New gTLD Update ICANN Meeting Sydney, Australia Monday 22 June 2009 >>KURT PRITZ: Hi, could you all take your seats, please, and we'll get started. We have got a lot of material. And if somebody could close the doors, please, that would be good. >>KURT PRITZ: Are we good to go? Could everybody take a seat and get ready to go. If you could stop your conversations in the back because the people in the front want to hear. Thanks. Okay, great. Thanks very much for coming to this session. It's about ICANN's new gTLD program, the program to open up the domain space and delegate a number of new top-level domains. It's a very significant effort. We're going to try to do three things here today in the hour and a half we have, three and a half. First we are going to review the programs that are going to take place during this week, so it's sort of a primer for everybody to help you select which sessions you want to attend. We'll have moderators and participants from various sessions come up and give a brief description about what's going on in that session to help inform you and help you make your decisions. Second, I'll take you through some of the details, some of the work that's been accomplished in progressing the program since the ICANN Mexico City meeting. Development, response to public comments, increase in the level of detail of the program and so on. And then finally, a brief few comments on the go-forward plan, timelines, and some additional detail with regard to how applications for new TLDs will be processed. And then the half-thing at the end, given the time we have, we'll be happy to take questions. So I'm really pleased to have joining me today some of the people that are going to help moderate the sessions and provide a lot of information to those who either want to participate in the new gTLD process or are looking to providing feedback into the final formulation of that process. So I want to welcome -- I'm going to welcome up on the stage a little later Mr. Lyman Chapin from Interisle Consulting, J. Scott Evans from Yahoo!, Rod Rasmussen, the co-chair of the Anti-Phishing Working Group, and Joe Sims, a partner at Jones Day. So what are we going to do? We are going to do these three things: Look at the workshops and activities at this meeting, then discuss the development work that has occurred in the new gTLD process so far, and that includes analysis of all the public comment that's occurred and the publications to date. Improvements and updates to the Applicant Guidebook. The Applicant Guidebook is sort of the instruction manual for those of you who wish to apply for new gTLDs or anybody else. A new set of explanatory memoranda, so as changes are made to the program or it's enhanced, sometimes those changes are best explained and the right amount of background is provided through additional memos. We also have published fairly recently a number of independent reports that we will briefly outline later. And then right at the end, as I said, we will go through some timelines, consultations that will occur with the goal of refining and then finishing the Applicant Guidebook and launching the new gTLD process and what those materials are going to be. So these are the sessions that are going to take place this week. There are quite a number of more sessions, but these are areas of particular interest to the community, I think, and ones for descriptions here, geographical names, how geographical names should be treated as a top-level domain and if protection should be afforded to geographical names. We have already had a couple of sessions on that with the GNSO Council, the GAC, and a joint GAC-GNSO meeting, and I know the ccNSO is taking up this issue on Wednesday also, so there will be a discussion there about that. There is also a technical session on the three-character requirement, and that is the current new gTLD program requires that all top-level domains in the generic space have at least three characters. A significant issue has been raised, especially by members of communities with languages. I am going to mess this up a little bit, but languages where participation in the new TLD may be hobbled somewhat by the three- character requirement. Think about Japanese, Chinese or Korean and the fact that single-characters in those languages often represent entire words, or two characters. So there's a balancing to be done between technical requirements, how to write bright-line rules, and encouraging full participation in the domain space. So that's a very important discussion that's ongoing. This overview, you are all sitting in right now so congratulations for attending one session. Then what we're going to do next is have a brief outline of these other sessions that are going to occur. Monday, essentially following this session immediately, there will be a panel discussion on vertical integration. And that is whether the current prohibition on combination of registry/registrar should be continued. Following that, later in the day there will be a session on root zone scaling, how the DNS root zone can accommodate new gTLDs, especially with a coincident introduction of IPv6, IDNs, DNSSEC, all burdening the DNS potentially. Following that -- or coincidentally, at the same time on Monday, in a few hours from now, there's a session on registry operations intended to provide potential new registry operators with some guidance and best practices. So it's chaired by ICANN staff and members of the registry constituencies. On Wednesday, there will be a session on trademark protections, the work of the IRT. The IRT, Implementation Recommendation Team, is giving several briefings to the community here, several of the constituency groups. This session on Wednesday is a general session intended to arrive at solutions for these issues. And then finally there's a session on malicious conduct; that is, will a multiplication of TLDs, of names on the top level, also afford a multiplication of malicious behavior or conduct, such as phishing, pharming, that sort of thing. So I hope I haven't talked too much about that but what I am going to do now is introduce each moderator and have them give a brief introduction of their session, so first I would like to introduce Joe Sims of Jones Day. >>JOE SIMS: Thank you, Kurt. Welcome, folks. Last time I was here at an Australian meeting in Australia was in Melbourne where I was announcing and trying to explain the new dot com agreement with VeriSign. They appear to have asked me to come back to talk only on subjects on which they expect figurative eggs to be thrown. The panel we are going to try to run here at 12:30 is to designed to give an educational basis for further discussion on this issue. And let's -- just to put it in context, go back to the beginning. Remember that when we started ICANN back in 1998, there was one registry operator, commercial registry operator, and one commercial registrar. Both of them were part of the same company, and so there was a -- literally a monopoly at both levels of the space. And on top of that it was government sanctioned, and most of you know that government sanction monopolies are the ones that operate the best and most effectively for the monopolist. So one big objective in creating ICANN was to introduce competition to the DNS with all the consumer benefits that are expected to flow from that. So we started with registrars, because that was easier to accomplish for a whole bunch of reasons. And in order to make sure that the introduction of competition at the registrar level was effective, we put in this ban on vertical integration to try to make sure that when new registrars were competing with the old entrenched, formally monopolist registrar which was still commonly owned by the only registry operator, that they would have relatively equal access to the registry. That process was pretty successful. We now have very robust registrar competition so I think we can say that that job was basically done. And now, for the past several years and particularly with the new gTLD process, what ICANN