GNSO Working Session Saturday, 1 November 2008 Afternoon Session >>AVRI DORIA: Okay. Let's start. On the agenda, we've got this for another hour. And then we get into the new gTLDs. Where is that agenda in front of me? Yes. So where we were at when we broke was, I had a list of at least one name. Mike, I had Jeff on it, but I think Jeff removed himself from the list. And what I was trying to get was an answer to the question of, as we start to go through the hypotheses, did we want to use the organizational method provided by the registry constituency, in other words, look at them, walk through study A, walk through study B, walk through study C, and basically look at each one, see if there are questions, see if there are comments, see if there are perhaps presumptions being made that people either understand or don't understand, are comfortable with, don't -- don't feel. So at the moment, though, I'm asking, do we want to use this organizational method as opposed to walking through them numerically? Does anyone object -- okay, Mike, I had you on the list. And then I'd like -- if anyone objects to following that methodology, please speak up. Otherwise -- >>MIKE RODENBAUGH: I'll pass, Avri. Let's move on. >>AVRI DORIA: Okay. Thanks. So does anyone object to basically looking at it chunked in the studies? In which case, why don't we start with Study A, question 1, and I'll turn it back over to you, I guess, to -- or to talk about these. Yes, Tim. >>TIM RUIZ: Okay. Just a question. So what do we -- so what is the goal of our discussion? >>AVRI DORIA: The goal of our discussion is to, A, make sure that people understand what these things are saying, the implications, the presumptions behind them, and to pick up any council members', other members', constituency members' viewpoints on these items. Understanding, understanding the things themselves, and understanding, as much as possible, constituency viewpoints specific to each of these questions. >>TIM RUIZ: So does that -- Tim, again, with the registrars. So does that include a decision by -- or a -- >>AVRI DORIA: No decision. >>TIM RUIZ: Topic-by-topic discussion -- a decision, okay. >>AVRI DORIA: No. In fact, these meetings, the Saturday/Sunday meetings, are never decisions other than, do we do this now, do we do that now. But, certainly -- And to do any decisions on this stuff, we will need a motion at least a week in advance of, you know, anytime we're going to vote on it. So we're not making any decisions about who is at this meeting at all. I'm hoping we can get a motion written before our next meeting, but that's a hope that I'm told might be a might optimistic. But.... >>CHUCK GOMES: All right. The first recommendation that the registries made was to combine from area one studies -- study recommendations 1, 14, 21, and GAC data set 2. You can see -- get my -- I've still got to get used to this. You can see that I have Study 1 and 14 shown right there. You can take a quick glance at. I won't read it for you. You can read. >>AVRI DORIA: So go through them one by one. So -- >>CHUCK GOMES: So Study 1 is, you know, that -- the hypothesis is that public access to WHOIS data is responsible for a material number of cases of misuse. Et cetera. You can read the rest. >>AVRI DORIA: Does anyone have any issues, discussion on 1 at this point? >>CHUCK GOMES: And at this point, Avri, it's not -- the recommendation is just that we combine these -- >>AVRI DORIA: Right. >>CHUCK GOMES: -- four into a study. >>AVRI DORIA: Right. No, at this point, what I was almost taking as given is that we're using this as a method. We're not actually talking about the combining them into a study. We're just using that study -- >>CHUCK GOMES: So you don't want to discuss whether or not this should be combined? >>AVRI DORIA: At this point, I think no. At this point -- >>CHUCK GOMES: Okay. >>AVRI DORIA: -- I think we're taking that we've taken it -- no, leaving it up -- I think we're taking that as a working model. >>CHUCK GOMES: Well, we're in the wrong place if we -- >>AVRI DORIA: Well, no. Because that's the organizational model we're using. >>CHUCK GOMES: Okay. >>AVRI DORIA: And we'll use that as a working organization model. In other words, take it as, yeah, we'll chunk them that way until we get to a decision point later. >>CHUCK GOMES: Right now, you want people to look at this A -- >>AVRI DORIA: To look at -- >>CHUCK GOMES: -- A, the four elements of it. >>AVRI DORIA: Mm-hmm. >>CHUCK GOMES: I need to go to another part of the document to talk about what we thought was the priority of this chunk. >>AVRI DORIA: No. But I wasn't getting into priorities, either. What I really want to do is make sure -- and I guess I -- >>CHUCK GOMES: I'm totally unclear of what you want. >>AVRI DORIA: I guess I didn't make myself clear. What I want to do is go through the list of hypotheses, but using your Study A, B, C as the working organizational principle for now. >>CHUCK GOMES: Please understand that A, B, C does not include all of the studies. There's only some of them that we suggested combining. >>AVRI DORIA: Okay. Then we'll go through the others afterwards. The other option that I'd asked people about was whether we just went to the hypotheses themselves and went through them in numerical order. But it looks like, and there had been some comments in favor of using the work you guys did as an organizing principle for understanding, for discussion purposes. >>CHUCK GOMES: So what is it that you're looking for as we go through these now? >>AVRI DORIA: I'm looking for, at first, to make sure that people understand 1. If they have any constituency viewpoints about 1, any issues, is there comments people want to make about these. Or do we have any other constituencies that want to sort of contribute content to -- or questions? Do they understand it. >>CHUCK GOMES: So you can't really -- I don't think you can really get into constituency comments unless you start talking about values and priorities and things like that. So I'm not following at all what your objective is. >>AVRI DORIA: My objective is to get as much information out on the table as possible at this point. And it looks -- since you guys had done so much work, it looked like a useful organizational principle to sort of just take what you have and sort of -- So it makes sense for you to talk about what your priorities were with this. >>CHUCK GOMES: Well, then I need to go down another part of the document. This doesn't -- all this was a -- just showing which ones we thought could be combined. >>MIKE RODENBAUGH: Why don't we start with your section 3. Does that make more sense, Chuck? >>CHUCK GOMES: Yeah, that's -- >>AVRI DORIA: Okay. >>CHUCK GOMES: The -- And that's the cost estimates for formal studies. >>AVRI DORIA: And that also lists the studies there, so that's good. >>CHUCK GOMES: Okay. So that's why I was saying. >>AVRI DORIA: Okay. >>CHUCK GOMES: And this will then -- Notice, this is Study A that we're just looking at, combining 1, 14, 21, and GAC data set 2. We thought that was a top priority. And, you know, that's what we -- that's what we put forward right there. Let me pull it up so you can see at least 1 and 14. And part of 21 and GAC data set 2. And I think rather than -- >>AVRI DORIA: So the -- you're basically saying these three are sort of in the same subject area? >>CHUCK GOMES: That's right. >>AVRI DORIA: 1 and 14 look like they come from opposite points of view on, essentially, the same subject. >>CHUCK GOMES: Mm-hmm. And so we thought if, in fact, a study could be designed to either substantiate or refute these hypotheses, that that could be very valuable information going forward. So I'll just leave that open for discussion, then. >>AVRI DORIA: Okay. I've got Claudio and then I've got Tim. >>CLAUDIO DE GANGI: Claudio, with the IPC. Chuck, I just had a question. As far as combining these, there was a study in area 1, Study 15, it seemed pretty related to this. And I was just curious how come that wasn't combined. >>CHUCK GOMES: How come 15 wasn't combined into this? >>CLAUDIO DE GANGI: Yeah. >>CHUCK GOMES: Let's go down. See, where do I have -- 15's in which -- whoops, I just passed it, didn't I. Here we go. So.... Well, that one's really oriented towards port-43 access, a little narrow focus, I think, is why. Now, whether we made the right decision there or not, keep in mind, we're not really dogmatic on -- like I said before, on all of these things. So if there was consensus to move 15 up into the other, I'm not sure that we would fight that, although we thought 15 was a little lower priority than those other things. But that -- you know, maybe the council disagrees with that. And we can discuss that. Okay? >>AVRI DORIA: Okay. Tim. >>TIM RUIZ: So the question I have is if these were combined, you know, they came from different sources, and so can we come up with a single definition, or is that even our job, for abuse? It talks about spam and other illegal activities, abuse of personal data, a lot of sort of ambiguous terms that mean different things to different people. So how do we come up with a single -- or is there a thought on how we might come up with a single definition of abuse or illegal use or whatever? >>CHUCK GOMES: Let me respond first, Avri, and then you can jump in. Keep in mind, too, that the hypotheses working group really stated -- were tasked with doing concise hypotheses. But we made the statement that anytime any cost estimates or study designs happen, that the people doing that need to go back to the full WHOIS working group report that gave a lot more detail and defined a lot of this. Now, whether it defines it enough, I'm sure it varies. And that will have to be done. You're absolutely right. >>AVRI DORIA: One of the things I was thinking -- and this is partly something that Chuck had said earlier -- was that we're not just talking about port -- I mean, we're not just talking about price estimates when we go out. So one thing that we could decide to do is to list the hypothesis information. And part of what we're asking for from a professional study doer, one that does them on a scientific basis, is to look at the hypotheses that we've collected -- and basically come up with which one of those is testable, how are they testable, and so on, to actually ask them to take that next step, if we don't do it ourselves. >>TIM RUIZ: So we would -- Tim again. I keep forgetting to do that. So you don't think we need to define that? I'm just concerned that if, depending on who gets this and is doing the feasibility study, you know, what will be, you know -- and it says another term, "undesirable activities," so as they're looking at that, how will they frame their feasibility study when it's something as ambiguous as, you know, "undesirable activities," because I believe in the report, I don't think it goes into any further detail than that. So is that a definition we'll need to come up with or is that part of what they'll look at and