 predictions, did you notice you skipped number four ?

Milton Mueller
11 years ago ·

Where have you been? There IS an alternative model to “morality and public order.” It was the product of a cross-community working group and the report was given the Board four days ago. See
More distressing, you seem to buy into the paradigm that after 3-4 years of so-called bottom up policy development by the GNSO, the GAC can simply re-write the policies that managed to win consensus or strong support by sending a letter to the Board. I hope you’re wrong – I hope ICANN’s board has more respect for the people who actually participate in its processes.

11 years ago ·

Oops. Will have to reorder. You’d think I’d be able to manage counting to seven, wouldn’t you? Cheers

11 years ago ·

When I wrote about the “dog’s breakfast” that was referring to what you seem to believe is a viable alternative model. Each to their own.

11 years ago ·

The GNSO comment is a little strange. Clearly some kind of resurgence in respect for the tablets of stone passed down from the mountain all those years ago.

But as the real world discovered long ago, the Ten Commandments, as good as they are, were not all that effective at producing actual live rules for society. They typically need interpretation and occasionally careful ignoring.

11 years ago ·

/* */ <- No comment

Leave a Reply

Your email address will not be published. Required fields are marked *