is trying to do is the second half of that original job, which is introduce competition and presumably consumer benefits to the registry space. So the question is, under these circumstances, which are obviously different circumstances than we faced in 1998, is vertical separation still necessary or desirable or appropriate? If so, why? If not, why? So that's what we will discuss in this panel. I'll lead the discussion and try to not participate any more than I can simply not resist, with the main work being done by two economists who both teach at law schools and are recognized as experts in the vertical area. And if you give me the next slide. Steve Salop is one of the best-known industrial organizations antitrust economists in the world. He has been an economist at the Federal Trade Commission, the Civil Aeronautics Board, at the Federal Reserve and is a prolific author of all sorts of publications and the creator, along with Tom Krattenmaker, of the notion of raising rivals' costs which is a very important and very influential post-Chicago school economic concept that has had very significant staying power in the antitrust community. Josh Wright is a younger professor of law and economics who is following a similar path to Steve. Also a prolific author. Steve is Georgetown, Josh is at George Mason. They represent both sides of the economic and political spectrum, so I think the panel is designed to give a -- as balanced a view on these issues as we can manage. And the -- This is our agenda. We'll try to lay that foundation. We'll try. We are going to be talking about economics, not about policy. The policy is your job. Our job is just to talk about the economics. We'll try to be as objective and neutral as possible, and we'll try to focus more on principles. We'll then have some time to talk about the application of those principles to ICANN-specific circumstances. We'll have some participation from registries and registrars. And then we'll take questions from the floor. I hope to see all there. Thanks. >>KURT PRITZ: Thanks, Joe. The next session that's also going to occur later today has to do with root zone scaling, a very important technical issue. Lyman Chapin, very familiar to all of you, is doing a lot of the work in this area, coordinating studies, and he had and his team are doing a lot of the work, so I will ask Lyman to come up and briefly describe what's going to happen here. >>LYMAN CHAPIN: Thank you, Kurt. We have a session on the root zone scaling study that's scheduled for later on this afternoon at 4:00. And the study itself was commissioned as a result of an ICANN board resolution in February. It was commissioned by a team of people from the SSAC, Security and Stability Advisory Committee, the RSSAC, the Root Server System Advisory Committee, and the ICANN staff. Do we have slides? Okay. How do I make them go forward? You have got the machine. Okay. Essentially the issue is that if you think about what we've had over the past ten years, maybe even going on a little further than that, but certainly over the lifetime of ICANN, the root zone of the DNS has remained relatively stable. On average, we have added about one new TLD a year, and the change rate per TLD has been on the order of roughly one significant change per year. That's not a lot of churn in the root zone. And what we're looking at today is a situation in which we're about to or have already begun to make a number of changes that are going to affect the root itself and the system of root servers and root server operators that support the DNS root. And these include not just the addition of new gTLDs, which is the one that I think is foremost in people's minds, certainly right now and at this meeting this week, but also includes adding IDN support, particularly ccTLDs, but also for new gTLDs, adding support for IPv6, and adding support for DNSSEC, for signed zones, the root zone itself, signing the root zone itself and signing the top-level domain zones that live within the root. And we're contemplating doing all of these things essentially at once. So the board resolution and the activity of the steering group formed by SSAC or RSSAC and the ICANN staff came up with a terms of reference, the URL is listed up there. The terms of reference essentially lays out the dimensions, the scope and the target deliverables of the study. Go on to the next. So we assembled a team of people to work on this. We have a very short time frame. We only have until the end of August to complete our work. The names of the folks that are on the team you see up on the slide. And you will also see the URL for the announcement and public comments. These are all good background pieces of information to familiarize yourselves with, if you are interested in this topic, before you come to the session this afternoon. Obviously, this afternoon we will have an opportunity to go into all of these things in greater detail. So finally, if we go to the last slide, the session will be in Function 4-5 this afternoon at 4:00. And the public comment site that has been opened for this, you can send public comments into scaling@icann.org. Thanks. >>KURT PRITZ: Thanks, Lyman. Next we want to talk -- the next session will be Wednesday to discuss solutions to potential trademark issues having to do with the introduction of new gTLDs. And I'll ask J. Scott to come up. >>J. SCOTT EVANS: Good morning, everyone. As many of you probably know, in Mexico City the ICANN Board of Directors issued a resolution asking the Intellectual Property Constituency to put together a team of trademark or intellectual property and consumer protection experts to come up with recommended solutions for the implementation of rights protection mechanisms in the new gTLDs. And that group was constituted -- next slide, please -- in March. And here we have there were over 15 experts who were on the committee from nine different countries. There was also a registry representative and a registrar representative that were present to also make sure that the technical feasibility of any solutions that were recommended would be scalable and would be implementable by registries and registrars in the current technical situations they find themselves in in the Domain Name System today. Next slide. The IRT modus operandi, that we met at weekly two-hour conference calls for two months. There were thousands of e-mails. We had two face-to-face meetings, one in April in Washington, D.C. that was a two- day meeting where we called out and designed the five areas that we would concentrate on based on the various public comments that were received during the first and second version of the DAG policy development periods. We then had a second round of consultations and meetings in San Francisco in May. And the first day of that meeting, it was a three-day meeting, there were parties that were invited to present who had either already been implementing solutions such as Nominet that has a solution similar to the rapid suspension system you will be hearing more about this week, eBay, who has a VeRO program, some trademark owners who had put in comments, some parties who had criticisms, and then a final draft was put out on May 29th. The IRT has already presented the -- an overview to the GNSO on Saturday. Then yesterday, there was a presentation to the government advisory. And then there will be constituency briefings tomorrow. There are various members of the implementation and recommendation team that will be making every constituency meeting to present this to the constituencies. The big community consultation will be June 24th, that's Wednesday afternoon. And you will see there there's a great deal of time that has been allotted. That is the time for everyone who is not in a constituency meeting or hasn't heard a presentation to come and hear the solutions that are being recommended and have an open discussion about them. The views, the criticisms, the supports, the proposals, so that we can hone these and get these workable for moving forward. There will be two future