World Wide Alliance of Top Level Domain-names

ccTLD-ICANN Meeting in Honolulu, Hawaii, USA, 1 Feb 2001

 Home | Meetings | AdCom | ccTLD | Participants 

ccTLD-ICANN Meeting in Honolulu, Hawaii, USA, 1 Feb 2001

Asia Room, East West Center, University of Hawaii, Honolulu, Hawaii

Agenda | Participants | Morning Session | Afternoon Session


9:00 -12:30

- ccTLD-ICANN-IANA and US DoC relations
- agreement principles
- elements and structure

2. IANA Database and Whois (principle issue)

3. Contracts

1:30 - 4:30

1. Financial Support for ICANN

2. Joint activities: Outreach/Meetings

3. Quality of Information: IANA Database, DNS

4. Others

List of participants

1. Mike Roberts
2. Louis Touton
3. Diane Schroder
4. Prof. S.H. Kyong (Board)
5. Alejandro Pisanty (Board)

6. Peter de Blanc (.vi)
7. Peter Dengate Thrush (.nz)
8. Kilnam Chon (.kr)
9. YJ Park (.kr)
10. BK Kim (.kr)
11. Keven Meynell (CENTR)
12. James Ross (.tv)
13. Bart MacKay (.cc)
14. Hiro Hotta (.jp)
15. Ohashi Yumi (.jp)
16. Makiko Matsumaru (.jp)
17. Kenny Chuang (.tw)
18. Albert Wang (.tw)
19. Christopher Roosa (.us)
20. Chhay Ea (.sg)
21. George Daniel (.ag)
22. Elisabeth Porteneuve (.fr)
23. Abhisak Chulya (.th and ccTLD Sec.)

Morning Session

Minutes Written by Abhisak Chulya

Session Chaired by Peter Dengate Thrush and Mike Roberts

Peter Dengate Thrush welcomed all ccTLD managers and representatives as well as ICANN's Mike Roberts and his staffs. The meeting was then proceeded to Mike Roberts presentation by giving out the history of why we were here today. He had five slides to present and here they are as follow.

Slide #1 (Mike Roberts) IANA and ccTLDs in the Jon Postel era - Country TLDs originally granted for researchers - Jon delegated the registries based on personal acquaintance - With growth of Internet, and more country registry requests, Jon adopted some semi-formal practices, e.g. use of the ISO 3166-1 table - In 1994, with further growth, Jon documented basic principles and practices in RFC-1591 - Subsequent evolution of policies described in ccTLD News memos

Slide #2 (Mike Roberts) ICANN History Relevant to ccTLDS - 7/97 - US Gov't decision to privatize DNS technical management and IANA - 6/98 - US Gov't White Paper published with call for written a agreements among all parties and adherence to four principles - stability, competition, representation and bottom up process - 11/98 - ICANN and US Gov't sign MOU - 6/99 - ICP - 1 issued to document Postel's ccTLD delegation policy - 11/99 - ICANN Board adopts TFF recommendations refunding responsibility and allocations - 2/00 - ICANN GAC publishes statement of principles for ccTLD delegations and administration

Slide #3 (Mike Roberts) ICANN History Relevant to ccTLDs - Cont'd - 5/00 - ICANN issues invoices to ccTLDs for financial contributions per TFF rec'd structure - 6/00 - ICANN posts ccTLD status quo agreement for public comment - 9/00 - MOU Amend. 2 confirms ICANN must achieve stable agreements with ccTLDs in order to complete transition - 10/00 - ICANN reaches agreement with CIRA and Canadian gov't on new structure for .ca - 11/00 - ICANN posts updated financial structure for public comment - 11/00 - ccTLDs propose alternate agreement form

Slide #4 (Mike Roberts) Principles for ICANN - ccTLD Relationship - Use original Postel - IANA ccTLD concept: a) ccTLD manager is trustee for Local Internet Community b) competent operation of registry & name servers c) consensus efforts to resolve disputes d) respect views of governments - Recognize ICANN as global consensus forum among Internet stakeholders, including ccTLDs and a means for implementing consensus

Slide #5 (Mike Roberts) Principles for ICANN - ccTLD Relationship - Cont'd - Promote transition from US government supervision by reaching stable agreements with ccTLDs as soon as possible - Use flexible agreement structure to accommodate varied circumstances: a) different ccTLD registry models b) different local needs c) different governmental situations

Louis Touton then continued after Mike Roberts's by presenting about Agreements between ccTLD and ICANN with the next 8 slides as follow.

Slide #1 (Louis Touton) Basic ICANN - ccTLD Agreement Structure - Written agreement between ICANN and ccTLD manage - Assist ccTLD manager in serving needs of local Internet community - Provide for appropriate monitoring of ccTLD manager's trust obligations - EnableICANN to ensure global DNS interests are protected (technical stability, transparency in registration, etc.)

Slide #2 (Louis Touton) Special Concerns to Be Addressed in ICANN-ccTLD Agreements - Appropriate application of local and global policies - Circumstances in which delegations are changed - ICANN's performance of the IANA function - Contributions to funding ICANN activities

Slide #3 (Louis Touton) ccTLD Agreements - Two Basic Situations "The designated manager is the trustee of the top-level domain for both the nation, in the case of a country code, and the global Internet community." -- J. Postel, RFC 1591 - Legacy situation: The IANA monitors the trust locally and globally - Trilateral situation: The national government, where interested, responsibly monitors the local interest; ICANN monitors the global interest

Slide #4 (Louis Touton) Determining Which Situation Applies - Does the national government want to be involved? If no --> legacy situation - If the national government wants to be involved, is it willing to agree: a) To afford the ccTLD operator security from arbitrary governmental action, with reference to reasonable stated performance requirements b) To responsibly monitor the ccTLD operator's activities in the interest of the local internet community c) To acknowledge ICANN's responsibility for ensuring sound operation of the global DNS If yes, yes, yes --> Trilateral situation

Slide #5 (Louis Touton) Documents for the Two Situations - Legacy situation: Agreement between ICANN and ccTLD manager - Trilateral situation: a) ccTLD manager and national government: agreement or equivalent (see .ca Umbrella Agreement) covering local interest b) Nation government and ICANN: letter of other communication c) ccTLD manager and ICANN: agreement covering global interest - Flexibility to accommodate intermediate situations

Slide #6 (Louis Touton) ICANN & ccTLD Manager Responsibilities in Trilateral Situation - ICANN a) Maintain authoritative root server system b) Maintain authoritative and publicly available ccTLD database c) Maintain audit trail regarding delegations - ccTLD Manager a) Provide stable and secure registry and name server operation b) Participate in development of and abide by ICANN consensus policies, e.g., global interoperability, Whois, data escrow, etc. c) Contribute to ICANN's cost of operation

Slide #7 (Louis Touton) ICANN & ccTLD Manager Responsibilities in Legacy Situation - Commitment to operate ccTLD in trust for the interest of the local Internet community - Acknowledgement that no property rights are acquired in the ccTLD - Compliance with ICANN-developed policies concerning domain-name-dispute resolution (to the extent applicable) - WHERE the ccTLD registration policies encourage or allow out-of-country registrations, agreement to observe all applicable ICANN-developed policies - Mechanism for conversion to trilateral situation in the event the national government agrees to the necessary items

Slide #8 (Louis Touton) - additional staffing for IANA root management - Root management database improvements o Automated preparation of emails o Pro-active tracking of tickets o Scripted checking of root zone for conformity - Improved root-zone Whois service, including port 43 service - Transfer of database maintenance responsibilities from NSI registrar - Streamlined USG approval procedures for ccTLDs with trilateral situation in place.

After Louis Touton finished his presentation, there was an open discussion between two sides. Peter Dengate Thrush (PDT) started off by mentioning that there were a number of elements of history that he disagreed with. Two things that we had not gotten important into that history are: first, ccTLD invitation into Task Force on Funding. Also there were no ccTLD representatives on the board to approve the budget. Second thing is related to ICANN view on GAC and role and increasing role that government would play in this process. There seems to be unnecessary divided between ccTLDs and GAC. Our perception on GAC has been dominated by one or two countries especially the view on delegation and redelegation.

