ICANN Nairobi New gTLD Update and EOI Panel Discussion Monday, 8 March 2010 >> All right. We are just about ready to get our meeting start. If you would be kind enough to take your seats. And it's a good suggestion to move forward. It's kind of hard for the panelists, who are making the presentation, to not see anybody that's interested sitting right in the front. So if you would be kind enough to move forward, if you would. Thank you. >> Ladies and gentlemen, please welcome Kurt Pritz, senior vice president, services, ICANN. [ Applause ] >>KURT PRITZ: Thank you. For those of you that want to attend, please take a seat, and we'll get started right away. We're here -- we're starting late but we are going to spend the next two hours talking about the new gTLD program. The first 30 minutes will be a description of the work that's occurred in the program to move it forward over the past few months, and sadly for you, that will be me describing the work that's been done and the progress made. And immediately following that, we are going to conduct a panel discussion regarding the proposed exercise to conduct an Expression of Interest or pre-registration exercise. And so we have a number of esteemed panelists that are joining us, and look forward to that. Thanks. Perfect timing. So if we're ready, thanks very much for attending. I hope you all enjoyed the opening ceremony. I thought it was very cool, and I am really pleased to be here. And I feel blessed to get to see places like this as part of this job. But we're here to do some business and the agenda is really pretty poor. There's a very finite number of open issues associated with finishing the applicant guidebook. So 99.9% of you know what that is. The applicant guidebook is essentially the instruction manual for those who wish to apply for a new top-level domain and new gTLD. Completion of that work, settling the remaining issues is the prerequisite to launching the process, and then at the close of that, ICANN will accept applications for new gTLDs. So what we want to talk about here is the remaining open issues. What's been done, how do we get to closure. And at the end of that, we'll have our panel discussion about the Expression of Interest round, very brief overview of that exercise by me, and then Dr. Bruce Tonkin will be leading the panel. So what are the open issues? Well, they are listed here. I am not going to read them, but we talk so much and focus so much on what's left to go. Before I talk about these issues, I just want to talk about what we're not going to talk about today. We're not going to talk about all this stuff. Because these were formally open issues that were closed. So what we feel that we're stagnating and not moving forward, really we have done quite a bit of work, and, you know, this is ICANN and I am not formally saying we are never going to talk about this stuff again or make changes to the guidebook before we're done, but these are issues that have been addressed and closed. Which way do I point this. There we go. Nor are we going to talk about these issues. This is another set of issues that, after we published the first guidebook, all were raised. And through community discussions and lots of meetings and work, these were addressed, too. So while we feel that we're not making progress because we have several issues left, really quite a lot has been done, and this community, working together, has moved the thing forward quite a bit. So a lot of work has been done in the time since the Seoul meeting, particularly on these issues. And what's been published is a proposal for a final model for trademark protections and community protection models. A solution paper for mitigating malicious conduct. We will talk a little bit more about that later. Conquering the IDN three- character restrictions and relaxing that. How we are going to handle variants, a model for that have been published, and might be final. A memo describing different proposals for amending the registry agreement, which is very different, so an explanatory memo, and that's been published. A registry benchmarking study that's going to inform the evaluation has been published. And then a lot of these comment analyses. The comment analysis on guidebook 3, on the EOI model, on the GNSO work on trademark protections, the so-called STI group, and an IDN working group report. I will say that -- I have said this in other meetings, but what ICANN is trying to demonstrate in this, in a great big fat document, which is really difficult to do, is that every comment that is posted is read and it's thought about, and discussed. And sometimes comments result in changes, sometimes they don't. And what we try to do in this big old document is demonstrate that there really was thought and a balancing that occurred before changes were made or changes were not made. So let's get to these open issues. I'm going to touch briefly on each one, but I am going to preface the discussion by saying that there's now a set of rights protection mechanisms, trademark protections, that are posted where, after guidebook 1, was a blank sheet of paper. And it's a set of recommendations that have been developed by the Intellectual Property Constituency through the IRT, for those of you who remember what that is. And then vetted at board direction by a GNSO team, the special trademark issues group, who vetted and reached consensus on this really solid set of trademark protections. So, you know, we're attempting to label this one done, and that's what we want to start talking about here, is is this one done. Similarly with malicious conduct. You know, we have posted a set of seven solutions, and we're working on a couple more. So there's a blank sheet of paper there after guidebook one. Now there's a set of nine proposed specific efforts that will serve to mitigate potential for malicious conduct in new gTLDs. Root zone scaling is still being worked on. Staff has contributed to the technical discussion there. An additional economic analysis has been undertaken, and that is now under way. Economists are -- there's a guy cleaning the window back there. I thought he was waving to me. Has been undertaken. Work, as I said, done in IDNs, and then probably the most difficult -- "difficult" is probably the wrong word, but the remaining issues that require the most work have to do with the registry agreement and how we are going to structure the marketplace with regard to vertical integration of registries and registrars, and also some other contract terms such as a process for amending the registry agreement going forward. So part of the purpose of this presentation is to forecast or provide you with information about other sessions this week. Trademark protections are one of those. There's going to be a session later today in this room -- right? So, you know, we start with the trial close. Are we essentially complete on this? I think so. The process most recently is that the GNSO completed its review work and delivered a paper on a trademark clearinghouse and uniform rapid suspension system. This GNSO group reached consensus in most areas. And based on that, versions of these models have been posted. Another -- also, other work has been done on other rights protection mechanisms, such as a post-delegation rights protection model. So all that work can be found -- we don't have time to describe it in detail, but all that work can be found at that link on the bottom of the page and this stuff is all posted, of course. So here is a simple graphic for describing the trademark protection mechanisms or rights protection mechanisms that are now part of the proposed guidebook. There's rights protection mechanisms to address registry operations prior to launch, the IP clearinghouse or trademark clearinghouse will support mandatory IP claims and sunrise processes. And then after the registry starts operations, the Uniform Rapid Suspension, a post-delegation process and thick WHOIS will protect the rights of others during the ongoing operation of registries. So these proposed rpms really cover the breadth of registry operations and provide some protections for right holders across that life cycle. With regard to malicious conduct, in the last meeting in Seoul, we posted -- ICANN posted a set of seven specific protections that have been incorporated into the guidebook. I'm not going to remember them right now, but some included background checks of applicants and the like. So there's two bits of work being done with regard to malicious conduct going forward. One has to do with zone file access, so zone file access is a tool used by many who combat abuses. And if you can imagine an environment where there's hundreds of new gTLDs and each has their own rules and process for acquiring the root zone -- not the root zone -- the zone file data, that would become very onerous to those who use this as a tool. So different constituency groups have teamed together in a group chaired by Rod Rasmussen of the Anti- Phishing Working Group to develop a consistent methodology across all new gTLDs for accessing this zone data. And that group has really made excellent progress. They have developed four different solutions, and I have got two favorites, and posted those for public discussion. And right after the close of that comment period, that team will work to select the final model for inclusion in the guidebook. So this is another small but significant victory, both in mitigating malicious conduct and moving the guidebook forward to completion. So I think that's good. Another bit of malicious conduct work or preventing malicious conduct work has to do with a model for high security zones. So a group of ICANN stakeholders, mainly in the finance and banking communities, identified a need for TLDs to be able to self-identify as high- security TLDs. And so this group is working to develop criteria and a methodology by which those TLDs who wish to be identified as this for their registrants can do so and be certified and hold themselves out as high-security TLDs. So that work is posted at that link below. There's a session on mitigating malicious conduct on Thursday in this room also that I hope you can attend. I guess we are going to watch this slide for the rest of the show. Can you just go to the next -- there we go. Okay. Where am I? I want to go back to the slide where I was. Back. Can you go back? Okay. Sorry. So ICANN has undertaken additional economic study at the behest of the community. There were several requests from several quarters that ICANN undertake additional work to, again, try to balance the cost and benefits of launching the new gTLD program, identifying potential costs of defensive registrations by rights holders and developing other mechanisms that might actually enhance the launch of new TLDs, providing additional benefits beyond those that have not been anticipated. So economists have been retained, and they are working away. The most significant part of this study is really going to be the second phase that's going to occur between this meeting and the meeting -- the next ICANN meeting in Brussels. And the deliverable for that second phase of the meeting is to perform analysis to estimate the cost of defensive registration. I said that a minute ago. To develop some sort of metric to assess the expected cost and benefits of the new gTLD program as a whole, but second to assess costs and benefits perhaps of individual applications, so that if a protest or if a criticism of the new gTLD program is that certain TLDs will be introduced that -- whose detriment outweighs the benefit, there may be an opportunity for someone to -- an entity to object, and there would be some way and an opportunity to prove up that that detriment outweighs the benefit. So they are trying to develop those three metrics and methodologies for us. There might be a third phase to the study after that where, depending on the outcome of the first two phases, where additional work is done to further enhance the benefits of the new gTLD program. So I'm not going to look at these, but ICANN staff is working with the economists on a regular basis, and they have started to develop a list of questions for a survey they are undertaking, that's part of the first phase, and they are collecting data. The last of the so-called overarching issues is root zone scaling and will the DNS, will the root zone continue to behave in a predictable fashion, in a secure and stable fashion, with the introduction of new gTLDs. And it's not just new gTLDs; right? Already, today, listening to Doug, we heard about the implementation of DNSSEC, the introduction of IDNs, and the deployment of IPv6. So all these things are affect being the root zone at the same time as all this exciting work is happening concurrently. And so RSSAC and SSAC are still working to you understand the effects of new gTLDs in addition to all those other new introductions. What's been done in the last few months is to create models for delegation rates. It's been determined -- you know, it's not just how many TLDs that are in the root that might affect performance. It's the rate at which they are introduced. So a lot of work has been done to model what the delegation rate into the root zone will be given different numbers of applications. And this is a graphic that you can look at that's online, but it indicates that the delegation rate is really spread out over time more than you might expect, because applications all take different paths. Some are delegated right away because they are immediately successful. Some are successful after an objection process or extended evaluation. All these effects add up to sort of spread out the delegation rate over time. And so this work is intended to inform the SSAC and RSSAC work on this matter. IDNs. So one exciting thing that's occurred, and we have come to this realization and plan quite some time ago, is that IDNs are going to be introduced -- are going to be available as part of the new gTLD program just at the same time the program is introduced. I think we kind of take that as a given now, but not too many months ago we were wondering about that. So I think that's really good. There's been two issues in particular that needed to be solved before IDNs could be introduced as part of the new gTLD program. One was the three-character restriction that gTLDs, as presently governed, require at least three characters. So they are not confusable with ccTLDs. There are certain languages -- Chinese, Japanese, Korean come to mind, other languages -- where single characters represent whole words. And so it's been brought up in public fora such as these as to whether the restriction, the requirement to have at least three characters, would really hobble the introduction of IDNs and new gTLDs. So a working group was formed to undertake this study. And the result of that group's work has led to a model in the applicant guidebook that, for new gTLDs in scripts that are not confusable with ASCII scripts, that two-character names will be allowed. So they are not single