consultations, so if you can't make it on Wednesday, and you are anywhere in the northeastern United States, you can come on July 13th in New York City there will be a consultation, and there will also another one in London, June -- July 15th. And we want everyone to participate. We are here to listen to everyone's concerns, criticisms, supports, because that's what we're trying to do is offer reasonable middle-of-the-road solutions that will protect consumers, your families, your friends. That's what we're looking to do. And we thank you. And we thank you very much. >>KURT PRITZ: Thanks, J. Scott. Finally, excuse me, I'd like to bring Rod Rasmussen up, the co-chair of the Antiphishing Working Group, who will help us conduct this session on the possibility of increased malicious conduct in the DNS with the introduction of new TLDs. >>ROD RASMUSSEN: Thank you, and good morning, everybody. I got a little bit of a promotion there. I'm actually the co-chair of the Internet policy committee, which is part of the Antiphishing Working Group. Just for background, the Antiphishing Working Group is a -- a fairly large organization. We have over 3,000 members, representing 1700 companies, organizations, governments worldwide. We are constituted to, you know, fight abuse on the Internet, criminal abuse, in particular, and phishing and malware and things like that. So I am -- we actually have a subcommittee called the Internet Policy Committee, which represents -- there's about, oh, 90 to 100 people on that group. That has a wide range of representation, including members of -- a lot of the members of this community, Internet service providers, registries, registrars, et cetera, law enforcement, and various others that have these interests. So in the Mexico City meeting, we had heard a lot about the new gTLDs, new TLDs to be introduced. And there was a lot of interest within our community and then there were some requests from various folks that are in the ICANN community for the APWG to make some sort of comments on this expansion in the TLD space and how that might affect the, you know, crime space and malicious use, et cetera. So I put together a subcommittee that will take a look at this. And as you can see in red, we are still working on this. There's a disclaimer that we still don't have full consensus on this. But we are working on this. And what we are going to talk about on Wednesday, though, are issues that have surfaced with regard to the expansion of the TLD space. It's fairly unprecedented, potentially a very large- scale expansion. So it brings up new areas of potential abuse. There's a lot of talk about different types of business models, about different types of strings, et cetera. So we are taking a look at all those kinds of issues. But as I mentioned, this is issues at this point, we are working on a final draft that will include various recommendations that will probably have some policy recommendations in there, but most likely a lot of best practices type recommendations as well. And there may be some dissent amongst our own membership about how to best do that. So we'll present both a balance as to what we think are, you know, universal things that should be done versus various ideas that we would like to get out in front of people. And one of the key concerns we have here is just raising security as an issue, because it's a lot harder to close the door after the bad guys have gotten in than to keep them out in the first place. So that's kind of an overview of that. I want to talk briefly about the categories of issues we'll be discussing. And, by the way, in this session, it's not going to be just the Antiphishing Working Group. There are some other organizations that are also presenting their views on issues here. But we've divided this up from the APWG IPC's perspective, we've divided this up into three areas. And that's potential for abuse based on the types of strings or the types of models that are being incorporated into creating the new TLDs, scale issues, if we're talking dozens, hundreds, thousands, however many new TLDs there are. That's a totally different prospect than we've seen in the past, where there's been one or two registries brought on at once. And then taking a look at some long-standing concerns, what was one of the very interesting things that I found in Mexico City is a lot of people who were potential new registry operators were coming up to me and others in the community and saying, "Hey, I want to run a really secure registry. How can we do that? What can we -- can the APWG tell us what are the best practices and things like that we need to be doing?" So one of the things we want to help do here is surface some of those things that we've seen in the past in various issues and attacks, et cetera, so that new registry operators can take advantage of that learning from the past to create a very secure TLD potentially going forward. So those are the issues we're going to talk about. And I thank you for having us down here. [ Applause ] >>KURT PRITZ: Thank you. So what's cool about ICANN is, you know, we get to work with people like that who I think are at the very top of our field. And it heartens me and I think everybody here that -- of course, we're all in ICANN -- that we're going to be successful in our endeavors, because as the need arises, people in the audience here stand up and volunteer and take part in the process in a very meaningful and passionate way. So it's very cool. So now you're stuck with me for a little while. I won't take too much of your time. I want to first review new developments in the new gTLD program that have occurred. I've got too much stuff for the time. But the idea is that this material will be posted. And you can review it after the meeting. And I'll kind of provide a guide as to what the material is. And then we'll talk about time frames going forward. So what have we done since -- let me page through a couple of slides. What have we done since the Mexico City meeting and the review book we've done there, the review we've done there? First of all, we've done a public comment analysis. And what's that? Well, we publish these versions of the applicant guidebook, the directions for how to apply for a new TLD. And we post it for public comment. And this is ICANN, and there's a lot of public comment. And so what we've done is analyzed that public comment. I'll talk a little bit more about that. We published now excerpts for the guidebook. So in certain areas, we've improved the guidebook. That's accompanied by a set of explanatory memoranda that I described earlier. Independent reports. And then how are we going to do this? How are we going to evaluate the applications? So there's been a lot of preparation work done organizationally and operationally to prepare for that. And we'll talk a little bit about that. We can make more material done later. This is what's posted on the Internet, a comment analysis where comments have been broken down into many subcategories. And those issues are addressed. The applicant guidebook excerpts, those areas of the guidebook that have changed. Several explanatory memoranda. And those go to the three-character issue that I described earlier, should so-called CJK scripts be allowed to register names in less than three characters to provide for full participation for those regions and communities? In past meetings, we were asked to publish the research that led to the guidebook recommendations on how objections to top- level domains based on morality and public order would be handled. So ICANN's published the research that provided the basis for those conclusions that are in the guidebook. There's also been considerable comment about post-delegation enforcement, particularly for community TLDs that are restricted in some way, self-restricted, and that they -- those TLDs continue to live up to their obligations to the communities they represent. There's another memoranda that supports the proposition that new gTLD registries be required to maintain thick WHOIS databases. So I