Peter Dengate Thrush: Do you see the US government revolution on this process to go to a point where there was only able to influence ICANN as a member of GAC?

Mike Roberts: No body in ICANN can speak for US government. The government actions in the white paper and MOU and subsequent amendments just said what it means. Therefore, the more important thing is when we will get to the end of the ICANN transition agreement and process with the U.S. Government. The government just wants us to get on with the agreement.

Peter Dengate Thrush: What law or policy do you see is likely to be developed through the ICANN process which will be binding on ccTLD managers?

Mike Roberts: Let's distinguish between GAC relationship to ICANN and GAC principles. In both of those cases, if cc community feels improvement needs to be made, then you should submit the paper through the GAC saying what kind of relation you want GAC to be. On GAC principles, there has been some inconclusive discussion between the GAC and representatives of ccTLDs which have not resolved into any revised principles. Our experience in talking with the governments that actively involve in redelegation and restructuring now such as Sweden, Japan, Belgium, US, Taiwan, Australia is that they wish to conclude agreements with ICANN that incorporate a trilateral arrangement because that is the way they are already operating. These governments are active partners with their ccTLD registries. If a registry has problems with the principles as applied to its situation, this should be made known to the GAC and possibly taken up on their next meeting agenda.

Louis Touton: Regarding the relationship with your government, GAC principles that are out there, if there are situation with particular country which other accommodation that want to be met like Canada for example, I don't see in principle any reason why the effected party with respect to that country can't come to the agreement and achieve on that basis. The GAC principle say something different and New Zealand government and New Zealand Registry can come to the understanding. There is a failure of communication on this trilateral situation. When the governments tell you and tell us, I know we are in sync for a variety of reasons. Part of the challenges to bring this relation to existence is to squeeze out those differences and communication and continue dialogue not only individual national government but also the ccTLDs.

Peter Dengate Thrush: I like to identify three things from your presentation today: Dispute Resolution, Data Escrow and Whois, that you see as being appropriate policy to make through in ICANN process. Are there any more?

Mike Roberts: Any thing that affects the operation of name server for all top level registries and jeopardizes the stability and integrity is a matter of consequence that account into our agreement. That is fundamental to our responsibilities.

Peter Dengate Thrush: With dispute resolution, who you think should take responsibility?

Louis Touton: The question of what a local or a global issue is itself a global issue. We are working on a frame work with one of the basis global policy that include delegation policy and that is the creation of two-letter country code to represent ccTLD that occurred back in the 80's with the motion that it is in the interest of all the internet. There will be the code that is assigned to individual country and territory that they can accommodate the legal, cultural and economic circumstances that each territory involved. They should be given some levels of economy of delegation of local community. This is reflected in 1591.

Peter Dengate Thrush: I asked about crucial principles for ccTLDs and what you are saying is that people who will decide about cc Global issue will not be cc but the global community. At the moment, people who decide cc are the cc.

Louis Touton: It's a global judgment made initially when 1591 was being set up. We can discuss some basic understanding about what kind of things are local and what kind of things are global.

Mike Roberts: When working out agreement with big registries, it's been taking for granted the integrity of the DNS required and the ability to reactivate name server within a short period of time. This is an operational management issue. Obviously, we have a lot of diversity out there. Some registries have under 100 domain names. Data escrow for these kind of registries would be floppy discs. The economic consequences of the failure of the name servers are less important than if you lost .uk or .de. One of the reasons why the Internet registries enjoy as much public trust of the community as we do is that our engineering is turned out to be marvelously reliable considering the amount of internet traffic that we have. We do not want to have major name servers failure that undermined the public trust in the integrity of the global internet.

Peter Dengate Thrush: The cc is going to be bound by policy developed for the ICANN process. The cc would need a considerably great participation in the discussion about policy including sega such as NSI has in relation to contract with ICANN or cc is going to divide themselves when ICANN developed a process.

Mike Roberts: In the evolution of an organization such as ICANN in transition from entirely government control environment, the expectation is that the community will regulate itself in the public interest. There will be stages along the way. The obvious question is on upper limits on financial contribution I expressed that to a number of people already and we are ready to discuss that as part of the interest and also the financial components in the agreement.

Peter de Blanc: I want to be clear what are you saying here is that some dispute resolution policy will come down and since signing up to this contract would be effectively agreeing to pass the dispute resolution policy into the contract individual domain names only for the cc.

Louis Touton: I didn't intend to say that at all. I would not agree with the uniform dispute policy would be one size fit all. I think that it's probably should be some basic level of dispute resolution that should be present. It ought to be a system that well suit with a particular ccTLD. For example, in case of CIRA, they have a different Canadian internet community to determine and I don't think any body guessing they would have to go to UDRP. Certainly, there is nothing in the documents that require that.

Mike Roberts: There are two different cases here. One has to do with dispute and delegation. That has to be dealt with exclusively between ccTLD manager and ICANN. Second clause has to do with the treatment of people at any level within country domain. The original motion that why we are adopting UDRP is that it is a nonjudicial way of dealing with dispute resolution and wide supporting community for ICANN adopting a structure for nonjudicial system. It is certainly permissive. This is a substantial consensus in the internet ICANN community that it is a good idea that there are non-judicial ways to solve this problem. If any ccTLD registry feels there should not be a non-judicial way of dealing with domain name registration disputes, I'd like to hear the rationale. I think the Board would be interested.

Peter de Blanc: I just want to make clear that in signing the contract, cc will agree on some type of dispute resolution policy.

Peter Dengate Thrush: The issue is who will decide on the adequacy of the process. ICANN is going to say it will be ICANN process to decide the dispute resolution or will it be entirely up to Japan to decide the type of dispute resolution process that suit Japan and not have their choice of process subject to anybody else process.

Mike Roberts: You are being hypothetical. The manager is the trustee for the interest of local community. It is expected to be tangible to go to supreme protected. I can't be responsible for the trustee. I assume that the manager and the local community care about resolving disputes. If you're saying that they don't, then I do not believe that is a fair representation of the situation.

Coffee Break

Peter de Blanc: There are some various countries concerned about Whois database whether it's legacy situation or trilateral situation. We would like to know where the determination of the quality of information appeared in Whois database will come from and if signing this contract, ccTLDs will be agreeing to a particular level of disclosure of the information.

Mike Roberts: The content of Whois is very old. ICANN inherited from Postel regime is binding on everybody. When I first took up ICANN job, questions started to pop up about Whois. The general position that this staff has is Whois database should not be changed except by through an ICANN policy process. The circumstances surround Whois have changed dramatically since it was put in place. The reason why you had this information was when that host stop answering, you picked up a phone and tell them that you are dead. We have a quite substantial difference and you just gave one of the number of examples where internet is expanded in an area that is creating the ambiguity and issues about Whois. Within European Commission staff, there has been debate going on now for two and a half year about what Mike Roberts is going to do with Whois problem. My immediate reaction was I was not going to do anything about that. Perhaps ICANN organization through a deliberate process will come to some new conclusion about updating Whois for the 21st century internet. Where we stand right now is the various degree of discomfort with the existing Whois. And the only reasonable way to undertake the changes is to very broadly base process. What should ICANN do to get the written agreement with all the name registries about this issue? There would be a very strong negative global reaction for any TLD registries to de-commit from the existing Whois. I can't give you any advice or opinion about what would happen as a result of negative reaction of the rest of internet community to any registries unilaterally doing that.

Bart MacKay of .cc: What is the scope of ICANN in this process as related to Registrants. Back in late November, I heard Andrew McLaughlin made a quote - Registrants in ccTLDs should be very cautious because of the fact that ccTLDs are governed by the changing wind of the governments that they are attributed to, but gTLDs is essentially a safe heaven. It raised a question in my mind about ICANN official policy. He made it clear that ICANN is not a consumer protection agency but at the same time he was willing to go that far.

Mike Roberts: I don't believe he was correctly quoted at all. Further more, none of staff make policy pronouncement on a behalf of ICANN.

Bart MacKay of .cc: The question is what is ICANN role in that part as far as consumers go.