character names being allowed yet, but two-character names will be allowed right away. And this group asked our policy-making bodies, the ccNSO and GNSO, to undertake further study to develop a mechanism where, for certain scripts or certain languages, single-character names could be introduced, too. So it's another area where we think we can draw a box around and call it done. The other issue having to do with IDNs are variants. Variant TLDs, variant characters -- jeez, I don't believe I started talking about this, but variant characters are representations, two different representations of the same character in a language script or even a cross script, and a top-level -- a variant top-level domain has these variant characters in it that often look almost exactly the same or do look exactly the same, but when typed into your browser will resolve to two different places. It's a very complex issue. A permanent solution is not in place but a temporary solution is in place for the applicant guidebook so it's not a bar to launching the process whereby an entity requesting variant strings will be delegated one. The other will be reserved pending the technical resolution of those issues. So I think that's good. There's a lot of discussion about vertical integration. How can or should registries and registrars be allowed to combine either organizationally or by providing vertically integrated services in the new marketplace. What benefits would accrue to users and registrants in the case of vertical integration, or what are the potential harms or abuses that could occur if vertical integration is allowed. Right now, nearly all registry contracts have a prohibition against co-ownership of a registrar and a registry. A lot -- Everybody in this room knows, a lot, a lot of work has gone into this. Debates at ICANN meetings, intersessional meetings, Webinars, other discussions. So we expect ICANN's retained economist to study this issue and other industry experts. We expect papers to be released concerning this almost right away, and to further inform the debate and to look forward to a solution on this. So I see the community kind of coming together around this issue that was formally very divisive. And I see a solution to this in the short term. You should know that the GNSO has undertaken a policy development process on vertical integration that's occurring in parallel with this integration work. It's not expected that either the implementation work or the PDP will interfere with one another, or delay one another. And then one of the last issues we'll discuss is the amendment process for the registry agreement. This is another issue that's still under discussion. The issue is whether all registries should have the same agreement, like registrars do now, or should they have individualized agreements, and how should those agreements be arrived at. Should there be individual negotiation of amendments? Should there be a mechanism whereby amendments can be implemented across all registry agreements? You know, the end, of course, is going to be somewhere in between there, that there's a variety of choices here. The registry stakeholder group recently posted a model, and ICANN incorporated that model into an explanatory memorandum that's been recently posted. So, you know, this is one of the two key issues remaining. So I encourage -- and I think it's an important issue, so I encourage everybody in this room who has been very active in this, there's not been a lot of discussion about this issue, and now is the time to get involved with it and make some contribution to it. So that's sort of a segue to the last slide of this section of the presentation. And that's a request for your continued participation. As a result of the publication of all of these documents, ICANN's opened ten comment fora or forums on various issues. So there's an opportunity to comment and still affect the process on many key areas: Trademark protections, registry operations, the registry agreement, IDN issues, and malicious conduct. And so I encourage you to read the material that's been posted. It's a riveting read. And get involved in the process. And, from my point of view, your contribution can urge closure, too, that this looks like a good set of recommendations and we're driving to closure. So that was a rapid-fire review of what's been done to date. And I don't know -- Nancy, what time is it? >> It is noon. >>KURT PRITZ: Excellent. I finished just on time. I want to ask our panelists to come to the stage. And we're going to have -- we're going to have a discussion on the proposed expression of interest round. I think it's good, too, because my voice is just about -- has just about given up. So who's coming up? For all you ICANN cognoscente, you know these people well. Dr. Bruce Tonkin, from Melbourne, I.T. is also a member of the board. He's going to moderate this session. Avri Doria, the former chair of the GNSO, is from the noncommercial stakeholder group. Bertrand de la Chapelle is from the government of France and his country's representative to the GAC. Zahid Jamil is -- he's in the DNDRC. I know he's from the former I.P. constituency, from Pakistan. And we've seen him already today. Dr. Olivier Crepin-Leblond is in the at-large. Richard Tindal. Is a new gTLD applicant, but a contributing ICANN community member that we all know. And Antony Van Couvering is a fellow New Yorker and works with Minds+Machines. So we're going to talk about the proposed EOI or preregistration process. So why are we doing this? Why is it in the public interest? Well, that's what these guys are going to talk about. So I'm not going to talk about it a long time. But we think it promotes the public interest because it's going to help to facilitate a secure and stable launch of the new gTLD program. And how's it going to do that? By -- these guys are going to talk about that -- ascertaining the number of applications for certain, providing knowledge for op- -- you know, getting operationally ready to administer the process, putting a stop in, identifying any unanticipated issues. And in the end of the day, probably hasten the launch of the new gTLD process by taking a lot of uncertainty out of the whole thing. There's been a lot of public discussion and comment about this, two postings, two complete public comment periods, publication of a draft model for public comment, and explanatory memos posted. So things have happened fairly quickly in this. But there's been a lot of public discussion. I think those of you who are interested in this know that the -- the key aspects of the proposed model, which is going to be discussed here, is that participation would be mandatory for participation in the first round. There seems to be a community agreement around a deposit on the full application fee that's about 30% of it. That application fee would be nonrefundable unless the round is not launched. Then, you know, maybe there's other circumstances where there would be refunds. That's something we might talk about here. Any information collected will be made public, including the string information. And, very importantly, a fully executed communications campaign would occur before the launch, equivalent in effort and depth and scope to the whole new gTLD program, because, you know, the goal, of course, is that at the end, we don't want somebody to raise their hand and say, "I didn't know about this." So it's a very important facet of it. And, finally, it should be known that it's not the launch of the new gTLD process. No evaluations will occur or be started as a result of the receipt of this information, not until the process is formally launched and applications are requested. So there's -- So that means there's a lot of prerequisites to this. We'll publish version 4 of the guidebook. So that's kind of shorthand talk. But what it means is a lot of very important issues should be settled, you know, who can apply. So vertical integration needs to be settled. What can they apply for? So the three-character issue needs to be settled. Corporations or brand holders, should they apply or not? So rights protection mechanisms need to be settled. So people who participate in this program can do so from a position of certainty. So I'm going to turn you -- the esteemed audience over to Bruce. A series of questions has been put to the panel. And I look forward to see what they have to say. We all look forward to see what everyone here has to say, too. And so I encourage your participation by coming to the microphone. So thank you very much for being here. And, Bruce, I'll -- I'm going to leave you -- I'll leave you to -- what? >>BRUCE TONKIN: Okay. Thank you, everybody. One of the things we will be doing here is using the chat forum within the Adobe connect. So if anyone wants to dial into that, I'll be monitoring that. And if people have questions during the session, they can submit them there. I think just to reinforce what Kurt has been saying, I've heard a couple of comments around the corridor saying, is EOI the start of the new gTLD process? I think it's probably fair to say that the new gTLD process actually started in the year 2000, when there were about seven names introduced. Then there was a second round in 2004. And one of the names out of that round, .Post, was only recently just signed, in December of last year. So it's been really a contiguous process since 2000. It does add additional step. And the question before this panel and the audience is, is there value in adding additional step into the process? Does that give more information? Is it useful? And are there ways that that additional step in the process could be further improved? So I'm going to ask each of the panelists to give a very brief two- minute opening remark on their general view of whether the EOI process is useful and how they might improve upon it. And then we will begin to address some of the individual questions that have been put forward by the community, issues such as, you know, should the process be mandatory before the first round of new gTLDs? Should the information be published? Et cetera. So first I'll probably go in order so I don't give anyone any preference, but I'll start with Richard Tindal on my right and then move across for the first comments. >>RICHARD TINDAL: Thanks, Bruce. So after much consideration of the EOI issue, I've come to the conclusion that I support the staff recommendation. I believe that it's a great example of the ICANN process at work. We've had a bottom-up proposal. It's been clearly presented to the community for comment on two separate occasions. There was extensive community comment on the proposal. The comment -- the concept was modified as a result of that community input. The staff have analyzed and summarized and presented a clear recommendation. And we're now in final debate on that recommendation. And the board finally will make a vote. So I think it's an example of the ICANN process at work. I support the EOI, because I think it generates valuable benefits. And I see no significant disadvantages. The benefits are: Valuable information will be provided for root server scaling resource planning. Valuable information will be provided for application review resource planning. It will generate more cooperation between prospective applicants, and it represents a safety net in the event that the EOI process reveals some unexpected issue in the new gTLD process that we haven't thought of so far. But I do feel the EOI is primarily a data-gathering exercise. An EOI, in my opinion, is not an application. Applications cannot and will not be accepted and reviewed until the applicant guidebook is approved by the board. I think the proposal that the staff have presented is clear. I think the pros and cons are on the table and are well-summarized in the document. And I urge the board to consider all of the input and vote on the proposal this week. I support the proposal, and I hope the board -- the board votes "yes." But if they choose to vote "no," I will accept that, too. What I would not like to see and what I hope does not happen is that we engage in another several months of discussion on this topic. In my view, the board should vote this week up or down. [ Applause ] >>ANTONY VAN COUVERING: I concur with Richard on that. I wish I had said what he said. But I'll say what I say. So at the ICANN meeting in Seoul last October, in a complete surprise to me, the gTLD timeline completely disappeared. And nothing was put in its place. ICANN provided no reason for the decision, no guidance, and no way forward. And for potential applicants, that was very difficult to swallow. But a group of us decided to look at the so-called overarching issues to see what we could do to solve them quickly. We got together a group drawn from the entire ICANN community that included applicants, intellectual property experts, registrars, governments, and people from the business community. We came up with a proposal that was taken up by the board, who asked the staff to come up with a model. They called it expression of interest. And this proposal solves two major issues facing new gTLDs so that we could get the process moving again. One, it takes many of the overarching issues out of the theoretical and makes them practical. It provides data for them. Second is, in the vacuum that ICANN left applicants in, the EOI program provides clarity, milestones, and the beginnings of a reestablishment of a timeline. But the question's been asked, can applicants know whether to apply before all the so-called overarching issues are solved? And the answer is yes. Because economic analysis, the EOI does not -- if you want to apply, that doesn't affect your decision to apply. Intellectual property, that question does not affect your decision to apply. Malicious conduct, the same. As far as root scaling goes, the EOI program provide a lot of data about it. According to the staff, the DAG is 95% finished, and the EOI seeks to know who the string and the applicants are so that ICANN can make a decision based on facts. It provides a huge number of benefits to ICANN and to applicants. And so here we are. The vote -- The board will vote on the staff proposal on Friday. Some of the people on this panel will be talking about the benefits of EOIs and some will talk about their doubts. And I would concur with Richard here, is that I really encourage the board to either vote it up or down. What we don't want and what the EOI proposal originally was not meant to do was to introduce any additional reasons for delay. It was meant, actually, to give applicants some certainty about what's happening. And the time has come for decisions and not half measures. Thank you. [ Applause ] >>BRUCE TONKIN: Zahid. >>ZAHID JAMIL: Thank you. I am a counselor to the GNSO from the business constituency. However, I am not representing them. I am in my personal capacity. However, I should state that the BC position on EOI is that it is opposed to this concept altogether. I know the International Chamber of Commerce has put out comments that