think that's significant, and the reasoning behind that recommendation is in that memoranda. And then finally there's a paper on predelegation testing. So just before a top-level domain is delegated, we want to do some testing to ensure that the applicant has lived up to some of the promises in the application that was made and to help ensure that, it launches a stable, secure top-level domain. There's also been several independent reports filed. J. Scott briefly alluded to the Implementation Recommendation Team report that's addressing intellectual property issues and trademark issues. So that final report's published. The GAC timely published the report on their position on geographical place names and their conclusions as to how geographical names should be hand at the top-level domain and at the second level. ICANN published an enhanced economic analysis and report and also a response to the AT&T report. And there were two reports having to do with malicious conduct, the Antiphishing Working Group report that Rod mentioned, and also a report from the gTLD security group, Internet security group, or the RISG. So those are posted. So the public comment analysis is this big old 200-page document where ICANN has taken, essentially, 200 comments, parsed it into subsections or subcategories so it could analyze the issues raised in the comment, and then come to some conclusions about that. Should we change the process? Should the process stay the same? So the point we're trying to make in this big old document is that -- and it's kind of tough to make a nuanced point in a big, fat document -- is that there's a lot of attention being paid to the public comment, that first the public comment's all read, and that then it's parsed into issues or categories so that issues can be addressed separately, taking all the comment into account. And, third, that there is some balancing of the issues. So how do you run a new gTLD process? There's a bunch of 60-40 decisions. So when there's a request saying, you know, a change should be made, it's often a really close issue. And the comment analysis attempts to demonstrate that a lot of this balancing is going on. So even if changes aren't made, there's some reasons why. So it's -- like I said, it's a big document. I would pick out -- It's annotated and it's got a table of contents and an index. So you can pick out the issues you're interested in. And then we published changes to the guidebook. So we didn't publish a third version of the guidebook, V3, primarily because of the issues that were just discussed, that some of these overarching issues are not settled yet. And we thought it inappropriate to publish a third version of the guidebook, that that might tend to indicate that these very important issues at the end are not being adequately considered. And so -- But we did want to have a meaningful discussion here in Sydney. And there's been a lot of work that's gone on on other issues. So it was decided to publish excerpts of the guidebook, areas where the guidebook's changed -- where the guidebook's changed, so we could have a discussion about that. The whole guidebook's still open for discussion; right? It's not approved by the board. It's not published in final version. So it's all for discussion. But it's -- I think it's also important when a position is taken to try to put it into that document, into the guidebook. 'Cause I think that really points up the discussion. We can publish memos on how geographical place names are handled and sort of agree on the content of a memo, but until it's in the stark black and white of the applicant guidebook with the pending implementation around the corner, that that debate can really be sharpened. So because ICANN puts something in the guidebook doesn't say -- doesn't mean to say that's the end of the discussion. We're trying to say, "Here's exactly what we think it would look like," and so we can have a really sharp argument instead of sort of an ephemeral argument. So that's kind of the why. The what I'm going to be talk in a little bit more detail about each of these things. But those are the sections where the guidebook's changed. And so I'm going to talk a little bit more about them. Here's a diagram that those who attend all the ICANN meetings have seen enough so they have it memorized. But this is a very simplified version of how an application for a TLD is processed. It's evaluated, and in most case, we think that evaluation will happen in fairly short order. The evaluation criteria are intended to be clear and objective. And the -- most applicants, we think, will pass the criteria, given the investment required and the caliber of applications we think we will receive. And the TLD will be delegated. So that's a pretty short form of the process. There are opportunities to object, for other third parties to object to top-level domain applications on certain limited grounds. Those entities that require standing. And also there's another process to determine what happens if we get applications for two identical TLDs. We have to resolve that contention. So the way the guidebook is laid out is, essentially, there's a module associated with each one of these things that. Module 1 of the guidebook is an overall description of the process. Module 2 is how these things are evaluated. Module 3 talks about these objections and disputes. Module 4 talks about string contention. And the fifth module talks about the delegation process. As Paul Twomey said earlier, there's a ton of stuff about that on the Web site. So what changes were made to the guidebook? With regard to string requirements, we -- string requirements, these are the technical requirements that a TLD string has to meet before it's inserted into the root zone. And the primary purpose is so it doesn't break the Internet; right? So, for example, a TLD of all numbers might be interpreted as an I.P. address and not a name, and so TLDs with all numbers are prohibited. So this section was reorganized slightly. It's made a tiny bit more conservative. So if you can find out how, you should win a prize. Also, the guidebook raises the issue of this three-character requirement that we've often talked about. Also in the application -- in the evaluation section of the guidebook, we've asked considerably more questions about the background of the applicant. The antiphishing group preliminary paper seemed to suggest this would be appropriate, also certain constituency groups, the intellectual property constituency, the business constituency suggested that there be certain background investigations made into applicant history in order to avoid participation by individuals who have not been constructive in the past. So we've put some questions in about that. There's also some enhancements to security questions and DNSSEC, for example, DNSSEC implementation or provision. Provision of DNSSEC is not required now, but there's a notice that it will be. And then there's another set of changes to the questions that align the questions better with the scoring methodology. And do this other stuff. So, for example, there -- applicants are required to provide some sort of instrument in order to guarantee that registry operations will continue at least for some time, some basic operations, so that names would resolve in the event the registry fails. So there's safeguards built into the process intended to protect registrants. So we've enhanced -- you know, we've increased the number of options for those and tried to put more particularity into the definition of what that security instrument is. There's also improvements to or changes in the definition of geographic names, so how will certain geographic names be handled. The ICANN board in Mexico City said that we should make that definition more specific so it's more clearly understood an objective, and more clearly understood by registrants. So that's been done. So that's essentially the changes to the evaluation criteria. With regard to the dispute resolution step, which is in pink, we've clarified standing