come back and say, "These need to be defined further," or "It's too broad. We can't really do a study on that," or -- >>AVRI DORIA: I think that we certainly have the option of defining it further if we feel that we should, if we feel that we can. But I think a viable alternative is, at a certain point, to sort of take what we've done -- if we decide to go further, to take what we've done and ask somebody who does studies -- does scientific studies as a profession to look at what we've done, to ask us the specific questions that need answering, to basically come back with a recommendation. And if they can't come back with a recommendation, to come back -- I mean, with a proposal, you know, for a study. And if they can't because we need to answer more questions, to get them from us, there's a certain part of getting a professional study done that none of us is really a professional study doer, you know. I mean, some of us may do them in other parts of our careers. But, by and large, doing, you know, scientifically valid social studies is not what we do. We have questions. We can form hypotheses. But we're really not study designers. And so we'll have to figure out at what point it's worth it for us to say, you know, we've gathered our hypotheses, we've gathered our opinions on them. We've done best. Now, we think this is worth following through with. Can we get a professional to sort of work with us, to tell us how much it would cost, and to ask us the questions they need to define a study where they have proper controls, where they can show validity for their results, et cetera. >>CHUCK GOMES: And -- this is Chuck again. Tim, I think probably -- and Avri hit on this a little bit there -- I think what I would predict is, is that if we took this, for example, this set, and said, "You know, give us some cost estimates," and they're going to have to do some other things to even get to the cost estimates, I would predict that there's going to be some interactivity between staff and any consultants they might use on this and us to get more clarity where they need it. Now, if we can decide where we need clarity in advance and provide that, that's probably good. But I think there's going to be some back and forth in the process of coming up with the estimates. Because, you're right, in some cases, it's just too vague for them probably to really be able to go forward. So they're going to need some more direction from us. >>TIM RUIZ: And I think, too, part of my point was, by combining them, which doesn't seem illogical, given the subject matter, by combining them, then what we're implying is that these -- there is a single definition across these combined studies for abuse or misuse or however it wants to be put. And, you know, is that something we need to look at so that we make sure we cover what was actually being considered by the proposers of those studies? >>CHUCK GOMES: That's kind of what I -- whoops. That's kind of why I started out with, you know, looking at even the suggestion to combine them. Does it make sense to combine them like we suggested? At some point, we need to decide that. Because you could be right. Maybe it doesn't. But, again, it's a decision that we have to make. >>AVRI DORIA: And I guess it is also a decision whether people feel we need to define "misuse" and "undesirable activities" more precisely -- if we need to. >>CHUCK GOMES: By the way, I would encourage staff to speak up, too, because you're the ones that are going to get tasked with this. So if you've got any input, please speak up. >>LIZ GASSTER: Just to say I really reinforce Avri's comments about the framing of the studies and the professionalism with which we would go about this. We recognize we're not experts in the design of these. So we would, in almost every case, except perhaps in a couple of the cases that, Chuck, you identified that didn't really seem to require, you know, a design, we would be bringing experts in to help us do that. >>AVRI DORIA: Philip. >>PHILIP SHEPPARD: Obvious question about definition of terms. It is not specific to this set of studies or any other issue. It's a very general issue. It is trivially easy to solve. When you have a study, you find somebody who you think is capable of doing the study. Those are the people who define the terminology for you. When you read the first draft, you either say, yeah, they got it right or you make a comment to them. It's dead easy. We can move on, I think. >>TIM RUIZ: I don't think agree -- >>PHILIP SHEPPARD: But you don't want the bloody studies, do you? >>TIM RUIZ: No, I don't want the bloody studies, but that's got nothing to do with it. The point is is that what they may define or something as important to this group as what do you consider abuse of WHOIS I don't think is something you can leave to some third-party expert who has not been involved in this process for the last seven years. That's the point I'm trying to make. In fact, each of these studies, every proposer of one of these studies may have had a different idea in mind of what abuse of WHOIS is, or what does it mean by "undesirable activities." What the heck does that mean? I don't think that a third party can read the minds of the individuals who wrote these -- proposed these reports and decide that. All I'm saying is, does it make sense for us, as a group, to have some more defined -- specific definition of those things so that when we hand this off to a third party, they can actually make a reasonable feasibility study of what we're asking them to do. That's all I'm proposing. And that, actually, will save us time from getting back a report that says you know, "You need to fine this, you need to tell me what you mean by that." And I think that's exactly what's going to happen. >>AVRI DORIA: I think we end up with two alternatives. One, we can certainly do it, and I recommend that anybody that believes that there are terms that need definition, that they propose definitions. And then we discuss them and we give them. But the other part is, I think anyone who's a professional social scientist who's going to be looking at doing these studies is going to read the massive amount of information and writings and output we've put out on all of these things over the years on various commentary that's been, and that they'll see the multiple definitions of "undesirable activities," of "abuse," because we've been talking about it and GAC's been talking about it, and it's been written about often. So I have a feeling that a social scientist who gets this as a problem will have a research to do as part of their bid to sort of look at these things and put that together. And then, as Philip says, they would come back to us with a first draft of a study proposal where they would have defined it. We would look at it and say, "huh-uh" or we would say, "Yes, it makes sense." >>ZAHID JAMIL: Zahid Jamil. I sort of propose something. Since we're trying to figure out what this means, if someone's concerned about definitions, I would first of all say I completely agree with you, Avri, that's one way to go, a social scientist would be able to understand, many people would probably be able to understand simply by the words "abuse and desirable." But, nonetheless, as a compromise, what we can do is go around and give either today or within the next 24 hours, have people come up with their views as to what it was, attach that as an appendix to whatever proposal it is, and say it includes but is not limited to these sort of things. And send it off as part of the proposal. So whoever is going to write this can actually use them as a guideline. >>AVRI DORIA: It's an interesting thing. One of the things I would suggest is that we would first actually point out as we were going through, what are all the terms that we would want to collect. I don't think we would then need to do it over the next day or so, because we're not moving quite that quickly. But, basically, we've now got two terms that we're saying we would collect the various opinions of in an appendix on the inter alia includes, but not limited appendix, and I don't know what people think about that. It's certainly a way to go about it. I think it's a fairly decent one. And then we can -- So what we would do today if we were going to follow that is, we would underline or collect the terms that, you know, we would want to do that for. And then we'd have to figure out, okay, "misuse" is probably one term. "Undesirable activity" is another. Is "commercial purpose" a term? I don't know how many of those terms would need definition. But that could be an interesting list to build up. What do other people think? (inaudible) I suggest that if we're going to do that, that, you know, we start to collect the terms. And at the moment, I think we've got at least two, "misuse" and "undesirable activities." Does that make sense, Tim? >>TIM RUIZ: Yeah, I think that's much better, yes. >>AVRI DORIA: Thank you. >>CHUCK GOMES: Okay. For formal studies, the next one was Study 15 that we assigned, we thought, a medium priority. And we're not sure how feasible that is. That doesn't mean it's not feasible, but just that we had some doubts there. How would you do that? Is it really possible to do that? And maybe some people have some thoughts there. >>AVRI DORIA: Okay. I'll ask a question, which is, what we're saying here is that we would go and ask the people -- I mean -- how do you determine that, I guess, is -- it's the same question I have. Do you go to everyone who's doing "undesirable activities" and ask them if they're using port-43? Yes, Paul, and then Claudio. >>PAUL STAHURA: Well, one way you could do it is ask all the registrars to turn off their speed bumps and their protections so that their WHOIS doesn't get mined via port-43. And then see if the spamming and abuse increases compared to when it's turned on. >>AVRI DORIA: Okay. Thank you. Claudio. >>CLAUDIO DE GANGI: And I think when the study was submitted, there was a suggestion how it could be completed. But I think one way would be to look at maybe TLDs that don't have port-43 access and make a comparison that way. To see if it makes a difference as far as spam. >>CHUCK GOMES: And that certainly is an alternative. But if you're not dealing with a comparable situation it might not mean too much. But it's debatable. >>TIM RUIZ: All I would do is leave this with arguable data. I thought that's what we were trying to get away from. There's so many other reasons or possible influences and effects that you can't assume or make the assumption that it's related to port-43. >>CHUCK GOMES: And I'm not sure we need to resolve the feasibility issue right now. In fact, if we try to do that on each one of these, we will take months. But the point is we had some doubts there. Maybe we're wrong, maybe there is a good way. And that may have to be pursued further. >>AVRI DORIA: Jeff. >>JEFF NEUMAN: I just wanted -- This is Jeff Neuman. Just on the comment of -- I'm not -- Which TLDs don't have port-43 access? I thought that was required. And certainly all the unsponsoreds do. And those are probably the ones that are at this point most susceptible to spam and abuses. So that would be kind of tough to do. >>CLAUDIO DE GANGI: I meant ccTLDs. >>JEFF NEUMAN: Yeah, you're kind of comparing apples and oranges there. >>CHUCK GOMES: Yeah, that's what I was thinking on that, is that -- I thought you were talking about ccTLDs. And many of them have very different policies with regard to WHOIS and so forth. So -- But, again, still something to be kept -- >>PAUL STAHURA: It occurred to me another way to do it is to have the registrars output differing information via their Web site than port-43, and then see, you know, which of those e-mail addresses get more spam. It's -- it seems like this hypothesis is trying to get to the difference between port-43 and Web-based. Web-based, it's easier to do, like, CAPTCHA codes and so on to prevent it being mined. Whereas, on port-43, it's hard to do that kind of CAPTCHA code thing. So I think that's what this Study 15 is trying to get to. Port-43 is -- it's harder to protect it against mining. >>AVRI DORIA: Okay. Thank you. >>CHUCK GOMES: All right. Going on, then, to study 2 is another one that we thought would probably require a formal study to do. We gave that a low priority. That's our opinion. I'm just curious what other people think. Okay. The next one was -- >>AVRI DORIA: By the way, no comments now doesn't mean anything other than no comments right now. >>CHUCK GOMES: Working group document, areas 4 and 5, Study B, this is another one that we suggested combining. Let me scroll down so that you can see more of that. All right. So we suggested -- this is one that we suggested combining some studies. And you can see study 17, Study 1, Study 13, GAC Study 11. And this one, we thought, was -- in our first consideration of this, we thought Study B would be a top priority. >>AVRI DORIA: Okay. Mike. >>MIKE RODENBAUGH: So I think our constituency's of the view that study 3 and 20, which basically are the same studies, should also be combined into this Study B. I'm wondering if you could explain a little bit why you left those studies different. I know you had said something to do with contractual compliance rather than study-worthy. But I'd just like to you explain that a little bit. >>CHUCK GOMES: And I'll have to go back to -- hold on. Let me go down to where I have -- it was 3 and 20 in the last grouping? Do you recall? Well, let's find them. Because I obviously haven't memorized these. There's 20. And by the way, Mike, we are not opposed to different combination. So I'm not necessarily saying -- I don't know I have a quick response on 20, why we didn't decide to combine it there. You can see, though, that we do have the note there that we had some doubts about feasibility of that, how would you do it, is it really doable. And maybe our doubts are unwarranted, I don't know, but that certainly needs to be considered. Yeah. Now, what was the other one? 20 and -- 3? >>MIKE RODENBAUGH: 3. >>CHUCK GOMES: Let me find 3 real quick here. Oh, that's area 3. Sorry about that. 3 may have been one of the ones we thought was compliance. Study 3. No, we didn't -- That one seems quite a bit different than those other ones that we combined, but I'm not sure it's out of the question. >>MIKE RODENBAUGH: That's GAC study 3 you are looking at. >>CHUCK GOMES: I'm sorry. Thank you. Keep me straight, please. Appreciate that. So, okay. Okay. So this is one we did think was a compliance issue. Now, whether we, as a council, believe that compliance will be able to deal with this is another issue. Assuming -- Our view on this one is that it is a compliance issue and should be handled there. Now, that's ideal; okay? And maybe some people disagree with that. So we shouldn't have to do a study if this is a compliance issue. At the same time, I don't know if we would necessarily be opposed if people wanted to do a study on this. Our thinking is we shouldn't need to do it. It should be handled by the compliance department. So I think that's open for discussion. I'm not hard line on that, or I don't think any of the registries are. Okay in so we can talk about that. G question. >>AVRI DORIA: Any other questions, comments at this point? So that was group B; right? If you can go back to group B. I think you are almost there. Study B. I think part of the problem with your report, one thing is areas, study -- overloaded terms. >>CHUCK GOMES: We were trying to keep the connection all the way back to -- >>AVRI DORIA: Right. >>CHUCK GOMES: -- the WHOIS working group for that reference, but you understand, you can see I have my challenges there, too. So I certainly empathize with that. Anything else on the study B? Okay. And -- So hold on a second. Was that all study -- that was all study B, wasn't it? >>AVRI DORIA: That was study B. >>CHUCK GOMES: Okay. Area 4 -- from area 4, studies 18 and 19 and GAC studies 9 and 10, this one we thought it was probably a top priority. You can see the hypotheses there and whether or not we should combine them. Any comments there? >>AVRI DORIA: Mike. >>MIKE RODENBAUGH: Yeah, I'm not really sure this is feasible, either, given -- we definitely have to take it -- try it, defining what commercial purposes are, natural person versus legal person and that sort of thing. I don't know, this seems very problematic. I'm sorry curious why you identify it as a top priority. >>CHUCK GOMES: Why we identify it as a top priority? Not jumping out at me right away, as you can tell. Well, the -- I mean, it seems to me that if we were able to substantiate or refute some of these hypotheses that that would be very useful information in terms of policy work going forward for WHOIS. I mean, if we had the answers to some of these questions verified, one way or another, my opinion, it's very helpful information. Why would you think it's not? >>MIKE RODENBAUGH: I guess I just come at it because obviously I view proxy services as more of a problem than benefit, and it doesn't really matter to me who is causing the problem, whether they are legal or natural, and whether they are commercial or noncommercial. >>CHUCK GOMES: But with regard to privacy interests, the distinction between commercial and private individual is huge. >>AVRI DORIA: And there is an untested version -- I mean, there could be many, many natural people who are using these proxy services, and you just don't see them because they are -- I mean -- so collecting the information here could be useful to sort of answer that question of who is using it and how and why. And you're right, and I did add two more terms, the "commercial purposes" and the "natural persons," because I know commercial purposes ranges all the way from me putting my resumé on my private Web site because I am looking for a job to somebody actually selling services to somebody selling illegal services. So there's a very wide definition what have that might include. Yes, Tim. >>TIM RUIZ: I think some of it is dependent, too, on -- maybe the way to put it is, another term for the list is proxy or privacy service. Because they are all over the map. And we tend to focus that in a certain area where it might be kind of obvious. But there are less obvious, basically, proxy/privacy services do the exact same thing as those that are obvious. They are just not as obvious. The other thing to consider is that, in some cases, you know, a proxy or privacy service, by definition, it's automatically a commercial registration; correct? I mean, if they become the registrant of record in which, in many cases, happens, then that is a commercial registration. So then what you are trying to get to is the beneficial user data behind it. And that might raise an issue that I will just put out as far as feasibility is concerned, is that just the very nature of a privacy service may make it difficult to get some of the information that you want to get to, unless ICANN is going to make, like, a blanket decision that this is a -- this falls under 3 point -- 7.3, I think it is. So you have to give up the data. But that's part of the feasibility that has to be done. But again, that's not going to be done by a third party. That's going to have to be done by a decision made by ICANN and staff. >>AVRI DORIA: Philip. >>PHILIP SHEPPARD: Thanks, Avri. It's probably just worth recording what I suspect the origin of these two recommendations from the GAC are, because it relates back to the original WHOIS working group report where we had made specific recommendations that were different, depending if you were a natural person or legal person, or commercial, noncommercial. So I suspect the request from the GAC here was to try to scale that difference, to understand the basis and rationale for that recommendation. >>AVRI DORIA: Thank you. Tony Harris. >>TONY HARRIS: Yes, I'm struggling with the wording of these two, because I have always, I suppose, erroneously assumed that people taking on proxy and privacy services are not normally a commercial firm who is anxious to have their presence and products known on the Internet. But obviously I must be confused. And as far as -- that's study 18. Study 19, the wording here I think indicates when it says that the use of proxy services is directed toward registrations made by natural persons, I mean, revealing the identity, well, when you have cyber crime proceedings of any sort or you are chasing somebody, normally they would be probably natural persons and not a registered company with an address, and you could easily go and ring their doorbell. So that's -- It's really logical, and why do you have to study that? I mean, that's.... It's by default. >>CHUCK GOMES: And Tony, that was one of the general statements in our document that we said. You know, if there's general agreement in the GNSO about some particular thing, let's not study it. So we could decide, we all agree with that, we don't think we need to study it, that would certainly be a reasonable outcome. But if, in fact, there are differing opinions on that, there's where the study becomes valuable because we get maybe a chance to see what the answer is from an independent set of data. >>AVRI DORIA: Anyone else wish to comment on this group? Yes, Mike. >>MIKE RODENBAUGH: Just to put a little more color on Tony's excellent point there, I mean, generally the people -- when people are trying to get WHOIS revealed through -- from a proxy service, it is when they are fighting crime. And normally, it's going to be either stolen credit card and, therefore, an individual, or potentially a business credit card but mainly it's individuals. Or it's actually individuals who don't know that their Web site's been hacked and it's being used for crime. So I agree, there's really not much need to study this issue. I don't, frankly, see the point. >>AVRI DORIA: Any other comments on that group? >>CHUCK GOMES: The next one we