Mike Roberts: Certainly true that we have no statutory authority in any jurisdiction to enforce or deal with retail consumer dispute. We make it clear at the beginning. On the other hand, within .com, .net and .org plus new seven registries, there is a accreditation language that essentially commits the registrar to behaving as a good corporate citizen and not doing engagement to permit anti-consumer behavior to occur. I would expect the operators of all the country registries essentially have the same view.

Peter Dengate Thrush: Can I ask the members of the ICANN board how I would react to an application of 7.5 of GAC principle which said the delegee should not be subject to discriminatory policy or procedure from ICANN or relevant government. How do you see in the situation involving territory governed by a national government or disagreement between those two, the citizen of territory and the national government.

Louis Touton: ICANN is not a governmental. The challenge of ICANN process is to try to manage ourselves in ways to prevent the situation from happening so that the government has the kind of confidence to work these things among ourselves without the risk of government intervention.

Peter Dengate Thrush: Are you saying that 7.5 is meaningless because ICANN can't do anything? Why would we have that clause that said the citizen manager should be protect from arbitrary government involvement if we can't do anything about arbitrary government involved?

Louis Touton: The GAC principles in fact at least have two roles. One is they do constitute advice to the ICANN community about GAC view on how delegation ought to work. They also serve as advice by the government who are members of the GAC to each other and to other governments. So I see that passage as largely a best practices for a government. That kind of thing can be encouraged because ICANN staff, ICANN board, ccTLDs manager will lose the fight with governments if we get into it. But if we get government on our side, the responsible government say this is how a responsible government deal with. I think this is a power tool we can use.

Peter Dengate Thrush: Can you see ICANN and the Board putting the submission to the GAC to take this back to the members using that as a way making international government bring pressure to the row of governments.

Louis Touton: Absolutely.

Mike Roberts: I think it's important not to try to argue the general case from the pathological case. We have some countries violate the international law and what are the rest of the world community do about that. And we use the economic sanction for the pathological case.

Bart MacKay: Isn't the position the ICANN is taking with the trilateral situation only acting to accelerate the problem with the government in many cases by making the statement that ICANN will not do anything with the ccTLD managers unless there is an approval from government? Is there a third way to approach rather than a tri-party situation whether there would an ICANN agreement with ccTLD manager and a letter of commitment going to local internet community saying this is what we commit to you?

Mike Roberts: Let me try to give a practical answer to your questions because I think it's good questions. Some 50 members of national government that appointed to participate in GAC and to set principles which they have not uniformly put into practice as we mentioned to you because governments take a long time to do these things. That is one reason why we have ICANN because government is so slow and internet is so fast. On the other hand, membership of the GAC accounts for something like 98-99% of all the name registrations in the internet. Although 50 is a small fraction of 250 of all registries but it almost all of the activities of the name resolution in the internet. I don't think a trilateral thing is really a model of genuine significant concern among the countries that are the major stakeholders in the internet.

Bart MacKay: What we don't want to get into for .cc is the trilateral situation because of the action of the Australian sovereign.

Mike Roberts: This is a perfect illustration of why we need GAC principles and why we need a trilateral arrangement. You obviously are in the situation where local government is changing the ground rules and also it has participated actively in the development and drafting on GAC principles. Presumably the question is are their activities consist with the principles. The question whether the delegee is a profit or non-profit corp. does not prejudice the issue of whether the registry is being operated in the public interest of the local internet community.

Bart MacKay: One of the concern that ICANN publicly stated is "will you sign the agreement if the government does not agree?"

Mike Roberts: We publicly said we will not and in fact we have dialogue with government and the GAC. But since the explicit interest of the government in the status of delegation and the person of the dispute over that, we would not sign agreement until that rectify. You essentially are saying your legal ability to execute the agreement under the jurisdiction of that country is being disputed by the government. No one in the right mind would sign agreement under those circumstances.

Peter Dengate Thrush: As far as question arises where delegation who is outside territory not subject to jurisdiction of that government, would that government interest has anything to do with it?

Mike Roberts: Delegations are not made outside country territories.

Bart MacKay: In this particular case for .cc, the delegation was done back in 1997. We got a trustee relationship there. The question comes up at what point in time does the extension of the principles really occur. If that local community in fact endorsed this delegation, but the national government said no. Who then is defined as the local community?

Mike Roberts: You have taken us into area of ambiguity which will take some time to sort it out. Let's talk about the main stream of 99% interest of internet. Situation like .cc will play out and get dealt with under regime of national authority and the law.

Other issues

Elisabeth Porteneuve: I would like to ask Louis and Mike on Global view of ccTLDs as of today. If I notice from the ccTLDs that participate in most recent meetings which is about 50-60 registries, but from your list of those ccTLDs who paid as contribution to ICANN there is 98 ccTLDs out of 242. There are still a lot of ccTLDs that you have not got in touch with. Could you provide the overall picture of how many ccTLDs are active or discussing with you or how many ccTLDs would like to have trilateral relationship or how many ccTLDs are in difficult redelagation situation? This kind of statistical approach would be helpful. Louis talks about US approved the redelegation and I would like to know how long it will take to resolve the redelegation case. Where is the bottle neck?

Louis Touton: I share your concern about participation by the 242 cc out there. And it's been concern for both ccTLD constituency and DNSO. ccTLD Secretariat can fill me on what kind of participation we had there. There are many of those that have not paid and still receive the service from IANA. We hope we can bring them into community as active participants. Now the reality of ccTLDs as Mike eluded to, there are 242 of them but once you get beyond 20 or 25 in term of number of registration, you are getting down to small number. They have financial challenge in many cases so they can not send people to a meeting like this. That is the real problem that we have to figure out how to deal with. What is lots of participation that you've found among ccTLD managers, Elisabeth?

Elisabeth Porteneuve: I have been observed from Names Council Election. We found 82 ccTLDs answer the first votes and a little less on the second ballot.

Alejandro Pisanty: I am a bit concern on number of participation.

Peter Dengate Thrush: The question that was reflected back to Elisabeth is not the question that she asked. Elisabeth's question to ICANN staff is " do they know how many ccTLDs are active", " and how many are discussing things with you", and "how many of them are in difficulties". I would like to support those questions. I strongly recommend to ICANN staff to prepare a position paper for the Geneva meeting which provide a lot more statistical information that we had not been able to get about IANA activities and cc activities with IANA. I had for example asked the Chairman of GAC on a number of occasions to list the countries that have delegation and redelegation difficulties because of the extreme focus on the GAC. There are a lot of countries. Mr. Roberts said this morning there are still rising tide of challenge of cc delegation. Where?

Mike Roberts: I said at any give time there are four or five delegations with some levels of dispute from small to majors.

Peter Dengate Thrush: But you actually said there are still rising tide of challenges from ccTLD delegation. Let's have the number. Which countries are being challenged? So finally we can put that question back to IANA. I think if you could prepare for Geneva some statistics on which countries had delegation, which countries had challenges delegation, what is the resolution about it? Maybe there are some confidentiality questions relating to those number I like you to provide data without breaking the confidentiality. But Elisabeth's question on some basis statistical information would be the materials of assistant to this debate.

Mike Roberts: Due to our staff resource and need to work on agreement with individual delegees, you will not see any significant statistical improvements in three weeks. But it's a commitment to raise the staff level that we have.

Louis Touton: There are concern in confidentiality area. Particularly because the principle policy in time is first applicable to any kind of dispute is encouragement of confrontation within the country. There are a large number of situation occur. We continue to take the view that Jon Postel did when he was alive that it's not productive to those discussions to have everybody in the whole world focus on them. Most by far 80 % plus of any kind of dispute get resolved at that level we don't have to get involved. There are probably on the range of 15 -20 dispute of one kind or another and nearly all of them are still in the process of encouraging the different parties with different view point about how this thing should be resolved. They are however at the stage where IANA might be viewed it at the time. We are hopeful that they will get resolved.

Peter Dengate Thrush: I suspect the reality of the hard core dispute is actually very small and you can get them out in the open and realize how few they were and what impact they really have on running the internet. A lot of heat will go off the delegation/redelegation debate.

Elisabeth Porteneuve: When you raise the heat of delegation/redelegation, who give the ultimate authority on that? Whether this authority is within ICANN or within the country.