requirements for morality and public order. That's been an open issue of debate in the community, and considerable research has been done and papers written about that. The board asked that ICANN create a mechanism to filter out frivolous objections in order to streamline the process. And so we're approaching that. On community objections, we received feedback from several constituency groups how that objection should work, how the standards for the objection should be written, how an applicant can more or less easily defend themselves against the objections. So we've changed some of the language there. And then the role of the independent objector has been talked about for quite some time in the community. What if there's an obviously objectionable TLD where there's no objection? So ICANN's seeking to create a narrow, limited role for an independent objector. And there's detail in the guidebook about that. If we talk about string contention, what happens when we get applications for two identical strings or strings that are so, so, so close that they're essentially identical, because they would result in user confusion? And so you know, the cognoscenti know, and those who are new do not. But policy recommendation, seeking to enhance opportunities for small, community-based top-level domains, provided that in the case of this contention, that a community-based top-level domain would get a preference. And so the guidebook seeks to do that through this scoring mechanism that's sort of difficult to implement. So testing can -- you know, testing goes on all the time where we try to hone the scoring and -- so that's -- that's what's been done here. There's been a lot of testing so far. And the testing will continue. So there are members of the community that are very interested in this issue. And I encourage you to read that and give feedback on it as we go through. And then, finally, we get to delegations. So what happens in the delegation process? You sign an agreement with ICANN. You go through the predelegation testing that demonstrates you have lived up to some of the promises in your application. And then you're delegated into the root zone. So out there for comment in this excerpt of the guidebook, we have a requirement to maintain a thick WHOIS database, that there's some reservation of some names, geographical names, at the second level. So we -- the board has asked the GAC to report back to the ICANN community on what protection should be afforded at the second level. The GAC kindly reported back. In an effort to create a bright-line clear rule of what geographical names should be afforded some sort of protections at the second level, which names might be reserved, they developed a list of names. So they're country and territory names that are explicitly written in those documents. So if the recommendation to the GAC is adopted, then there will be a clear list of countries that are protected. And then I've already talked about delegation. So that's the -- that's sort of all the changes to the guidebook. There's been a lot of work done trying to determine how we're going to process these applications. So ICANN is in the process of building a separate standalone evaluation organization. The ICANN staff will be kind of skinny, because most of the work will be done by evaluation panels. And the hiring and letting go of those panels is going to give us the flexibility we need to grow and shrink with varying numbers of applications when that application work gets done. So we've been doing a lot of work to determine how long it will take to process applications. So, for example, based on quite a wide range of applications, but this is a set of 200 applications, we think it would take about a month to process all the applications through and get -- once the closing date is arrived at, and then about four or five months to evaluate, four and a half, five months to evaluate all those applications. Why does it take that long? Well, we think -- there's a number of panelists that will be evaluating those applications. And after those are done and scored, they need to be spread out on a table to ensure that one evaluator didn't give a different score to, essentially, identical answers from another. So there has to be some sort of normalization process. And then the initial evaluation also includes a step for communication with the applicant. So you can imagine a couple- week process where applicants are asked questions and answer them. So given that, we think the evaluation time frame is about four and a half or five months. And then the predelegation testing, signing an agreement, submission and approval to the ICANN board, we think that would take about two months. So we think about seven and a half months after the receipt of applications, we'd receive that. If applications are objected to and go down separate tracks or there's string contention, that would add more time. So I think those are clear. And these will be posted so you can stare at them and ask staff questions during the week. So how are we going to -- ICANN anticipates receiving about 4- or 500 applications, you know, plus or minus, 4- or 500. There's a great deal of uncertainty as to the number of applications we receive. When I was giving a briefing to the board on this, Paul Twomey raised his hand and said, "What if we get 2,000?" Which he should have asked me before I was standing in front of the board. But ICANN is planning for all those contingencies. And what does the - - what does the scaling work we've done so far mean? You know, we've done our own internal buildup of organization and analysis, and we've also commissioned an outside firm to mirror that work to sort of test it. And what do they say? They say, well, for -- well, what's it say here in for a volume of 500 applications, we'll probably have 175 people touching all of those at one time. So if you think about the six different reviews that have to take place, about 175 applicants. The most intensive evaluation is the technical evaluation, where we'll have, essentially, one evaluator for every ten applications. So if we want to process all the applications in five months, we'll need four -- four and a half or five months -- we'll need one evaluator for every ten applications to do the technical part of the evaluations. So think about if you receive 500 applications, you'll need 50 technical evaluators. That's a lot of technical evaluators across which to have consistency and normalized results. So this seems to point to that if we receive more -- you know, 300, 400, 500 applications, instead of processing all the applications at once and announcing the results, we'll probably have to process them in batches in some way, and a fair mechanism for determining which applications be taken first has to be developed. And you can think through all the issues there. Everybody's been thinking about this for a long time. So what's going to happen next? J. Scott talked a little bit about consultations going forward. So this Sydney meeting is -- the Wednesday session is the first of three sessions ICANN's going to hold primarily to drive to solutions on trademark issues. So we're going to talk about the IRT results, the proposals other groups are making with regard to trademark protections. We're going to weave into that concerns about malicious conduct, because those two issues are really, really closely aligned, and in these three sessions, attempt to drive at what the final version of the guidebook will be before the end of July. We're also going to conduct additional outreach sessions in Hong Kong and Abu Dhabi that will address many of these same issues and also include other issues such as IDN implementation and focus on policy work. So we'll have more information about all these sessions, but you can learn about them on the ICANN Web site. So the schedule, with the caveat that the community controls a lot of what we do here, says that we'd wind up these issues and the remaining issues in the guidebook around the August time frame and be able to publish