recommended for a formal study was study 6, and we thought that was a top priority. There is statistically significant correlation between more restrictive ccTLD WHOIS policies and levels of cyber crime in a domain. But we're really not sure whether that's feasible to do. >>AVRI DORIA: Okay. Paul. >>PAUL STAHURA: Even if that was true, which I think it might be, that doesn't mean that the WHOIS policy had anything to do with the correlation. It could just be that dot com has opened WHOIS or, you know, has WHOIS compared to some ccTLD in dot com that attracts more cyber criminals than dot FR, let's say. So I think studying that -- the way it's worded in this hypothesis doesn't make sense to make a study out of this. >>CHUCK GOMES: Paul, that's what we were getting at in our general statement at the beginning of our recommendations that if, in fact, the study is not going to produce anything that will be useful for future purposes, then we shouldn't do it. And in that case our priority here would be way off, and I think that's something we need to decide before we move forward and spend a much of money on a study. >>AVRI DORIA: Any other comments? Okay. Next. >>CHUCK GOMES: Okay. Area 6 was another one. We gave it a medium priority. It was study 20. This was one brought up earlier in our discussion. Again, we're not sure about feasibility on this, but that would have to be evaluated. >>AVRI DORIA: Yes, Paul. >>PAUL STAHURA: I would add relay requests -- relay information requests to the list of definitions. Because, you know, there's telephone requests, there's e-mail requests, there's fax requests. So I would add that to the list of what does that exactly mean. >>AVRI DORIA: So you are saying that you think that the response differs in the method of request using -- being used. >>PAUL STAHURA: Yes. Like I would guess -- >>AVRI DORIA: Yes. I was just looking for clarification. I wasn't arguing a point. >>TIM RUIZ: So this is the one where B actually is the same as the one that the registry's felt was not -- was actually policy issue. So if we were going to leave this grouping, we might consider taking only A here, and then B would be -- the B part would be grouped with the other one that's on the exact same thing. >>CHUCK GOMES: I'm not sure, are you talking about the combination study A and combination -- oh, okay. >>AVRI DORIA: You have A and B there, and B you equate with study suggestion 3, which was the one that -- >>CHUCK GOMES: Exactly. >>AVRI DORIA: -- you marked as compliance. >>CHUCK GOMES: And -- >>TIM RUIZ: So just the B part would be consolidated with 3, not the whole -- >>CHUCK GOMES: Could be, yeah. >>TIM RUIZ: Got it. >>CHUCK GOMES: Anything else on 20? Study 12 we gave a low priority to. Again, we questioned feasibility on this one. >>AVRI DORIA: Comments? Yes, Paul. >>PAUL STAHURA: Sorry, but falsify WHOIS information. I know that if you asked both sides of this issue what that meant, there would be a bunch of different opinions on that, too. Because, you know, if -- This question I think is getting to there might be some registrants who have their right street address but have a wrong e-mail address because they don't want to get Spam, if they think they are going to get Spam because of the WHOIS. So they might falsify their WHOIS information with just the e-mail address where everything else is correct. So is that really falsified information? Or are you talking about all of the information being false? And false is a question too, because the WHOIS information could be self-referentially correct, like Dayton is in Ohio and has a certain ZIP Code, so that's true. Dayton is in Ohio and has a certain ZIP Code, but that information is not for the person who registered the name. So in my opinion, that would be false information. But also, if Dayton had a different ZIP Code, that would be false, too. So I think it needs to be more defined what false WHOIS means. >>AVRI DORIA: Thank you. Tony Harris. >>TONY HARRIS: I would just like to say in my opinion, this whole concept is really saying let's put accuracy in the goodwill of the registrant and not worry about it. We'll give him a disclaimer saying your data won't be published or available and be a good boy, and I mean, why would he -- why would he change his behavior? And isn't that really getting away from the fact that we should be making sure that the data should be accurate? I mean, it's just leaving it up to the registrant. >>AVRI DORIA: Thank you. Okay. Moving on. >>CHUCK GOMES: Okay. Study 4. We gave it a medium priority. Any comments on that? Okay. Study 7, which is what we combined -- that's our study C, combining GAC studies 5 and 6, and again, this is the one where the hypothesis working group actually thought that these two could be combined as well. Again, the registries question whether the studies are feasible. Any comments? >>AVRI DORIA: Any comment? Okay. >>CHUCK GOMES: I think probably that's a good breaking point because we're just about out of time here on this part of our agenda, and we have another very important part of the agenda coming up. I encourage you to look through the rest of the report. Whether this is useful for the Council or not, we can decide as a council and decide how to move forward. >>AVRI DORIA: A couple things as closing it. So first of all, we will have another discussion on where we go with the WHOIS studies at our open meeting to get community input on it, and I guess you will be going through some part of this presentation again -- or you will be presenting something. I'm not sure what. I collected six terms, and if we are going to follow the suggestion for collecting people's definitions, which I think is an interesting exercise, I'm curious to see how it works out and curious to see how the editors of the document would cut it in, but the terms I had were misuse, undesirable activities, commercial purposes, natural person, proxy or privacy service, and falsify information. Yes, did I miss one? >>PAUL STAHURA: Fraudulent domain name registration. >>AVRI DORIA: Fraudulent domain name registration. >>PAUL STAHURA: If you scroll up a little bit, it was on the last one. >>AVRI DORIA: Fraudulent domain name registration. Okay. And I saw a couple hands while I was writing. I saw Tim and -- yes. Okay, I'm not sure who had their hand up first, Eric or Tim, because I didn't see either, but.... Okay. Go ahead, Tim, and then Eric. You had another term to add to the list? >>TIM RUIZ: Yeah. If I could find it. >>AVRI DORIA: Eric, are you ready with yours while Tim is looking for his? >>ERIC BRUNNER WILLIAMS: Sure. This is Eric Brunner-Williams from CORE. During the Fast Flux working group starting period, we had several weeks of struggling with definitional issues, what actually constituted Fast Flux. So my first comment to you, Avri, is that the immediate collection of terms that may appear to be poorly defined or in need of some examination is unlikely to be exhaustive. Or at least our experience with the Fast Flux working group is we found the substantive issues -- the substantive definitional issues took us quite a bit of time to actually determine from what we thought was our original presenting problem. So I suggest you won't immediately find things like ease of attack through automated means as being one of the undefined terms. You won't find business models of the attackers as one of the undefined terms. People aren't going to give you these kinds of things, which turns out actually are the fundamental questions to be asking. Thank you. >>AVRI DORIA: Thank you. Tim. >>TIM RUIZ: Yeah, the other one was adequate corrective measures as it relates to taking action when there's fraudulent WHOIS. It's interesting, we have been going through some work with ICANN in regards to the WHOIS data protection policy and how that's been working, and interesting that ICANN can't even tell us clearly what the corrective measures are that registrars are supposed to have taken. So I think that -- And another example where I think some work needs to be done before this ends up in a third party's hands, because if I can't tell this right now, a third party isn't going to be able to figure it out. >>AVRI DORIA: No, I think the suggestion of collecting the definition among others, definition includes but is not limited to collection of views on this, and then perhaps asking Liz or someone to sort of collate it all without providing a consensus definition, you know, because the next step would be -- and that would be a hard step -- would be to move from the collection of includes but is not limited to, then that magic step to a working definition, and then one never gets beyond a working definition to an actual definition. But perhaps the scientist doing the study proposals can take our collection of "including but not limited to" definitions and produce a proposed work definition that would be definition used for the purposes of the study. And then, still, it wouldn't have the approval of being absolute truth but it would be when the study says X, this is what is meant by it. As I say, a working definition. Okay. Thanks. As I say, we will come back to this, our favorite subject, many more times before it goes away. I think it will go away as soon as the world figures out how to balance privacy and law and order. And when the world figures out how to do that balance, we will solve out WHOIS. Now the next thing is do we have the folks to talk about new gTLDs? Is there break first? We have a coffee break? >> Yeah, 3:00. >>AVRI DORIA: Why would we do that? Do we have coffee for a coffee break? We have coffee for a coffee -- I don't see coffee. >> Yeah, there is back there. >>AVRI DORIA: There is? Sorry. I never thought that we actually scheduled a break, even though I must have done it. Okay. Break. (Break). [ Laughter ] [ Applause ] >>AVRI DORIA: Okay. Welcome back. Coffee break is over. And we now come to the next part of the meeting. And this one starts today and continues through to tomorrow morning with a bit of a break in the middle. Basically, Kurt will be sort of leading this session. I'll be helping him and keeping queues as we talk. Is your hand up? >>ADRIAN KINDERIS: No. >>AVRI DORIA: Okay. I thought you were going -- I see. 1 through 2 November. So Kurt is going to go through it. He'll tell us how. I'll basically be keeping a queue list for him, as necessary. But, Kurt. Thank you for coming. >>KURT PRITZ: Thank you, Avri. Hi, everybody. >> Hi. >>KURT PRITZ: We have talked many times about this. And the last, formally, was in Paris, where many of you attended. We've had a few brief updates during GNSO Council calls since then. And so this is a continuation of an ongoing dialogue between the council and the staff regarding the implementation of the consensus policy that was approved by the board for the introduction of new gTLDs. So there's some new material here, but not -- I don't think many surprises. So the way we're going to organize this, if it's all