GAC / ICANN Board Joint Session ICANN Meeting Sydney, Australia 23 June 2009 >>CHAIR KARKLINS: Good evening. Good evening, ladies and gentlemen. It is my great pleasure to welcome members of the board to the meeting of the GAC. This year -- This meeting, 39 members participate in the meeting in Sydney. We have one invited country -- namely, Russia -- and we have two observers taking part in our deliberations. I have big, big pleasure to inform board members that China has returned to the GAC after a certain time, and we are very happy to welcome delegation of that country coming back and contributing to the deliberations of the GAC at the moment when we are at a crucial, crucial point in reaching agreement on important issues which may determine further development of Internet for a long, long time. I sent you a suggested agenda, and I hope that what we suggested is acceptable to you, and we can proceed accordingly, addressing issues starting with the fast track, and then new gTLDs, with all the questions we have around it, then we would talk through the institutional confidence, way forward proposal, and in that segment Paul Twomey would dwell more on the proposed tribunal idea. And then if time will permit, then we would address issue of the state of play with ICANN and UPU contract on dot post. And at the very end, we would lick to talk about meeting policies and organization of ICANN meetings and how to make sure that interests of all constituencies and organizations and committees are taken into account in preparing agenda of the ICANN meeting. >>PETER DENGATE THRUSH: Thank you, Mr. Chairman. And on behalf of the board, thank you very much to all of the GAC members for hosting us. This is one of the most important sessions we have at these regular international meetings, and I look forward to and I know the rest of the board does. Janis, thank you for the courtesy of the agenda topics. We are going to talk about each one of those, as suggested. You may be pleased to know that we have added one. We're going to have a further report on something that we have reported on a couple of times, and that is the progress with the WHOIS accuracy studies. We've taken the opportunity of preparing some slides, which we are in the process of trying to put up on your screen. But while that's happening, let me take the first opportunity I have had to say what fantastic news it is to me, and I think to the rest of the board, that China has returned to the Governmental Advisory Committee. And I extend a very, very warm welcome to the Chinese delegation. You are very welcome. It's my personal belief, and I think shared by members of the board, that the strength of ICANN is crucially dependent on having a strong representative GAC, and the return of the Chinese delegation to the Governmental Advisory Committee is a great stride towards that. So welcome. And now, if we can move to the agenda topics. And I am having difficulty reading that at this distance. But I see the first one is the IDN ccTLD fast track. Paul, perhaps you could take us through the recent developments on that. >>PAUL TWOMEY: Thank you, Peter. And thank you, Janis. It's my pleasure to address what is now my 34th Governmental Advisory Committee meeting that I have spoken in. And we have one of the first agenda items is to talk a little bit about the IDN fast track and where we are on IDN fast track. You will be aware, and I know you have had briefings this morning on the revised Document of Responsibilities and the associated explanatory memorandum that was posted before this meeting. The key points, and the document has been a matter of some discussion, obviously with ICANN staff but, importantly, with the ccNSO and also with members of this committee. The content specifically directed towards adherence to technical standards. It's very similar in content to the Country Code Top Level Domain accountability framework program that we have had in place. There is a dispute resolution mechanism, which I think is very important for reasons we will come to. And very importantly, in the context of the Security and Stability Advisory Committee's upcoming report, or report on redirection, there is a specific prohibition, including prohibition of wildcards, per their resolution. And their resolution, of course, also looks at other issues of redirection, which are an important issue. It's proposed that the form could be one of more of either a document of responsibilities and an exchange of letters or in terms of an application, an online form. And again, this has been developed in collaboration very closely with the ccNSO. The next question becomes a question of enforcement. And the community -- there are still divergent views on whether this should be obligatory or not. I think there is a sense from the board, but there are community views, divergent. We are going to need to have dispute management and enforcement mechanisms, and that is very clear. I think the report from the Security and Stability Advisory Committee makes that quite clear, that we will have, particularly, potential technical issues that may emerge in the implementation of IDNs, not only just wildcards but potentially other mechanisms used for redirection, which may require either dispute management or enforcement, or probably, even more importantly, put in the hands of ccTLD administrators, the mechanism of available for them to say to others, well, this is an issue which is of dispute with the ICANN community and we need to be in discussion with them. Because there will be times in this environment, I expect, as there has been in the ASCII environment, where domestic pressures will be brought to bear for such things to be introduced. The stability problem is very important across IDN TLDs if the standards are not followed. And that remains one of our great concerns. So that's -- I know you have had discussions about this today, but that's where we are in terms of the status. One of the issues that was discussed in terms of the IDN ccTLDs is some financial aspects. And there are actually three memoranda that have been prepared for this meeting and have been in active discussion with the ccNSO and others today. And they are -- the first four, there is memorandum on cost associated with ongoing ccTLD support. And this applies, overall -- it would apply to all ccTLDs, IDN and ASCII ccTLDs. And it's part of the expenditure area group report. Quite a detailed report that's been prepared on sources of costs. And that report reports that ICANN costs associated with the support of Country Code Top Level Domains is approximately 17% of the expense budget of ICANN, and around $9 million. And there is great detail behind that report, and I know that was shared with the ccNSO this morning. So that's sort of the overall costs associated with ccTLD support. The particular costs, there was a question asked here by this committee previously. The specific costs of development of the IDN ccTLD program, a similar methodology for costing has shown that the IDN program development costs have been about $6 million. And with the expectation of what would be a potential number of requests for an IDN ccTLD, it is proposed that the expense per request, assuming that these should be cost neutral, that the cost of actually implementing this new program should be borne by those who benefit from it, is about $27,000 per request. And in discussion, important discussion, I think both with members of this committee but very importantly ICANN staff with the ccNSO, and very importantly with the ICANN board, ICANN board members, the proposed -- and this next wording is very careful, prearranged recommended financial contributions would be a request processing fee of around $27,000, which is processing costs only. And in the discussion, which I think factors into the first point of the overall discussion with the ccNSO around fees contribution from the CCs towards the costs of the ccTLD support, that the annual contribution be somewhere between 1 to 3% of revenue based on the registry size in the local currency to recover the costs described. So that's a proposal that's on the table. And I can report to you that there was detailed discussion with this with the ccNSO this morning. It was not just around the issues of IDN ccTLDs, but broadly around starting the process of consideration of contributions from the ccTLDs as a whole towards those costs. Of. The IDN technical standards is an important topic. I sometimes use the analogy that the introduction of IDNs, after TLDs being in ASCII for as long as they have, is a little bit like taking a 15-story building, in which the columns in the building are built -- are all in red brick, and then replacing all those red bricks with multi-colored bricks. You have to do that very carefully. You have to make certain that all the doors still open in the rooms, that the lifts still work, the windows still open. And I can't underestimate or overstate the importance of adherence to technical standards to ensure that everybody's window is open, everybody's door is open, everybody's lifts work for it to ensure global stability and usability of IDNs. The IDNA protocol is under revision in the Internet Engineering Task Force. Indeed the former chair of ICANN, Vint Cerf, is chairing that IDN protocol group. It is anticipated that the revision will be completed prior to the launch of IDN TLDs. But as we have made clear, that's a contingency that ICANN has no ability to control. We don't get to set that. And I know many of you have either yourselves or you have members of the Internet community involved in that work. If the IDN protocol is not completed by the time everything else is ready for the launch of a fast track, there will need to be some very careful consideration made to determine the use of either the draft revised protocol or the 2003 version of the protocol. I have to say, if it was my personal -- one of my personal view, which by this stage will be very clearly a personal view by the time this comes up, I would really hope to think that we had the protocol completed by the IETF. But where the IETF is presently up to in terms of its draft revision is clearly, I think at least in terms of the view of ICANN staff, is clearly a big improvement upon the 2003 version. So we may have to think very carefully when we get too close -- when we get close to, hopefully, Seoul, what is the status of the protocol and think through what are the implications of us adopting a standard which has not been fully improved. Of course the IDN guidelines are well in place also for the second level. There are, in the technical standards, I think, potentially two very new topics of consideration we Ned to keep working through. One I have already mentioned, which is wildcards and redirection. And the second one is -- and I know this has been a topic of discussion today -- enforcement and validation of IDN tables content. And again, the issues of how do you actually -- how are we going to ensure some degree of enforcement and validation and discussion and cooperation is an important topic for us to think through in that context. So, Chair, that might be -- Do you want to take questions on the topic first before we move on? >>CHAIR KARKLINS: Yeah, thank you. We had intensive discussions on the issue on Sunday, and this morning we had a briefing given by Patrik Fältström through a video conference from Stockholm on technical standards, in particular on IDN tables. And I would like to ask Manal, from Egypt, maybe to walk through the current views of the GAC which came out from these discussions. >>EGYPT: Thank you, Janis. And thank you all for having the time to jointly have this session with us every meeting. Allow me first to welcome that the new version of the document on proposed documentation on responsibility has taken into account the option of an application form, which has been discussed in Mexico. And we all agreed on the importance of adherence to standards. And the GAC believes that this should satisfy the need for commitment to standards. And at the same time, leave the door open for other registries who may would like to proceed with a more formalized agreement with ICANN. Regarding the financial contributions, the GAC would like to reiterate the importance of keeping financial contributions voluntary, similar to what's going on with the ASCII ccTLD managers, as this was not clear from the current posted documents. The GAC also believes that the costs estimated between 25- to $50,000 for processing of each new IDN ccTLD is prohibitively high and would introduce a financial barrier to IDN ccTLD managers, especially from developing countries. I would like also here to add it was mentioned yesterday during the open session that it's an investment on both sides. I mean, registries also has invested in this. And in terms of readiness testing, capacity building, regional