this third draft version of the guidebook, the last draft version of the guidebook, in September that. Public comment period would wrap up right after the meeting in Seoul. And that would enable ICANN to publish a final draft -- not a draft -- a final guidebook. Let's get "draft" out of our lexicon -- in December, with the idea of then accepting applications in February. Which about two months after that, two and a half months after that. So that's essentially the time frame. And I think that's it. So thanks very much for coming. I'm happy to stay for a little while and take questions. I think we have, essentially, a half hour left in this session for those that want to stay and ask questions. So if anybody has one, do we have -- Nancy, do we have, like, coordinators? >> We have microphones in each of the aisles that people can go to, as well as my microphone. So please feel free. >>KURT PRITZ: Howard. >>HOWARD LI: I'll be the first one. Actually, I don't quite have a question, but I do have recommendations. Actually, you know me already. I'm from CNNIC. So in public comments analysis, it mentioned that to -- addressing the variants issue, it's to use an alias to resolve it. But -- and in the comments, it says it's not possible; right? So, actually, we do appreciate all the efforts that Oak has put into researching a feasible way to resolve these issues through the technical way. But from all the years of experience that we are working with IDNs in CNNIC, we admit that's extremely difficult to resolve that problem using the aliasing, either DNAME or CNAMEs. So -- But this issue has to be addressed. And we propose that we can do it through the policy way. So right now, what we are doing, the current successful practices are creating multiple records that have the same registration information to realize the equalization of the variants. And although they are different entrants in the rule technically, but by policy definition, in the policy, per se, they are all one, all one name belong to one single registrant. So I think it's recommended that ICANN's new gTLD policy to accommodate these needs by following six steps, which is -- I kind of pulled it out of my head. So one is, an IDN TLD application must indicate whether the characters in the string, whether they have variants. If yes, what they are, what the variants are. Second is, are all variant strings really reserved in the first place? Then second, the applicant must apply if they want the variant string to be active. If yes, which string? Because variants may have more than two or three. So if they want two or three, they have to indicate it. The fourth, that the variant string should be delegated to the applicant at no cost. I mean, when I say no cost, no application fee, due to the fact that the major evaluation is already done. And then the fifth is the applicant TLD operator must distribute the TLD, like, when they're doing the name, and all its variants as one, they can only sell it as one. And the last step is to -- since activating the variants will actually create extra cost, extra load for the DNS server, so the cost should be recovered through a reasonable raise of the registry-level fee at a variant style. And how much it should be raised and -- we need to do a cost analysis on that. And that's my proposal. Thank you. >>KURT PRITZ: So, just to demonstrate my ignorance, each variant is essentially another delegation, because it cannot be addressed through aliasing? >>HOWARD LI: Yeah. >>KURT PRITZ: And each variant, wouldn't it be confusingly similar with the existing string? >>HOWARD LI: In the user's mind, yeah. >>KURT PRITZ: So it would be, essentially, reserved? >>HOWARD LI: Yeah. But if you want it to be active that people can use it, then we can -- we have to, the applicant has to indicate that, which one they want to do. >>KURT PRITZ: Right. And so you're our -- one of the things you're advocating is that should be at a lower rate because it's a necessary step? >>HOWARD LI: Yeah, because ICANN given the applicant as one and the applicant can only sell them as one. It's for the benefit of the user. >>KURT PRITZ: Yeah. Okay. Amadeu. >>AMADEU ABRIL i ABRIL: First of all, thanks for publishing excerpts of what will be the new guidebook instead of waiting for the whole thing to be completed, as was the first two. I think it's very helpful when -- you know, if we have some advance where we can comment on that. I have one simple question regarding the procedures. I am told that now it's Santa Claus that will bring the final guidebook, or perhaps the (inaudible), or who knows, Saint Johns in June 2010. But, anyway, let's suppose it's Santa Claus, just for planning. We are having these sessions in New York, London, Dubai, and Hong Kong. That's sounds like an interesting holiday program except if you are not on holidays. Are they supposed to be about the same issues with more or less the same problem? You mentioned that in Hong Kong and Abu Dhabi probably they will be dealing with IDN issues. But my impression was that for the first two, you simply said it is about the famous reports, IRT, et cetera, but not about the guidebook itself, not about what's changed in the guidebook, not about what are the issues regarding, I don't know, contention, disputes, technical evaluation, et cetera, that we are still discussing? >>KURT PRITZ: Thank you. Two points. One is that I think if you -- interested parties should aspire to attend one session. In other words, if you give input into this Sydney session, you need not attend the session in New York. If you attend the session in New York, you need not attend in London. They're meant -- >>AMADEU ABRIL I ABRIL: You don't want to see me around? >>KURT PRITZ: Just you. Just you. And the primary purpose of the sessions is to drive toward closure on trademark issues, so the trademark issues in the guidebook. So the sessions will certainly affect the guidebook with regard to that issue, but they're not -- and they will also necessarily affect the guidebook with regard to malicious conduct and avoiding that or preventing that. And so changes to the process that will help prevent malicious conduct. But they're more targeted at those than all the other -- all the other outstanding issues in the guidebook. >>AMADEU ABRIL I ABRIL: Okay. Thanks a lot. Just a small nonsolicited commercial advertising. For those that may believe that perhaps Santa Claus will not bring the final thing, at 1:30, we have a session where cultural, linguistic initiatives and city initiatives, but it's not restricted to that, try to discuss and present a proposal for how to deal with an early window in case Santa Claus doesn't bring the gift. So just join us at 1:30. >>KURT PRITZ: I hope you don't get a lump of coal. >> I have a question online, Kurt, would you like me to read it to you so it can be answered? >>KURT PRITZ: Do you want -- I'll -- okay, sure. >> All right. All right. The question online is as follows: What will the process end timetable for resolving the overarching issue dealing with demand and CC? And this comes from Steve Metalitz. >>KURT PRITZ: Steve Metalitz. >> Thank you. >>KURT PRITZ: No, I'm sorry he couldn't be here to ask that question. So I don't know if you can hear me, Steve. ICANN's -- ICANN published, in preparation for this meeting, a new version of the economic study that was released earlier that was enhanced and also a response to a different economic study. So, you know, I'd rather answer Howard's technical questions than economic study questions, because I'm more out of my depth. But the study, as I understand it, attempts to do a couple of things. It attempts to demonstrate that benefit will accrue from the introduction of new top-level domains -- excuse me -- benefit will accrue from the introduction of new top-level domains, that the benefit is somewhat independent of demand. ICANN has done introductions in the past, and each one, I think seven TLDs were introduced, so the paper demonstrates that there will be benefit from community-based TLDs, from TLDs that offer