Mike Roberts: The issue on who has the ultimate authority or not will be resolved in the context of current discussion. We operate under the policy pronouncement. The US government took that position after substantial consultation with other governments. The U.S. government just completed Amendment 2 to the MOU in September which makes its intentions regarding ccTLD agreements very clear. I understand perhaps there are questions about it but those questions are productive to the process we engaged in. The government made it very clear and show the willingness to consider the transition completed and intention on reaching the reasonable agreement. In stead of looking at what US government would might do, we should look at what US government HAS done.

Abhisak: Regarding the number of participants, we have 95 participations on the second ballot of NC election and this is 95 out of 242 ccTLDs that we sent the ballots out. Out of these, we have about 30 ballots bounced back that means emails are no longer valid. So we may have more participations if we have the right emails. And our 95 numbers is actually close to 98 numbers of people that paid to ICANN. Basically IANA database must be changed or modified to reflect the right information.

Mike Roberts: We agree with that and we have a new staff member that will give half of a time to work on that.

Abhisak: Another point that I want to make is the number of participations of 95 or 98 is still far from 242 ccTLDs. I would like to propose on the outreach program between ICANN and ccTLD. Currently, ccTLD Sec. is working jointly with APTLD within Asian region. I like to do more outreach to other part of the world. We should discuss more this afternoon.

Louis Touton: Has anybody gone through and compare your list of 95 and our list of 98? That would probably be a good exercise.

Mike Roberts: There are two aspects on this. For those of us that involved in countries that actively developed the internet are continued to help those who have not got there yet to come along. If you look at this stage is pretty obvious that when you get down to 75 or so, you are down to fewer than five thousand names registration or maybe under fewer than 1000. This means that work at stake from the economic point of view is pretty small. It is pretty obvious that the people with active interest to participate in ICANN are the ones that active in the internet substantial registration and substantial economic and business questions that need to be resolved that fall under ICANN policy umbrella. Actually 90% level of all cc registration is about the number 24 - 25. I don't want to say that the only denominator. It's a steep curve here extend to which the local top level domain registry is a significant level of high activities. I doubt beyond the number 50 or 60, there are much of full time professional paid staff in these registries.

George Daniel of .ag: We have been looking at the ccTLD agreement with ICANN. What is the level of procedure to facilitate the agreement with ICANN?

Mike Roberts: The ones that are giving the first priority right now are the places that wish to complete these agreements under the trilateral arrangement. But we are perfectly agreeable to discuss with managers of any registries in agreement today.

George Daniel: How would you proceed?

Mike Roberts: The person that has necessary financial and management responsibility at the registry needs to get in contact with Louis.

Contracts issue

Peter Dengate Thrush: As for legacy situation, do you see as a possibility for cc to sign a thin contract to keep getting us the root server that we want but we don't want to take part in any of the rest. We just want to continue as we were. We are not going to involve in our government. Would this be a way or option for some cc instead of buying the whole package.

Mike Roberts: The general intention of staff and legal council is to have all the our agreement be as lightweight as possible. Louis has sketch out for you as legacy situation. It's up to all of you what you like to do. However, we should not ignore the fact that we have had specific advise from our GAC, not once but twice, that we should not execute any further registry agreement with anybody without the specific knowledge of government involved. This raises the question with me about do you have to send out letter to the local government involved. We have not sent the letter yet because we try to be as flexible as we can about it. On the other hand we simply can't take the position that the unsupported statement of the cc manager that their government has no interest in the delegation is adequate. That would expose us to a great deal of criticism. Any registry agreement we sign with any cc manager is going to be made with explicit awareness of the government involved.

Peter Dengate Thrush: Two quick comments. First before you adopted the practices as the start, the board should recommend FBI can practice and for the staff to take GAC advise to the board exposed you to further challenge. I suggest that part of ICANN bylaws and put the policy substitute just described you are carrying out through the ICANN process. Second thing, APTLD has been working on the draft letter to you recommending strongly that you not send that letter in the current form.

Mike Roberts: We are perfectly happy to discuss the future of the letter with you. And the fact that we have not sent the letter is the reflective of our concern that the letter accomplishes its stated goal. The question of whether the diligent of the staff insuring interested parties to agreement have been consulted is not a policy matter for the board. It's a matter of straight forward managerial and legal diligence. But before we sign a contract with someone, we have done some necessary diligence to determine the circumstances that the agreement being signed and the awareness of the government of a territory with regarding to that taking place is clearly a question of managerial diligence.

Louis Touton: In the way you (Peter Dengate Thrush) express in the interview and I wasn't on that particular phone call but I do know what resulted and turned into white paper. Yes, ICANN followed the principles that were in place at the time it started until things changes though the process. But one of those changes was inherited in the white paper itself and that is that the notion there would now be ICANN to be the forum toward determining the policies and you surely must adjust the status quo to that circumstances.

Mike Roberts: What government are we talking about that would not be aware of the fact that we are about to sign registry agreement? Are there any government representative in the room that would not be aware of our intention to sign a registry agreement and a draft document? I don't think so.

Elisabeth Porteneuve: At the very beginning, some delegations by Jon Postel and the issue was contemptuous and never taken into account. You can not just say later on that because it is the status quo it shall be for other this way. Some of delegations at the beginning were contemptuous and please keep this in mind.

George Daniel: What is the status of the ccTLD and how long can that status be maintained?

Mike Roberts: I think you are asking my previous statement that we would not sign the agreement in the present of outstanding dispute over the delegations.

George Daniel: That is correct.

Mike Roberts: Since disputes come in a lot of different flavor, I don't know disputes will get resolved in a week or in a year. It would depend on why there was dispute, what was circumstances and how did propose execution registry agreement related to the disputes.

George Daniel: In the future process, do you have cctld with all the agreements within the ICANN process and how long can that be maintained?

Louis Touton: I am just predicting here. With 242 ccTLDs with some not even operating, there were privately always be some ccTLDs with no agreement at least for the foreseeable future. We should focus on finding the trouble-free ccTLDs and try to work on agreements there. Then, try to find those situations that do have disputes and can be resolved and get agreements. This process will take years to get through but hopefully we can begin making some progress then.

Bart MacKay: Awareness and actual signing on the contracts is two different things. Please clarify what you mean by that.

Mike Roberts: The question was asked in the context of Legacy agreement and our position is that were prepared to execute the Legacy agreement provided that the government is aware what is being signed. My position on this is an awareness can be posting on cc web page of cc in question which is the most frequent form of reference for the local internet community including the government. It seems to me that if my manager or organization was about to execute the agreement, you put it up for public comment. That would be adequate evidence of all parties including government who is aware of the manager intention to execute the Legacy agreement with ICANN. The question of whether, when and in what manner a national government chooses to become interested and involved in the operational of its national registry is a complex one. This is not an area where ICANN either has a resource or the interest in getting deeply involved. But we do have an obligation under a reasonable application of Jon Postel to make sure all the parties including government are aware of what is going on.

Kilnam Chon: Regarding Outreach, discussion we are going through now only cover minority in term of global population. Like out of 90 countries, how many real countries and how big is the population and who are among the 90 that participate in this agreement process. The problem is a population wise. For example, India has 1 billion population but did not participate at all in this process. ICANN has deadline and want to come up with something by this summer. ICANN try to come up with major countries to meet the deadline. Can we make agreement flexible enough for the late comer? Can we accommodate those late comer? In Asia, much less than half participate in this discussion. Of course we try to do outreach but we don't have enough resource.

Louis Touton: The countries that are not participate now, the African and India. They probably feel some resentment that they weren't involved now. Basic structure got set and now they stuck with it. And I don't have good answer to solve that resentment problem other than just try to be as flexible as you can.

Mike Roberts: Kilnam, you are one of the internet pioneers and you participate in and watch these all grow up. ICANN is trying to do its work in a more sophisticated and difficult Internet world while preserving the values which you helped to put in the Internet in the first place. Is the internet in the value we are proposing to confirm in the written document reflecting mutual obligation between the country registry operators and ICANN? Is there a reason to doubt the history to bring us to this point in a way that those people feel significantly disadvantage when they finally get organized to come to the party. I like to think there isn't but obviously it's important question. I think you should have faith in your own history and background in this and particularly the depth of your experience representing Pacific Rim to speak to that issue.

