Minds + Machines

I regret that the posts on our blog are getting more and more specialist, but I don’t apologize. As things come to a head for new gTLDs, we are getting through the final details, which have become proxies for the larger battle. So it is with Expressions of Interest.

Here at the ICANN Nairobi meeting, I’ve heard some talk from people in the non-commercial constituency — usually an ally in getting new gTLDs approved — about how Expressions of Interest (EOIs) may be harmful to small registries. To be honest, I have a hard time fathoming the objections, because it seems to me that EOIs are extremely helpful to small registries. In fact, EOIs may be the only way they can survive the ICANN process.

(Expressions of Interest is the process where companies who want to apply for new gTLDs will have to submit their new TLD name, their contact details, and a check for $55,000, which will be counted toward the final $185,000 application fee. The process is mandatory, and the purpose is gather information on the applicant pool so that ICANN (and the technical operators of the Internet) can gather the data they need to take applications and prepare to delegate the new gTLDs. At the end of the process, all the information will be published, and only those who put in an EOI submission will be allowed to apply.)

“Small registries” mean primarily cultural-linguistic communities, for instance the already-announced .BZH (Brittany), .GAL (Galicia), and .EUS (Basque community). (It does not include city TLDs, which are “geographical TLDs” with special rules.) Each of these small registries has a community of several million, which during the first few years of operations will translate into a registry that contains less than 100,000 names. This is based on .CAT (Catalonia), which started several years ago and at present has about 40,000 names in their zone. There is an assumption, which may or may not be warranted but which we can all accept, that there may be others of a similar type. There is also general agreement that these sorts of registries are valuable to the DNS and should be helped wherever possible. Therefore the argument that EOIs would hurt these registries is of special concern.

So what are the arguments being advanced on behalf of these registries against EOIs?

1. That the $55,000 is too expensive and disproportionally hurts small registries
2. It will be hard for small registries to raise $55,000 and then later go raise another $130,000, and it will be hard to raise it by the EOI start date (August 2010 at the earliest)
3. That having to reveal one’s identity will give powerful actors more time to object
4. That EOIs, by adding a step to the process, will slow down the gTLD process

Of these, only the last one has any merit, but even if it’s true (which I don’t believe), it still helps small registries. Let me address the objections and show why they don’t make sense.

1. $55,000 is expensive, to be sure, but that’s a “feature” of the new gTLD program as a whole, not the EOIs. Overall, the $185,000 filing fee has not changed, it’s just that some of it is due earlier. Considering that ICANN had originally contemplated introducing new gTLDs in early 2009, it cannot be considered “early” at all. Furthermore, the real cost to small registries is keeping their operations going while they wait for the seemingly endless ICANN process to finish, not to mention the extremely demoralizing uncertainty about whether it will happen at all. When the EOI process finishes, and small registries know that they have a guaranteed and protected application “slot,” they will be much better off, because they will know that the process is happening and they will have some clarity around milestones and (hopefully) the re-establishment of a timeline.

2. Perhaps we can chalk it up the fact that non-commercial interests almost by definition aren’t experienced in raising money for commercial ventures, but almost any business person will tell you that the less risk you have, the easier it is to raise money. EOIs remove a very significant element of risk. When the EOI results are announced, most of the small registries will be guaranteed that they are the only applicant for their string (it is unlikely that there will be two applications for .GAL, for instance). That makes it *much* easier to raise money. The EOI submission date will also be a deadline – another very helpful element in raising money. Finally, because the entire $185,000 fee is known in advance, they will not need to go back to their funding source with a “surprise” second round, it will be plain to see from the outset.

3. Suppose that the Tamil people wanted to do .TAMIL, and faced the objection of the Sri Lankan government, who have just concluded a terrible and bloody war against Tamil separatists. Because EOIs occur earlier (the argument goes), it would give the objectors more time to prepare their case. This argument flies in the face of reality — and the ICANN process. The objection process (which will occur after the TLD application is submitted) will already take many many months, plenty of time for objectors to marshall their arguments. Whatever additional time is given to objectors, is also given to applicants — to either counter the objection, or to work out an arrangement. Finally, the ICANN process is not set up to put in “hidden” applications: for an application to succeed, it will have to overcome objections, it can’t try to sneak past them.

(I should note that brands who oppose new gTLDs have seized on this last argument and are evincing a a new and touching solicitude for the fate of politically sensitive small registries. Would that have anything to do with the fact that brands themselves don’t want to be forced to show to the world that they are busy planning their new gTLD applications even as they try to wreck the process?)

4. EOIs do add a step in the process, but it shouldn’t take more time, since things are all proceeding concurrently. Even if it does add months to the process, which I doubt, that is more than compensated for by the fact that EOIs add clarity, certainty, and something concrete that applicants can show their investors, their community, and their shareholders.