collaboration in developing language tables, script tables, variant tables and things like that. And this also should be -- we should bear this also in mind. I think that's it. Sorry, one last thing. And this is -- we might have not discussed intensively here because we have just heard from Patrik this morning. But I feel that the tables -- you are looking at me, actually, saying that ICANN should judge the content of the table. And this is not what I meant. I didn't mean the content of the table itself, but I meant at least there should be criteria for the submitted tables. I mean, for consistency, for checking, for comparing. I mean, IANA should not be accepting any tables that are being submitted. In fact, from a very personal opinion, again, maybe separating the submission of tables from the application thing, so that the application could either refer to one of the submitted tables or submit a new one. I mean, from a repository perspective, it should be structured, it should be -- because if we start this process, it is irreversible. Thank you. >>PAUL TWOMEY: So Manal, I'm glad you gave that clarification, because this is a complex area. And I think I would feel more comfortable with the way you have just positioned it. I'm not going to -- We'll have to take that under consideration, obviously. But as I think you are well aware, the degree of care or specifics in a particular script for one CC compared, for instance, to a gTLD who has to use the same script across multiple languages, you might find differences. And we have always followed the proposal that the variants table should come from the TLD. It should not come -- it should come bottom- up, not top-down. But I think the idea of criteria and specifics, that I think is very sensible. >>EGYPT: Again -- Sorry. Just to make myself clear again, because I have been misunderstood more than once. I was just proposing a criteria, and it's mainly housekeeping for IANA and not judging the content of the tables. And at the same time, again, some registries might not stick to the criteria, but then it's going to be their fault. Again, there should be some criteria. >>PETER DENGATE THRUSH: If I could just respond to the point about the GAC feeling that these costs might be prohibitive for some developing countries. I know you are aware that the bulk ICANN revenue comes gTLD registrants. Historically, the vast bulk of those have lived in the United States. And I'm sure no one is suggesting that those registrants have any obligation to subsidize the Internet in developing countries. I'm not sure whether you are suggesting that the ccTLDs should be subsidizing other ccTLDs. So I would be interested in any GAC proposals as to how the GAC thinks that cost might be shared or ameliorated -- the effects ameliorated. >>CHAIR KARKLINS: Peter, thank you for your question. I think there is a fundamental principle, underlying principle. I heard you -- you saying that during ccNSO session, a question of CC contribution to ICANN budget was raised. And it may happen that some process will be launched soon or -- sooner or later. And I think that this is the right way to do. Because no matter whether that is historical reason, personal, objective, subjective, these new IDN ccTLDs should be treated in the same way as ASCII ccTLDs. From political perspective, it is very important for ICANN's standing in the world to make this launch happen very smoothly and rapidly as soon as it is technically possible. And those 25,000 we should wait and balance very carefully whether international standing of this organization costs $25,000. That's my point. I think that standing costs much more than that. And we can find those 25,000 if that's the prohibitive for a country using Arabic, Slavic, or whatever script to launch a TLD under fast track. I don't know whether you noticed, but I would like to repeat what Manal said. In different regions of the world, preparation for IDN, launch of IDN, also has cost millions. And those millions and maybe tens and thousands of hours, working hours, the people who coordinated among the countries these variant tables and language tables and so on, these millions and thousands of working hours are not factored in in the cost analysis paper you presented. So maybe if they would be taken into account, then this would even out also some of those millions ICANN has spent on the preparation of this issue. So we need to make maybe an important political judgment in this respect. And I think that these 25,000 also may be recovered over time if we introduce this annual fees based on number of registrations. So this would be maybe GAC's answer to your question. And if I may ask you one precision. We don't really understand what stands behind terms "prearranged" and "recommended financial contributions." If you would be so kind to dwell on this, whether that is voluntary. If that's so, we are fine with this formulation. >>PAUL TWOMEY: I'm looking at my French colleague. Bertrand, no, not you. I said "colleague." You fell for that one. You fell for it. >>PETER DENGATE THRUSH: Janis, just while Jean-Jacques is collecting his thoughts, I don't think he was prepared for the question, so let me come back to -- I appreciate the matter of understanding. But that can cut in a number of ways, some which may have consequences that are less than desirable. And the suggestion that they be treated the same as the ASCII, of course, the origins of the ASCII one come from a number of U.S. government contracts. And it's U.S. funding that developed the Internet to the extent that it was and that led to the development of the ASCII. One of the consequences of that is that we live with a very interesting political and contractual relationship now with the United States government. If you're suggesting a continuation of the arrangement so that U.S. gTLD registrants fund a further round of Internet expansion, including IDN ccTLDs, think of the consequences of the political relationship with the U.S. as a result. So Jean-Jacques, are you able to help with the origins of where the prearranged wording came from? >>JEAN-JACQUES SUBRENAT: Are you attributing to me the ownership or the paternity of the term "recommended financial contributions"? I don't have a very clear memory because of jet lag, but -- [ Laughter ] >>JEAN-JACQUES SUBRENAT: In any case, it's an honor to assume something as original as that, and I hope as constructive. So the idea of voluntary contributions come from the fact or the finding that in international organizations in the multilateral world, actually, payments due by states on a contract basis or on a treaty basis are generally called voluntary contributions. So it was a question of vocabulary. "Recommended," because I think that it is clear that no organization, especially not a treaty organization, we are not a treaty organization, cannot dictate to sovereign states whether they should or should not pay, in fact, this or that fee. And that's why I think the idea of "recommended" came into play. So if you add all this and cook it carefully for about two hours, you come with "recommended financial contributions." And I don't think there is any element of mutually self-destructive elements in those three expressions. They can live together perfectly well. There's no internal contradiction between the three terms. And, therefore, I would, in my turn, recommend "recommended financial contributions." I tried to explain why it's termed that way. I think it's fair, because it recognizes the sovereign nature of sovereign states, but at the same time, it recognizes the reality that some sort of contribution would be necessary and in fact quite fairly called for. This being said, Peter, could I come back to an argument you produced a few minutes ago? I'd like to make a comment on that, about standing, et cetera. I think that whatever the past contracts may have been between ICANN and this or that sovereign state and the companies located in that state, I think that the future could be a slightly different picture altogether. And I think that for the purposes of the real and continuing internationalization of ICANN, we should, we could look at this in another way and perhaps suggest that if certain candidates find that it is above their means to make such a recommended financial contribution, then perhaps some financial assistance or method could be found. So I don't know, honestly, I haven't thought about this right through. And forgive me if I use a word which is not appropriate in everyone's ears, but some sort of fund or arrangement which would allow to meet the expenses of this or that candidate which the GAC considers worthy of that help or that assistance and would go towards meeting those costs. So I don't know how you want to call it, but that's the idea. >>PAUL TWOMEY: I think to pick up -- thank you, Jean-Jacques. I think that's a good explanation of the strongly recommended nature without the compulsory element. I think the second thing is, we are open to the discussion with the ccNSO as well, where there's a discussion potentially about how these ongoing revenues may be -- ongoing costs may be met. So I think we're open to further discussion on that. >>PETER DENGATE THRUSH: Just before Roberto, I just say this is at the moment a proposed position. It's not a position that's been adopted by the board. There are a number of members of the board who feel very strongly that there should be compulsory financial contributions. However, I think probably the thing that there is agreement on is adherence to the technical standards. There's a lot more room, I think, to discuss the financial contributions. Roberto. >>ROBERTO GAETANO: Thank you, Peter. In fact, you are introducing the subject that I wanted to bring to the table. There's a lot of discussion about -- as usual, I would say, about money. But I don't have the feeling that the issue about money is the more important that we have to face. We have the adherence to technical standards. We are going into a situation where we have IDN TLDs, which are far more complicated issue technically, where we don't know exactly the deployment to a large scale, what kind of potential problems could bring if there's no complete adherence to standards. I would remind that if it was a very simple technical thing, a completely predictable and no sweat to deploy, we would have had IDNs in the root since quite a long time. The reason why this deployment is carefully made and only after complete agreement on standards is because the technical community feels that adherence to standard is a necessity and it is extremely dangerous not to adhere to that. So I hear that the IDN ccTLD operators would voluntarily agree to adhere to the standards. I'm very pleased to hear that. But I think that we have also to figure out what exactly to do if one or more voluntarily decide to break the standard. What is the action that can be taken? Who takes it? Why is this entity taking the action? What is the authority that the entity has to take the action? And so on and so forth. So I think it's fair to say there is the willingness to adhere and we are willing to coexist specifically in this world and to share the common resource. That's fine. But I think that we have also to figure out what to do in case we have some misbehavior. And I think that that is the main issue. And I would like to hear, you know, opinion on that, what we can do on that. Going back to the money issue, I think that there is -- whereas for the technical standards, which is an absolute obligation to have mechanisms by which we ensure that everybody adheres to the standards. For the payment, I think that, as usual for the questions of money, there's a latitude and different approaches, and we can get to some agreement. I would like to point out a few things. We are talking here as if it was a member state that has to pay ICANN. It's not a member state that has to pay ICANN. It is an organization that is doing business. And for this -- now, that organization might be mandated and might do that under the authority of a government. Absolutely. But I think that there are cases and cases. We have seen the situation, for instance, in the ASCII TLDs in which we have had ccTLDs that are freely competing with gTLDs by using the ccTLD code in a way that is not related to the significance that this code has for the country. For instance, dot TV, dot ME. There are lots of cases. We cannot say that in that situation, we have an issue that is related to the country. Those are organizations that are operating commercially in a market, that are making profits. And that has nothing to do with the fact that -- with the relationship that they have with the country that has that code delegated. Now, there is a potential for this situation also in the IDN world. And I think that we need to find at least some provisions by which if we want to evaluate the case of an IDN