Alejandro Pisanty: It would be a valuable output from WG-E of the DNSO regarding outreach. I don't remember the details. Prof. Chon, please look into this. It may be helpful.

Kilnam Chon: This is a complain to ICANN. Mike Roberts, you visited so many countries in Europe, most of them are small. How many time you visit China or India which has one billion people each. In a long term, population issue is much more important. I think you should cover 70-80 % of population as a CEO to have a contact.

Mike Roberts: Your point has been well taken. I think we look at all of our resource for outreach. The question of commitment to a lot of travel was specifically discussed by the search committee with my successor Stuart Lynn. There are improved ways to help our outreach.

Close for lunch.

Afternoon Session:

Financial Support Issue:

This session is kicked off by Peter de Blanc (ccTLD AdCom Chairman) presentation on "thin" and "thick" model to the audience.

Peter de Blanc:

Let me preface these remarks with a few caveats. First, this is not a formal presentation. There are no slides or spreadsheets, and any suggestions of fact are merely approximations. These remarks are my personal remarks, and do not carry the consensus or endorsement of my (ccTLD) constituency.

There are a few ideas that have been circulating within ccTLD community. The ccTLD contribution to ICANN for the year 1999 - 2000, the information is public information. Approximately 98 ccTLDs paid for the fiscal year 1999 - 2000. This adds up to 907,709.16 USD. The invoices requested by ICANN amounted to $1,496,000.00. There is apparently willingness on the part of those who paid to do so. We probably can say that with a little persuasion, not demand, 1 million dollars is not unreasonable. I should not say not unreasonable. I think it's all perfectly reasonable. Just on the face of Status Quo as it is, ccTLDs have not paid anything for the current fiscal year but we really have not published our position on paying. The general feeling is that payment will come with a contract for services.

In the past, 1.5 million was requested and only 1 million was actually received by ICANN. Out of these payments/invoices, 60 bills are not over $1,000, and 42 bills are $1,000 or more (There are some overlapping).

We are now starting to get down to the category that Mike and Louis talked about earlier where approximately 50 GAC member countries have 98 - 99% of all the (non COM, ORG, NET and EDU) domain names. About 90% of the registrations fall within 25 ccTLDs who are GAC members.

Calculating 90% of $1.5 million requested is $1.35 million. If we look at the remaining 2 % for all other ccTLDs that are not in the GAC agreed that is only about $30,000.

I would like to explore here a "thin model" and a "thick model".

Here is how to distinguish the difference. The thin model would imply that ICANN would be only technical and coordination body and that there is no policy involved. This is just a theory. As related to root server, there is no cost to pay for root server services proposed in ICANN budget group (which I am a part of). Some information from usually reliable but yet unconfirmed sources such as David Conrad and Paul Vixey suggest a reasonable amount of money to pay for a single root server operation, i.e. suitable housing, electricity, would be approx. 150 - 175,000 USD per server. There could be maximum of 15 servers I believe under current structure. So this will get us to about 1.5 - 2.2 million USD to pay for caring for these root servers.

Going with the thin model where the contract for services that ICANN offered services to ccTLDs and such servers being basically detached from policy and having only to do with root server services. (Putting aside the current ICANN budget which has to do with policy and meetings), I am going to the dollar amount for root servers services which is not a current ICANN budget or current ccTLD contribution level, let say we come up with arbitrary two million USD.

If it was strictly contracts for services under the thin model, that would mean 90% of the registrations (25 ccTLDs who are part of the GAC) would be paying 90% of the 2 million USD (1.8 million), leaving $200,000 to be paid by the rest of the 25 ccTLDs and it's very small amount paid by various domains that never show up at these meetings.

That is the thin model carried to the extreme. That is truly what some people involved in ccTLD business are looking at as how much should my ccTLD pay for these services. There are ccTLDs looking at nickel and dimes and value received for money paid and looking at that potential 2 million dollars divided by all these GAC members — to come up with whatever their reasonable and equitable share should be. And that would be a thin model. Who are the GAC members who are arguing for that position and holding to that position?

Five years ago, people who are running their name servers did a lot of these without any policies and mostly supported by whatever academic or technical admin. of ISP.

For thick model, if there is policy involved, once we introduce the concept of policy, then we have left (abandoned) the thin model (buying a service). We now enter into a new realm. We are talking about policy, ccTLDs agreeing to the policy or consensus setting policy and process of consensus formation, etc. By introducing the concept of policy that basically commits us to pay for the cost of policy formation, now we add on to 5 million USD plus 2 million USD for root server services that have not added in there yet. We now have gone into the realm of policy and into the realm of least 7 million dollars budget.

Practically speaking, the ICANN budget is too low without even getting into an issue of outreach. The budget should include at least 25 % reserve in case whatever things happen. In order to be a viable business, there must be a sufficient reserve. Also, the budget should provide for paying for all necessary services, and NOT relying on volunteers.

Now we are in a thick model. The budget is now 7million dollars which is going to be 30%-40% increase from last year budget. Naturally, we probably will get 40% increase in resistance of ccTLDs to pay that money. A few ccTLDs are not willing to pay what their assessment is. They have shown these by not paying in the past such as Germany and UK. They have paid $100,000 or $200,000 less than the original suggested invoices.

Interestingly enough a number of ccTLDs paid more. Under what circumstances they are willing to do so? It's a question of value received for moneypaid.

Let us look at the small African countries, for example. If there is any potential contact to begin building the concept of Internet development in those countries, it will be the ccTLD manager. In some case we cannot even find or know who that person is. First of all, we really need to reach these people.

If we are going to thick model, why not go with the REALLY thick model, a big fat model where we actually provide the service to the community and to the world, a service that goes beyond creating a little bit of policy. Let us provide some serious outreach. We only provide lip service to policy development if we only develop policy with 25 members of GAC. There is the rest of the world. They may not know enough the policy is being developed that may affect them at some time. China is only starting to know and I have met people from China ( and discussed this with them). So we have to outreach to these people. We have to maximize the thick model.

We could add $500,000 a year to ICANN budget for the first two years. $200,000 of that would be used for transportation allowance to bring these ccTLD managers to at least one meeting that happen somewhere around the world in one year. Another $200,000 would be used to send volunteers to travel to other ccTLDs that we just can't connect to them, maybe they can't get visa.

We want to know if that ccTLD manager in Africa has no back up power or other necessary things. We want to know who is actually operating it. What kind of stability inside that country and what kind of relationship is there between government and people. We can get them more involved in the policy development so they do not rebel later.

The remaining $100,000 out of $500,000 would be used to assure, perhaps under joint operation between ccTLD Secretariat and IANA, the rigorous maintenance and updating of ccTLD database. I have to congratulate ICANN for hiring more people to work on this. This activity is best done under the umbrella of ccTLD Secretariat.

One thing it (working within the ccTLD umbrella) does is to defuse the political down side with personal contact, contact that is not from the American company. My personal argument is for the thick model. My presentation is simply looking at the different models and what they are and what we would be doing within ccTLD organization, developing some position papers relating to these various models. In this case we are not looking for consensus. I think we will come up with a paper that has two or three models and signature of those managers who subscribe to which of these models and then present that to ICANN staff.

James Ross: It's very interesting discussion. The 10 cent per name using the figure ICANN has put out — 30 million names out there, I don't think how you can get to 7 or 10million dollars Peter de Blanc is talking about. With regard to the gap between that and the additional required, would that be also factored in per name basis?

Peter de Blanc: Two factors: one, probably double the 10 cent. Two, the fact that a large number of ccTLDs basically wanted to contribute to the development. They would get the social satisfaction. It's worth a lot.

Peter Dengate-Thrush: Can I ask Mike and Louis regarding $5 million? Is this $5 million been used up in startup expenditure, legal cost, NSI negotiation, new gTLDs formation? You can't be expecting that to be continued.

Mike Robert: I prefer to go through my presentation with revised structure and implication and will be happy to answer the question.

Elisabeth Porteneuve: Regarding contract with root server people, I have the same figure that Peter de Blanc got about 120,000 to 140,000USD.