are in opposition to this. All of the comments I have read on the Web site, there are substantial comments that say they don't agree with the EOI going forward. Let's first talk, this is not an expression of interest. It is a preregistration process, because it is mandatory. It does not solve all the outstanding issues that are required to be solved under 9.3 of the Affirmation of Commitments. We have no shortcuts in life. We cannot solve the outstanding issues by saying, "This is a solution." It doesn't. Those issues are still on the table. They have ton resolved. Once they're resolved, you go directly to the application stage. So I think that if asking people to apply when the rules are a moving target is not -- is unfair and creates uncertainty. My last point is towards the issue of whether or not this process will actually give us the information that apparently the EOI is supposed to give us. Well, I think it probably won't for the following reason. It excludes some types of applicants which would otherwise apply. And that impacts the purpose and the value of the data. One. Those who do not want to disclose their string and for it to become nonconfidential, those who are undecided due to the uncertainty of the rules, startups or innovators, not-for-profits, and developing economies, and those who want to keep their business model and plan or government conflict between, say, dot Kurd, K-U-R-D, confidential until -- and they don't want to give it an unlimited objections period so that people can try and conflict with them. So if we aren't going to get these applicants into this process, and let's not forget the developing countries -- if we don't get them into this process, then what is the value of the data we gather under the current model? I'm not sure. Are we really solving the data problem? Or is this really just what it is, the beginning of the new gTLD process? [ Applause ] >>BRUCE TONKIN: Thank you, Zahid. Just also just for clarification, the -- you mentioned the rules are a moving target. There will be at least another draft applicant guidebook which hopes to deal with most of the issues that are on the table today. So I guess what you're referring to there is if there are changes subsequent to that, which would probably be new things that would be identified as a result of the EOI rather than any other new things. Because the purpose of the process today is that we've got a number of issues on the table, including trademarks, vertical integration. All those issues will be resolved in the next version of the guidebook in June, before the EOI process starts. So just being careful, the rules you're talking about are potential new issues that arise and the rules associated with those; is that correct? >>ZAHID JAMIL: Bruce, I disagree. The reason I disagree is that if you look at the December document, for instance, I can count at least 13 or 14 heads where it says that the rules will change. And it says the way we're going to protect ICANN is going to protect itself is, we'll just make it clear to everybody that the rules can change so they can't sue us. The only issues that are identified in that document as being outstanding are, one, vertical integration; two, IDN 3 character. Nothing else has been identified. It doesn't mention trademarks, it doesn't mention economic study, it doesn't mention root scale studies, it doesn't mention many other aspects. So I would slightly disagree with that. My reading shows that these are problems, but we'll let them know and it's caveat emptor. >>BRUCE TONKIN: Kurt might be able to clarify that for us. Because I do believe those other issues that you mentioned for the applicant guidebook are going to be addressed in the applicant guidebook. Certainly trademarks, vertical integration. I'm not sure which the other one was that you mentioned. But that is -- economic studies. So that is actually in Kurt's timeline to be addressed by June. So just want to clarify that. Avri, please. >>AVRI DORIA: Okay. As you said, my name is Avri. I want to say also that I am speaking for myself. And while I think I'd find a lot of agreement in the NCSG on this, this is my own personal view at this point. In terms of doing these things, I think it's interesting that ICANN, which is a process-generating organization, has decided to take a process that seems to be faltering and failing and add yet another process to it to sort of let's solve the old process, the not-working process, by inventing yet a new process. I always find that problematic in that, yes, it does slow things down. Yes, it does take new resources. So in that respect, I do think we have a problem. I can't have spoken already for that one. I think that the EOI compounds unfairness problems that still exist in the original process. The original process does not take into account the -- a notion of variable fees for those in developing areas, for those that have cultural needs for new gTLDs, IDN gTLDs, who have not been able to enter this market yet. And by imposing on them a very large expense at a time that is probably a year before they would actually have to come up with the other prohibitive expense, and not only a capital expenditure at that point, but then a continuing operating expenditure, you've put extra burden on people who can't even handle the burden that's currently on them. So in that respect -- Certainly this may help some of the early insiders who are now in desperate need of a marketing method to try and raise more funny -- more money, to try and meet their current issues and problems, but it really doesn't help those who, at the end of the day, are the ones that need the new thing. So I think it's duplicative in effort. I think that one of the main dangers that come out of it is that we will find new overriding issues. As we start to see these and we start to see these names, we've gone up from four or five of them, I've heard people talking about nine overarching issues, and new ones coming in all the time. So I do agree with the speakers that spoke at the beginning. The board should vote up and down on this. The board should vote "no" on this. And the board should just tell us to fix our timeline and get moving on actually starting the process that has been delayed for 2.5 years now, going on 5 years, if you consider when we started. Thank you. [ Applause ] >>BRUCE TONKIN: Thank you, Avri. So we've -- just for the reminder of the panelists, we now do have a time, just to monitor time, so just on the top right of that screen, for those who are looking down there. I'll hand it down to Bertrand. Avri was pretty much exactly two minutes. >>BERTRAND DE LA CHAPELLE: Thank you. My name is Bertrand de la Chapelle. I'm the French foreign affairs ministry representative in the GAC. In this position, I will not speak on behalf of the GAC, although we had a very interesting discussion yesterday, and I will speak from some of the elements that were mentioned. I will speak here as a representative of the French government. Very quickly, the first thing that I want to raise and to highlight is how happy I am that after the basically three and a half months of community comments, we have the first community interaction. The mechanism by which ICANN is asking the community to comment online on draft is very beneficial to raise issues. It is incredibly inefficient to solve problems. The reality is that the EOI was a very interesting idea interjected to try to solve a problem that was the disappearance of timeline in Seoul. The discussion since then has not allowed the community to reach a consensus for those who have taken the pain to read all the comments in the online system. And thanks to the staff who has made an incredible work of synthesizing those comments, there clearly is no consensus. And the key fundamental distinction is about what is this beast, whether it is just a data-gathering, or whether it is the actual start of the process. And the more the discussion has evolved, and especially the -- already, the last two days of interaction among the community is showing that we're clearly heading into something that is becoming more and more the beginning of the round. And as the EOI, the way it is proposed, is actually not only addressing how many strings or what strings, but also who will have the right to apply in the first round, then it is prejudging a certain number of things. And the reality is, in the current function, it is actually exacerbating some of the problems that the DAG will already pose. And so, in conclusion, the question is, at this moment, the question is either we go on a data-gathering, and then we just need the strings, not the applicants, because this is where the root scaling problem is. If we go in the other direction, why not rather -- and it's just an open question -- get into finalizing the documents and the process to launch it correctly. But we'll discuss it further. [ Applause ] >>BRUCE TONKIN: Thank you, Bertrand. I couldn't quite see Olivier behind there. So next, please, Olivier. >>OLIVIER CREPIN-LEBLOND: Thank you, Bruce. I'm going to speak on my own behalf. And I'll draw some of the ideas from the at-large -- well, ALAC document, which was submitted for the consultation, but also some from my own convictions. I'm not a person to speak for extended lengths of time. I'd rather have a dialogue with you, the audience. So I'll try and make this as brief as possible. The main concern that I have with regards to this process, the new gTLD process and the EOI, is one that favors big applicants, big-time applicants, large corporations, and doesn't really cater for the small-time applicants, for the communities, for the small applications which will cater specifically to smaller applicants. I think the starting fee is way too large. It's got to be reduced. And I agree with Avri on this. I think it will deter a lot of community applications. And what I propose is to have a -- and that's drawn from the at-large document -- is to have a multiple number of different types of application. That's our main concern. Thank you. >>BRUCE TONKIN: Thank you, Olivier. [ Applause ] >>BRUCE TONKIN: So I think of the questions that we've had for the panel to address, I think most speakers have spoken about whether they think there are benefits or not of the EOI. And a couple of panelists have spoken about the issue around fairness and a particular concern around, I guess, people from developing countries and small businesses. Do any other panelists want to make any further comments on those two items first? On, perhaps, specifically the best way to ensure that it is global and fair? Antony. >>ANTONY VAN COUVERING: Certainly. It should be fair. It should include the most number of applicants possible. That is the point of this. We've waited a very long time to get to this point. The key thing there is the communications plan. ICANN has one drawn up. We've certainly given Karla suggestions for other ways to fully inform everyone. So the key is to let everyone know. As regards the fee, I understand there can be reasons why you'd want it higher or lower. There are specific reasons why it's $55,000. And certainly, given the very, very heavyweight structure that ICANN thinks is necessary to make sure that registries don't fail, it's not a big sum. So if people are very concerned about this being too expensive, then I suggest we look at the whole program. We know that ccTLDs can operate on very much less than what's required by ICANN. But that's the way the program has gone. So given that, it's really about letting everyone know. And, you know, ICANN's been doing that for years now. And an intensive process following the vote, I think, would ensure that everyone has time to make a decision. Thank you. >>BRUCE TONKIN: Okay. Any other comments? Olivier. >>OLIVIER CREPIN-LEBLOND: Thank you, Bruce. Coming back to the fee, if there is no class of application in the application process, then it's going to really deter some small communities, because $55,000 is a lot of money for some small communities. You're looking at this system in a purely commercial way. But a number of organizations wish to have a gTLD just for their own community. And that's a big deterrent. That said, I also think that the system will not be global unless ICANN as a whole does not make an effort into outreach. And, you know, outside these walls, ICANN is very little known. And if you do a Google search on domain names, you'll find it's not even on the first page. So I would say it really needs to publicize this process and go further with that. >>BRUCE TONKIN: Thanks. Yeah, Richard. >>RICHARD TINDAL: I'd like to comment on the fee issue as well. I know it's drawn a lot of comment and concern. And, of course, I think everyone is sympathetic. But I would like to put some facts on the table. If we don't have an EOI, it will cost $185,000 to apply for a new TLD. That's the application fee. And its cost recovery, there's been a year and a half discussion on that. I understand it's too expensive for some. It's cost-recovery based. My main point is if we don't have an EOI, the fee will be $185,000. If we do have an EOI, the fee will be $55,000 initially, and then some months later, another $130,000. So the fee is, in fact, equivalent under both models. I think the issue that we're discussing about fees is not really an issue about the EOI per se. It's a general concern about the fee structure of applying for a TLD. That's the first point I'd like to make. Second point, running a registry has always been an expensive proposition. Running a registry for a new gTLD is going to be an extraordinarily expensive proposition. The new DAG is loaded up with extensive new requirements for a registry operator. We've got financial continuity instruments, we've got an extensive raft of trademark and malicious activity protections. We've got all sorts of requirements built in here. And I'm not making a judgment on the importance of those. We've got DNSSEC in there as well. Running a new registry is going to be extraordinarily expensive. If an applicant can't afford to meet the application fees, in my view, it's going to be extremely challenging for that applicant to be able to meet the requirements of the contract to run the registry. The application fees for the new TLD program, in my view, are the tip of an iceberg for the cost of an applicant. [ Applause ] >>BRUCE TONKIN: Bertrand and then Zahid. >>BERTRAND DE LA CHAPELLE: I would like to piggyback on what Richard just said. In a large way, he is right. The issue of the fee, the issue of fairness, the issue of whether the EOI or this process is favoring some actors or disfavoring some actors is not only an EOI issue. It is a DAG issue. And if there is one benefit of the EOI discussion, it is to show that there was not so many problems on the strings, but that there are very important problems still in the DAG regarding fairness on applicants. >>ZAHID JAMIL: Developing country, if they have to put up 185,000, and I know that this is the process, it's certain, I can get my government, if it's a