ccTLDs for a country in the development world that is run on for-profit and for the benefit of the local community and so on and so forth, I think that we can discuss, then, the issue about money. But if we have situations in which then the IDN ccTLDs is run in a commercial way, competing with IDN gTLDs, making profits, accumulating profits, I wonder whether it is morally acceptable that that specific IDN ccTLD says, "No, actually, I voluntarily decide not to pay my share of costs" -- not fees -- "my share of costs for the maintenance of the infrastructure," which simply means that somebody else has to pay those costs. So I think that we -- I personally -- I don't know what the decision of the board will be at the end. But I am personally, I am extremely open to see situations that can favor the development of the Internet in countries for the local community that has been on the other side of the digital divide, so to speak, specifically for reasons that are related to the script that they use. So I think this is fine. But I would like also to warn on the fact that this is not the general case. So I think that we have to be able if we want to make a serious discussion about money to leave a bit the question, the matter of principle and to get also in some specific cases or provisions in which to run a commercial activity, making profit, compete in an international market with organizations that pay, that have contracts and pay fees, whether this -- leaving somebody else to bear the cost of the infrastructure, whether this is something that the GAC will consider an acceptable situation. >>PETER DENGATE THRUSH: Just a quick response, Roberto. The matter of the ccTLDs that are competing with the gTLDs is a matter for the ccNSO. Not really something that the board can deal with. >>CHAIR KARKLINS: Bill Dee, European Commission. >>EUROPEAN COMMISSION: Thank you. Just a couple of quick points. The first one is a bit of a reaction to Roberto's comments, which I think are fair enough, actually. But they do raise other questions in my mind. I mean, in terms of fairness, in terms of competition, we have a situation at the moment where most of the ASCII ccTLDs do not pay fees to ICANN. So what's the solution there, actually, in terms of fair competition between the ASCII ccTLDs and IDN ccTLDs that may be in competition with them? I just leave that question hanging in the air. The point I wanted to make and the reason I asked for the microphone is in response to the issue of enforcement of technical standards, which is something you mentioned, it's something that came up in our last exchange in Mexico. And I just want to share there with the board information that we received this morning from Patrik Fältström, who's, I think, a very well respected member of the technical community. And I suspect he may even be a member of the IETF Working Group on the IDNA. And we took the opportunity during our exchange with Patrik to ask him directly, actually, for his views on enforcement. And I think he was very clear that enforcement by ICANN would not be appropriate. But more interestingly, would not be possible through a contract. Now, I wouldn't think of giving you a detailed explanation of Patrik's arguments for that, because they are his arguments. But we do have a transcript, I think, available for that session. I'm sure you could have a look at it. But for the record, actually, it does reinforce the doubts that some of us had, actually, already in previous meetings about that particular argument for needing some kind of binding arrangement with the registry, IDN registry. And therefore, I would encourage the board to reexamine that particular issue. Thank you. >>CHAIR KARKLINS: Thank you, Bill. France, Bertrand. >>FRANCE: Thank you, Janis. I don't want to take too much time, but.... [ Laughter ] >>FRANCE: I'd like to replace this in a broader picture following Peter's comment. I, we, do not think that it is still possible in 2009 to say that the Internet is being funded by the U.S. actors, and that the budget of ICANN is mostly funded by U.S.-based registrants. And that is the rationale for not cross subsidizing. If I misunderstood your statement, I apologize. But I wanted to make that for the record. The consequence is that cross subsidizing cannot be treated as a competition matter, as if a TLD in IDN were a product, and where the main concern of ICANN were to not have the consumers of one product -- i.e., gTLDs in Roman ASCII -- subsidizing the users of another product, the new one, which is the IDN ccTLDs. This is not what it is about. As a matter of fact, IDN ccTLDs and IDNs in general are not a product. They are a long overdue right. They are not a benevolent gift from ICANN. They are a duty for ICANN to implement and to serve the communities that have been waiting for that, long. And as a final comment in that respect, on the business and competition between the new entrants that would be given preferential treatment versus the existing TLDs, can't we consider that a little cross subsidizing on a duty is an acceptable counter balance for 20-year head start in a market in. Latin TLDs of all sorts have been on those markets and serving customers for the last 20 years and more. If the guys and people who are introducing IDN TLDs when it becomes available are the ones who will compete against the others, they are the disadvantaged. And I think on a moral basis, this should be at least part of the picture that we should take into account. Now, covering those costs and discussing how they should be covered is a very important thing. But I'm sorry, I felt I needed to place those elements in the picture, because I'm afraid they are a little bit forgotten. >>PAUL TWOMEY: Thank you, Bertrand. And I can understand your sentiment. I think, probably just to clarify this point, a couple observations. I just think the discussion -- you should see the discussion as a reinforcement that the ICANN board members -- it often gets talked about ICANN should do this and ICANN should do that and ICANN should pay for this and ICANN should pay for that. I think what you have heard is a reinforcement of the culture of the ICANN board where the ICANN board members are very clearly understanding it's not their money. It is the money from the community. And so I think there's always a discussion in the ICANN board and a sensitivity that this is not -- you know, it's not our personal money or we inherited the money or something or other. This is the community's contribution and it's from registrants. So I think that's basically. I think the second one, which is this issue of fairness of contribution towards costs, has been an ongoing discussion inside of ICANN. And, indeed, there are strong requests in the strategic plan and operational plan for alternative sources of revenue, and such things like that, which have all been forces around this. And I think, frankly, and I was very pleased in the reports I received this morning from the reports of the ccNSO from my staff, I think the basic proposition is understood. And I think potentially, the timings might align. So when we were in Mexico we were having a discussion where the IDNs -- there was a sense that the IDNs were separable to ASCII when we were talking about this financial space. I think we are get to go a position -- I don't want to put any words in the mouth of the ccNSO, but I had the sense today that we were closer to a dialogue with the ccNSO where the two were going to be considered by them as -- in a both ASCII and IDN as being sort of a together issue. And secondly, that at least from some members, there's been some discussion of potentially, amongst their own community, they would help make decisions about how needs to be, quote, "subsidized" or not. So I think that's probably -- I don't think any of that is contrary to the point you are making, but I just thought I would put those points on the table. >>FRANCE: Sorry, just to clarify. This was in no way an indication in one way or the other against discussing contributions or to say it should not be contributions. It was the fundamental approach to why there should be a contribution that I wanted to highlight. >>PETER DENGATE THRUSH: Just a quick response, Bertrand. Thank you. It's always interesting to have these debates with you, and I appreciate this one. I am really interested, and we may take this off-line, in what the basis of this so-called right to an IDN is. From where comes a right to an IDN? And how can there be a duty? It seems to me what you are suggesting is all those volunteer engineers working at the IETF to solve technical problems somehow have some duty to continue doing that and somehow they owe people an obligation to complete this work and make this technical solution, this brand-new innovative thing available. You obviously don't see it this way. You think they are obliged to carry on, for some reason, because they morally owe a duty to deliver. >>FRANCE: If I may answer this very pertinent question on the question of right, why did I mention right, it's because it's a question of equality of cultures. It's fundamentally an argument that I think the ICANN board and yourself on many occasions have very forcefully argued in favor, that at the current stage -- and I don't mind that much, because I, like all Latin character users, I am relatively favored by the current situation, but I can understand what the situation would be if those who use Latin characters had been forced to use characters that they don't master for the last 20 years, and they would be waiting to get the Latin characters. So I don't want to make a long answer. The question of the right is a question of balance. It is a responsibility -- let me put it this way. It is a benchmark of the commitment of this organization as a regulator, de facto, of this space to provide an equitable access to the domain name space as quickly as possible. I think it is a duty, but I may be wrong. But I agree, it's an off-line discussion. >>CHAIR KARKLINS: I think that we should cut this discussion on fast track. >>PETER DENGATE THRUSH: I see Steve Crocker at the back. Steve. >>STEVE CROCKER: If I might, I have been listening to Bertrand -- >> Microphone. >>STEVE CROCKER: Is there one available? >>CHAIR KARKLINS: While Steve is coming to the mic, maybe, Brenton, you can say. >>AUSTRALIA: Thanks, Janis. Perhaps maybe we might actually be in agreement on this issue, and perhaps what we need to do is separate the idea of pricing principles -- broad, cost-based pricing principles -- recognizing that that's the most efficient outcome, from the issue of development. Recognizing that there is going to be a gain, especially to developing countries from getting as many people onto the Internet as possible. And the telecommunications industry has done this for a hell of a long time. They look at efficient pricing through various systems that are cost based, that allow international roaming that allow sharing of networks, and they also have an approach that allows for development of various countries that are disadvantaged. And the most sensible way may simply be to actually look at broad pricing, cost-based principles, and then actually identify those countries that we may be able to help to encourage their development through some sort of discounted pricing arrangement, or perhaps an exemption from fees, recognizing that we are all going to benefit from getting as many people onto the Internet as possible. Thanks. >>CHAIR KARKLINS: Thank you, Brenton. Steve. >>STEVE CROCKER: Thank you. And I appreciate the contribution, and I want to follow up on this. So I listened very carefully to Bertrand, and it was very moving. It echoes similar sorts of arguments that I have lived through in my life in the U.S. and the distinction that I think is important is equality of access. The opportunity. It would be absolutely unacceptable if we passed a rule that said unless you only use English, for example, or to exclude certain groups. But that's considerably different from therefore saying, "And we'll do it for free," and allocate resources and so forth. So I think the underlying idea that you have, I think we all are strongly supportive of. That moves, then, to the point that Roberto so eloquently said. Somebody has to pay for something. And to cover that up and to say under the umbrella of "but there's a principle and a right" and so forth I think is not quite the right -- it conflates two things. So we have to have a businesslike discussion about where the resources are going to come from and who has to pay. We can make some allocations in favor of those who have not had the resources in the past, we can cross the digital divide, but we need to do that explicitly. And I would want not to join that with the separate kind of argument that we do favor equality of and access for all cultures and we do favor bridging those gaps. But that does not translate