Mike Roberts: I am not sure how the rumor got started. Actuality is the root server committee has reported to the board that the root server operators feel fairly strongly that root server should not be monitized to commercialize. The status quo on financing which is from the beginning should continue and the existing hosting organization where the operators have indicated that they supported that. Louis is very close to having completed those agreements but those agreements are for mutual set of obligations an unmonitized arrangement to continue on the basis that the root server was run today as a voluntary public service. There are funds that added to ICANN budget to support the cost of the new distribution master and some related things. Some of those were funded in our mid-year revision. They are part of documents that Peter just reviews this week. Those costs which were only for fund ending and six months of this year will have to be turned in as annualized level and incorporate under the new budget year starting in July. They do not include monitizing. This is the same fundamental principle that were in task force funding on the address registry. The address registry does not believe to monitize addresses. There are few scatter people who don't agree with that. But that seems to be a very strong consensus within that community. And on that basis, they have relatively minor share of the total revenue obligation to support ICANN at the present time.

Louis Touton: We have first stage agreement with them. We are near the second stage agreement. Financially, I don't hear any call from any operator for funding of their year-to-year operations. However, in root server system there will be a number of significant upgrade needed both distribution master system and after that for support of IPV6 , DNSsec improvement, Key management. ICANN will be expected to fund those improvement in some way.

Peter de Blanc: Mike, please clarify on monitizing the address registry. As an ISP myself, I was purchasing a connectivity from providers which came with certain blocks of address space that they already have. Towards the end of my operation with ISP, I was beginning to receive letters from the upstream suppliers that address space in excess of hundred of addresses that ISP would be expected to pay extra. At the same time, as a consultant on behalf of certain clients I was supplying the address space for organizations that were not ISPs. It requires advance payment of $10,000. I was reviewing the requirements if you are going to be commercialized ISPs or you are going to be end users. For commercialized ISPs, there is an ongoing annual fees. I have a little bit of difficulty in saying how the address space not monitized. Even now the company that I sold my ISP to now said you can get domain names come with one IP address. If you want more, it's five dollars a month per IP number.

Mike Roberts: This is not a time and place to discuss the issue between address registry and name registry. The history of this is task force funding spent a considerable time talking about it. The address registry pointed out that all three of them are nonprofit corporations that operate on cost recovery basis by charging the member dues. They try to keep the dues down. They feel it's in the interest of internet not to try to make money of allocation of address space because it would lead to tenacious behavior among the ISPs. On that basis they took a relatively strong position that they felt their contribution to the revenue budget ought to be on the order of 10%. I want to stress that they were successful in convincing the other registries people who are there. ICANN doesn't make this decision. ICANN serves as a forum for people writing a check to sort out perception of fairness. You pointed out that we don't have an issue of not having money, we have anissue of fair contribution level. The sense of that was that 10% in a continuation of an unmonitized address space was the place to go. If it becomes a policy issue with ICANN that ought to be equal treatment and equal monitization of names and addresses. Then, that will get to the board and would obviously have fallen out on the budget. But it's probably not a good idea to say the budget is going to drive an issue whether address space fall under new approach to how member dues and so forth.

Bart Mackay of .cc: We are firm believer of outreach but too much emphasis were placed on the fees being charged to ccTLDs being set aside for outreach program where it would include a bunch of travel for people. In most cases of ccTLDs that need assistants probably would be better served by strategic relationship with existing managers where they can actually help invest in the technology that they need, not just giving money for them to learn about things.

Mike Roberts: Before I make this presentation, I want to let you know that this is not an accounting seminar. I am not trying to get down to any budget detail. What I want to do is to give you some sort of overview of financial structure and particularly the revenue structure of ICANN because the issue of fairness of revenue allocation to the cc has been the matter of some contention through the whole history of ICANN and remains that situation today.

Mike Roberts's presentation on financial matter. Please read his slides carefully on financial structure.

SlideSlide #1 (Mike Roberts) Revised Financial Structure Proposal GOALS - Produce the funding necessary to support ICANN's mission and programs - Allocate fairly the responsibility for providing funding - Adopt structure and supporting algorithms that are perceived as fair in their application - Ensure the revenue system is efficient and administratively workable

SlideSlide #2 (Mike Roberts) Revised Financial Structure Proposal PRINCIPLES - Treat all registries in similar economic situation the same way - Where feasible, align financial support responsibility with registry manager - Continue fixed and variable components of algorithm for calculating contribution - For variable component, continue to use registry size - Provide for special circumstances

SlideSlide #3 (Mike Roberts) Revised Financial Structure Proposal REGISTRY CLASSIFICATIONS - Class 1a - Commercial, worldwide, unrestricted (.com, .net, .org, .biz, .info) - Class 1b - Commercial, worldwide, restricted (.pro and .name) - Class 1c - Commercial, country restricted (.tv, .cc, .ca, etc.) - Class 2a - Noncommercial, worldwide, unrestr. (nothing exists yet under this class) - Class 2b - Noncommercial, worldwide, restr. (.aero, .museum and .coop) - Class 2c - Noncommercial, country restricted (many cc fall under this class) - Class 3 - Special purpose (very small cc registries)

SlideSlide #4 (Mike Roberts) Revised Financial Structure Proposal REGISTRY CLASSIFICATION FEES - FIXED - EXAMPLE - Class 1a — $100,000/yr - Class 1b — $50,000/yr - Class 1c — $10,000/yr - Class 2a — $50,000/yr - Class 2b — $25,000/yr - Class 2c - $5,000/yr - Class 3 - $500/yr (current level)

SlideSlide #5 (Mike Roberts) Revised Financial Structure Proposal REGISTRY FEES - VARIABLE - EXAMPLE - At worldwide total name registrations level of 30 million, and requirement for variable component to raise $3 million towards ICANN core annual revenue budget, the cost per name registered would be close to 10 cents/year.

Open discussion

Kevin Meynell of CENTR: Firstly, You are suggesting that with respect to variable component you are charging you will leave that up to ccTLDs to decide among themselves, how much they should each contribute. So you calculate the overall charge for the entire ccTLD Constituency.

Mike Roberts: That approach would be abandoned under this structure. It was found to generate a lot of argument. It was not particularly favorer by gTLD registries. And when the cc community attempted to come up with an alternate to what we proposed, nothing could be agreed upon.

Louis Touton: I am not I exactly understood your question.

Kevin: Would you then bill ccTLDs on per registry basis or would you bill the whole constituency as a whole and allow them to break up the charge to constituency?

Mike Roberts: We have said we will prepare to consider aggregation proposal that will reasonable on their face and will propose by the registry. For other reason we have to have an individual agreement with every registry and we expect those agreements to include price caps in other protections. Then the questions about the utility to you as a community of aggregating it opens to some questions.

Kevin: You are saying you will do it on per registry basis.

Mike Roberts: We have to have a per registry agreement and agreement will have language about responsibility for financial contribution in it.

Louis Touton: At one end of spectrum, ICANN just could say I need 5 million USD and everybody sends us money. We learn that will not meet the basic requirement of providing stable source of adequate funding. So we do really feel a much higher degree of individual registry accountability in the form of here what you think you owe and you pay it sort of arrangement and the variable notion that in this particular proposal is basically a registry side proposal.

Kevin: How you propose to calculate a number of domain registered in each ccTLD?

Mike Roberts: We need to work with community to do that. We use information from some of our friends in the registry search last year. We did get our feedback that they did not feed us all the right numbers. It seems to me that community obviously has to the extent that where ever the proportion of variable fall out and how significant it is, you want it fairly a portion presumably you have incentive to help us the system to measure registry size accurately.

Kevin: If the registry would not willing to cooperate, how would you assess their contribution to ICANN?

Mike Roberts: I will give a legal answer. Since they obligation to support ICANN in the agreement with us according to some statement about the consensus financial structure and if they fail to do that, they would be breach the agreement.

Kevin: You would not make an opportunity assessment. You would just consider that a breach of contract.

Mike Roberts: The proposal revenue structure and assessment would automatically be part of agree-upon annual budget. Everybody would know at the beginning of the budget year where they stood. The question of how frequently you fix the registry sizes for the purpose of this calculation is again one of those efficiency and feasibility issues. Right now it fixes quarterly and dot com doesn't have to be that way and considering that with a handful of exceptions most of rest of the registry sizes are very modest and don't have a lot of utility to them. It may be perfectly adequate to do this once a year. In fact to do once a year in advance so that the people know fore warning exactly where they stand.