The proof of the pudding? In the ICANN EOI comment period, small registries supported EOIs. And all applicants, large or small, rich or poor, community-based or entrepreneurial, want this process to be predictable, orderly, and certain. For some small registries, their survival depends on it. EOIs provide all of that and more.

Yesterday, at the .ORG Forum, ICANN CEO Rod Beckstrom said,

There’s a clash of models going on in the world. It’s a clash of this decentralized multi-stakerholder model versus traditional government top-down model or centralized models. And this model we’re working on is different, it’s a mix. Governments are stakeholders, but they’re not the only stakeholders. They’re participants, but they’re not dominant. And trying to maintain that balance is one of the great challenges all of us face, particularly when there are those who would that seek to control things. And the question we should always be asking is ‘What’s best for the public?’

And at the ICANN Studienkreis last week in Barcelona, I asked a panel that included Fiona Alexander from the U.S. NTIA how the Government Advisory Committee (GAC) saw its responsibilities to listen to the ICANN community now that the GAC has an effective veto on ICANN policy. Ms. Alexander told me that I was entirely mistaken to think that the GAC had a veto.

Officially, then, governments are just one group of many that participate at ICANN. If so, ICANN and the GAC need to get the word out, because the rest of the domain name world is treating a letter from GAC head Janis Karklins as if it were the thunderous voice of God.

This letter, which warned the ICANN Board not to consider the Expressions of Interest proposal until the ICANN meeting in Nairobi, has been greeted with such headlines as Governments Deliver Another Blow to New Top Level Domain Timeline, and privately ICANN Board members have told us that it’s now “impossible” to support Expressions of Interest prior to the Nairobi meeting for fear of annoying the GAC.

In contrast, the unanimous vote of the At Large Advisory Committee (ALAC) in favor of Espressions of Interest has attracted no notice at all. The ALAC, which represents individual users of the Internet, has in in principle the same weight as any other advisory committee (such as the GAC), but apparently some equals are more equal than others.

The new Affirmation of Commitments, the “charter” for ICANN, clearly sets up an expanded role for the GAC. It is responsible (in part) for choosing the people who will conduct reviews of ICANN, and it is repeatedly mentioned in the Affirmation, while other ICANN groups are not.

The question is, does the new role of the GAC give them a veto power over ICANN? Formally, the answer is no. In practical terms, however, judging from the reaction to their “advice,” mere grumbling from the GAC can upset ICANN timelines.

It’s up to the ICANN Board and the CEO to determine where they are going to draw the lines with GAC. It’s up to the ICANN community to insist that as an important part of that community, the GAC not only injects its opinion into the debate, but listens as well. The stakes are high, because as Rod Beckstrom correctly notes:

The Internet has not been successful because one company or five companies got together and formed a cartel, and said “this is going to be the standard,” or a government said “this is going to be the standard”…. Mankind is facing global issues that have to be managed on a global basis. What we’re doing here [at ICANN] is an exciting and important new model for what can be used for addressing and solving many of these problems.

Whether the new model is “exciting,” as Mr. Beckstrom says, or depressing, as many fear, will turn in large part on determining the influence of governments within ICANN. The first indications will come from ICANN’s Board of Directors at their next meeting in February.

A group of people met at the ICANN meeting in Seoul and called themselves the Expressions of Interest Working Group — EOI-WG in ICANNese.

We were trying to come up with a way to get the ball rolling on new gTLDs, and we thought that if we could just get a list of the new TLDs and the people who were going to apply for them — well, then we would know a lot.

We’d replace the theorizing with practical knowledge, with many benefits.

• We’d know how many new TLDs there would be, so we could stop seeing the phrase: “…a potentially unlimited number of names.” We’d also be able to address root scaling issues on an empirical basis.
• We’d know which new TLDs would eventually be applied for, and who would be applying for them, giving lots of time for brand owners and governments and others with jealous privileges to prepare their objections, and relieve us from the theoretical evils that today are forecast with ease and assurance.
• ICANN would know enough to assign resources in response to actual data, which would result in a cheaper, faster, and clearer process.
• Applicants would know if they were the only ones, or if they needed to prepare for an auction or a settlement. Clarity, which has been a rare element so far in the process, would be theirs.

So the Working Group continued on from Seoul to hold several meetings and produce several drafts, and the Report of the Expressions of Interest Working Group has now been submitted to ICANN.

Here’s the cover letter I wrote when I submitted the Report, to give you a flavor of the contents.

On behalf of the Expressions of Interest Working Group (EOI-WG), I am pleased
to submit our final report to ICANN, attached here as a PDF file. In
addition, I have attached the EOI-WG charter, which sets our composition and
our remit.

The EOI-WG makes a series of specific recommendations about the EOI process, as
well as providing point-by-point answers to the questions asked by ICANN staff
for this comment period. In each case, our recommendations are accompanied by
the reasons for them. Where applicable, the level of consensus is also noted
(e.g., “unanimous consensus” or “strong consensus”).