right with everybody here, is, I'm going to give a brief review of the posting of the applicant guidebook. So you see that we've sort of changed the name from "RFP" to "applicant guidebook," which we think is more descriptive of what it's about. It's intended to help the applicant understand how to apply for a gTLD and help the applicant make a strong application for one that will, in the end, be accepted and result in the delegation of a new TLD into the root zone. So we'll briefly review the material that's been posted, sort of an overview for it. And then what we'll do, as those of you who have read it notice that it's organized by modules, so there's six modules there, five with substance. So we'll do a module-by-module review of the applicant guidebook. If possible, then, what we'll do is take questions and provide the best answers we can. At the end of each module. And in that way, we can work through the whole sum of the material in four and a half hours we'll have together two, and a half today and two tomorrow. So as a target, we'll try for getting through modules 1 through 3 today. Then we can have -- tomorrow morning, we can start with a brief review of what was said for those who were unable to make it today and then continue on with modules 4 and 5. And then what we'll do is take questions after each module. And because so many staff members worked on this and worked with the council on this and that there's so many different subject matter experts in each of the subjects addressed by the applicant guidebook, there's a number of staff members here that I will just deflect the questions aimed towards me over to them. And so we're all here to provide the very best information we can for you. So I'll do my best to take us through a little stroll through the Web site in a minute. And, Karla, when do you that, go through the Web site, help me, if I forget to point something out that's meaningful on the Web site, poke me on the shoulder or something like that. So what was published was the applicant guidebook itself, the six modules that apply, essentially, the directions for how to fill out an application for a new gTLD. Because they're merely directions for that, we thought some explanatory information that would annotate that guidebook would be important, either to explain the thinking that went on behind it, or some of the development work. And so leading up to the publication of the guidebook and then coincident with it, and then I think in one case after it, we published a series of memoranda that provide more description for what's going on. And then there's some other supporting information there that you will see. And there's what we think is a nice comment section. So the modules are organized, and then the supporting memoranda are next to them where they're linked up. And then there's comment for associated with each -- with each module, with each subject matter, and then some additional background material. So that's what we published last week. So this is sort of the disclaimer. But the applicant guidebook, the RFP is clearly an initial draft. So it's for public comment. And we expect it to be different when it's a final draft. As you have read, there's many facets to this thing. It's pretty darned deep. And -- so we're expecting a lot of community discussion, a lot of input into the implementation model, and a revision is expected. So to those of you who are going to participate in this process, too, this isn't the definitive version, so there's going to be revised versions after this. If there's substantial revision, it's foreseen that there will be another draft version of the guidebook before the final version is posted. So it kind of depends on the comment and the amount of change, the timing of the process going forward. And we'll talk about that, too. So this is -- I think this is for all of us. Those of you that developed the policy recommendations over near a two-year period, and then as we work together in creating this implementation model, these are, you know, what we think the program themes are for launching new TLDs. So when you go out there and help -- you know, as part of ICANN, help explain this, these are some of the themes that we think are important. Certainly the competition and choice aspects. And when the council first considered the very first term of reference as to whether or not there should be new gTLDs and answered that in the affirmative, that there would be -- the benefits and opportunities provided by new TLDs would clearly outweigh the risks. So that's one of the messages. But then, in doing the implementation, we have principles of conservatism. Especially in this first round, we want to ensure the ongoing stability and security of the domain name system, of course. That's our paramount mission. And also, we want to be conservative in other items. So we want to be fiscally conservative in launching a process where there's expected to be hundreds of applications. And we want to be conservative in how names are introduced into the root zone, because we -- even though there's been learning in the past, we know that there's going to be a lot of learning in this round, and there will be adjustments to the process thereafter. And we think, as you have told staff, we think the process is going to become more liberal as it graduates from rounds to perhaps someday being just an open process for applying for new TLDs. There's certainly a big emphasis on registrant protection. And we'll talk more about that as it's embedded in the modules themselves and in the criteria by which the applications are measured. For those on the outside that want to know how we went about this and how we decided to launch new TLDs, as you know, as part of ICANN's founding documents, and described at length in there. But then there was the nearly two-year consultation process that you guys had, the policy development process, where all the other constituencies were consulted. And there were many public sessions and comment taken about how best to develop policy for doing this. And since then, there's been sessions about the implementation. So I've kind of already made the last point about there being -- there being an extended public comment section. So, Avri, you know, Mike has a question. What I'd really like to do is get through a bunch of material and then take questions. But if you have, like, a point of order or something like that, then -- >>MIKE RODENBAUGH: (inaudible). >>KURT PRITZ: Is that all right with you? All right. Don't forget it. >>AVRI DORIA: I'll write you down on the list. >>KURT PRITZ: We're cutting off the queue at three. >>AVRI DORIA: You mean, Mike gets to talk three times. >>KURT PRITZ: So what do we think another important aspect of the process that we think is, on one hand, uncomplicated, but on the other hand, robust when it needs to be. Most of the applications, we think, will go through a straightforward process, what we've all in the room come to know as the initial evaluation, where there's -- where the string is evaluated -- well, we'll talk more about the details later, but the string is evaluated for security and stability concerns and confusing similarity, and then the applicant is evaluated for the technical and financial wherewithal to operate a registry. So we think, in the majority of situations, the process isn't simple, but it's uncomplicated. But then, you know, it's got built into it robustness when it needs to be. And that follows the policy recommendations that provide for the protection of, you know, what you all thought and we think, too, are important interests. And the grounds for objection to a TLD or methods by which we resolve contending strings if there's applications for identical or very, very, very similar strings. So that would be a good mantra for us to have, too, I think as we go forward that this process is uncomplicated, but it's robust when it needs to be. So how did we organize the application guidebook? Well, we -- it was organized by the evaluation process itself. And the evaluation process was organized into what we call modules and I think this is fairly self-explanatory, but as the preceding slide stated, all the applications go through initial evaluation. And then most, we think, will go through transition to delegation, in certain instances, an extended evaluation will be required. Or there will be a formal objection to the process, or there will be contention over strings. And then the transition. So the modules are -- well, I can't say it again, so I'll go on to the next slide. Oh, I guess I will say it again. So this is essentially the titles of each of the modules. I'll just augment this slide with saying that the questions to the applicant and the scoring criteria for those questions and how the applications are evaluated is embedded as an attachment to Module 2, the evaluation procedures. And the base agreement and the associated specifications, formally known as appendices, are appended to Module 5. And then also published along with this, like I said, were explanatory memoranda. And they attach to various modules, as, you know, I could have done that for this slide, but we'll see that later. So -- are we on the phone? Avri. >>AVRI DORIA: Yes, we are on the phone, but we don't have any people on the phone. >>KURT PRITZ: Well, for those of you listening to the recording, the explanatory memoranda are -- address protection of rights issues, cost considerations for new gTLDs, how the project and process, evaluation process, were costed out, because the fees are based on a cost recovery mechanism. There's an explanatory memoranda regarding geographical names and handling the application for certain geographical names. There's an update on DNS stability criteria. So ICANN published a memo regarding the criteria that each string must comply with. Earlier, that update has been made to include IDNs. So it's fairly complete. And it's different than the first. So it's a criteria each string must follow that's fairly straightforward. There's a deep -- There's a deep memorandum on string contention and the entire life cycle of string contention, how strings that are identical or confusingly similar get identified early in the process and how they're handled through the whole process. So it really touches on a number of modules. There's a recently published memoranda on morality and public order objections, and touches on what the standards for resolving those objections and disputes might be. And, finally, there's a document that outlines the changes to the base agreement. Here are some links. We're going to look at them later. But besides the memoranda, there's a cross-reference, a mapping of the GNSO policy recommendations to the applicant guidebook. There's the CRAI report on the separation of registries and registrars that has been read, I think, with interest. There's a preproduction version of the algorithm that assigns scores to two strings to