Kevin: My final question. With respect to 10 cents per name charge, I appreciate opportunity for negotiation as a round figure. Seeming increase registration over cost of time, this would yield the increase budget over time, you propose the charge per name would reduce in proportion to the number of names registered.

Mike Roberts: Here how the way the algorithm works right now in dot com for the sake of example. It's five hundred thirty five thousand dollars a quarter. It is 25% of the agreed upon expense dollar to be recovered for the year, fixed in advance. We get the registry figures of the actual volumes by registrar and each of those numbers is converted into a percentage and that percentage is multiplied $535,000 to create a quarterly invoice. This paper assumed something very close to that would be used because one of the fundamental changes and the TFF report is ICANN budget which is fixed by the board at the expense level. It is not driven by the revenue side. It's driven by the expense side. There is no provision for ICANN adopting a financial structure that is free running on the upside. In fact we are prepared to guarantee to every registry a ceiling on the annual contribution within some range that is also tied to ICANN long term consensus policy about financial support.

Peter Dengate-Thrush: I think you can see, Kevin, compared to the original proposal which is 30 cent a name to now 10 cent per name. We hope that one day it will come down to 1 cent.

Bart MacKay: One of the principles went into the proposal is having a mixture of fixed and variable expenses. We would possibly end up, for example our organization (.cc) in class 1A category, on par with dot com with fixed component and a number of registration per year will make it substantially higher than Verisign is paying. The variable aspect is actually much more than fair system to everyone but the same fixed component with dot com should be sub-categorized.

Mike Roberts: We are not searching for complexity. We are searching for simplicity. Since they strongly held view on fixed and variable, it seems consensus will involve some of both. If name registries can come to some agreement on fairness to get rid of one of those, you certainly will not have argument from us. There are pro "fixed" folks and pro "variable" folks out there and you all have to live together.

James Ross of .tv: This system doesn't scream out as a clear system. For example, if you work out 5 million USD divided by 30 million names, it works out to be 6 cent per name across the board. Maybe some confusion about 10 cent. This is not the whole cost. It's only variable cost. If we were to go to 6 cent per name, dot com would come out about 10 cent. They would do nicely about this deal. As for .tv, we pay about 60 cent a name. I think people need to do arithmetic on their own model. If you do break down by name, you will see significant discrepancy.

Peter Dengate-Thrush: Louis said you are agnostic about how the cc dealt with the split up of money. The response that has to be that cc as a group has been quite agnostic about the idea of making any contribution at all and why. There has been no ccTLD response to TFF report for this. We have to solve contract issue. Are there any room for legacy contract for those who want only thin model as Peter called it? We just want the confirmation of the current contractual relationship with ICANN and don't want to pay for others. How would you see that being further them in the process?

Mike Roberts: Obviously in principles, all alternatives are on the table. The creation of ICANN and expectation of the government and expectation of the major stakeholders and players are the people will participate on a common basis and they will be treated on common basis. I don't believe that the community or the board would say no. We are flexible but our flexibility is limited by the basic charter and bylaws of ICANN and the white paper, all of which assumed a common set of endeavor based on consensus agreement.

Peter Dengate-Thrush: But if the consensus agreement by ccTLDs is thin model, what would ICANN do?

Mike Roberts: That issue is a question whether it is a financial issue or policy issue. If it's the policy issue, the board will have to determine to prepare to partition any of the policy responsibilities. I would personally put the likelihood pretty low. On the financial side, the question of whether cc proposal would be acceptable to the board is basically a question whether other registries believe that it's fair and equitable to all the registries. We don't any issue of whether we have enough money but we have an issue of fairness. And no subgroup of the ICANN constituency gets to decide all by themselves what is fair. You are subject to peer judgment of what we are proposing or asking is fair across the DNSO and ICANN community. I am assuming that when you come forward with your proposal, you will be prepared to argue with your peer registries that this is as fair as it needs to be and suitable to your need.

Abhisak: You said the old model compared to this new model we might have to pay a little bit more extra. I look at .th, for example, we have to pay $1,167 on the old model. For the new model, we are under Class 2C and we have to pay $5,000 for fixed component and if we have 10,000 names with 10 cent each we will have to pay for variable component $1,000. So we have to pay $6,000 total compared to $1,167. This is not a bit extra.

Louis Touton: The discrepancy is perhaps just a data problem. A figure of $1,167 is based on 3,580 domain names at the beginning of the year. You record a much larger number here.

Abhisak: There are a lot of countries that were invoiced for $500 on this list of ccTLD Contribution for FY 1999-2000. Will they have to pay $5,000 minimum now?

Mike Roberts: No, the assumption of Class 3 threshold ($500 for fixed component) would continue on some basis that were satisfactory to the community. It's really up to you to decide what state of low economic activity you start counting on other things and send them bill for a small amount of money that makes them part of the community and makes them feel that their voice counts and they have relationship with us as justify by written agreement. Whether it's $500 or $5,000 based on what number of names and what kind of status, you all have to decide what is a fair thing to do and tell us.

Kevin: With respect to fixed band, this is more sub selection process row than ICANN selective? I mean Classification.

Mike Roberts: It is based on your self chosen economic model. A number of formerly nonprofit, noncommercial registries have been converted to commercial and this intend to take that business decision into account and to some extent a business decision to do that move you into a category of a higher obligation that pay because you are now proceeding on a basis that you will run a commercial registry and make money of the domain name system. It implied judgment of your peers that if you are doing that, you have a higher ability to pay and ought to pay more.

Kevin: This will be discussed within CENTR and will be commented in Geneva meeting.

Alejandro Pisanty: The cost of taxation should be balanced with the concept of the case running totally by the government. That would be significantly against the community and that argument would be backfired on those who lose.

George of .ag: If we are small country restricted, the fixed fee should be proportional.

Mike Roberts: The reason we published the staff paper and did the example is just to have the people and community do the analysis. If you want to rejuggle the things both by the number of layers and the discounts from high to small, it's sort of like doing the spreadsheet but the bottom line you still have to come up with two million dollars.

The end of Financial discussion.

Joint Activity: Outreach Issue

Abhisak: In term of Outreach program, currently we have the outreach program within ccTLD. We work with APTLD which also has an Outreach program by sharing cost. We are looking to ICANN for sources of funding to help this program cover in other regions like Latin America and African. Possibly paying through ccTLD Secretariat to perform the outreach. This would in turn help ICANN collecting more contributions. Let's get those who have not participated in the process.

Mike Roberts: We just start a budget group and we will publish a baseline budget in a couple of weeks. The floor is open for proposal for budget changes and allocation for next year budget.

Louis Touton: Generally speaking, for ccTLD outreach and ccTLDs from nearby countries maybe more effective to do that and get them involved in the ICANN process.

Abhisak: That is why we want to do this with regional organizations like AFTLD, CENTR or APTLD. For example, APTLD is doing the outreach in Asia by going to Lao, Vietnam, Cambodia, Myanmar, better than ICANN doing it. ccTLD Sec. can coordinate and possibly cut the cost down. CENTR is also doing outreach. I am not sure the scope of its program.

Louis Touton: CENTR got a pretty good penetration of Europe already. But you should definitely talk to Peter de Blanc and provide input to the budget group on that. In term of cost, from what I am hearing the kind of things you are saying would not be that expensive.

Kilnam Chon: First, DNSO working group reached an outreach and sent the final report to Names Council. Somehow NC decided not to take any action whatsoever. They don't even bring up to the board. Based on this final report on WG-E which is outreach and awareness, it probably is a good idea to set up a judge committee of the ICANN secretariat. We should try to identify not just in the budget group on what we suppose to do such as final report of WG is still valid or not or we have to revise. One obvious area is it may cost more money depends on what we want, who is going to develop materials. These materials need to be updated quite often due to the nature of Internet. Then, we should translate these materials into major languages. The next area of activities is actually doing the outreach in two ways. We have to go out to those countries. We also have to ask them to come to central location like we have APRICOT at the end of this month. APTLD has a jump start on outreach doing it two years ago. APNIC is doing very well now. CENTR is starting to picking up. Other areas we need to put more efforts like AFRICA and Central Asia. There will be a series of activities coming up. This is a very challenging work which we should do. Money is not really the issue but rather the people and commitment will be the major issue.