into forced labor for the IETF, just to make that point rather strongly, or forced subsidies and so forth. Thanks. >>PETER DENGATE THRUSH: We're ready to move on. >>CHAIR KARKLINS: Thank you, Steve. That leads us to the next question, new gTLDs. >>PETER DENGATE THRUSH: Now I'm going to talk a little bit about just the overarching issues and progress on overarching that have been identified in the new gTLD program. Obviously this is an enormous exercise and there's things going on everywhere, but if we could just click through this. And because we have taken, I see, Mr. Chairman, over half of the time on just the opening topic, I will go through this reasonably quickly, if that's all right. First of all, we have a root zone scaling study to report, studying the expansion of the root zone. And this is designed to study the effects of the new resource records for DNSSEC, handling the installation, introduction to the root of the internationalized top- level domains themselves, and the effect of new address records and glue for the IPv6 solution, and then new gTLDs in general. The timeline for the study. The team has been assembled as of May. The terms of reference are available on the Web site. They have had working meetings here in Sydney. A draft report is expected by 30th of August, and there will be public review and comment through the ICANN annual meeting in Seoul in October. Another one of the overarching issues, the trademark protection. You will know that the board in Mexico formed a team, the implementation response team, who have gone and worked extraordinarily hard and come back with some solutions that they proposed to the community in relation to protection of rights. Some of those measures are set out in IP clearinghouse, Globally Protected Marks List, so forth. A mechanism of uniform rapid suspension. A new post-delegation dispute resolution mechanism. They have addressed the WHOIS requirements for new gTLDs, and worked on the use of the algorithm in string confusion. The timeline for discussing this, well, they are working -- they began in Mexico. The draft report was posted on the 24th of April. The final report was available through the 6th of July for public comment, and there will be further consultations here in Sydney already, and more to come. And then in New York and London in the next few weeks. There's parallel work going on in the implementability of those -- sorry? >>PAUL TWOMEY: New York, London and Hong Kong. >>PETER DENGATE THRUSH: Sorry, I understand we have added a consultation in Hong Kong. So Sydney, New York, London, and Hong Kong. And there is further groundwork going on in relation to implementability which is going to go into draft three of the guidebook. A lot of focus on malicious conduct and obviously preventing it. Phishing, financial fraud, identity, et cetera. This has now been separated out in the discussions from issues of branding and trademark protection. Different processes are required. And there's outreach to the Anti-Phishing Working Group, the Internet registry, Internet security group, FIRST CERT community, and other financial institutions. Recommendations to come from the Anti-Phishing Working Group and SSAC and others, is there should be enhanced vetting processes for applicants, must be a greater level of scrutiny for the strings themselves, they provide a single abuse point for contact, so there's a new way of contacting registrars. Improvements to the WHOIS data -- it's a reporting system. It's a reporting system that we have, and there's been improvements to that know when there's complaints about WHOIS. How the follow-up is working to check whether there has been any progress. There's increased reporting, auditing and compliance requirements for the new registrars. And verification of registrant data is being improved. There will be sanctions for registries. There is going to be thick WHOIS and there is going to be a new rapid takedown process for discussion. >>PAUL TWOMEY: Peter, I wonder if I can just add on that rapid takedown, which is often a topic raised by people concerned by malicious behavior, is I think in this discussion, we'll see the usual sort of pendulum process in an ICANN discussion about something like this. The pendulum is first going to move to how can we put in place rapid takedown, how can we have good mechanisms dealing with this concern. The pendulum will need to move back, in some respects, to deal with two issues, one of which is handling of legal liability issues for inadvertent mistakes. So how do you deal with the error rate issue. And one also to ensure that the request was done for the right purpose, that this really was about malicious attack. It wasn't somebody using the mechanism for another purpose. And there is going to be some -- You should just be aware that we are already making that point to the intellectual property community and others that what looks strong to have rapid takedown at the beginning of the pendulum, we will need to have some, also, safeguards to make certain we don't do it with errors or inadvertent of purpose. >>PETER DENGATE THRUSH: Thanks, Paul. Checks and balances. Very important. So the timeline for that, again, consultations around the Sydney meetings here, then again, New York, London and Hong Kong. Work is going on with the banking and finance sector on the security- specific TLD concerns, and we are developing appropriate mitigation measures for version 3. Next topic -- Janis, if there are questions on new gTLD? Steve. >>STEVE CROCKER: I just want to make a very quick comment. A small clarification on the timeline for the root scaling study. We set a very aggressive schedule, and that date of end of August is actually for the delivery from the contractor to the steering group. We did build into the schedule a very explicit period for public comment following that and prior to Seoul. But I think the date to have in mind for when you will see something is probably close to two weeks later, maybe a little faster than that. But mid-September is probably a better date to socialize rather than end of August which is an internal date. >>PETER DENGATE THRUSH: That's for discussion at Seoul? >>STEVE CROCKER: Absolutely. >>PETER DENGATE THRUSH: So my question was, that's for discussion at Seoul? And the answer is absolutely. Thank you, Steve. So questions on the overarching issues in relation to new gTLDs? >>CHAIR KARKLINS: Peter, maybe Paul can walk us through economic analysis, and then we will go on all subjects, because we have one person who will introduce list of big issues from our side. >>PETER DENGATE THRUSH: Mr. Chairman, I'm in your hands. >>PAUL TWOMEY: Thank you, Chairman. So the issue of economic analysis as relates to new gTLDs has been underway for some time. It has a number of foci, one of which has been around the issue of consumer benefits and -- of introduction of new gTLDs, and the other one has been about issues relating to structure and particularly around registrar and registry separation. We've had a series of reports, we had a CRAI report early this year, in June, we had a report by Dennis Carlton also on the proposed mechanism. I think probably most importantly, we had a dialogue yesterday with a series of experts, economic and legal experts, around particularly -- some discussion around consumer benefits, which I think, frankly, they had -- took a fairly straightforward introduction of competition is good for consumers approach. But there will certainly be a fair amount of more work done on that. And secondly also then on issues of industry structure. We expect, as of this afternoon, or at least in terms of what's been happening in this meeting, we expect that there will be a further working group of -- with at least significant economist expertise convened to deal with potentially issues around industry structure, and that that group will also obviously be further expanding on issues of consumer benefit issues around introduction of competition. The consumer benefit aspects, of course, are at the heart of their discussion around the structure issues. So that's -- (inaudible) the 30th of July. But, I think, frankly from everything we've heard today in the various -- what we have been in consultation on this, but we expect that panel will be coming together as part of a working group. >>CHAIR KARKLINS: Thank you, Paul. And now I'm turning mike to Mark, United Kingdom. >>UNITED KINGDOM: Thank you, Janis. All of us in the GAC have been following the gTLD project with close interest, of course. And as a project that has significant potential to increase competition, promote innovation, and increase access worldwide. And we've been discussing the gTLD round, proposed round for next year at great length and at this meeting and the previous meetings and on conference calls in the intervening periods between meetings. And we'll be submitting comments on version 2 of the draft applicant guidebook and looking forward to contributing to the finalization of the third draft. As Janis indicated, we've also been compiling a list of the -- what we see as the key public policy issues and questions that are of concern to us or appear to be unresolved, appear to us to be unresolved. And we'll be finalizing that list in our discussions tomorrow for inputting to you and for others in the community to see. Ask we thought it would be useful now to signal to you some of the key issues that we have identified. And the first one, as you probably would expect, is one related to the paramount concerns of government about the impact on the stability of the Internet. And we will certainly look forward to looking carefully at the root zone scaling report in mid-September as providing reassurance to us that the scaling up of the root will not impair the stability of the Internet and that the technical safeguards are sufficient. Economic studies. We registered our concern at the Mexico City meeting that the reports available to us then did not provide appropriate answers to the board questions you yourselves put. And we're also concerned that the second report that's just emerged does not really provide a comprehensive economic study that we feel would ensure that the whole GTL project is soundly based in terms of its economic impacts and so on. So we really urge that further studies be undertaken on the economic impacts. And one of the key issues that discussions in the GAC has highlighted is the lack of empirical research on defensive registrations. We feel that still is an area that needs further study, with particular, with regard to the cost burdens likely on business and how they balance out with the potential benefits to consumers. A third issue that we feel has received not enough attention is the risk of end user confusion. Our expectations are that the average Internet user will want to see ease of navigation around the DNS remain unimpaired with the expansion in the number of gTLDs. And so we really feel there's a need for more studies to be commissioned which assess the impacts of a radically changed gTLD environment, the impacts that that will have on end users and the extent to which expected proliferation of domains may cause confusion. A fourth issue, again, a kind of preparedness issue that we feel is one that ICANN needs to address, is the low level of awareness amongst stakeholders and the business community, in particular amongst small and medium-sized businesses outside the established Internet industry and community. We impress upon you really to actively promote the opportunity for business that the gTLD round creates, to undertake that promotional activity in the period running up to the launch of the round. And that brings me on to some of the key issues with regard to the implementation of the round. And I should say at this point, we're very grateful for all the information and advice that Kurt has given to us here, both here and also on the conference calls we've had with him. He's been very frank and very full and very helpful in the advice and comments that he's given to us, and we really appreciate that. We are likely to make proposals on gTLD categories. We feel that ICANN has not really fully considered whether a more category-based approach to the introduction of new gTLDs should be considered. And we have very much in mind the situation of noncommercial, cultural, linguistic, and regional gTLDs, which governments really look to to stem some cultural diversity on the Internet and the creation of local content and so on. They are not going to be in a -- They're in a different position, really, to the commercial operators. And we're also mindful of governmental bodies, such as the Universal Postal Union. We'll touch on a separate issue with regard to the UPU later on, I hope. But governmental bodies are obviously in a different position from purely commercial registry applications. We've also been looking closely at the objection