measure how similar they are, how visually similar they are to one another. And we'll talk more about that in a second. These links are different colors, because I clicked on one earlier, not the other. And finally, there's an interactive process flow where, you know, you wave your your arrow over the various steps in the application process and it provides -- it provides detail or you can drill down into the process to get more detail. So to a really short extent, we will take a look at that later. This slide came out looking better in this room than I thought it would be, but it's a map that is also posted along with all this other material of how this applicant guidebook is organized and the associated material. So you can see there's the modules posted there. They are in a couple different places on the ICANN Web site. And then each of the explanatory memoranda are linked to it in this map, and they are also linked to it on the Web site, as you have probably already seen. And then the other resources I described are down in the lower right, in that somewhat unattractive green color. And then to the left are the attachments to the module, so you see the applicant questions and evaluation criteria are attached to Module 2 and the agreements attached to Module 5. So if you are looking for a certain subject matter, this map will be useful, and you will be able to identify this map clearly when you take a look at the Web site. So if I were going to try to do this (indicating), this is the -- this is if you go to the ICANN front page and click on this, you will come to this Web site. Where should I go from here, Karla? I think I want to go to comments and show them. So the comment fora is organized sort of along the lines of the process map that you saw. So you see -- And if I click on this green arrow, will this help me? Dan, if I click on this green arrow, is it going to help me? Or I'll just scooch over here a little bit like this. So over here (indicating) there's each module. Module will I module that's organized and you will see there will be a table of contents of each module on the gutter on the left if I click on that, but I'm not gonna. And associated with each one are the explanatory memoranda. So for example, on Module 1, there is a description of the fees associated with the application, the evaluation processing fee. And this memo describes how the process, the evaluation process, was costed, and provides detail behind that. So in Module 1, you will just see a number. In the explanatory memorandum, you will see how that number was developed. And then finally, here is the comment forum and here is where you can see the comments. So you can either comment on the whole thing up there in the top line or comment module by module. Adrian, I know it's hard for you to resist. >>ADRIAN KINDERIS: But -- >>KURT PRITZ: Is this a good question for holding to the end or is this sort of a point-of-order thing? >>ADRIAN KINDERIS: It's a point-of-order thing, if I can. I'm sorry to annoy you, Kurt. Kurt, are you going to be doing this presentation to the public throughout this week? >>KURT PRITZ: So, what you are seeing here contains most of what you will see on Monday when you come to the "Understanding the Draft RFP Guidebook," and you will listen with interest on Monday even though you have seen this material. >>ADRIAN KINDERIS: With all due respect, and I really appreciate the work you are doing -- this may be out of line here, but this is a GNSO council session, and I am conscious of the time we are using to do this background information which I'm fairly sure everybody at this table knows. Is there a chance we can speed through this and if anyone wants to go back to understand this, they can attend one of the public sessions? And then we can get into some of the nitty-gritty of the council discussion. >>KURT PRITZ: Actually, I thought we would have more time here. Then the public session is really kind of abbreviated, it's an hour and 30 minutes or something like that. But I am almost done with this part of it. >>ADRIAN KINDERIS: I have made my point. Thanks. >>KURT PRITZ: I think the only other place I want to go is to the -- Yeah that, would be good. So there's an interactive process flow that -- at least you don't see a picture of my boat anymore. So if you scan over these, you can get descriptions of each one or drill down into different parts, and it will show the process flow associated with each part of the process flow diagram that you have seen so often before. And then finally, I want to go to -- well, this is the bottom of the page. So here you see the full applicant guidebook, the explanatory memoranda again, and these additional resources. So that's essentially everything in the.... Things got dark, huh? I think it will come back over time. So in accordance with Adrian's wishes, that's the end of the introductory material. And now I am going to take up Module 1, which is really an introduction and overview of the entire process. >>AVRI DORIA: Hold on. Since you stopped there, is it a good point to let Mike get this question in? >>KURT PRITZ: Go ahead, Mike. >>MIKE RODENBAUGH: They are very brief questions. One of your slides mentioned one round of public comment. Wouldn't there actually be two rounds of public comment after the next iteration of the guidebook is published? >>KURT PRITZ: Yeah, there will. >>MIKE RODENBAUGH: Thanks. And then -- >>KURT PRITZ: Only if the praise is so effusive and the review so glowing that we think we can go right to the final RFP or guidebook, then we'll do that, but if there is another draft version, there will be certainly be another comment period. >>MIKE RODENBAUGH: Thanks. It would be very, very helpful, and I appreciate it might be difficult, that this is 97 pages. It would be really nice to have an executive summary that is four or five pages or so. >>KURT PRITZ: Okay. I think that's underway. >>AVRI DORIA: Chuck. >>CHUCK GOMES: A follow-up question to what Mike asked there. If there is a second public comment period, wouldn't that significantly affect the schedule? >>KURT PRITZ: We'll go over the schedule later, but the answer is that if there's a second version of the draft RFP, that would probably be posted in February, and the final RFP would probably be posted in May. And then if there's -- So -- And then there's a couple of ways of going about it. If there's a four-month communications period of that, then applications wouldn't be accepted until sometime in the third quarter. But if there's a way to either accelerate that or go straight to a final RFP, then we can accept applications in the second quarter. >>CHUCK GOMES: But it would be later in the third quarter than under the first scenario. >>KURT PRITZ: That's right. >>AVRI DORIA: Eric. >>ERIC BRUNNER WILLIAMS: Thank, Avri. It's Eric from CORE. Kurt, I just want to have a sense of how much time every revision will cost, not what the end of the process is, but what's the incremental cost. What's the delta that an additional revision costs? Is it one month? Three months? What is it? >>KURT PRITZ: Of the RFP? Probably it will be posted for a 30-day comment period, and then, you know, there's 35 or 45 days, probably, to revise -- you know, write the next version of the revision. So just sitting here, I think it would be two- to three-month cost. When you said "cost," I thought you were talking about dollars and I didn't know how to figure that out, but then I remembered time is money. >>AVRI DORIA: Okay. I guess it's fine to move on to Module 1 seeing -- Oh, okay. Please. >>ZAHID JAMIL: Two quick questions. Maybe this is too soon because this is a draft stage. One is, I am looking at the period of objections and it says that the period of objections will be as long as the evaluation period. Both of them are undefined. So it raises the question as to what is the period in which you can file an objection? Is it one month? Two months? Three months? I don't see it defined. Maybe I missed it. That would be one question. And would there be set criteria for that? Are we thinking about something? Second, I just used the algorithm. >>AVRI DORIA: I actually think these questions are from the actual modules and not on the introduction. So if you could -- >>ZAHID JAMIL: I'll wait. >>AVRI DORIA: Yes, Jeff, yes, Paul. >>JEFF NEUMAN: This is Jeff Neuman. If the process is extended does that raise the historical cost of the application? Laugh. >>JEFF NEUMAN: I mean, was that baked in, I guess is the question. >>KURT PRITZ: That was essentially baked in. >>AVRI DORIA: And Paul. >>PAUL STAHURA: Speaking of schedule, the four-month period, when will we know when that four-month period starts? Is it going to be now? February? May? When will we know when the four months starts? >>KURT PRITZ: Well, it will clearly be announced. So the default position is it starts with a publication of the final applicant guidebook publication, at that moment, unless we sort of decide that there's a path for starting that period sooner than that. >>AVRI DORIA: Yeah. >>RON ANDRUFF: Ron Andruff. Just a question, Kurt. What is the logic to go in sequence rather than parallel? Because in terms of the announcement, as you put up the new schedule now, the new time line, you see that there is a kind of noise happening already on a global scale. So why isn't ICANN putting out a big notice right now instead of running this in serial? >>KURT PRITZ: Actually, the communications period has started. So there are two kinds of communication periods, I think. One was embedded in the documentation associated with the policy recommendations. I think it was an implementation guideline, wasn't it? That there be a four-month period of time from the application of the final -- is that me being so annoying? -- that starts with the posting of the draft RFP. So that's that. But in the meantime we are certainly communicating globally about it. Letters were sent from Paul Twomey to ministers at all of the governments in the world describing the effort. E-mails are going out to each of the managers of a TLD that describe it, and that's part of a global communication strategy. So we are communicating right now. It's just that aspect of the policy implementation guideline. >>CHUCK GOMES: Kurt, if I could just follow-up on what you said there. Back to the new gTLD process, the concern on the committee -- and you probably remember this, Ron, when I say it -- was, is that granted, all of us that are an active part of ICANN are aware of this. And they have done some communication, but the concern was to make sure that those that are not normally a part of this have plenty of time to react, and also have time to prepare an RFP if they want. >>AVRI DORIA: Thanks. I have got Mike and then I have got Tony. >>MICHAEL PALAGE: Thank you. >>AVRI DORIA: And I have got Paul. >>MICHAEL PALAGE: Mike Palage. Kurt, as you can appreciate, there are a number of attendees here today that have a vested financial interest