Mike Roberts: We would like to have more conversation with you. We are in the middle of trying very hard to hire full time staff to work on this area. If we were very lucky, we should have this person by Melbourne and we can have more conversation about it. But I would not count on that.

Louis Touton: It's a very challenging problem. I know NC are talking about this issue at the end. I don't remember when WG-E report is finalized and you are right that nothing really happens and I don't know why.

YJ Park: When I propose to WG-E, some members of NC commented that we didn't go through enough public comment period. But my understanding was at the time, it already went through all the public comment period. And this was about last June.

Peter de Blanc: This is obviously an idea that the time has come.

The end of Outreach Issue.

IANA database Issue:

Mike Roberts: I want to repeat a little bit structure on what I said in Marina del Rey. We are in the process of creating management database for registries that is flexible to use in variety of purposes. There is a severe legal limitation on us related to delegation database Whois. We are basically going to get ourselves in more flexible position where both your need and our need can be met in what we call "the registry management database". We are very much open to your suggestions about information that ought to be in there to help everybody communicate effectively and handle the administration business. But as you know any changes to delegation database may involve government action and we just simply can not be flexible about that.

Louis Touton: Some of the correction to IANA database actually turn out to be redelegation of matter under disputes at the time. It may not be appearance to you at the time. It's a mixture of things and we are looking ways out this. We move toward the regime where we are not relying so much on formal delegation but instead on the business contact people operating.

Peter Dengate-Thrush: We will get ready to talk about it in Geneva.

Kilnam Chon: I think we should spend ten minutes so several people here can make some good comments. Today we should really try to come up with issues and we can then elaborate in Geneva. Elisabeth and Abhisak can comment very much on this database through their DNSO and ccTLD efforts. This way Geneva meeting would be more productive.

Elisabeth Porteneuve: I am not prepared to talk about it today. I made a snapshot overtime on IANA Whois database. I put an engineer on the subject and I hope to have something from him to present in Geneva. We may learn from this on how to proceed next. When Mike said you are working on the new database, I would like to be involved in that process. We are partner with you. We need to protect the public database. We need to work together. Exchange information and updating it, some parts are contemptuous which I agreed perfectly with Louis. Even small changes may lead to redelegation. If we know how many contemptuous situations we have, we are able to manage it in good understanding all together.

Mike Roberts: We are sensitive to your point of view and we will do what we can to respect it. It is impossible for ICANN to delegation any legal responsibility associated with the delegation database. We do not contemplate any technical changes because they are not needed. The question of the extent to which that database is completely public versus partly private is I think you put on the table and we are sensitive to that and we are already dealt with it in a matter when there is disputes. The issue of whether the delegee in a given situation feels that some of the administrative history delegation changes is privilege with regard to that country and that registry is something need to be considered and we have no information about that. Any changes in amount of public information regarding to delegation database are quite sensitive. They would require us to consult with the board before we undertook them. And we are open very much to the views of the registries either individually or collectively about that. With respect to whatever software system and support we use to meet our mutual needs with regard to business data and billing and administrative notification and so on, we are happy to work with you.

Abhisak: The issue is we now will have a new database that can be changed or updated quickly.

Mike Roberts: It's pretty obvious our ongoing negotiation with a number of registries and governments and soon that they want something that has many of attributes files and that put a special burden on every body. For instance, any things significantly effect delegation record with .ca since they are operating under the umbrella agreement and we are operating under a letter from industry of Canada and we don't just do that on the phone call. There will be many more of those as we go forward. These are binding legal obligations that have governmental significant in some cases and they has to be dealt with carefully.

Kevin: I personally conducted a number of mail shots to the RIPE area and used to contact IANA database. I found 30% of contacts to be invalid either bounce back or contact just move on to other organizations.

Mike Roberts: We are aware of the problem there. We are gaining on the problem and we do not claim a perfection by a lot. You have to understand a lot of registries are pretty quiet and they don't respond emails all that fast either.

The end of IANA database issue.

Other issue:

Peter Dengate-Thrush: Consequence of the discussion we had today and that is we discuss a lot of issues about cc involvement, three or four different major areas with ICANN. I think you agree with the proposition that cc ask to come to a role they sort of changes and protection for decision making process. Do you have any thought about protecting cc interest with ICANN under the new regime dealt with all the issues we cover today.

Mike Roberts: One of the issue is the manner the supporting organization and at large have nominated and selected director ought to be reviewed. And I just finished couple days ago responding to that on the behalf of the board executive committee (we got two members of the board with us today) and proposing the meeting with the board in Melbourne, small enough group so that some conversation can take place to discuss those issues. Somewhat in the same sense of the question of money, development of ICANN bylaws took six months and went through seven drafts even before ICANN was created and you know that bylaws has been amended several times since. A lot of bylaws drafting and redrafting had to do with attempting to find consensus on the issue how to best represent community interest on the board. I think the board is perfectly happy and want to hear from you. Quite a number of people has direct connection with cc registries on the board already. They are in the position to comment on having experience on both sides.

Louis Touton: Because of the nature of ICANN being organization intended to be the forum for development of consensus, probably the best kinds of protection are participatory mechanism rather than restraint mechanism on ICANN although inevitably it will be some mixture. Over one quarter of current ICANN board are ccTLD people. We have Jun Murai with admin. contact for .jp. We have Jonathan Cohen who is on .ca for the trustee. We have Rob Blokzijl who is on .uk board. We got Ivan Moura Campos who is heavily involved with Brazil. There are at least five with very strong ccTLD connection. Of course, all the board members wherever they come from are legally bound. They really represent the whole community. They are perspective on the ICANN board with ccTLD experience.

Peter de Blanc: People on the board will have a ccTLD perspective. However, they got there by accident. They didn't get there because there is some mechanism in place to put them there. There is no guarantee for certainty that in the future there will be anyone that will end up on the board that will have at least a ccTLD perspective.

Peter Dengate-Thrush: Legally and emotionally, a director has to abandon as much as they can on human term where they come from to be adirector. A reason why we had a conversation with you previously suggested that supporting organization is in charge of policy for that area and there is a presumption that the board will follow. That is the key aspect of the supporting organization.

Louis Touton: How would you see such a supporting organization relating to domain names organization and cc is part of domain names?

Peter Dengate-Thrush: My personal view is that we would continue a representation in DNSO and very much lesser legal. It would reflect our current participation in DNSO. And the reason for that is we have so many issues to discuss and some many other issues in relationship with the rest of ICANN that has nothing to do with DNSO issue. Relationship between trademark and NSI and other members of DNSO actually take a small part of our role in managing cc. We would not abandon DNSO but I would expect that area would be somewhat reduced.

Elisabeth Porteneuve: I agree with you, Peter DT, I believe two channels is better than one. And like Peter de Blanc said that the chances that ccTLD had elected director to the board through DNSO is nonexistent. We need something to discuss or debate just like we are doing today. It is not of interest of people from DNSO.

Louis Touton: Have you thought about mechanism of GAC?

Peter Dengate-Thrush: We have and we excluded it so far because the greater intimacy that we actually has as registries than the government does. The ICANN does not apply to follow any government advisory policy. It seems to be a more distant anticipation than we actually have and actually won't.

Mike Roberts: I think the appropriate form to explore is the opportunity to sit down with executive committee in Melbourne and go over your view.

Peter de Blanc: At least there is a fact that there is a willingness on the part of ICANN staff to explore this. So we can in fact continue on with our constituency and say how we can get a situation where the necessary funding guaranteed. The necessary outreach happens so that those who interested in outreach, billion Indians and billion Chinese and all the people in Africa, can have their needs and satisfaction. Thank you for invitation.

Peter Dengate Thrush: I have great pleasure in thanking all of you for attending, a long way, a travel difficulties. We appreciate contributions you all made. I would like to thank directors from the ICANN board, Mike, Louis, Diane. We appreciate the quality of presentation that you have made. I think we move the relationship substantially in this last day. Thank you very much.

Meeting adjourn at 16:00.

© ccTLD Managers
Page updated : 2001-02-16 18:21:01