government run one, to fork up the money. I'll need the time, but I can do it. If I am a startup business, I go for V.C. and I say here there rules are set, I would like to do a round of financing. If it is certain, maybe I can do it. But when it is uncertain, the governments are going to say sorry. When you are uncertain, startups are going to say, look, it's just the wrong time for you to do this. It's premature. So you are excluding them. So I think that's not fair. And it's not global, either. And I think that's an issue for governments as well, especially developing countries governments, developing country economies and businesses. So are we going to say only those who have the pocket of money to be able to do this are going to be part of the new gTLD or not? I think that's an important issue. The reason is we need to be sure that we don't get to be perceived as a land grab by incumbents, and that's important because there are certain processes outside of ICANN that are taking place which do impact this. If we get seen as an ICANN which is only favoring the incumbent and western economies, what will happen is there will be some sort of negative feedback in those forums. And I think that's a perception we do want to avoid. And so I think the communication aspect, which is a separate issue, is a very important thing. I want to give you an example of the IRT communication. There was one in New York, one in London, one in Hong Kong, a sort of one in Sydney, and there was one small one in Abu Dhabi. I know that the communication group within ICANN did the best they could, but they have resource problems. But you can't actually have a proper global outreach unless you have boots on the ground or you are reaching out to the relevant people in various countries. So that's the second issue of communication. For it to be fair, for it to include everybody. And the last point, I would just like to comment something about the exchange that me and Bruce had. If the rules are going to be final before the EOI is launched, then let's just go for the application stage. >>BRUCE TONKIN: Antony and then Richard, and then we will move on to the next topic. >>ANTONY VAN COUVERING: So let's talk about fairness. We are in the middle, hopefully nearing the end of a very long process. If you look at the comments to the EOI process from community and linguistic applicants, they are like, "Go ahead, please." There are those of us who have relied -- and I don't speak just as a commercial one but if you look at the comments you will see there are others as well, who have relied on what ICANN has said, who have relied on staff's statements about probable timelines, who have announced their TLD applications, who have gone to their municipal governments or provincial governments and said, look, ICANN is doing this. Now, these people are out of money, so we are really weighing differences of fairness. On the one hand, you have people who have announced, who have gone out, talked to their community, who have gone forward based on -- relying on what has been said in rooms like this. On the other hand, you have fairness to, as yet, invisible applicants who may or may not exist. So when we're talking about fairness, we really have two sides of the coin here. Certainly, to be fair to those who, like the Basque community, like the (saying name) community, like other community linguistic groups have come forward, we really need to move forward. That's what they need, because they are wasting -- they have staffs. They are waiting, they are waiting, they are waiting, they are waiting, they are waiting. So let's move forward. >>RICHARD TINDAL: One quick last thought on the fee issue. And that is, again, if we don't have an EOI, it's going to be $195,000 out front. If we do have an EOI, there will be this bifurcated step of $55,000. The EOI is going to give us valuable data, I think we probably all agree, valuable data for ICANN to make judgments about how it should manage the process and the resources. Can you not hear me? So it will give ICANN valuable information about how they manage that sort of second part of the process. So something to think about is that, in fact, one outcome of the EOI might be that we learn that the second phase of fees does not need to be $130,000. When we look at the number of contention sets, the number of objections likely and just the overall volume and shape of activity, we may find, in fact, by bifurcating the process that we can reduce the second part of the fee. [ Applause ] >>BRUCE TONKIN: Okay. Avri and then Bertrand. And then on to the next topic. >>AVRI DORIA: I wasn't going to comment until a sentence included the comment I think we can all agree. And then I have to point out the fact that no, I actually do not agree -- oh, okay -- that this is going to give us valuable information without which we cannot continue. I think that we have collected sufficient information. Yes, I know that there's always the possibility of more information. There's always the possibility of better information. And we can continue to collect it. So I don't see this as providing any necessary set of information. It may be useful, but there's certainly nothing necessary about it. I see the look. I see the batching notions that they have come through have gotten us beyond the scaling problems people have talked about. The other thing I wanted to bring up now that I am already talking is the whole notion of this money is but the tip of the iceberg, this money is but a drop of the bucket for starting a new registry. I think one of the things that we talked about in the GNSO was that we had to be careful and not assume that the L.A. expense model for what it costs to do a registry would be the same as the cost model for doing a registry in Kenya. Certainly, there are certain obligations to be met but the cost models will be different. So proportionate to those cost models, this money is significant. Proportionate to those models, this 55,000 a year before you need to provide the 130 is a significant amount of money. And it becomes a prohibitive entry cost for those who don't happen to have it and can't use this for further marketing to raise further investment on their multi-year process. [ Applause ] >>BRUCE TONKIN: Bertrand. >>BERTRAND DE LA CHAPELLE: I think Antony is right to raise the question of fairness. And I actually would completely agree with what he said on that point. I want to make one thing very clear. The discussion we are having is not and should not be about whether there should be new gTLDs or not. I repeat, this discussion is not about whether there should be new gTLDs or not. This discussion has been taken by the community in June 2008. So I have heard an expression that -- I don't take it for myself, but I wouldn't like one of the gaming being procedure gaming. And the discussion we are having here is a sincere discussion to see whether the currently proposed EOI is actually helping moving forward or is potentially raising new problems. Regarding fairness, quickly, Antony is absolutely right. There is, at the moment, something very, very unfair. During the last two years, we have had what I call an EOI on the strings. A certain number of actors have come forward, have invested money, they have invested time. They have been public with their strings, which means that when the application run opens, with or without an EOI, they will have competitors. Other applicants for good and, other times, may be not so good, we can judge reasons, didn't come out with their strings. And the situation is that at the moment, those who have played transparently, who have accepted competition, who have invested and trusted the processes of ICANN are actually penalized because we are trying to find the perfect solution that will solve all the problems for all the strings in all the rounds in all the future. This is unfairness. I agree. At the same time, the current EOI, as I said, is perhaps creating some kind of unfair situation or unbalanced situation. Particularly, and I won't get into details, actors who apply for geographic TLDs today in the EOI would be disfavored as compared to other applicants who would apply for dictionary key words. I don't get into the details. At the same time, the current EOI clearly favors actors had are going to apply for large range of TLDs as key words. And so the key question that is underlying my remark earlier is we have been talking about categories or track differentiation. And the GAC has issued comments on that. The reality is that the last two years, and I am willing to explain that to anyone who is interested, has shown that the categories have self-emerged, as Avri had announced, actually, a couple of years ago. So we must take into account whether, in the discussion of the actual launch with or without the EOI, the differentiation of treatment is taken into account, because applying a single regime to different situations is not fairness. >>BRUCE TONKIN: Thanks, Bertrand. [ Applause ] >>BRUCE TONKIN: So just picking up some of the other issues, I think we have covered certainly the issues around the fees and the issues of whether there should be categories for different types of applicants. And that's been a general discussion around the overall application process as well, so it's a similar discussion on the application fee as well. On the topic of the mandatory nature, I'd be interested in views from the panelists of whether an EOI should be a mandatory step prior to putting in a final application. Are there any comments on that aspect? Richard. >>RICHARD TINDAL: Did you want to go first? We're talking about mandatory. >>BRUCE TONKIN: Yes. >>RICHARD TINDAL: So the problem with the voluntary method is we simply won't know how accurate the data is. Even if we incented the EOI participants in some way, let's say we made it cheaper in some way for them to participate, we still wouldn't know how many people participated in the EOI and how many didn't participate in the EOI. If we could somehow know that 10% or 20% or whatever percentage did participate, then we could make some extrapolation from that, but we are just not going to know. If we have a voluntary EOI and we have 50 applicants, what does that tell us? I think it tells us nothing. It doesn't tell us that we are not going to have 300 applicants when the real application opens. It doesn't tell us how many contention sets we may have. And in fact, on issues where we're concerned about potential objectionable strings in terms of morality or geographic infringement or whatever, we will have no idea whether those folks participated in the process or not. So I don't think a voluntary EOI really adds any value to the process because we're not getting data that we can rely on. >>BRUCE TONKIN: Okay. Zahid and then Bertrand. >>ZAHID JAMIL: Right. Having it mandatory with disclosure will, as I said earlier, exclude certain parties. So that raises the question of are we actually getting those parties in. And if we aren't, then that data is not coming through. So you have the issue of those who don't want to disclose. You have those, as I said, who are uncertain about the moving target rules. Those who find it too expensive, Avri mentioned many, municipal governments, and those who cannot get V.C. funding if the rules are not finalized. And then there is another aspect, those who say that basically this is going to be a de facto objections period. It's something that the staff document also mentions. For those means, because they don't want a P.R. nightmare, those that don't want to disclose their business models may not participate. So do we actually get them into this process? If we don't, then it goes to the issue of are we collecting the right data or the data we think we were collecting. In addition, it also forces people to defensively register. Some may actually say, well, hell with it. I may not want to but let's defensively try to register this. This, I think, is a serious problem. I also want to come back to the issue of governments who may say that this is either forcing me to register or gaming me out of the system, like dot Kashmir and dot Macedonia. So I still think that's an issue. >>BRUCE TONKIN: Bertrand and then Olivier. >>BERTRAND DE LA CHAPELLE: I think it's important when we talk about whether things are mandatory or not to make the distinction between whether applying the string is mandatory or applying as an applicant is mandatory. We could have different modalities. France has made a comment on the online consultation regarding what we call, let's say, dual call EOI. We believe that the string can be mandatory, be it in an EOI or as the first step of any round, whether there is an EOI or not. Because it has the benefit of setting the numbers of strings which is the only question that we really care about in terms of root scaling. On the other hand, making the applicant mandatory is actually determining in an EOI in particular a right to apply that is limited necessarily to a really small subset of actors. And, therefore, it is limiting competition as well. So we would favor and we proposed, and unfortunately the staff has not wanted to integrate it in the proposal, the notion of mandatory for the string, and then the round opens on the list of strings that has been proposed through the first phase EOI or first step. This distinction is important. >>BRUCE TONKIN: So Bertrand, so I clearly understand that, your suggestion is that people nominate the strings they want, but you allow, during the application process, anybody to apply for those strings. Is that correct? >>BERTRAND DE LA CHAPELLE: Let me illustrate it with the strings that we all know. As I said earlier, people who have said we intend to apply for dot music, we intend to apply for dot sport or for Barcelona or for whatever, when the round opens, EOI or not, there will be competitors for Barcelona, for sports, for music. And sometimes it's good because it allows a choice for the delegate. But the string has been somehow identified before. We could do each round the following way: Opening up the Expression of Interest on the string. People say I want to do dot shoe. And if I don't say I want to do dot shoe in this round, it will not be taken into account in the round. And then when the round opens, this is the list of strings that have been submitted, and maybe prefiltered because some may be eliminated just on the string, and then it is open for the different actors. >>BRUCE TONKIN: Okay. Olivier. >>OLIVIER CREPIN LEBLOND: Thank you, Bruce. I will draw my answer from the ALAC position which was published in January. The At-Large believes that the participation in the EOI should not be a prerequisite -- should be a prerequisite to file a full application later. And this will allow potential applicants to develop their proposal without running the risk of being outplayed by larger players later on. Effectively, you show you have an interest in this, and you are not going to have someone just jumping in front of you at the last moment. Thank you. >>BRUCE TONKIN: Okay. Avri, did you want to comment? Yeah. >>AVRI DORIA: Yeah, on the question of mandatory, I would argue that it shouldn't be mandatory. I think the main purposes, as I sit here and listen, for doing this is, one, as a marketing and fund-raising device for those who need that. And also, as some sort of illusion of forward movement while we are waiting for everything to get solved. So those two things do not require mandatory. Now, certainly if it is of marketing or fund-raising value for you to have declared yourself and put yourself out there and shown how committed you are to your kind of fund-raising, then if we have to do one, then it might as well be voluntary and not force those who have other plans, other fund-raising issues, or whatever, to have to also participate in that game. >>BRUCE TONKIN: So that depends on your purpose. So if the purpose, as the staff has stated, to gather definitive information on how many strings there are going to be and to try to use that to inform their process, that's the staff purpose that they have stated. So that's what we are discussing. So I agree individual motivations might be different, yeah. >>AVRI DORIA: I understand that, but as I have tried to sort of point out, is that that string information, collect being the number of strings there, is not a necessary activity. If you look at the plan that they have put together in terms of batching, you look at the work that's come out on the root scaling in terms of moving in a deliberate way, moving bit by bit, we could see that they have already put in a plan that takes into account the fact that there's an unknown number, but we proceeded in a step-wise way so that if issues turn up at some point, certainly we can deal with those at that point. So the need to collect an absolute number up front has really been obviated by the solutions they have already come up with. So that's why it's not a necessary purpose for this. >>BRUCE TONKIN: Okay. I think I'll jump onto the next topic because it's kind of interrelated, and if others want to make a further comment on this topic, they can, because they are interrelated between the mandatory is the publishing. I think Zahid had mentioned he was concerned if it was mandatory, somebody would be required to declare what their string is in advance. And I guess that's an issue if the string becomes public or not, and that really comes down to the issue, is should the strings be public and what's the public benefit in doing that. I will start from the other end. Olivier, did you want to comment on that? This is the issue related to mandatory but assuming it's mandatory, as the starting point -- or it doesn't matter, actually. Should the information be public? Just be clear on what the information is. The information is the string and the name of the legal entity applying to the string, and I think a purpose. But there's -- and contact information. So there's no business plan, there's no technical information. It's really just the string that they are applying for, and who they are, essentially. >>OLIVIER CREPIN LEBLOND: Ultimately what you are asking for is are the strings to be kept confidential or whether they are actually going to be publicly available. I think that ICANN's culture is not one where you keep things confidential. There's plenty of transparency, or there should be as much transparency as possible, and it goes directly against ICANN's own culture. And because of that, I don't think that it will be possible to keep strings confidential. You know, no disrespect to anyone working at ICANN, but it's just not the sort of thing that happens here. >>BRUCE TONKIN: Anyone else want to comment? Do you want to comment on that, Bertrand? >>BERTRAND DE LA CHAPELLE: In most of the questions that we are facing, I'm always in a delicate situation because the question I am asked is always do you agree with black or do you agree with white. And usually my answer is it's maybe not the right question. And I posted on the second comment a post that was titled, rather than black and white, have shades of blue. In this case, the answer is maybe we can explore if we are open to differentiated treatments, cases where the string can be made public and not the applicant. Or cases where the applicant can be made public but not the string, because it -- >>BRUCE TONKIN: Yeah, yeah. So you are basically saying it should be flexibility. >>BERTRAND DE LA CHAPELLE: -- the whole thing is we are stuck into those mechanisms where there is one proposal, and you either adopt it en bloc or you reject it. And it's very hard. >>BRUCE TONKIN: So you are saying that some information can be made public and some not, and it might depend on the type of string. >>BERTRAND DE LA CHAPELLE: Yeah. >>BRUCE TONKIN: In some cases you might say the organization is applying, in some cases you might say the string. So you are looking for flexibility, essentially. >>BERTRAND DE LA CHAPELLE: Let's say we can have situations where -- there are cases where if it's a certain type of string, it must be public. Other cases where it may be confidential provided there are some special criteria or requests. We have to discuss that. But at the moment, we are all stuck into a one-size-fits-all system. >>BRUCE TONKIN: Okay. Avri. >>AVRI DORIA: Yeah, in this case I think I can take what would be an NCSG opinion on it, which is that any information that's collected anytime should be made public. >>BRUCE TONKIN: Okay. That makes it easy. Zahid. >>ZAHID JAMIL: I think that the whole purpose of making a (inaudible) disclose, the string may also be the business plan itself in some circumstances and the string itself could also raise controversy between governments. And that extended period, which is a de facto challenge period, raises a lot of concerns. There's a lot of politics, a lot of business challenges and a whole bunch of things going on there. Let's move beyond that. Practices for tenders and bidding in RFPs do have a process of sealed bids, things in closed envelopes, depending on the sensitivity of information. So there is precedent in best practice. Also, there is a precedent within ICANN itself. Noncom applicants, when they apply, their application never becomes public. Nobody ever know who applied if they don't get through. So that precedent is already there. Let me suggest a solution to the ICANN confidentiality repository of this information. Demand Media made an interesting comment. It said we have confidence in the confidentiality maintaining purposes of ICANN, but if this is going to be an issue, you could actually appoint a third-party auditing firm. And I speak personally when I say this. So if you did that, you would solve many of the issues of many people who aren't able to apply simply because of the issue of the confidentiality. That's one solution to that particular problem. However, I speak as an individual when I say that. >>BRUCE TONKIN: Thanks, Zahid. Antony. >>ANTONY VAN COUVERING: Yeah, I have full faith in the ability of ICANN to keep things secret. As many of us have been in this process for years know, many decisions are taken in absolute secrecy, and only then revealed. Cheap shot says Richard, who is right. But surely the burden has to be on those who wish to keep things confidential. ICANN is an open -- supposed to be an open and transparent organization. So there may be some cases, I can't actually think of any, but there may be some cases where confidentiality makes sense. Surely that should be done on a case-by-case basis by applying to ICANN. So that's my comment. >>BRUCE TONKIN: Okay. Richard. >>RICHARD TINDAL: I don't have strong views on this issue. I think I can see the validity of the pros and cons on both sides of the privacy issue. I would make the point that I think that the -- that the method proposed in the staff paper I think works whether the EOI information is public or private. I don't think it impacts the effectiveness of what the staff have proposed. On balance, I would probably endorse the public disclosure of the data, because I think in general, that transparency will add more credibility to the whole process. And I think it will share more information, which I think is the primary purpose of the EOI. >>BRUCE TONKIN: Zahid and then Bertrand. >>ZAHID JAMIL: Thank you. We were having certain discussions off-line when the group got together and there were examples of gamings that were discussed. And I think we haven't really discussed the gaming on this panel. >>BRUCE TONKIN: That's the next topic. >>ZAHID JAMIL: I'm sorry. I am stop now. >>BRUCE TONKIN: That's okay. So now we can enter -- Bertrand, did you want to have one more comment on the previous? Yeah. >>BERTRAND DE LA CHAPELLE: Just to make the position clear. The position that I was expressing was an expression of openness for compromise if there is a problem with certain actors who clearly want some cases being confidential. More generally speaking, I'm fully in favor of openness in publication. For us, the main question is in the mandatory nature and the openness is a rather clear question. >>BRUCE TONKIN: So one of the other concerns that people have raised about whether it's the Expression of Interest process or the full application process is the concept that some parties might try and predict valuable names that they think they could sell to another party, and so that would either enter an EOI process or they might even do a full application with the sole intent to try and sell the business to another party. And one of the concerns people are raised is would the EOI encourage that or are there any mechanisms to -- is that a problem in the first place. So I will hand it over, starting from this direction. Richard, do you want to -- >>RICHARD TINDAL: So I don't think we have a problem here. I think trading any EOI slots is going to be infrequent. I think it's not going to happen more than if we did not have an EOI. And to the extent that it does occur, if at all, I don't think it would harm the public interest. And let me explain why I think that. First of all, the price of a place at the table has always been $55,000. If we don't have an EOI, the application fee is $185,000. But you can very quickly get a $135,000 refund. If your intent is to speculate and trade in slots, you have the ability to do that at the same price under the non-EOI model. Now, you might say, that's not quite true, because if I've only got $185,000 cash, I can only speculate in one slot, under the other model. But if we have an EOI, and it's only 55, then I can speculate in three slots. And that is true, to some extent, that is true. But it's an incredibly risky thing for any applicant to do. Let me tell you why it's risky, three reasons, quickly. First of all, trading of this sort is not going to happen in geographic community trademark or morality, public order strings. That's simply not going to happen. Those strings -- the certainty of those strings getting through the process is entirely unknown at the EOI stage. And so no rational person is going to buy an EOI slot on a string that's of that nature. So we're really talking about generic strings here. That's the issue. Second, no one's going to speculate in attractive generic strings, because, by definition almost, there's going to be more than one applicant. So if you have multiple applicants and you're a pure speculator, you're in trouble. If I apply for dot blog because I want that and intend to proceed and want to launch dot blog. And if I face someone at the EOI stage who is a speculator, there's no incentive for me to negotiate. They're going to come to me and try to trade in some way or negotiate a deal. I'm simply going to wait until the full application process. There's no benefit for me to negotiate with that person early. I will wait for the full application process. I'll wait for them to put up the extra $135,000. I'll wait for them to give a full application so I can see all their information. And at that point, I'll start to negotiate with them as a colliding applicant. But at this point, we're in exactly the same scenario as if we hadn't had an EOI. So my point being, for attractive generic strings, we're not going to see this sort of trading. So now we have the third and last category, which is, I won't call them unattractive generic strings, but sort of less-attractive generic strings. This is where it may make sense for someone to speculate and trade in EOI slots. An unusual generic term. That may happen. Again, it's incredibly, incredibly risky for that person. By definition -- and, of course, at this point, remember, we've had a full communications campaign. So anyone who might want to apply for that string at that point has chosen not to. We've got this speculator that we're concerned about who's put in this low-end, if you like, generic string. And now that speculator is in the position of having to find someone who wants to buy that slot. Now, they may find someone. It's possible they may persuade someone who didn't -- who knew about the process but didn't choose to put an EOI in. It may be that they could put a string in for 55 and sell it to someone for $70,000. It's possible, very low probability. It's extremely risky for that speculator. If they don't find someone to buy, they've lost $55,000. If they do find someone to buy it, at the end of the day, has the public interest been harmed? I don't think it has. >>ANTONY VAN COUVERING: I'll take another take on what Richard is saying. Suppose I'm Bertrand and I'm a speculator. And I want to speculate on a string. And I apply for it. Now, at the end of the EOI period, two things will happen on that string. Either I'm the only one or I'm one of two. If I'm one of two, I really don't have a good chance to sell that, because anyone who would buy it would have to contend with the other guy. If I'm only one, then I haven't achieved the full value of that. I would rather wait until I've done the application and then gone through the evaluation. And then I have something that's much more tangible, if I'm Bertrand, to sell. Because now I have a fully delegated top-level domain that may be worth something. And at that point, I might decide to sell it. And then I would be in the position of, say, public companies like variance or NeuStar, who sell their shares all the time, or, indeed, other companies who run TLDs and can freely buy and sell their company. So as a speculator, it just wouldn't make sense to do it. Thank you. >>ZAHID JAMIL: Well, there's -- on type of gaming -- this is just a thought. If somebody made an application and you wanted to target that person 'cause you know what their string is and the rules are still not set yet, maybe you could do something about the lobbying process and change thresholds, et cetera. One possibility. I don't know. We'll have to think about that. I think some thought needs to be given to that. Obviously, there's the issue we're discussing, EOI squatting or even proxy services. It's a good business to have. But if you're going to squat on a new gTLD application, there are various ways