procedures and costs issues with regard to the introduction of the gTLD round. Our overarching concern is that complexity and cost of objection procedures do not amount to a barrier to legitimate objections by interested parties. We've also been considering the proposals for handling moral and public order objections. All of us I'm sure believe that great care needs to be taken to respect the sensitivities of words and terminology in relation to national, cultural, geographical, and religious significance, and we in the GAC seek further guidance from ICANN on the practical modalities for addressing objection that will be lodged on -- inevitably will be lodged on these grounds, including the proposal to establish a panel of -- the panel of three judicial experts. How would such a panel really take full can account of all the cultural and different legal interpretations that exist around the world as to what is morally offensive or threatening to public order? The draft applicant guidebook appears to require governments to follow the same procedures and pay the same costs as other objectors. For example, to pay for objections to applications with city names. And there may be other public-policy issues that governments may wish to lodge objections on. But, obviously, the time and cost implications of the formal objection process may not actually meet the requirements of governments. So we really would look to ICANN to consider the position of governments in terms of lodging objections. We note that there is the proposal for an independent objector, and that may prove -- may provide one avenue for governments to object at no cost. But we're wondering if consideration could be given to the creation of a separate special procedure for government objections. Sorry, it was a long list, and I'm only selecting a few of the key ones. But I'll finish off on one long-standing issue of concern for us, and that is the single-fee structure, which we see as creating limitations in those that are going to apply for gTLDs and is likely to skew the market in favor of applications from those who have the financial resources to lodge applications. And many of those will obviously come from advanced economies in the developed world, so there's a real -- a kind of a digital divide angle here that has to be taken into account. And, generally, we feel that there needs to be a greater transparency about how the fees are going to be arrived at and the kind of costs on which they're based. So there will be a fuller list for you to enjoy which we'll finalize tomorrow, as I said earlier. But those are some of the key issues. And I'm sure colleagues here will want to chip in on some of them. And we look forward to some of your comments today on issues that really are deeply of interest and concern for us. And we're looking at these issues very much in a positive light, because as I said right at the beginning, the gTLD round has the potential for making a remarkable contribution to the development of the Internet. And we just need to be sure that some of the policy approaches are firmly based and empirically supported, as appropriate. So thanks very much for what I hope is not a tirade, but one I hope to trigger some productive discussion. >>PAUL TWOMEY: Thanks for that. Maybe it's not appropriate to respond to all of them now. We'll wait to see what you present tomorrow. Perhaps just a couple of observations. Empirical work, as required, I think you said, or as appropriate. I think the second two words are actually quite important, because I think a lot of economics around the way markets should operate has been around for several hundred years and there are some fairly basic principles and doesn't always require that you go out ab initio every time as to whether you want to have a new Skype or whether you want to have a new whatever. So I think the nature of the studies, we'll wait to see the request. I would make this observation, and forgive me for having said it. But we went through -- This issue of competition and choice for new gTLDs is a ten-year-old discussion inside ICANN. And I can report as a former chair of the GAC, we had quite a lot of discussions around it. And I recently had reason to read back through the communiqués when I was the chair of the GAC and see the number of times we said in the GAC, "Competition and choice for new gTLDs is a good thing." Indeed, we passed resolutions -- I see my former colleague from New Zealand nodding -- we passed resolutions actually calling for ICANN to accelerate the work on them. So that's the first point I'd make. And the second point I'd make is that in the last -- leading up to June last year, there was, I think, three and a half years of work done inside the GNSO process, again, then, with consultation, obviously, with other supporting organizations, including the GAC, where some of the issues that you've raised were actually discussed in quite a lot of detail. The one that struck me particularly was the category-based approach. And there was quite a discussion, quite a long discussion, about category-based approaches and challenges they faced, and eventually the decision made to go with the present proposal. So I just want to flag that, because while, obviously, it's valid and you have a concern to put up, but you must expect that potentially some of our responses will be, "Well, we've actually had a long discussion about this and this is what was discussed at the time and here's what the decision was." And I do appreciate this has been a long process and you people come to the table -- And having served in government, I do understand also that there are often issues you tend to look at the issue the closer they get to reality because of the time demands that everybody has. But let's look at what the list is. But as I said, there are aspects there that certainly are -- have been addressed over the period of time or have been discussed over the period of time. There are some areas that are quite clearly very pertinent to now. And so let's see what it is and we'll certainly respond. >>PETER DENGATE THRUSH: I'd like to join and thank you for the care of the prep- -- that's gone into the preparation of that. And it's comforting that the GAC is paying such attention to something so important. And I look forward to the full list as well. I -- when you were preparing that list, I just wonder if you could help me and clarify what it is about the government -- about governments as objectors and their requirements that might be different from any other objector. >>CHAIR KARKLINS: European Commission. >>EUROPEAN COMMISSION: Thank you. I'll take that one, because I think I was a main motivator in adding that to the list. I was surprised and I think some of my colleagues were surprised to see that we would have to get in line with everyone else. Because I think our understanding to date has been that the GAC gives advice to the board. That's what the bylaw says. And in previous rounds actually, if there have been concerns about particular gTLD applications, it would have been given -- advice would have been given to the board along those lines in a communiqué or whatever. And so, to be blunt, it looked as though we were going to be charged for giving advice to the board, which seemed rather strange to me. But seriously if one looks at the bylaws, the normal avenue for governments and the GAC or the GAC as a body to express concern or support even, I suppose, for a particular policy or implementation of a policy would be in the form of GAC advice to the board. So it seems surprising to me, actually, that governments will be asked to circumvent that bylaw provision, effectively, and to take part in the new procedure. >>PETER DENGATE THRUSH: Well, this is when you give policy advice to ICANN on new gTLDs. You don't -- and we're not expecting you -- to get involved in any individual applications, which will be an implementation decision. We don't look to the GAC for any advice on operational matters. So once we have sorted out what the rules for new gTLDs are, and we -- and that's the process that we're going through right now on all of those matters that you raised -- excuse me -- once that's done and the policy is in place, subject to all the usual input from the GAC, it becomes an operational matter. >>EUROPEAN COMMISSION: Well, I take that under advisement. I don't want to start reading the bylaws to you, actually, or finding them in my file. But my understanding was that the GAC had the right of initiative itself to give advice to the board. We don't just wait and we don't have this kind of -- these kind of parameters set by the board for us. If we decide it's a public-policy issue and we want to advise the board, the language is fairly generic, actually. So I'm surprised to hear you say that these red lines exist. I certainly wasn't aware of them. I thought it was a mutual understanding that you would seek advice from us, actually, when you felt you wanted it and we would give it to you when we felt you needed it. But perhaps there's something that requires further discussion in context of the relationship between the GAC and the board. Thank you. >>UNITED KINGDOM: If I could just chip in as well. I think a lot of us have the expectation that we would be able to intervene on individual applications. But it's the manner, the modality of how we actually do that. I mean, I think it was explained to us that there would be a public comment phase where individual applications would be opened for others to comment. But would governments have to lodge -- have to pay to lodge particular objections in the course of that comment period, or through the independent objector? That was the issue of concern to us. >>PAUL TWOMEY: Well, I think those -- you have just outlined, I think, several different processes. And I think the processes really do matter. So I don't dispute at all Bill's analysis of the bylaws, but say that the GAC can initiate an advice to the ICANN board on matters of public policy. And the question becomes is the determination of an application for a particular TLD string a matter of public policy. Question mark. I am not giving you an answer. I am just saying question mark. If there's input on a particular -- a particular objection or a particular concern under existing -- on existing approved policy by the ICANN board, which is the full objection processes that presently sit in front of us that came up June last year, what's the procedure if a government or -- wished to input on that particular one, either directly, or as you say, potentially through the objector or some process. I think that's a separable process discussion. And the reason why I think it's worth us passing that and talking it through, frankly, is that a large amount of the advice from the GNSO process -- indeed, I think Bruce Tonkin might be here to speak to it, but a large amount of thinking for that three-year period was learning from the previous rounds and saying as much as possible the ICANN board should not anyone the position where it is determining -- that as much as possible, there should be third-party -- specialist third-party arbitrators or adjudicators who make decisions on particular points rather than each one of those matters comes to the ICANN board for a decision. And for us it's actually a very important point because it also brings into play aspects of compounding legal liabilities. And about clarity and transparency of process. And so there's been a lot of thought put into that. So there's not a - - I don't wish to indicate any -- in no way would that indicate that the thought has been how to exclude governments from being able to participate in that process. There is an expectation that governments should be able to be included in that process. What is important as we talk this through together is these issues of transparency of process, of compilation or compounding of legal liability concerns for ICANN as a decision-maker. Now, I don't want to give you the sense that -- Bill's statement about the bylaws is absolutely right. It does come down to your views as whether you think a particular string or whatever you might have a concern about, is that a matter of public policy or does that fall under these other processes, and perhaps that's something we should really work through. >>CHAIR KARKLINS: New Zealand. >>NEW ZEALAND: Thank you, Janis. I just wanted to return to a point that Paul was making, and it's very good that you go back over old GAC communiqués. I suppose we should (inaudible) from time to time. Part of the issues behind the expansion of the gTLD space is, in fact - - was, has always been, a need to represent minority cultures and so on. So I just return to the question of multiple categories of top-level domain. There's been a lot of discussion at this meeting, and I think in Mexico, and prior to the Mexico meeting, within the GAC about the need for recognition of social and cultural top-level domains, usually -- not always, but usually in completely noncommercial applications. Within New Zealand we have Maori as an official language. Thomas has been talking about the Swiss situation with Romansh. 