in seeing the launch of this new TLD process. I was wondering if you could perhaps answer the question, are there any staff bonuses or performance clauses that are tied to when this process happens? >>KURT PRITZ: We have essentially a management by objective program at ICANN, and so each trimester, where -- each trimester, we each have individual goals to meet. The trimesters run along with the meeting schedules. So each trimester, for example, I have a set of goals to meet. My bonus is based on meeting those goals. So one of my goals this trimester was to publish the draft RFP. And so part of my bonus will be based on that. But the percentage of bonus is preset as a maximum, and then you get a numerical score for how you did on your goals, and then you get a little bit of a bonus. >>MICHAEL PALAGE: So just a follow-up question. To understand what the motivation is, obviously a lot of people here want to see this happen sooner as opposed to later. But there are some in the community that are not as, shall we say, opportunistic towards the launch of new gTLDs. So on behalf of that community, I think it would be important if you could perhaps share with the community what the targets are for bonuses. When do the new TLD process have to start? What is the date? And if, in fact, that date slips, when does staff begin to suffer economically for missing that date? >>KURT PRITZ: I don't know. I think I have given you a lot of information that -- you know, I don't know how to go down the path on this question. >>MICHAEL PALAGE: Thanks. >>AVRI DORIA: Tony Harris and then Paul. >>TONY HARRIS: My question is for Kurt. Kurt -- >>KURT PRITZ: Where are we? >>TONY HARRIS: Over here. >>KURT PRITZ: Tony, how are you doing? >>TONY HARRIS: I don't know if I understood you correctly, but I think you said a few minutes ago, in describing the time line with the second draft for comments, that even so, there might be an opportunity for new applications to be presented in the second quarter. Did I hear rightly or not? >>KURT PRITZ: Yes. So if there's a second draft, we probably won't hit the second quarter date. >>TONY HARRIS: We probably -- I'm sorry? >>KURT PRITZ: Will not hit the second quarter date. >>TONY HARRIS: Then it would definitely go to the third quarter. Thank you, I misunderstood. >>AVRI DORIA: Paul. >>PAUL STAHURA: Back to the four-month thing. I just want to follow-up on Chuck's comments. You know, there's already global outreach to notify the planet that this is coming. At the Paris meeting, there were a lot of press releases. There are press releases now. It seems like every ICANN meeting, it is delayed by another three or four months. I would push forward with starting that four-month clock either now, so that the world has a chance to comment on these RFPs, or at the latest, at the second draft RFP. It doesn't make sense to start that four-month clock when we have the final RFP, because then the world can't change it. Let's start that four-month clock while there is still an opportunity for the people who get notified to actually have input into this process. >>AVRI DORIA: Thank you. I want to remind people to give their names when they start speaking. Werner, you wanted to speak? >>WERNER STAUB: Werner Staub. I would like to add to that that we have a change with respect to the expectation in Paris in the current draft in that there is no clear announcement of the follow-on rounds. And of course this is an interpretation of the signals that the community here has given, and I think we should take into account that most of us who are present here, have been present all these times, are focused on the first round. But we make a big mistake if we just focus on the first round. We are going to overload it unless we offer a follow-on series of rounds. And the first one of those should not be any further than six months away. We have to accept that the rounds will overlap. If they are further than six months away, nobody will believe it. Everybody is going to jump in the first round anyway. >>AVRI DORIA: Thank you. I guess you can move on to Module 1 now. Thank you for taking the questions. >>KURT PRITZ: Great. Karen, where is the -- The next round is described in what module? In Module 1. So the next round -- the fact that there is going to be a succeeding round is mentioned in Module 1, and we'll hone that date more when we post the succeeding modules. Thanks, Werner. So Module 1 is really an overview, and describes the whole process. So you read Module 1 first, and you had a number of questions. And hopefully a lot of those questions were resolved in the ongoing rounds. There's a process ma of the application life cycle. There's a description of the two types of TLD for which you can apply. One is an open TLD, which is essentially what we would call unrestricted, and the other is a community-based top-level domain, which we would call restricted, and we'll talk more about community-based TLDs in succeeding modules, but the selection is made at the time of application. Also list of requirements of all applicants. Requirements specific to IDN applicants and fee information. So this is the exact same slide, but it has got those little boxes with red numbers that indicate which policy recommendation numbers are covered in them. So I thought -- This will be posted probably tomorrow, and I thought it would be useful for people to see which policy recommendations are associated with each module. So policy recommendation Module 1 is intended that the whole process is intended to reflect principles of fairness, transparency and nondiscrimination. It also aspires to be a clear and pre-published process. And to the extent -- the very extent possible using objective and measurable criteria. So one goal is to ensure that all the requirements are in this RFP, so that the applicant has complete and full knowledge of the -- of what faces her or him as they apply. Actually, "it," because it has to be an entity. And then regarding objective and measurable cry tear yeah, we will talk more about this in Module 2, but there is a balancing that goes on between making the process flexible to accommodate different-sized business models, different goals, different sizes, and making the criteria purely objective. So there's that sort of balancing that goes on. And as the recommendation says, this is a round. It pertains to the initial round. And there's some information on follow-up rounds. The adherence to IDN guidelines and protocols is described and what has to be there. Certainly a topic of conversation at this meeting, and the message I would like you all to carry forward, is that the fees are calculated on a cost recovery basis. So the $185,000 U.S. fee is the estimated sum 6 evaluation, development and risk. And Doug and I, and I don't know if Kevin, Kevin Wilson, who spent a lot of time on this can wax eloquently on the methodology used to arrive at this number. We shared some of that thinking in previous discussions we have had where we have shown some of the structure behind some of the costing. And then additional fees are paid usually to outside panels if necessary, so dispute resolution providers are paid directly, extended evaluation panels in certain cases where there's an additional fee would be paid directly so ICANN is sort of out of that game. So that's essentially everything there is to say about Module 1. But there will be more. >>AVRI DORIA: Let me get the list here. I have got Adrian, I have got -- I am getting them down in the order in which I saw them. Tony and Mike and (inaudible). Tony, Mike, (inaudible) Jeff, Eric. >>KURT PRITZ: I just wanted to make one comment. I think this is going to be great. I would encourage you to repeat your comments and questions in the comment fora and not -- while we're taking notes it, would really be good for the record to have the comments you are making here on the comment fora also. I think the comments made here are going to be the most meaningful, so I would appreciate that. >>AVRI DORIA: Remember to give your name. Adrian. >>ADRIAN KINDERIS: Adrian Kinderis. Kurt, where do you see corporate domains fitting in your open and community-based definitions? Corporate domains being -- I will give you one example, a dot IBM for example. Maybe I will answer my question and you can help me out. I kind of see them fitting under two. One, if it's IBM, it has a very defined community and a restricted population, as such. So it could fit under that. Yet someone like eBay who might want to give out a domain name under their TLD to every one of their clients, that's somewhat unrestricted and open. So is it just that it will be just on their basis that a corporate domain can be -- a corporate top-level domain can be allocated? >>KURT PRITZ: So, first, a community-based TLD could still elect to be open if it wants to retain the freedom to be unrestricted. Second -- >>ADRIAN KINDERIS: No, it -- sorry. Just to pick up on that, I'm just reading from the definition here. It says for purposes of this RFP, a community-based TLD is a gTLD that's operated for the benefit of a defined community that's consisting of a restricted population. >>KURT PRITZ: Right. >>ADRIAN KINDERIS: Could you repeat what you said, then? Sorry. >>KURT PRITZ: An open TLD can -- if you elect -- anyone can elect to be an open TLD. >>ADRIAN KINDERIS: Correct. And not representative of a community. >>KURT PRITZ: That's right. So then a community-based TLD would be measured against those exact criteria. And the only time a community-based TLD is really measured against criteria is in the comparative evaluation process. So it would be measured against that as a way to resolve string contention. >>ADRIAN KINDERIS: Thank you. That helps. >>AVRI DORIA: I want to mention if you are actually directing your question to a specific page reference in the document, please mention that as part of your comments so others can find it quickly. >>PHILIP SHEPPARD: This is Philip Sheppard. I am referring to page 1-20, and this is section 151. Dispute resolution filing fees, I understand of course there are going to be third parties dealing with that. You have currently provided, usefully, some orders of magnitude. Is it the intention by the time we get to final document that you would have certainty from all of the providers you are talking to? >>KURT PRITZ: I was looking for our subject matter Amy Stathos on this. I think in certain -- I think there will still be ranges for all disputes. In certain disputes, types of disputes, certain objections, it will be more or less nailed down, the flat-fee disputes. And I describe those later in the presentation. But some disputes will still be settled by an hourly rate, but it will be a smaller band, and that's in concert with the current practice that these dispute resolution providers currently provide. >>AVRI DORIA: Okay. Tony Harris. >>KURT PRITZ: You know -- >>AVRI DORIA: Sorry. >>KURT PRITZ: Philip, Karla reminds me we will be