where people will not have the same level of understanding or sophistication about how to not negotiate, like Richard and others do. For instance, somebody applies for a string and says, "I've got your string." That person is either from a developing country, is uninformed, or gets very bad legal advice. And he says, let's negotiate. This is really serious. I know we can't save people from themselves. But speaking from developing countries, I know that even cybersquatting at the moment is something that they don't get the best advice for. And they tend to not even file dispute resolutions or arbitrations sometimes. So when you're talking about something like this, which is greater, I know we can't save people from themselves, but this is a risk and maybe something we just need to think about. >>BRUCE TONKIN: Thanks, Zahid. Avri, do you want to comment? >>AVRI DORIA: Yes, brief comments, two brief comments. One is, I really do have every faith in the ability of the implementation team to think out all the possible gaming options and possibilities that may already be in the air. And I have complete faith in this community that we will discover new ways to game whatever they come up with. >>BRUCE TONKIN: Bertrand. >>BERTRAND DE LA CHAPELLE: I want to thank Richard for having used a very interesting word to describe the type of strings that are music and the rest. He said it's not for those strings, but it's for the generic. Hello. This is called the new gTLD program. He could never have explained better that, actually, the generics are these ones and that the geographic, the community ones, and the brands, excuse me, they're not generic. So that's a very interesting element. Maybe we need to think about these categories. Second thing is, I agree with Antony that it's probably not the kind of gaming that he's alluding to that is our major concern. The major concern is the risk of land grab. It's the risk of, in spite of all the complaints that you can make, there is a possibility that deep- pocketed actors -- I even heard there's this expression "the mills." I don't know if it's appropriate. And it's attributed to different categories of actors. But people who would want to apply for a large number of strings and basically snatch in the EOI the most interesting ones. And I want to tell you one thing, especially in the generics: The short and semantically meaningful words are a scarce resource. It is a public interest duty to manage them in the right way. And, finally, I want to get back to the bottom line, almost as I always thought it, what turns out is that the EOI itself is bringing exactly the same problems as the DAG itself. As a matter of fact, it highlights them, exacerbates them, and in a certain way, the EOI, as such, is less a solution than an additional problem. However, all the questions that have been raised and the pressure that has been brought back into the debate are excellent, because I think everybody now understands that there is a need to move forward. The question is, EOI or not, if the EOI goes this way, it is the actual launch. Let's not hide ourselves behind words. It is a de facto launch. And if it is launched before the DAG is finalized, we are in a problem. So I'm wondering -- >>BRUCE TONKIN: One of the things -- cover just a little bit of the topics were. Just one more comment on the gaming, and then we'll review what will be in the next guidebook. Because that's obviously critical to most people, is understanding what is in the guidebook before an EOI can launch. Olivier, yeah. >>OLIVIER CREPIN-LEBLOND: Thank you, Bruce. Like Avri, I totally agree with her that it's impossible to minimize gaming. You can try and reduce it as much as you can. But, you know, the word wide recession was brought in by gaming. Whatever you're going to think of, someone is going to have more creativity and think of another way to get around it. Thank you. >>BRUCE TONKIN: Thank you. So I just want to come back to Zahid's question early on. And certainly when I read the EOI, it says that it's going to resolve some issues, and then it's got in brackets, i.e., and then it says, three- letter names and vertical integration. Whoops, I've just taken the one slide off that I want. Can we have that slide back that was up there a second ago? Okay. That slide is a project plan from the staff for producing the next version of the guidebook. And it lists the issues that are to be resolved before that guidebook is published. And that guidebook is the precursor to before an EOI can start. So the first one is the economic study that will be produced before that guidebook is complete. The issue of vertical integration between registries and registrars. I'm not sure I can read the next bit underneath that. The -- amendment process for changing the registry agreement. That's right. So one of the issues that the registries have been talking about, once ICANN signs a contract with them, what mechanisms have ICANN got to make changes to that contract. So that's the registry agreement process. The next one down from that is the clearinghouse for trademarks, trademark clearinghouse. The next one is the universal rapid suspension system, URS. The next one below that is the three-character one. The one below that, again, finding it hard to read. Variant management. That's the issue where, say, within a language like Chinese, where you might have simplified and traditional, hey, you might manage those two variants of the same language. And then the final one is malicious conduct. So those were awesome the issues that have been raised as overarching issues. And are intended to be resolved before the guidebook is produced, or before the EOI starts. So as Avri and others have said, if we have all those issues resolved, what new information does the EOI give? That's a valid comment. But I just want to be clear on what is resolved before the EOI starts. Is that a question, Olivier, or -- >>OLIVIER CREPIN-LEBLOND: Yeah, I just want to find out, what is your starting time for this? Because I thought DAG v4 was going to take place before Brussels? >>BRUCE TONKIN: Yeah, that's a good question. The starting time of the timeline is intended to be roughly 1st of January. So this is a staff project planned for this year. The key thing is, up on the top right, it says, "Shortest path." So some of the -- it may be longer and it may be the guidebook gets pushed out. And we don't know that. But that's -- that's the best result the staff believe they can achieve in their planning process. But I think the key thing I'm trying to establish there is that's a precursor before -- if an EOI is or is not launched, is done. >>OLIVIER CREPIN-LEBLOND: So it is before June? Before June? >>BRUCE TONKIN: That -- No. Let's be clear. That's the best -- that's what the staff are aiming to do, is by June, by Brussels. What I'm saying is that that's their best -- that's their project plan to achieve that. It's not achieved yet. So I just don't want to be sort of locking in dates. Zahid, and Avri. Again, just questions. I don't really want to debate on it. Go ahead, Zahid. >>ZAHID JAMIL: It seems like everything will be done by that stage. That's the plan. And if the plan is to get everything done and finalized before the EOI is done. Why don't we just go straight -- I have asked this question -- why don't we go straight to an application process? This is a question. >>BRUCE TONKIN: That's for -- what you're really saying is, do you believe there's benefit in EOI. Assuming that's all done, what we're trying to determine today is, is there a benefit of doing an EOI. That's what we're trying to determine. Yeah. Avri, do you have any other questions? Yeah. Okay. I want to really open it up for questions from the audience. We have some online questions. But we'll start with an online question. If someone wants to come to the mike, they can do so. Starting with an online question. We're going to find those. And I guess I want to distinguish the questions -- some of these online questions are partly questions for the ICANN staff. So I'll just try and sort of identify ones that are really for the panel. Any questions? Yeah, I can't see a question that looks like it's for the panel. Perhaps if you can identify, just write down on a piece of paper a question. And I'll start here with this mike here. Just speak loudly. They'll turn the mike on eventually. Maybe come up here, actually. Okay. >> (saying name) with a question and comment as well. As a legacy stakeholder in the root zone with that hat on, I would say I would like none of this to occur. Taking that hat off, I will say I believe this process makes sense. It's apparent that since the Vancouver meeting, the ICANN process and gTLD approval has been somewhat whacked, as I found out recently. Putting that hat back on again with regards to root stability, we are about to embark on several measures that jeopardize that stability, the first being, of course, signing of the root. We are told -- we have been told, we've been led to believe that this is going to work just fine. It has a nonzero probability of being just fine. With regards to additional entries in the root, we may find that that works just fine for 30, 40, 60. We may start seeing, like, little wisps of smoke out on the edge of the Net with those additional entries. And my question for you guys is, what happens if we do run into some problems here? And we discover that, oh, well, 40's okay, but maybe '50s not okay? And how do you back them out? Who's going to decide -- Who gets to, like, get taken out once they're in, once they've built a business around it? That's my question, not as a legacy stakeholder, but just as, what are you guys -- work that one -- I don't see anything here in any of this other than, like, it's all going to work. And maybe it's not all going to work. Maybe you're going to hit a limit. >>BRUCE TONKIN: Can I just comment on that. Firstly, it's not a question about whether to have an EOI or not. But let me just identify one of the reasons the staff have for an EOI. And one of those reasons is, there is not confidence in the number of top-level names that people are going to apply for. So the staff have done some analysis. They've talked in the community. They've come up with a community. And they've said that they think, you know, maybe it's up to 500. But people in the community are afraid that it may be a much bigger number. And the term that's been used in the community is "unlimited." And "unlimited" is equated to -- there is a current limit in the com zone of around about 80 million records. So technical people say unlimited equals 80 million. That's a massive change to the system. So part of the staff reason for the EOI is, we want to get a number. And then we want to be able to communicate with the root operators what that number is and get their feedback on, you know, how they're going to handle that number and what the issues are. The second thing that's a discussion, certainly, in the -- some of the security committees and also ICANN, is a mechanism of saying stop or pause, and what's the criteria for when that stop or pause happens. There's general agreement that if there is any smoke, as you say, you know, it's a bit like driving a car and you see smoke. The first thing you actually do is stop the car. And then have a look and work out what's caused that smoke. And it could just be, you know, that you have run out of water or whatever and you put water in and it fixes it. So it could be that we see some smoke, we pause or we stop, and then we address what that smoke is, which could be additional infrastructure, improving the process before we move forward. So that's certainly the intent. And the idea of an EOI was to actually give us a number that would allow us to make some rules there before we got everybody to put in a full application. So I just want to be clear there. So it's not a question -- This panel's addressing whether there's an EOI or not. You're addressing the issue of how do we deal with an unexpected number of top-level names and how do we deal with stress in the system and how do we stop. Is that right? >> Yes. But I think -- also, you guys -- I think you need to consider this question at this point in time. >>BRUCE TONKIN: Yes, agreed. And it is -- it is a question. >> I believe the EOI is a good process going forward. >>BRUCE TONKIN: Yes. >> Because it will give you that number, where we -- ICANN staff can really start scratching their head, going, "Oh, my gosh, how many are we really going to have?" >>BRUCE TONKIN: There's two steps in that. The first step was, before the EOI is launched, there will be a view from the board or the staff that says, "What do we do when there's stress in the system?" So there will be some guidance as to what we plan to do if there's a problem. So that's before the EOI. And then after the EOI, then there's also an analysis of what the impact is, depending on just the raw numbers. But just to give another concept of time line, DNSSEC is happening, starting to be rolled out this year. It goes live in the root in, roughly, July of 2010. Let's say the EOI happens toward the end of this year. Let's say the full application is sort of middlish next year. Nothing would go in the root until -- from gTLDs probably until about 2012. So just understand that the time frames of DNSSEC and adding new gTLDs are about two years different. So I just want to be clear there that this is not all happening at once. It's actually dragged out over years. DNSSEC's happening this year. IDN ccTLDs, some of those will probably happen this year, not -- you know, there's not huge numbers. And then new gTLDs probably in 2012. So we do have time to plan. Just want to address that. >> I just want to make that concern -- >>BRUCE TONKIN: That's very valid. >> I want that concern on the table. >>BRUCE TONKIN: Jumping across here. >> My name is Fabian. I'm from AFNIC. I am going to speak in my personal capacity, but I just want, for transparency purposes, to tell you that it is -- this is public -- AFNIC has formed a partnership with CORE, which has been selected to operate dot Paris. I just would like to make one quick comment on the fee issue and ask a question, a process question. As for the fee, Richard, we've talked about it, and I appreciate your argument and your perspective on the fee, which is the cost recovery, recovering the cost of ICANN in the selection of the applications. But I would be very interested to have information on the justification of those costs and see what it really costs to ICANN. And also, I'm thinking from a business point of view, I'm -- I believe profitable business usually relies on economies of scale. So if there are hundreds and hundreds of applications, how does that work for the process at ICANN? That's just my question. >>BRUCE TONKIN: Hang on. Just be clear. What is your question? >> It's just a comment. >>BRUCE TONKIN: Comment, sorry. >> But I'm saying, I could agree with this perspective that it is cost recovery. But I would be interested to know what are the actual