50,000 people. That's never going to be commercial. And I think -- I think ICANN does have a responsibility to recognize these groups. There was another extremely effective workshop that I was able to partake in here talking about linguistics in cultural domain. There's obviously a huge overlap between cultural, social and linguistic. And I just urge the process of consideration. I appreciate the pressures that are there. The enormous number of questions that have to be answered. It seemed very complicated ten years ago, Paul. It seems much more complicated now in all of these areas. But this question of noncommercial, cultural, linguistic and social top-level domains must not be overlooked. >>PAUL TWOMEY: Frank, I mean, I agree. And you and I have spoken about this for a long time, so it's a.... And you can see that that was reflected in the work done by the GNSO Council, particularly on an objection basis, and the objection for a community group is specifically directed toward that concern. Indeed, I always thought it was somewhat progressive because it was trying to give the same sort of outcomes -- I am not going to say rights but the same sort of outcomes to community groups, including the people you are talking about, as potentially the TLD operators or the intellectual property holders would get. We all know in previous parts, earlier times in the Internet, the concerns in parts of Africa and other places of the sense that names were taken that were community -- of high community value but there was no way of actually defending them when they were then trademarked by somebody in another part of the world. So I think there's quite specific recognition of the point you are making. If you have concerns about the proposed implementation that you think puts barriers to that where you are, I think that's certainly something we are in the process, quite rightly, of listening to the concerns. I would share with you and hope that you would somewhat share our pain of two aspects, one of which is obviously things do cost money. But probably more importantly, what we are concerned about is that complexity will drive those costs up. The cost of processing, the cost of dealing with the more complexities in terms of having to make subjective decisions, the more your costs go up. Also, frankly, your liability issues go up, but that's part of the costs. We can manage it in the costs. And probably more importantly -- I know certainly it's been my concern, much more than the latter, is that you are very well aware that the Internet community, and particularly the domain name community, is very entrepreneurial. And so the concern is how to ensure we do not set up mechanisms for unexpected gaming, and they are just expert at that. Now, that's all I am doing is outlining aspects of the environment. I'm not trying to say no to the point you're making. I am just saying they are the aspects of the environment we have been trying to manage for at a staff level, and very happy to engage to see if we have got that balance wrong. But let's -- I am just trying to flag the various things that we have to -- we are trying to manage for. >>PETER DENGATE THRUSH: Frank, can I help just by giving an indication of a presentation we got at the board from the staff at our retreat when we explored this concept. And we very quickly got to seven or eight different categories -- large, small, language, tribal, ethnic, all sorts of different categories. And the consequence of that is all of those required a number of changes to the contract. Every one of them had to have -- and then we got to about seven or eight different terms in the contract that had to be treated differently. Different dispute resolution mechanisms, different costs, different prices, different terms. The sort of things you can imagine. So all of a sudden your matrix has got 56 to 64 different features that we have to manage cost for. So I am just explaining, if we go down this track, we start adding enormously to the costs, and the delays and other opportunities for -- as Paul says, for people to start manipulating those and picking the best, and it makes compliance for difficult. So there is no problem at all. I am not saying we are not at all interested in your point. I'm just sharing some of the difficulties of creating the system and in managing it and securing it. Of. >>CHAIR KARKLINS: I will give floor to Switzerland, Thomas, and then France, Bertrand. And then maybe we should move on to the next questions. >>SWITZERLAND: Thank you. I think that this is a very interesting discussion. And just to support what has been said by New Zealand. In addition to that, we realized in the discussion yesterday or the day before yesterday that also the people presenting the IRT to us already, without having defined them, thinking in categories because half of the discussion was talking about dot brand. So dot brand, in their minds, is already a category. And the more and more I have the feeling that maybe it might be a little more complicated in the beginning, but it might be more effective and less cost intensive to really think those categories through because you might be faster in resolving the problems than if you have one big hole. So maybe there's also reasons why there are gTLDs and ccTLDs. And I think it might also help to go online faster with some categories, whether it might be less risk, less problems, less problems to be solved than others. So this might also help to maybe instead of waiting with everything until every problem is solved, that you have some categories who are easier or less risky, dangerous, to put them online first and show there are new TLDs that are working. Thank you. >>CHAIR KARKLINS: Thank you, Thomas. Bertrand. >>FRANCE: I would like to share with the board some interesting discussion that took place between the GAC and the GNSO on that very specific topic, and especially the categories. It is absolutely true that in the process of the PDP in the GNSO, the decision was made at one point to consider whether categories should be taken into account or not. The decision that was made then was not, no, it's not applicable. It was, as the GNSO itself explained to us, it was at that stage, we won't do it, and after the rounds, maybe we will consider in the future whether categories are emerging in the process, which is different. The reality is that today, instead of having the categories that emerge in the first stages of the process, the categories have emerged in the delayed process. And so the problem is that, at the moment, it seems more and more evident that it is the single one-size-fits-all approach that is provoking the delays, because you need to fix every single button of the whole suit before you can use it. Whereas, as Thomas just said, at least two categories that are the key topic of discussion. Geo domains and brand TLDs seem to be very coherent categories, with the consequence that instead of having a huge matrix where you multiple all the different elements, you actually have a certain number of regimes, with no variations. You would have typical regimes with rules for geo domains that have to be clarified, for brand TLDs. But if you think about it, this is already what we are discussing when we talk about registrar/registry vertical integration that could be altered for dot brands. That's one element of a regime. I think we have all the building blocks. The fact is, we're not treating them as simple regimes. And as Thomas said, imagine today if we had only one type of TLD to handle national country names and generate top-level domains, we would have a great difficulty finding a single regime. And it's very good that we have two regimes. So why not keep it? >>CHAIR KARKLINS: Suzanne. >>UNITED STATES OF AMERICA: Thank you. And I wanted to follow on a comment I think that Paul had made about to the extent possible, the GNSO was recommending to you that you use neutral third parties to help make your decisions. I wanted to flag a U.S. concern with one of them. The issue of how to understand your approach to public order and morality objections is on our GAC list, and we haven't quite gotten to a solution. But I did want to flag a concern that we have identified in that the terminology comes from a particular treaty, but it is not defined. Public order and morality are not terms that have an agreed international definition. Rather, they have been interpreted country by country, which is certainly what we are accustomed to doing and an approach which we would expect to continue to be able to do. The proposed approach in the guidebook however would suggest that we will not be in a position of doing that, because you have chosen three juridical experts who will make those decisions. So it is a bit confusing for us to understand. And it goes a little bit to the issue of an objection if we are not in a position of being able to interpret public order and morality from our national perspectives and we are expected then to defer to a panel of experts. So I did want to flag that. And in fairness, I will admit that we do not have an alternative proposal handy. So I do not have plan B just today. But I think it's important to flag this contradiction here between a panel of experts that you would then assume that governments who are quite accustomed to interpreting these words on a case-by-case basis, on a national basis, we would be asked instead to defer. So we're not entirely sure we understand how that is meant to work. So I did think that was important to flag. >>PETER DENGATE THRUSH: Thank you, Suzanne. There is a lot of work to go in that. But the principle is reasonably coherent. And that is, we have defined the standards by which those words are to be measured by reference to three international treaties. So it doesn't matter what it means in my country or your country, because the panel I've are going to apply what public order and morality means in relation to -- now I've forgotten the names of the three conventions, but the -- or the four that we've given. Universal Declaration of Human Rights, for example. So, Kurt, remind me what -- or, John, who are the -- the treaties that we're referring to as setting the standard by which those words are judged? >> The Rights of the Child. >>PETER DENGATE THRUSH: The Rights of the Child. Elimination of racial -- We're referring to the actual international treaties. And it's those definitions that we expect the panelists to apply. >>UNITED STATES OF AMERICA: If I may. I believe, Peter, I have to take issue with your statement that it doesn't matter how it is determined in your country or mine. From our perspective, it does, indeed, matter. And that's the rub. >>PETER DENGATE THRUSH: Harald. >>HARALD ALVESTRAND: Just one comment. Not being very well versed in diplomacy, I will only point out that there's only one DNS, and, I'm sorry, we have to make the decisions once for the DNS. And in that case, what -- which country's standard we apply is problematic. >>CHAIR KARKLINS: I think we have reached as far as we could. We hope to be in a position tomorrow to finalize our input and comments to version 2. And I would like to suggest, taking into account that we have only 12 minutes for this session, that you we move on. >>PAUL TWOMEY: Chairman, I wonder if I can propose an important -- an important process point. I think we found at board level and at staff level that it was very valuable the interaction that took place under your guidance with members of the GAC on the geographic terms issue early this year, that we were able to share further thinking, practical proposals, see what was the intention, raise issues that perhaps had not been considered. As you think through your 13 points or whatever, I wonder whether it might be feasible for you to leave open opportunity for further dialogue on institutional operational issues. If these points come at any stage to become formal advice on public policy, then we have to, you know, accept them or not accept them and write back and follow the full bylaw rules. And I think to -- on a number of issues you've raised, I think it might be valuable for -- I would think it would be valuable for my colleagues and staff, I should say carefully, that they be able to engage you in a practical -- we've already done a little bit this afternoon. So perhaps just at least at this stage, if you are able to think about how feasibly you could express those so that we could have an ongoing, you know, quite -- formal but nevertheless constructive sort of exploration of the issues, that would be, I think, valuable. >>CHAIR KARKLINS: Though, Paul, I think that there hasn't been any single