Our report also contains discussion of suggestions that we did not make as
recommendations, and the reasons why they were not adopted as consensus
recommendations.

The Working Group met first during the ICANN meeting in Seoul and then, with
expanded ranks, by telephone call. We also set up a mailing list for our
discussions. The Working Group was comprised of a diverse group of people,
acting on their on behalf and not as representatives of any stakeholder group
or constituency. Our goal was to see if we could come up with recommendations
to help ICANN implement an Expressions of Interest process for new gTLDs as
called for by the resolution of ICANN’s Board of Directors on October 29, 2009
at their meeting in Seoul. Although ICANN did offer to provide us with
resources, we decided to privately fund the entire effort in order to avoid
squabbles about whether we were official or not. We’re not official — we’re
just some people who got together to make some recommendations.

In brief, our report recommends the following:

1. An Expressions of Interest procedure is desirable and should be implemented
as soon as possible
2. The Expressions of Interest procedure should be mandatory for anyone seeking
to apply for a new generic top-level domain
3. A $55,000 fee must accompany the submission of an Expression of Interest for
each string sought, refundable only in very limited circumstances
4. At the close of the Expressions of Interest submission window, ICANN should
publish at least the name of the submitter and the string submitted
5. ICANN should publicize the Expressions of Interest procedure in a manner
sufficient to assure fairness to those who may as yet be unaware of the new
gTLD program, but this communications period should be as short as possible
within this constraint

Questions about the EOI-WG and its recommendations will be answered by reading
the attached report and charter, as well as by reviewing our deliberations.
The archives of our mailing list may be found at
http://lists.pra.im/pipermail/eoi-wg/. The archives also contain links to MP3
recordings of our two phone calls, held on Nov. 9 2009 and on Nov. 13 2009, as
well as the various drafts that we considered, in clean and redlined versions.
The EOI-WG members are listed in both the report and our charter.

We hope and expect that our work will be helpful to ICANN, to the ICANN
community, and to the new gTLD process.

Respectfully,

Antony Van Couvering
EOI-WG co-ordinator

I urge everyone who wants new top-level domains to read the report, and to make your voice heard on the ICANN Expressions of Interest comments site.

What’s ICANN without acronyms? Here’s a new one: IRT. That’s not Interborough Rapid Transit, the subway on the west side of Manhattan, it’s the Implementation Recommendation Team. Their job is to come up with a plan to address the concerns of trademark holders in connection with the introduction of new TLDs.

What’s brilliant about this move is that the team is comprised of people who suggested solutions to the trademark quandary during the public comments to the last version of the guidebook. Those who simply said “hurrah” or “harumph” will not be asked for any more of their opinions.

Nicely played, ICANN.

Mexico City, Mexico… March 7, 2009: ICANN’s 34th International public meeting in Mexico City has drawn to a close after the organization’s Board of Directors approved the establishment by staff of an Implementation Recommendation Team (IRT) comprised of an internationally diverse group of people to develop and propose solutions to the over-arching issue of trademark protection in connection with the introduction of new generic top level domain names (gTLDs).

“The Board has clearly heard and believes strongly that the concerns of trademark holders must be addressed before this process is opened for applications,” said Peter Dengate Thrush, Chairman of the Board of Directors of ICANN. “The establishment of this team, is an attempt to get proposed solutions from the people with skill in trademark protection and other issues.”

The IRT will be comprised of people who put forward solutions in the first public comment period on the new GTLD Applicant Guidebook. The IRT has been asked to draft a report by 24 April for comment and to produce a final report no later than 24 May so it can be considered at ICANN’s Sydney meeting in June.

With huge thanks to the folks at Dynamic Network Services here are interviews with Elaine Pruis and Jothan Frakes. Good questions, good answers, lots of information.

Elaine Pruis

Jothan Frakes

ICANN has released preliminary reports on competition and pricing for new TLDs. The reports done by Professor of Economics Dennis Carlton of the University of Chicago cover the impact of new gTLDs on consumer welfare and price caps for new gTLD Internet registries.

In the report “Preliminary Report of Dennis Carlton Regarding Impact of New gTLDs on Consumer Welfare,” Professor Carlton states:

“I conclude that ICANN’s proposed framework for introducing new TLDs is likely to improve consumer welfare by facilitating entry and creating new competition to the major gTLDs such as .com, .net, and .org. Like other actions that remove artificial restrictions on entry, the likely effect of ICANN’s proposal is to increase output, lower price and increase innovation. This conclusion is based on the fundamental principles that competition promotes consumer welfare and restrictions on entry impede competition.”