more competition and choice or lower prices. The study also states that this is a cost-benefit decision, right, that there's benefits, but there's also costs in the form of, you know, possible risk to rights holder issues, possible increased malicious conduct, all of the other studies that are going on. And so in order -- and says those costs should be addressed. So what -- what the economic study is necessarily interwoven with these other things, where the IRT is seeking to find solutions to other trademark issues. And other groups are studying to ameliorate potential malicious conduct. So I think that having said, this study is out for public comment, and the ICANN method is to put these pieces out for public comment, and listen to the public comment and then iterate next. So I think that's the next step. Actually, this gentleman has been having his hand raised for a while, if that's okay. >>PHILIP ARGY: Thank you, Kurt. Philip Argy from ArgyStar.com, and I'm moderating the 1:30 session that Amadeu promoted. Can I just urge people to come along and make their contribution because I think it has to be recognized there is a cohort of the ICANN community that is getting a little impatient. And this is an attempt to address that impatience in a methodical and objective way, and I think it is worth getting people's input. So it's not solely confined to the city and cultural TLDs, although that's the opportunity, if you like, in which that process can best be discussed. But the logic is there will be some TLDs in respect of which there is no rational objection, and they can really be fast tracked. So comment on how that process could work is quite important. And I notice that constituency session was inadvertently omitted from your schedule preview there. >>KURT PRITZ: So when is it set -- Can you tell us.... >>PHILIP ARGY: 1:30 today in Function Room 3, which is one level below here. >>KURT PRITZ: Thank you. >>PAUL FOODY: Hello, Kurt. Paul Foody. I think we met in Mexico. Am I correct in thinking that you said you are expecting 2- to 300 applications for new gTLDs? >>KURT PRITZ: That's a -- what I said is we're expecting 4- to 500 applications plus or minus 4- to 500. >>PAUL FOODY: And what is the maximum number of gTLDs you are going to permit? >>KURT PRITZ: Well, we are going to process all the applications. All the applications that meet the criteria will be delegated into the root zone. >>PAUL FOODY: But is there a maximum number? Is there going to be 1,000, 10,000, a hundred thousand? >>KURT PRITZ: So I think the limitation would be determined -- if there is a limitation, one of -- that's one of the reasons that we're doing this root scaling study. So the group that's headed by Lyman, that's been commissioned by the RSSAC and SSAC, has asked that we study, given the coincident introduction of new gTLDs and IDNs and DNSSEC and IPv6, whether there are issues or limitations on the number of TLDs that can be introduced. And I think that's -- Lyman can answer more if he wants, but I think that's one of the goals of the study. There are others. >>PAUL FOODY: But given the relationship between supply and demand, if there's only going to be 10,000 new gTLDs, then obviously there is a scarcity value, whereas if there are just going to be as many as dot com registrations right now, there's 80 million dot coms, if you anticipate there are going to be 80 million dot TLDs in the future, all of which will be permitted, that makes a very big difference as to whether is going to stump up 185,000. >>KURT PRITZ: That's right. >>PAUL FOODY: So when will you be able to give that information? >>KURT PRITZ: This information? Well, I believe a preliminary version of the study is due at the end of August. >>PAUL FOODY: Thank you. >>KURT PRITZ: Are we going back and forth? >>ALEXANDER SCHUBERT: Hello, Kurt. My name is Alexander Schubert. I am CEO of dotGAY LLC. For the last ten ICANN meetings, the timeline had been very stable. It was always like 9 to 12 months at any given point of time. Do you think the timeline will still look like this in, let's say, half a year or do you think we can really apply, let's say, before June next year? >>KURT PRITZ: Yeah. So I -- So ICANN is trying to do a couple things; right? There's a lot of people in this room and outside this room are very interested in applying for TLDs and want certainty, and they've invested a large part of themselves and their businesses in this process. So I think it's very important to provide predictability for those people. And then the balancing for that, as you know, as being part of the ICANN community, is doing -- and especially since what we're doing is going to affect the DNS significantly, is to do things with appropriate consideration. And I think one of the benefits and drawbacks to this whole process is that many important entities that have really good things to say have not been energized until, you know, the clarifying glare of impending implementation is staring us in the face. But some of the issues have not been addressed in the way that those entities would like. And I think they are raising good questions. So I think the right decisions are being made to do this additional study. You know, having said that, you know, you and I, I think, are on the same side. I would like to see this process through. A lot of work has gone into it, and there's a lot of expectations, and the whole ICANN community has been talking about it for ten years, you know. It was in the founding papers. And the significant policy development work that went into the 19 policy recommendations that are pretty neat when you read them carefully, all indicate that we should push forward as fast as we can. So ICANN staff is and a lot of people in here are really working to push this forward according to the time frame you saw up there. And there's a lot of partners in this. Lyman talked about the study he is trying to get done in time. The IRT team did an incredible amount of work in a short period of time. Those people invested at least half time, and half time being half of 16 hour days in order to do the work they did. So there's a lot of people in ICANN staff and in ICANN that are really invested into getting it done. >> Thank you. >>KATHRYN KLEIMAN: Hi, Kurt. Kathy Kleiman. As you know, the IRT report was written by a committee that was comprised largely, the vast majority, from the intellectual property constituency. Not a single representative of the registrants was on that committee, not a commercial registrant, not a noncommercial registrant, not an individual registrant. My question for you -- And so this is very unusual, in my experience, in ten years of the ICANN process, to have a group comprise that much from one constituency and presenting a report with ICANN support. So my question for you is, who's processing the comments from these public meetings? The meetings here and around the world? Will it be the IRT committee or will it be ICANN staff? >>KURT PRITZ: So I think -- Well, I don't think it's either. People think ICANN staff do things. But I think ICANN staff are channelers of everybody's opinion. So let me go back to first principles a little bit. First Mary Wong was on the committee. So even though she's an I.P. attorney, she's also a member of the noncommercial constituency; right? So that's -- but that's a technicality. >>KATHRYN KLEIMAN: And, of course, everyone was representing themselves in their individual capacities and was not -- without the transparency of the ICANN process, which is traditionally ICANN process. They were not allowed to report back or share with their constituencies what was going on. So there was no process of back and forth with Mary when she was on that committee. >>KURT PRITZ: Right. I grant you that. I was just making a small point. So -- so this constituency, the I.P. constituency, raised a number of serious concerns during this implementation process. And the board, seeking to find a path to arrive at some solutions in the best way, asked the I.P. constituency to form a group and do some