moment when our dialogue has not been constructive. I think always our dialogue is constructive. So shall we move on? Go ahead. >>PETER DENGATE THRUSH: Allow me to update you on a matter that I know is of concern to the GAC, and that is the data accuracy studies in relation to WHOIS. This is a study that's now underway to determine the accuracy of WHOIS data of the population -- all the gTLDs. It's a statistical survey. And the information we have that will have a plus or minus error of 5%, with 95% confidence. The items that we are looking for are in relation -- stated in the RAA contracts. We will be looking at the name of the individual or entity, and we will be looking at the postal address and those requirements there. There's a timeline for the study. It's being done by the National Opinion Research Center, a very successful -- prestigious -- thank you, Paul -- Chicago-based organization that does this kind of survey and research work. The process means -- and I'm not going to go through the timetable in view of the time -- will be largely completed just after our Seoul meeting. There will be a final report submitted in November, and we'll be releasing that to the Internet community around December. >>PAUL TWOMEY: One other -- Is that fine we continue? One other topic obviously of concern, has already been raised several times, is ongoing discussions between ICANN and the Universal Postal Union on -- in the previous sponsored top-level domain process. Just to report back to you that there's been -- there were further discussions after Mexico City. That those discussions came to something of an impasse. That the chairman of the board wrote to the UPU again on the 10th of July 2008 pointing out aspects and welcoming suggestions for ways for moving forward. He wrote that specifically to the director general of the postal union, UPU. And that we have received, the chairman has received from the director general of the international bureau of the UPU a response to that letter. We have received that letter response on the 19th of June, with a point of reaching out, looking for further discussion. So we are looking forward to having discussion, I think even here this week, with representatives of the UPU to see whether we can try to bridge this particular point about the application of ICANN contentious policy in a gTLD vis-a-vis the UPU's own acts or public international law. So that's the rub of the discussion, and we're hoping to be able to further that this week. And that's posted on the Web site. The next item is improving institutional confidence. And you will recall that the President's Strategy Committee finalized its work in Mexico and put forward proposals to the board of ICANN. The board of ICANN -- and that President's Strategy Committee had been -- basic area of work had been undertaken for two years with multiple consultations. I have forgotten how many consultations were involved in the process. I see my colleagues at the table. The board asked the ICANN staff to look at implementation, very similar to the new gTLD process really, to go and come back with proposals for potential implementation. That report was posted 15 days or so before the start of the ICANN -- this ICANN meeting. And there's a number of revisions within them. They include a series of recommendations from the PSC where the staff statement is this work is being undertaken within various existing reviews. And a further set of recommendations where the -- around certain procedures in ICANN where the staff are recommending that key committees of the ICANN board -- for instance, the Public Participation Committee and other committees -- take up those recommendations. And this has all been outlined in the response. It's actually appendix A to the report. Specifically on the role of the GAC, the staff recommendation is that the GAC and the board start a specific process together on looking at this issue of improved or further inventing the role of the GAC. There were preferred certain ideas put forward by the PSC which the staff recommend that the board and the GAC themselves discuss. For instance, meetings held in different cities, levels of meetings, attendance at meetings by the board and other supporting organizations. A number of propositions were put forward which I think we -- our recommendation is that needs to be a public discussion between the board and the GAC. The third area which was in the PSC recommendations was on further internationalization of ICANN, specifically the suggestion of an alternative -- not an alternative, an additional legal entity. And the board's recommendation was very, very clear. The board -- the PSC's recommendation was very clear, which was that the board should consider getting more information on that and sharing that information with the community before any step was taken along those lines. The staff report back to the board is that initial discussions have been commenced, which representatives of the two jurisdictions that were pointed out to have international nonprofit organization legal status, that was the kingdom of Belgium and the Swiss federation, that those discussions have not reached any stage yet where we could report anything back. They may not for quite some time. So this, I think, will be a slow position. A slow element. The final area, which is where I think we really should direct our attention, is the areas around accountability and further accountability mechanisms. And there were two quite specific proposals put forward. The PSC recommended that there be two forms of community vote vis-a- vis the board. One was that there be a recall -- that there be a mechanism whereby a decision of the board -- members of the community could move to have supporting organizations and Advisory Committees vote to actually have the board reconsider that vote. And so the proposition from the staff, language had been put forward for the board to consider, would enable that within 90 days or the next international meeting of ICANN, whichever is the longer, that there be a vote of two-thirds of the supporting organizations and advisory committees, and that that vote would then require that the board reconsider a decision it previously had made. It specifically says that in the case of the GAC, it would be a consensus vote of the members of the GAC then in attendance, which was a good recommendation to us. There has been some discussion about whether two-thirds of two-thirds is too high a standard from some people. But that proposal is up. There was a third -- second proposal that there be some mechanism whereby the community could call for the full board to resign, then for the board to be reconstituted. And even though the PSC took that advice and eventually did propose that, such feedback from the community, from this community in particular but other communities, that that would be a fully destabilizing option, in fact it became known as the nuclear option. That the staff recommendation to the board is that that option should not be pursued. The third -- it's on the table, but it's been recommended it not be pursued because of the amount of feedback concern that was raised. And I think even the PSC recognized, over time, that the enthusiasm for that particular proposition amongst the community had begun to fade during its deliberations. The third proposition I think is a very important one. The context of which is that the-- that there be a four -- we have a tiered mechanism for review of ICANN decision-making. The ombudsman, the reconsideration committee of the board, and then we have an independent review panel. That certainly the first two can be considered to be reviewed -- possibilities for reviewing them within the existing mechanisms for institutional review under the bylaws. But particularly on the independent review panel, the proposal is that that be augmented and expanded in its context. And consultation was made with a series of international legal firms and others to give us advice on some of this. But the proposal is that there be established a standing independent review tribunal. That that tribunal have standing members who would draw from multiple legal jurisdictions and legal systems, more importantly. That they be of senior appellate experience. The classic being a recently retired judge from the top appellate court, the European Court of Justice, the U.S. Supreme Court, the Brazilian Final Court, whatever. Also, that there be a panel that include senior technical experts. The wording has -- is probably not clear enough in the actual documentation, but the intention was that that would be convened by an independent review -- an independent dispute convenor, and potentially with advice of, say, the IETF or the IAB. In other words, it wouldn't be sort of ICANN staff saying, "Here's a list of names." It would be some mechanism through the independent review body to help do that. The independent -- sorry, I said "review." Independent dispute organization or convenor, of which there are numbers in the world. We use them now for Independent Review Panel. We propose that we continue to use them because they have existing mechanisms for running arbitrations and all the processes of running arbitrations that we don't think it's appropriate to duplicate, it's best to just outsource to that. The grounds under which such a -- what would this tribunal hear, that they would be from any materially affected party. So that's a very broad statement. They have to be material affected. But any materially affected party, so it's not just a contracted party. It's not just a party inside the ICANN community. It can be a party outside of the ICANN community who's been affected. And that the basic principles of what are known in some parts of the world as administrative law, but the principles of that the board made a decision that was not within -- that did not follow due process or fairness of process, that it did not make -- it made a decision where it was not faithful to its bylaws and mission, that it acted outside its powers; o thirdly, that the decision was manifestly irrational, it was not done in a way which was reasonable or based on the right sort of evidence. That is the proposal up for -- The board has not yet approved this one. It's not board language. This is a staff recommendation. But it's certainly something that we're looking for feedback on. I would point out to you that the chairman has made the point many times that what the ICANN board is looking for is not less accountability, but more, and more accountability to its community globally, and more accountability to its community globally and the people who will be part of its community. This is a mechanism which gives affected parties around the world an opportunity to bring an action against an ICANN board decision in front of a standing panel of -- across -- from people drawn from a number of jurisdictions under a set -- clear set of rules. So we would welcome your consideration of this at the appropriate time. >>CHAIR KARKLINS: Thank you, Paul, for this introduction. And we did look into the document briefly. It was clear that we needed more information. And we're thankful to you for this explanation of the meaning of the panel but equally, the composition and how it could be constituted, because I think this is a very important element which was -- which is missing in the written proposal. I see that Australia wants to intervene at this point. We don't have much time. But what I want to say is the following: In the document, this proposal is made, namely, the way forward. I noticed the section of "The GAC Role." And I must say that that section needs further work and improvement. And I believe that the GAC will be in a position to contribute some -- significantly to that section. We are working on our own document. For the time being, it is considered internal, but I cannot exclude that at one point we will be able to share the document with the board in order to provide you with our thinking on the GAC role in ICANN and potential evolution. By saying that, I would like to give the floor to Australia and then maybe draw this meeting to the end. But before that, I will ask also Thomas Narten to intervene very briefly. Brenton, please. >>AUSTRALIA: It's a very, very quick question. Administrative and merit review? Because we were having a discussion within GAC about positions on obligations of directors. >>PAUL TWOMEY: Quite right. It's not merit review beyond the test of reasonableness. And you've put your finger on one of the most important aspects that we've had to deal with is that corporations law throughout the world, particularly in the Anglo-Saxon world, but throughout the world make it clear that the board of the organization has to be the final decision maker. That's not to say that you would not -- you may not have a circumstance where the board may pass a resolution where