costs. >>BRUCE TONKIN: There is a document that we'll give you afterwards which is a document that explains the cost calculation. So the staff have produced an extensive document on that. So we'll give that to you afterwards. But do you have a question for the panel? >> Yeah. Actually, what I'm -- I just want to share a comment. With talking with potential applicants in France, for instance, you hear that the -- the ICANN fee is not so much seen as a necessity by -- on the part of the process of ICANN, but more as a tax, like an entry barrier. >>BRUCE TONKIN: Yeah. So, again, so what's the question, then? >> I just want to share this perspective of those applicants which are -- which might have a business case for a TLD that is subsidized. >>BRUCE TONKIN: Yes. >> But to obtain a subsidy to pay such a high fee to ICANN is a problem. >>BRUCE TONKIN: Yeah. So, basically, you're -- >> If it could be justified. >>BRUCE TONKIN: Just to be clear, I want to get questions for the panel on the EOI. So the cost question, though, there is a paper that's documented how ICANN has calculated that. And start by reading that. And, absolutely, the number's higher than what we would like it to be. Yeah. >> Okay. I just wanted to share comments on this. My process question is, if the comment periods have only been able to raise issues rather than solve problems, as I've heard on the panel, what's the process going forward to solve the problems? I've heard very interesting propositions at the GAC meeting yesterday. I'd just be interested to know how that can move forward. >>BRUCE TONKIN: So the process is that when public comments are received, the staff takes those comments into account, and then make a recommendation based on those comments. And then the board ultimately makes a decision. And typically, we go through several iterations of those public comments and try and address the major issues that are being raised. So that's the process. >> But there were propositions. >>BRUCE TONKIN: Propositions, yeah. >> -- from the GAC yesterday in terms of process. >>BRUCE TONKIN: Yeah. >> I just want to say that. And also just another comment, if you may allow me. In terms of differentiation, I've heard about differentiation, natural differentiation between categories of application. I've heard about low-end generic, high-end generic, geographical, and other models. How can that be taken into account, especially in light of the presentation that was given at the GAC yesterday of the LC TLDs? >>BRUCE TONKIN: On classification. So that's -- >> Which showed specifically that if the DAG had very precise recommendations for certain type of applications, the process could be simplified and it could cost less to process those applications. Thank you. >>BRUCE TONKIN: Yeah, thank you. Bertrand. >> Alexander. >>BRUCE TONKIN: Sorry. >>ALEXANDER SCHUBERT: My name is Alexander Schubert. I am vice president, business development of Dot Gay, I am not anymore CEO. I have a question to you regarding the EOI. If I were going to apply for Dot Gay for $55,000, I don't have to specify whether I am a community TLD or not; right? Okay. So if I'm the only one, there's no one else, then I can skip all those community stuff. I simply say I am for happy people, and I get this top-level domain probably. Whereas if -- >> That's not correct. >>ALEXANDER SCHUBERT: Yeah, whereas if I am applying in the -- >>RICHARD TINDAL: Let me jump in real quick. I don't believe that that's correct because you still -- An EOI is just a data gathering exercise. You do have to submit a full application later, and you could be successfully objected to on community grounds subsequently. So the fact that you are in there and the only one with an EOI doesn't mean that you get the TLD. >>ALEXANDER SCHUBERT: This is exactly what I want to say. >>BRUCE TONKIN: I think what Alexander is saying he wouldn't have to put in his application because you can apply as a standard or a community. I think what Alexander is saying is he could make that choice based on the information he receives. If there were multiple applications for a particular string, those applicants could make a decision whether they are going to put in a community application or a standard application. And Richard is absolutely right that the objection process can still come from the community. >>ALEXANDER SCHUBERT: And then there is more, but do you want to answer? >>RICHARD TINDAL: I was going to say, yes, you could make that choice, but if you decide not to then pursue a community path, you very much are taking the risk that if that is truly a community that they may object to you. >>ALEXANDER SCHUBERT: Okay. But the other thing is, if I -- if there is no EOI and I have to apply as a community TLD to not run the risk that someone else is applying, then I have to get the endorsement. And while I get endorsement, I inform the community that there is this new top-level domain thing around, and then the community can choose to, as well, to apply as well. Which they can't if there is this EOI, because then I don't need the endorsement anymore. >>BRUCE TONKIN: Okay. Any other -- Well, I think that's a decision that you need to make. The concept of the EOI is about getting information for how many people are going to be applying, what the strings are, and currently the staff proposes those things are published. You can decide if you want to have a TLD, you can participate in that round, and then you can use that information for deciding your strategy, you know, whether you want to form a community or not, but that's not the EOI. The EOI is purely collect the information. about the string and your organization. I have got an online question from Steve Metalitz which says if all the remaining issues will be resolved in the publication, what is the purpose of the public comment on DAG version 4? That is a good question. I think that essentially -- I think that that's right. The public comment -- the DAG version 4 is still a draft, to be clear, so I -- let me just reiterate what I said earlier. The issues that we have been talking about so far, we have been going through a number of public comment periods on trademarks on various things. So they are trying to converge on the final version. We're currently in public comments on the trademark issues. We're partly in public comments on some of the other issues like malicious conduct. So by DAG 4, let's hope that they are as final as possible. But they are still subject to further improvement because the DAG is only a draft, and the final guidebook would be produced prior to the actual application process. So I hope that's clear. We are still taking public comment on the draft applicant version 4. But the hope is that the issues are 99% complete as opposed to being completely open. >>BERTRAND DE LA CHAPELLE: Bruce, may I interject one second? I would just like to call attention to the fact that this hour and a half is the total one hour and a half for community interaction on that topic. Could I please kindly ask you to allow the questions to come rather than.... It was the statement (inaudible), but in general terms, I think it's important to get all the questions. >>BRUCE TONKIN: Next question. >>DIRK KRISCHENOWSKI: Dirk Krischenowski from dot Berlin. We all want to make this gTLD process come forward is not true since we have seen in the last years that there are strong forces that want to skip the process, stop the process, or delay it forever. In this light, I see -- I would be happy if I would be on this side, on the opponent's side of the whole process, that the EOI is coming because it's a perfect opportunity to absorb all the power and all the resources of ICANN focused on the EOI and that prolongs the gTLD, it already prolongs the gTLD process. In Seoul, when the proposal for the EOI was made public, we were a big fan of it because we thought now having a simple and quick solution for collecting data about potential applicants bring the whole process forward. I think that that could have made, already. But bringing in this pre- application procedure complicates everything and delays everything, instead of focusing on this data collection. So in this light, it's my question now, who believes that the timeline was version 4 of the DAG is really in June -- publication in June, it was the shortest pass, and we have seen in the past of ICANN that the shortest pass means already six months, nine months or whatever delay. So who believes that the dropping of the EOI applications may happen during this year? Or if it's delayed, next year. Who of you? >>BRUCE TONKIN: Any comment? Yep, Antony. >>ANTONY VAN COUVERING: Dirk, I completely, as I hope you understand, share your frustration they delays. And I agree with you that there are a number of people who will exercise any amount of ingenuity to find some new issue to deal with. And the purpose of bringing forward the EOI was to actually put some of those to bed, and also to introduce clarity for people like yourself and for me to know that we can move forward. Naturally, this being ICANN, we have seen all kinds of elaborations of the EOI process and so on. I actually don't think it will create delay because at the end of this week, the board is going to vote on it, and I encourage them highly to vote yes or no for exactly that reason. If they vote yes, we'll all be putting in our EOI chit, so to speak, our name, address, and a check, and then there will be no more discussion about the virtually unlimited number of gTLDs which is going to affect trademark this or that. And there will be no further discussion of a virtually unlimited discussion of gTLDs that might affect root scaling and so on. So I still believe that the primary purpose, which was to get rid of those theoretical fears and replace them with empirical data will be served. The fact that there are a lot of people who have -- are interested in the ramifications of the EOI and potentially are not in favor of gTLDs, even though they say they might be, is limited to this week. So that's my view on it. >>BRUCE TONKIN: Any other comments? I will take one more online one and then we will go across. David Taylor has asked a question I have also heard come up, can we really start the communication outreach on the Expression of Interest before the next draft of the guidebook is published? Antony and then others want to comment on that? This is really about when the communication plan should start. >>ANTONY VAN COUVERING: Well, I think the communication plan started a long time ago. It started in Paris when it was voted that we would do this. And ICANN has done a great amount of outreach until then. So what we are really talking about now is a communications campaign that is more concentrated prior to the EOI submission. So ICANN needs to let people know that this stuff is happening. I hope, I hope very strongly that they do not attempt to communicate the entire DAG to the world. What they are going to be doing is saying, hey, gTLDs are happening, and here is where you go to find out more about it to decide if you want to do it. So I don't really see the EOI as affecting it that much. ICANN should be talking to anyone they can to let them know that this is happening. And what "this" is is what we have all spent years talking about. And, you know, that's certainly not going to be communicated in a -- by ICANN to the world in one statement. It's going to be an indication to people that they should come and learn more. >>BRUCE TONKIN: Any other comments on communication? Zahid and then Bertrand. >>ZAHID JAMIL: I don't think you can start early enough in the communications from a developing country outreach point of view. The reason is as I gave examples, one, I think we need to look at the communications plan. I think it's a very good document that's been put out there. I think certain improvements which require or maybe will try to put into place certain plans to go down into certain countries, and I am not saying every country, but we can't cover it by continent, we can't just do it online. So here is a suggestion. Whenever we do this and the board says launch the communication plan, the idea should be that many countries actually have local participants who create a demand, get the relevant sectors together, the I.T. associations, the bankers, whoever, whoever would be affected by this, who have no idea this is going to affect them. I can speak to my country, Pakistan. If I were to tell the bankers this, they would immediately want to come and attend a session. But you need to outreach in that respect and have certain mechanisms by which this is done locally on the ground. This doesn't necessarily mean ICANN staff flying everywhere. And one of the things we were discussing with some of the communication team maybe you could have Webinars but organized physical events within countries. And I think that is very, very important to get the awareness and getting people in those areas really interested in this exercise, and maybe you will get more applications, which I think is a good thing from those sectors. >>BRUCE TONKIN: Bertrand on the topic of communication plan. >>BERTRAND DE LA CHAPELLE: I want to reiterate basically that under the current model, the EOI is the actual launch, and as such, potentially contradictory with the provisions of the AoC if the rest of the elements are not sold, one. Second, in any case, communication campaign or not, the actual EOI or the actual opening of submissions will not happen before Brussels, in any case, given the fact that it is under the condition that a certain number of things have to be solved before, knowing that at the current stage, we don't know how many of those things have to be solved before for the DAG 4 to be considered sufficiently final. So if the condition and the thing I want to raise is, if the board votes on something at the end of the week, it cannot be on launching the EOI because there is a need for another decision in Brussels. So the key question is how does the community work together between Nairobi and Brussels to make sure that what is accepted in the DAG 4 is sufficiently yellowed and fixed to allow the first step to move forward. >>BRUCE TONKIN: So, Bertrand, if I paraphrase what I am hearing, the earlier you can get the message out about new gTLDs, the better. But you would need to know that the draft version 4 was pretty final before you really communicated to people they should apply; is that right? >>BERTRAND DE LA CHAPELLE: No. My point is the only way the DAG 4 will be considered by the community sufficiently final to launch something else, whatever that is -- >>BRUCE TONKIN: Yes. >>BERTRAND DE LA CHAPELLE: -- is by allowing a cross-community working group to be set up to help identify what are the portions that need to be -- >>BRUCE TONKIN: But that's what -- but what about the communication plan? When does the communication start, is what a lot of people are talking about. When do you start telling people about new gTLDs? >>BERTRAND DE LA CHAPELLE: If the trend which is emerging which is actually the EOI is