balancing and develop a set of proposed solutions for consideration. So now we have the report of the IRT. And what are the next steps? So the next steps are these consultations. So now -- you know, I would see the role of the IRT now as sort of a first among equals. So now we have IRT recommendations, we have recommendations from others, we have other community participants that can come together. The idea is to plop these things -- You know, it's hard in a big room, but we want to plop these things on the table and all talk about them and have a community discussion about which ones find their way into the applicant guidebook. You know, we have the clear writing of the IRT about what the recommendations are, a lot of detail work, the idea that it's a tapestry. But we also have the comments already of many others in the community. While the -- another point I want to make is, while it has been a restricted group, exactly how you point out, that group, you know, as an attendee at the teleconferences, they strived to publish every bit of information they could. So they published minutes, unvarnished minutes of meetings. They published every report. They tried to be transparent so that their progress could be monitored. But now it's the time where the proposals are out on the table and they are, you know, a member of all of us that are to discuss these things and decide which ones get into the guidebook. >>KATHRYN KLEIMAN: But who's decided? Who makes the final decision? >>KURT PRITZ: Well, so I think it's the same as any other ICANN issue. So we've got the IRT report. There's going to be comment about it. We're going to have three consultation sessions, you know, London, New York, and here at least. We're all going to talk about it. And then to the extent ICANN makes the decisions, you know, ICANN will make a decision. But as I said, you know, ICANN, I think, is a channeler of the community discussions. We posted the first discussion of the guidebook. Maybe 60% of that is okay. 70%. There's a lot of comment about 40%. So we changed 40% of the guidebook. Don't write those numbers down. I'm just making them up. But there's a lot of red ink in the second version of the guidebook; right? And -- so then we had another iteration, and we're rewriting the guidebook. So it's -- ICANN is sort of this dammed harmonic oscillator kind of thing. I don't think staff is the decider. Staff is the monitor and then putting out for feedback again. >> We have one more remote question, Kurt. This is from Fatima Cambronero. Does ICANN expect the new gTLDs to cause a decrease in dot com registrations? >>KURT PRITZ: I don't know. We'll have to ask ICANN. I really don't -- I really don't know. The -- I don't know. I'm sorry. Is there any other questions? >>WENDY TANWHITE: Hi, sorry, Wendy Tanwhite, marketing director for Gandi dot net. We did a report recently about the liberalization of the space. And the -- two of the biggest things, I think, that came out for us, which is surprising, is, you know, for the actual consumer, as in consumers, small businesses, corporates, how are they going to be involved in a broader way in this process? Because I think the thing that's going to change is many more stakeholders are going to be involved in this process. Each domain, whether it's moral, brand-led, community-based, you know, you start to bring in much -- many more stakeholders. So how will you start to involve more people in that process, I think? I commend the way it's been managed to date. I can see, you know, you're trying hard to take multiple views. But this is going to become much larger and a much, much larger space. So I think that's what we're interested in and our consumers are interested in. >>KURT PRITZ: So I think a couple ways. And the couple ways I'm going to describe are not the total answer. But, you know, in a small scale, there's a heck of a lot more people in ICANN meetings now than there used to be. Participation by entities has increased as companies and governments and other groups realize, and individuals realize, that there's going to be a significant change to the domain name system, they're looking -- looking for where to participate. So we've seen an influx of entities that are involved in the ICANN community now that weren't involved before, because they have -- because they have a dog in the fight. And then there's a couple other significant efforts that are under way. Certainly the at-large group has expanded significantly and become much more important in the ICANN community with the formation of the RALOs that were planned for years and now have become a reality, the At- Large Summit that occurred in Mexico City where at-large had a parallel meeting and made several significant pieces of advice to ICANN. And then -- so there's at-large, and then there's a restructuring of the policy-making bodies through the review and improvement process. So those groups are being analyzed by themselves and others. New constituencies are forming and the group is being reconfigured to make participation more effective. So I think -- I don't know, that's sort of a general answer to your question and I don't know if it's helpful or not. >>WENDY TANWHITE: I think it is, and I think it will be a growing issue over -- well, once applications actually start happening, I think you will get more reality, as you will start to see. The closer you get to it, you get real bodies involved. >>KURT PRITZ: Yeah, so I thought the paper was good. So if anybody wants to read the Gandhi paper, I'm sure it will be on the ICANN Web site someplace. >>WENDY TANWHITE: Great. Thank you. >>KURT PRITZ: Next. >>SIVASUBRAMANIAN MUTHUSAMY: Yeah, I'm Sivasubramanian Muthusamy from ALAC. My question is at the moment, most of the gTLDs are concentrated -- the applicants for most of the gTLD registries are concentrated in a certain geographic region. With 4- or 500 new gTLDs coming up, do you think there will be a geographic distribution of registries? >>KURT PRITZ: Yes, I do. I think the reason for being for the new gTLD program, and especially the IDN program, is to encourage regional participation where there isn't any now. And I think that will foster, you know, not just new gTLDs in those regions but new registrars and new at-large groups and more participation here. So.... >>SIVASUBRAMANIAN MUTHUSAMY: No, the question more concerns about gTLDs, not IDNs. And the cost structure, the application process structure is kept very high. The cost structure is kept very high, and so -- But that kind of cost level, do you have an equal distribution of applicants from different geographic regions with respect to new gTLDs? >>KURT PRITZ: Yeah, so nobody knows. When I talk about IDNs, I'm talking about IDN gTLDs, not ccTLDs. So I think that's meant to foster participation in regions where there aren't any. The application cost right now or the evaluation processing cost is just that. It's a cost. It's ICANN calculating how much it's going to cost to go through a fairly complex process of evaluation. And just recovering those costs back. And I think that -- So I think two things. Over time, we'll get more certainty of cost, become more effective, and those costs will go down and the application fee will go down. As we go our legs under us a little bit, become more certain of the application environment, there will probably be an opportunity for scholarships or for reduced fees for media applicants. And I forgot what the third thing was. Anyway, so I -- Oh, and the fact that we want to announce a second round with potentially lower fees. You know, at the same time as the first round. So those that can't afford to apply in the first round might see light at the end of the tunnel, a chance for reduced fees and an application period only months from then instead of potentially years. So that's it. So is that -- Does anybody have any more questions? Because you all have been patiently sitting here listening to me. Okay, well, thanks very much. I hope we have a great meeting. Attend the sessions; okay? Thank you. [ Applause ]