The other report, “http://www.icann.org/en/topics/new-gtlds/prelim-report-registry-price-caps-04mar09-en.pdf,” addresses pricing issues associated with new TLDs. He concludes as follows:

“I conclude that price caps or ceilings on prices charged by operators of new gTLD registries are unnecessary to insure competitive benefits of the proposed process for introducing new gTLDs. I further conclude that imposing price caps on the registries for new gTLDs could inhibit the development and marketplace acceptance of new gTLDs by limiting the pricing flexibility of entrants to the provision of new registry services without generating significant benefits to registrants of the new gTLDs.”

These are among the first of many reports we will see on new TLDs in the coming months which are sure to spark heated discussions.

For those of you who think the domain space is still blissfully under the radar, you are wrong. The Pope himself is getting in on the action with a concern about religious TLDs.

According to thedomains.com

The Vatican warned ICANN of the “perils” of allowing new internet domains such as “.catholic, .anglican, .orthodox, .hindu, .islam, .muslim, and .Buddhist.”
ICANN, could find itself having to decide who gets to represent an entire religion on the internet, His Holiness pointed out, in a letter from Monsignor Carlo Maria Polvani.

As if ICANN does not have enough on its hands at the moment. To quote one ICANN attendee overheard on the conference floor, “Why do people hate the future?”

The reason for the latest delay to ICANN’s plan to introduce new TLDs — and the reason it’s been 10 years in coming — is the fact that domain names and trademarks don’t work together that well.

Now a plan to protect trademark holders (Powerpoint 2004, 1.2MB) across all gTLDs for about $150/yr (final price not yet set) might solve the complaint of trademark holders that they are being held for ransom by the introduction of new extensions. As an example, Xerox would pay about $150/yr to protect its brand “xerox” no matter how many gTLDs were introduced in the new ICANN round. This paltry cost takes away the largest objection from trademark holders.

On March 2 at the 4 pm session called “New TLDs and Branding,” Bart Lieben of the Belgian law firm Laga presented a plan to make trademark protection in new gTLDs workable. The plan is based on the over 50,000 validated trademark records that Bart has collected in doing the .EU, .ME, and five other Sunrise periods.

The trademark database, which would be continually re-verified, and to which new records would be added on an ongoing basis, would include information on 37 attributes such as trademark class, date, jurisdiction. Registries could implement their own policies (such as whitelisting, blacklisting, blocking, etc.) using the records, and registrars, who would also have access, would be able to submit error-free Sunrise apps. Currently, according to Bart, the average error rate of submissions from trademark holders is in excess of 50% and has reached as high as 80%, which has added substantially to the cost of new TLD Sunrise periods.

The response from those in the room, which included well-known trademark attorneys and TLD-denying consultants, was an excellent indication that the plan was sound. One criticized the process, saying that it should have gone through various ICANN working groups. Another criticized Bart’s affiliated company Deloitte as untrustworthy. No-one criticized the substance of the plan.

I think we may have a winner.

The GAC and ccNSO met this afternoon in response to the earlier ccNSO meeting in which ICANN announced that a contract between themselves and ccTLDs was a precondition for acceptance of fast-track IDN applications.

The GAC responded in an email following the meeting, regarding the relationship between ICANN and ccTLD operators, as well as ccTLD financial contributions.

Relationship between ICANN and IDN ccTLD Operator

• IDN ccTLDs should be similarly treated as ASCII ccTLDs.
• A documented relationship between ICANN and IDN ccTLD operators should be kept voluntary.
• A documented relationship on the basis of the proposed “Documentation of Responsibilities”, either as it stands today or in a modified format, may be encouraged but should not be a condition for IDN ccTLD delegations.
• As it has always been the case, it’s in the best interest of ccTLDs operators and the entire IDN community to adhere to all relevant IETF standards including IDNA protocol, IDN Guidelines and commit to complying with future protocol updates.

Financial Contributions

• IDN ccTLDs should be similarly treated as ASCII ccTLDs.
• Financial contributions should be calculated on a cost recovery basis. Full disclosure and breakdown of the costs involved in the IDN program would be desirable for better understanding of possible cost recovery models.
• Financial contributions should be kept voluntary and should not be a condition for IDN ccTLD delegations.
• Further information, from ICANN staff, on the different possible cost recovery mechanisms and concrete proposals would help advance positions on the subject.

Tensions ran high in today’s ccNSO meeting at the ICANN meeting Mexico City when CEO Paul Twomey and Peter Dengate-Thrush announced that ICANN expects the ccTLD operators who are interested in running IDN ccTLDs to sign a contract with ICANN as a condition of approval of their IDN applications. This is a new development in the ccTLD IDN fast-track process, and took the ccTLDs by surprise.

While ccTLDs have been encouraged to sign contracts with ICANN, many have not, and the ccTLDs see this move as extortion. The Governmental Advisory Committee (GAC), composed of governments for the territories represented by ccTLDs, are not very happy about this either.