so correlated with the whole program that we might skip it as a separate process but consider it as the first step in the actual application, it might work a little bit later. But then communication becomes simpler and can start relatively rapidly, saying these are the steps that will be taken, and by Brussels, this is what we anticipate to do. And start the communication campaign on a very simple level first, outreach large, and grow from them. >>BRUCE TONKIN: We will take Mr. Foody -- >> Sorry, Bruce. >>BRUCE TONKIN: Sorry. I keep losing you behind there. >>OLIVIER CREPIN LEBLOND: I kind of hide behind Bertrand. I think the communication plan is something particularly important. I mentioned it earlier. In fact, I think we should have started it yesterday. But we also need to communicate on two things: Brand and product. So some people say it should be done through the RIRs and through the registrars/registries. Some people think of a wider campaign. We need to look at that as soon as possible. >>BRUCE TONKIN: Okay. Mr. Foody. >>PAUL FOODY: Thanks very much, Bruce. Bertrand has made the point that this was voted on back in June 2008, and it was passed then, the decision for the new gTLDs was made then. But at that stage, it was envisioned that the new gTLDs would complement the existing TLDs, and they wouldn't replace them. Now, what we have seen is that given the ICANN's decision to allow companies to register their own gTLD, that inevitably that is going to cause a system whereby the new gTLDs completely replace the dot com. And we will see a situation where rather than registering a dot com and getting an online presence for a matter of dollars, it's going to be a couple of million dollars, perhaps. So I think we have got to go back to brass tacks and say what was the goal the introducing the new gTLDs. Was it, as ICANN claims, to improve competition, or is it an underhand way of replacing dot com, and basically gifting the Internet to the -- a very small number of very wealthy corporations who can afford to buy it out? Now, given that, to go out with an Expression of Interest now, before there has been the economic review that ICANN has implemented, before any decision on vertical separation has been decided, is naive at best. >>BRUCE TONKIN: Yeah. There's two issues there. The communication plan and the EOI. So the EOI doesn't go out until those have been addressed, bun of the questions the previous question was about the communication plan. So is your concern about communicating that before those -- in other words, you would want the communication to start after the draft guidebook deals with those issues; correct? >>PAUL FOODY: Given the issues being introduced with trademarks and the eligibility for these new gTLDs, very frankly it should have started ten years ago when ICANN is claiming that the decision to introduce new gTLDs was actually first started. If, for example, -- I have been coming to these meetings now for a year, and if the EOI comes out, for me to get into position and get ready to apply for a new gTLD, not that I want one, I think morally it's unconscionable to go this route, but if I were Joe Blow and I received notification in four months' time that this is what's going on, it's going to take me probably a year to get to a situation where I am in a position whereby I can submit an application that is -- that stands a better than 50% chance of going through. Most of the people in this room are in a situation where they can put up their $55,000 -- >>BRUCE TONKIN: Okay, okay. >>PAUL FOODY: -- with a fairly realistic belief that it's not going to be just money down the toilet. Most people in the world, that $55,000 is just money they are chucking away because their application will get rejected. >>BRUCE TONKIN: Right. Okay. Zahid wants to comment. >>ZAHID JAMIL: It seems I may be in danger of agreeing with Mr. Foody on this one. I think if you are going to have a communication plan which is unsure about certain aspects -- I am rethinking what my previous statement was. If you are going to have a communication plan which doesn't really know whether (inaudible) sorted out who the applicants even are, so when you go out and you are outreaching to these people, you're maybe not be giving them complete information they can actually use. And there is a time gap, I agree, for them to be able to say I just learned about this. Let me go out and try to get funding, et cetera. On another note and in addition to that, I am concerned about what the results of the EOI might be. It's one thing to say the EOI comes out and says you know what? We can go ahead with this. What if the EOI comes out with certain results, because it has to be (inaudible) to know the results, which actually raise further questions about and generates concerns about more overarching issues. I don't know, do we want that situation to develop so the EOI is further delayed? If June is when we are going to get all of this thing done, let's just straight go to the application stage. >>BRUCE TONKIN: Okay. Werner. And if your questions can be one minute in length so we have a chance for everyone to speak. >>WERNER STAUB: Thanks, Bruce. My name is Werner Staub. I work for CORE. My question is whether we can still prevent the current EOI plan from being schizophrenic as it is, and that it reflects actually our own attitude. We are being schizophrenic. It is impossible to say that we want to learn from the EOI, but then again, we say we want to predefine everything before the EOI runs. And then we don't want to learn from the EOI anymore. Moreover, we want to hand with the EOI privileges to people who, by that time, by whatever chance they have in the form, got some money together and applied. Of course it is going to be very much in favor of the incumbents. We can easily change it, I believe, and I wonder if other people agree, I would like to follow the suggestion that Bertrand made to rely on the string-based EOI. Just the string, no exclusivity for the person or organization that brought it forward, might then also have a very small fee. Maybe $5,000 to launch it. It would just be enough to know who it came from. From there on, we can do the communications on the basis of an existing list of strings. And that is going to be understood by people. There has been a half-page ad by ICANN about the year ago, it was totally incomprehensible. Nobody could understand what this possibly could mean. If you spend money on the communications plan, people don't even know what it is because there's no string mentioned, nobody will understand. Moreover, if you do that kind of EOI, nothing stops us from doing one in six months, another one six months later and so on. It would just kind of get an idea of time priority which could, by consensus, if need be, change. >>BRUCE TONKIN: I think what we should try and do is perhaps get as many of the comments or questions. Because I think if we each try and answer, we won't -- I've only got ten minutes left in this session. If we can go across to this mike, please. >> My name is Joe Cady, from STG Interactive. We are among those companies that you don't know about generally, and are applying on applying for a dot Frogens gTLD. And I'm speaking as an individual. Given that what we're trying to achieve with the EOI is to get an idea, to try and sort of bring those strings out of the woodwork that haven't been up -- haven't been lobbied heavily, let's say, for example. It's a pretty good guess, I'd say, that these are not community-based gTLDs and that for that matter, do we really need these to be involved in process? I would suggest that perhaps we make the EOI process only binding and obligatory for non-community gTLD strings and that, also, for these entities to have the option to -- of confidentiality, after all, we want the number. If we get a lot of false positives for the community -- for community-type strings, we can still get a pretty good idea of how many there are, based on how many we have already heard about, given their lobbying. A community string who has not been lobbying heavily up to this point is going to have a lot more difficult time getting through the process. >>BRUCE TONKIN: Okay. >> Hello, my name is (saying name). I'm an ICANN fellowship. I come from Pakistan, low-income developing country. I believe that ICANN is on the verge of a great growth opportunity of the Internet that can enable a humongous opportunity for participation for people like us in the developing world in the gTLD process. But aren't we dealing with gTLDs too narrowly? We are in the developing world, though still in small number in terms of gTLD-aware Internet consumers. And producers are really looking forward for the launch of the program, because it has great opportunity for us to participate in the global knowledge (inaudible). But is the gTLD really ready to engage us? I see the gTLD program in the future to fuse with ccTLDs. And we are all important stakeholders of this new program. Why are we underestimating the capability of the developing world for conducting EOI evaluations within the developing world for their set of applications, and thus reducing the financial resource burden in ICANN. ICANN has its own set of resource problems. And with this outreach into the developing world, encouraging the developing world, people's participation in ICANN, we can offer our services both voluntarily, as well as low cost as compared to what is done in the U.S. For example, $55,000 can be just 55,000 rupees, which is less than a thousand dollars in Pakistan. And I would like to volunteer myself to form the gTLD evaluation committee, maybe a test, for Pakistan. And I would like to ask, is the Internet community and ICANN willing to engage us fully, but in a meaningful way? And one last question. This is to this half of the table. A word like desi, D-E-S-I, is a generic name to represent the culture from Pakistan and India. And if anyone wanted to get that registered in the gTLD, this is going to be a totally desi thing as well. >>BRUCE TONKIN: It's going to be what? >> Total desi thing, which means a cultural thing for Pakistan and India. You won't be able to evaluate it in the U.S., because the importance isn't in the U.S. It has more importance for our regions. So these kinds of evaluations can also come through the developing world countries. Their own words and Roman script through the gTLD process. Thank you. >>BRUCE TONKIN: Bertrand. >> (Speaking French, no translation). [ Applause ] >>BRUCE TONKIN: Okay. Thank you. >> Good afternoon. My name is Joe Alagna. In the interest of full disclosure, I'm with Central NIC, and I'm also involved in Central Registry Solutions. I just wanted to make a comment and ask a question. An EOI or an application for a gTLD is just, by its very nature, very public. So I think there's really not a need to decide whether, you know, there's any secrecy involved in this. And, of course, the EOI was born out of the need for more information on behalf of ICANN. So I do think it has to be public. But I also think that it has to be mandatory, because it's not fair to ask someone to publicly identify an idea and then open it up to competitors. And so I just wanted to enhance an idea that I heard brought up on the panel earlier and ask this question. And that is, if the board does vote for an EOI later this week, is it being considered in that process that if somebody -- that this idea of a closed bid, where, in other words, many people can apply for the same application -- for the same string, but it would be a closed bid, and then once all the applications or EOIs were received, then it would be made public, and then all the details can be worked out? >>BRUCE TONKIN: That's the current plan. >> Okay. That is the plan? >>BRUCE TONKIN: Yeah, that's the staff proposal. And the panel was considering some different ideas. But that is the current staff proposal, that the results are published at the end of the expression of interest process, yeah. >> Right. I think that's a good idea. >>BRUCE TONKIN: (inaudible). >>TONY HARRIS: Yes, my name is Tony Harris, I'm speaking not on behalf of my constituency, but of the federation I represent, Latin American federation of the Internet and electronic commerce. We are known as a potential applicant or -- for a new gTLD string. And we have supported the EOI project as it has been formulated. I do not think it's a good idea to tinker with it. As has been said, it should be voted for or against on the merits as it stands today. Doing this based on strings does not seem like a good idea to me. It's opening up the field for me to applicants with deep pockets to latch on to ideas that other people already have. And as far as communications and outreach, I think we need to avoid being concerned with every last peasant on the face of the earth being aware of this, although it's a good intention, of course. I think people who are interested in any sort of enterprise or entrepreneurship take the necessary steps to find out where these things happen. And by the nature of gravity of their interest, they gravitate towards where these things are happening. That's just my simple opinion. Thank you. >>BRUCE TONKIN: Thank you. Last speaker. We do have to close the session down. So -- yeah, because we need to be ready for the next one. So last speaker, please. >> Thank you, Thomas (saying name) from Switzerland, GAC member from Switzerland. Since this is supposed to be a cross-community or constituency discussion, I thought it would be good if some of the government people would make take the floor. First a comment. I haven't heard any GAC member opposing the new TLD introduction, basically. But government representatives have some concerns. For instance, things like what is the appropriate legal basis for a ccTLD to operate on? What do users in the city expect to be the legal basis. What is the funding system that would allow as many linguistic and cultural TLDs as possible? These are things governments are concerned with. So my question to you is, do you think that the EOI process the way it's drafted now is the right solution to speed up the process towards implementing new TLDs rather than taking effort maybe in the time from now to Brussels to sit together across constituencies and try to work on the categories as has been mentioned and see what are the different challenges for different categories in order to allow for a more flexible regime that would speed up this process. And I would tend to support Bertrand's idea of having a working group working on this issue in the hope that this would speed up the process. >>BRUCE TONKIN: So looking at categories to speed up the process. Okay. >> Thank you. >>BRUCE TONKIN: I need to close the session, because I do need to prepare for the next session at whatever time that is. But I'd like to thank you all for attending, and especially thank the panelists who have put a lot of thought into their responses. And also for those what participated with comments and suggestions for the process. Thank you. Thank you. [ Applause ]