it says it undertakes to implement the decision of the panel. But you still have to leave the issue of the board quite firmly -- we got very clear legal advice on this, that the board has to be the final decision maker under the corporations codes. >>PETER DENGATE THRUSH: Can I just as a lawyer add the obvious point perhaps that may not need to be said, but perhaps lawyers see these things differently. These are additional accountability mechanisms to all the obvious ones. ICANN can be taken to court for breach of contract. And this has nothing to do with criminal prosecution for illegal acts. All the ordinary, if you like, mechanisms, of course, exist. And then, ultimately, we're subject to the laws of California in relation to the operation of the company. So these are the additional, if you like, special ICANN accountability mechanisms for making a mistake or refusing to change our policies in the face of overwhelming evidence from the community that something different should be done. In other words, these are the ICANN accountability mechanisms, not -- not the contractual or criminal or financial or California code violations. >>CHAIR KARKLINS: So I -- maybe I will suggest that now Thomas Narten will comment on issue of policing of IDNA standards, and then we will continue with the question of meeting agenda for one minute, and then we will end our meeting. >>THOMAS NARTEN: Thank you, Janis. I just wanted to go back and clarify a comment that was made earlier about IDNs and the need for adherence to technical standards. And as background, I'm the IETF liaison to the board and follow the technical work related to IDNs pretty closely, including the work in the IDNA. And Patrik Fältström made a presentation earlier today to the GAC which I've just had a chance to review in the last 20 minutes. I think there may be some confusion about one of the points that he made. First of all, the IDNA protocol itself does enforcement for some things. So requiring adherence to the protocols is one form of enforcement. And ICANN certainly can, through contracts, for example, require that TLD operators adhere to the relevant technical standards. However, there's also some aspects related to IDNs that the IDNA protocols themselves cannot enforce. For example, IDNA will accept what they call variant characters that are, for all practical purposes, look the same to the end user but are treated as distinct characters at the protocol level. If we want to avoid user confusion, registry policy will need to specify how to handle such variants. So hence, that's where we have the need for having variant tables that are defined. So again, ICANN can enforce through contracts, for example that, variant tables be defined where that's appropriate, and used at the TLD -- at the top level and secondary level in the domains. And ICANN can also require adherence to the IDNA protocols by TLD operators. So what is true is that further down the tree, like if you get down to the third and fourth and fifth levels, at that point ICANN cannot effectively enforce compliance anymore. But it is still extremely helpful to be able to enforce compliance at the top level as that is really where the problems can sort of cause the most damage because they impact all of the names that are underneath that are covered by that top-level domain. >>CHAIR KARKLINS: Thank you, Thomas, for this clarification. We certainly will take into account in our further discussions. And I believe that Bill, who raised the issue, will be able to read the transcript of this meeting and this particular part and will take into account. Peter, if you would continue, then, to the meetings. >>PETER DENGATE THRUSH: Just for a few minutes, Janis, on the last item, the meeting preparation topic. This is a continuation of a process that I have been very keen on since I joined the board, and we have, for some time, have made sure now that the board agendas are published and all papers are distributed to the board, et cetera, in advance. It's been a very difficult exercise to bring that same kind of thinking, although most people have wanted it, in relation to the meetings because of the volunteer nature of the community where much of the work gets done and the huge pressure of hosting three meetings a year. However, we are at the stage now of having some rules, and we have a two-stage document deadline introduced for this meeting. 15 working days for GAC material and 10 working days for all other documents. And I'm pleased to report that all the consultative documents required for this meeting have met that deadline. The Public Participation Committee is working on a full policy, which is being developed and introduced for Seoul. My hope, and I think the committee's hope, is that there will be a single document deadline so we don't have to distinguish the GAC from other parts of the community. So those issues are going to be discussed again at the board Public Participation Committee's public meeting on Wednesday, 9:00. So if you are interested in the aspects such as this, the deadlines for publishing documents and other materials for -- I know issues such as remote participation and such are going to be discussed. So we are working, as you can see, to make sure these meetings are better prepared and better attended. Thanks. >>CHAIR KARKLINS: Thank you, Peter. And we really appreciate the decision which the board took last time -- last meeting in Mexico to introduce the deadline for submission of documents which should be considered at the next face-to-face meeting. Equally, we appreciate all technical facilities which are put at our disposal and the disposal of other organizations; namely, video streaming. Patrik Fältström today addressed us from Stockholm from his office or home. And our meetings are audio streamed and those GAC members who were not able to come here, potentially may follow or listen to audio record afterwards. All this is good. What was not satisfactory in preparation for this meeting was the late information about number of workshops and sessions, and the in fact that they are running in parallel with our meeting. The one you referred to, Public Participation Committee meeting, is taking place at the moment when GAC is sitting in plenary. And we cannot -- unfortunately, we cannot be able to be present in that meeting because of that fact. Information about the DNS malicious forum came extremely late, and there was a lot of confusion about that particular session, and it was clarified at a very late stage. Saying that, I would like to also say that we appreciate initiative which was -- which came from the staff to initiate consultations of all organizations with Nick Tomasso and Paul Levins, and we found a way to share information at early stage. And we will certainly do that in preparation for Korea meeting, and we'll see how that works. I think that you should draw somewhere the line how many workshops are taking place during one meeting. You have ten years' experience of preparing meetings, 35 meetings. You can calculate the average amount or hours of workshops. And I believe that there is some limit. And understanding that this time there are many issues on the table and maybe there was a need for more workshops than usually. But I would suggest that one line should be drawn. For instance, 60 hours and not a single hour more. And the kind of deadline for submission requests for the workshops also should be taken, maybe 45 before the meeting and not any day -- day after that deadline. So that will facilitate preparation of the meetings. And maybe, Jean-Jacques, you can take these suggestions as input for your tomorrow session where we will not be able to participate. >>JEAN-JACQUES SUBRENAT: Oh, there's a trick. Yes. Thank you, Janis. I will be happy to. But I must point out, of course, that everything which is related to the GAC is not, per se, public participation. I mean, it's something else. We have to establish the same mechanisms, the same tools especially, for instance video conferencing, things like that. But what you are mentioning is something which is beyond the purview, beyond the charter of the Public Participation Committee as such. Although there are common elements to both. >>PETER DENGATE THRUSH: Janis, can I just respond by saying that we hear you, we understand the problem, and that we all want to try and find solutions to this for very obvious reasons, that the commitment is to have these meetings so that we can meet and so that we can exchange ideas and do the work that we have. And we also appreciate a number of matters which certainly are important to the outside community, the non- GAC community, and that is simple matters like participants have to apply for leave to their bosses to come to these meetings some time in advance. And if they can't show them a full outline of the work program and how this is work related to the corporation, for example, then it's difficult for them to get approval to come. We understand these things. We also know that we will have better meetings if there is more time to prepare and more materials produced on time. So I acknowledge all that you say. This is exactly what we are working towards. How we get there, I'm not sure. Each time we move forward, we get a little bit better. And then suddenly we find we're introducing new gTLDs, IDNs, dealing with the Joint Project Agreement, and improving institutional confidence all at once. And so to a certain extent, we are subject to those sorts of conditions. And perhaps we will pass through this bottleneck and things may be a little bit calmer. I doubt that, in fact. I think every year, we think this has been a busy year. Next year will be quieter. And it's 12 years coming up, and the graph still looks the same, which makes it all the more important that we get to grips with this issue and that we do work and do more work along these lines. So, Janis, thank you for your contribution. >>CHAIR KARKLINS: Yeah. Thank you. Thank you, Peter. I think another contribution which I could refer to is the outcome and the summary of the joint ACSO session where it clearly said that it is not just GAC's frustration, that it's community frustration and volunteer fatigue, if I remember correctly. So all of this should be factored in. Otherwise, we cannot proceed further. By saying this, I would like to thank from our part board members who came to this meeting and listened to us. And I -- from my side, I found that very enriching, this dialogue. And I believe that these two hours gave board members a better understanding of the concerns of the GAC, and also some proposals. Thank you. And if you want to say something, I see that Jean-Jacques is asking for -- >>JEAN-JACQUES SUBRENAT: Yes. I should have said this earlier, Janis. There is one element which is common to all the parts of ICANN, and certainly both to GAC and to public participation, Janis, and that is that we are trying to set in advance a calendar of main meetings as far ahead now as 2012, 2013. So this, I think, will at least help clarify a certain number of things and avoid shoving things at the last moment. Thank you. >>CHAIR KARKLINS: Thank you, Jean-Jacques. Paul. >>PAUL TWOMEY: Thank you, Janis. I just wanted to thank the members of the committee for the chance to speak here today. And I was somewhat remiss. I particularly wanted to thank on behalf of the ICANN board the government of Australia and the government of New South Wales for their hospitality in this meeting and their cooperation in helping get this meeting established. I thank you as a member of the ICANN board. I also thank you as an Australian taxpayer. >>CHAIR KARKLINS: Thank you, Paul. And seems to me that this is your last appearance in front of the GAC in your capacity of president and CEO. I understand that you may show up next time in Korea in capacity of senior president. So it's always -- so you're most welcome. But just ending on sentimental base, thank you very much. I think that those 35 appearances in the GAC in different capacities, wearing different hats, did not disappoint you completely. So once again, on behalf of all GAC members, current, present here, those who are not present here, thank you very much, indeed. We really appreciate your dedicated service to the GAC and to ICANN. Thank you very much. [ Applause ] >>STEFANO TRUMPY: May I add a final -- also my compliments to Paul. I met Paul in such meetings at least 30 times, more than a dozen as chairman of the GAC in the place of Janis Karklins, and the rest as the CEO. And it was really a pleasure. We learned a lot from you. Thank you. >>CHAIR KARKLINS: So just a reminder for the GAC members. Tomorrow, we are starting at 8:30 with the planning committee meeting. Those who are interested in planning questions, please come at 8:30. For those who are interested in other issues, please come at 9:00. Thank you. Meeting stands adjourned. (6:50 p.m.)