negotiating a fixed filing fee, so that will certainly be nailed down with specificity, and then either a flat fee that will be published or a range based on an hourly fee after that depending on the dispute side. >>AVRI DORIA: Tony Harris. >>TONY HARRIS: Tony Harris here. I am referring to page 3 of Module 1, and specifically to one to two, administrative completeness check. And what I am interested in is a statement about ICANN will post a list of applications considered complete and ready for evaluation as soon as practical after the application -- the close of the application period. My questions are two. There will be -- Then there will be no publications, I hope, during the application period itself. >>KURT PRITZ: That's true. >>TONY HARRIS: And the second question is about -- I don't think it's mentioned here, but in other modules you do talk about when you publish this information for everybody to see, certain parts of the applications will be confidential. My question is, what is considered confidential and who decides what's confidential, ICANN or the applicant? >>KURT PRITZ: Yeah, so we -- somewhere in here, it says that the answers to the financial questions will be confidential, and the rest of it won't. >>TONY HARRIS: Oh, okay. Thank you. >>KURT PRITZ: So that's a first guess. >>TONY HARRIS: I must have missed it. I didn't see it. Thank you. >>KURT PRITZ: That's all right. >>AVRI DORIA: Thank you. Mike. >>MICHAEL PALAGE: Thank you. Mike Palage. Three questions. The first deals with section 1.1.3 on page 1-8. And the specific sentence, it says: "ICANN will open a public comment forum at the time the applications are publicly posted on the ICANN Web site, and will remain open through the application round." My question is, what is going to be the dynamics of this forum? Will it be one forum in which all applicants will have to talk about 500 applications? Will there be a specific forum for each application? Or will there be a forum for a particular string in which there are multiple applicants? Could you provide some detail on how you see that forum playing out. >>KARLA VALENTE: So we're still looking at how we're going to structure there. Right now, our vision is that in the public forum, you would have the ability to comment on an application, or if it's multiple applications for a string, on that string itself. You know. So we envision, you know, giving some flexibility on how things can be commented on. That would depend, of course, on the resources that we get when we get there, basically, and the feedback that we get during this public comment period. But if it's going to be useful for the community to have the public comments done per applicant or per string, we can certainly accommodate that. >>MICHAEL PALAGE: I think that would help. And would probably, if you will, circumvent a lot of criticism later on by having that. So I think that's constructive. Thank you. >>AVRI DORIA: Tony, you have a follow-up on that one? >>TONY HARRIS: Yeah. It's very short. I'm sorry, are you finished? >>MICHAEL PALAGE: I have more. But not on this. >>TONY HARRIS: Just on this point, I do understand, Kurt, as you said, that the actual applications will not be published until the application period closes. But will there be a notification to the community of which strings are being applied for during the application process? Will there be an awareness -- will the community, let's say, know that somebody has applied for dot Africa, for example? Without any details. >>KURT PRITZ: After the close. No, after the close of the process. >>TONY HARRIS: After the close. >>KURT PRITZ: And so Dan reminded me to say that this is draft, and I'm answering questions with declarative sentences as if they were fact. So this is envisioned or as planned presently. >>TONY HARRIS: Thank you. Sorry to interrupt. >>MICHAEL PALAGE: No problem. The second question is on section 1.1.5. And that appears on page 1-11. And this has to do with subsequent application rounds. The current wording says that the goal is for the next application round to begin within one year of the close of the application submission period for this round. Based upon how things -- >>KURT PRITZ: It's not my objective goal on which my bonus is based. [ Laughter ] >>MICHAEL PALAGE: I appreciate that disclosure. Openness and transparency is always welcomed, Kurt. So with regard to this goal, I think one of the things during the initial drafting was, we focused on making sure that that second round was hard-coded in. And I think the reason that hard-coding needs to be there, and not just a goal, is based upon the price points and stuff like this that we've talked about, you know, if there are -- Without that hard coding, there is going to be additional pressure for torch get into this round. >>KURT PRITZ: I think you're right. And I think one of the -- one of the less than fully fleshed-out aspects of this document are the precise time frames that, you know, we had a previous question on. And the next draft or interim publications will indicate what the -- what the time frames or timelines associated with each module will be. And that -- you know, as that's hardened out, that will allow us to, you know, nail that down with a better number, I think you're right. >>MICHAEL PALAGE: Thank you. Next is section 1.2.2.2. And it appears on page 113. And this deals with contract execution and post delegation regarding community applications. It states: "ICANN must approve material changes to community-based nature of the gTLD and any associated contractual changes." So what I'm trying to reconcile is, today, the ICANN board approves all contractual changes on all gTLDs. So is staff envisioning not going to the board for certain contractual changes? >>KURT PRITZ: No. It's -- no, it says "ICANN." But what that was meant to convey was that if you want -- Talking in shorthand, if you want to get a chit for being a community-based applicant, then that imposes certain restrictions on the TLD. And it will not be -- then once you have the TLD, relaxation of those community restrictions won't be lightly given, because, you know, they were used to gain an advantage. >>MICHAEL PALAGE: And I fully agree with that. I guess I was -- The corollary was, if you're not community-based, you can get a contractual change without the ICANN board approval. >>KURT PRITZ: Oh, no. >>MICHAEL PALAGE: Because that's kind of how you read it. So I agree with what you said there. My point was the opposite. >>KURT PRITZ: Got it. >>MICHAEL PALAGE: So you may want to specify this. One final point, it's on section 1.2.2.2 -- actually, I'm sorry, 1.5.5, appearing on page 122. And this has to do with withdrawals. It says: "Refunds may be available to applicants who choose to withdraw at certain stages of the process." In reading this document, I have not seen when those different stages will be and what the percentages of the refunds will be. I think -- So that will be coming? >>KURT PRITZ: Yes. >>MICHAEL PALAGE: Thank you. >>AVRI DORIA: Okay. Thank you. I was ambiguous when I said "Mike." And now Mike. >>MIKE RODENBAUGH: Mike Rodenbaugh. Thanks. I had four questions, but three of them were just asked and answered by others, so that's great. Back to 1-13, though, on community TLDs. How long will new community-based TLD operators be bound to the restrictions in their contracts? >>KURT PRITZ: Well, that's just -- that's just it. Relaxation of those restrictions are not to be lightly given. And that's what that term is meant to convey. So advice you might have on the bar for that must be reached in order to get a relaxation, you know, would be an interesting discussion. So the intent here is to make that somewhat difficult. But as you pointed out, it's not specific here. >>MIKE RODENBAUGH: The concern here is in the past, dot pro and dot name and some others have really materially changed. And I guess there's just a lot of concern that that can be too easily obtained by the ICANN board. Maybe there ought to be a limit stated, that there will be no changes in the first term or there will be no changes in the first three years, or something like that. >>KURT PRITZ: That's right. And, you know, with that sort of analysis goes with what's the cost or benefit to registrants in order to keep the registry viable. And also, in those cases, those are restricted TLDs and not sponsored in the older lexicon. But I get your point exactly. And we're together on this in trying to figure out the best way to implement what you're describing. So any advice you would have would be appreciated. >>AVRI DORIA: Before you continue, Tim, you wanted to make a comment on this point? >>TIM RUIZ: Yeah. I think Mike's making the point, but I just want to even make it a little bit stronger, at least from my view, that that's really crucial. Because in this round, now, what's going to happen is they're going to be competing possibly against others and getting preferential treatment based on the fact that they're community-based. So, you know, it should be much more difficult than what it has been in the past for those types of TLDs to change. >>KURT PRITZ: Thank you. >>AVRI DORIA: Mike, you had a couple more or had you finished your list? >>MIKE RODENBAUGH: I'm finished. Thanks. >>AVRI DORIA: Ute. >>UTE DECKER: I had only one question, and it has been answered. >>AVRI DORIA: Thank you. Jeff. >>JEFF NEUMAN: Okay. I'm -- I might have two parts to this question. So I'm looking at the cost considerations memorandum, and, Kurt, there's a lot of information in that memorandum that is really left out, and it's kind of conclusory in nature. But you're using it to base, let's say, $26,000 for historical cost. So two parts to the question are, one, are you going to make that data available as to how you came up with that number? Or do you -- you know, -- yeah, I guess I'll leave it at that, and then I have a follow up. Like, for example, it says you spent $7.5 million in personnel that you're attributing to the new TLD process. Is there -- I'm assuming you have backup for this? >>KURT PRITZ: Sure. Yeah, we have -- we have backup by person, by hours. >>JEFF NEUMAN: Right. So the follow-up to that -- I mean, I think we should all be able to see that for openness, because it's kind of a large number to throw out there. And to just kind of get people to accept, okay, $26,000 an application, I think that's something you guys should publish. And then the other part to that is, this is the part I'm kind of struggling with the historical cost. Normally, a nonprofit organization sets a budget, they collect the money based on that budget and then they kind of start out the next year as, like, a zero sum. I'm not sure how a nonprofit goes back and charges for historical cost, especially when your 2006/2007 budget, your 2007/2008 budget, and your 2008/2009 budget all has new TLD components to it, yet there's $12 million that the current round of applicants are expected to subsidize. And I'm just trying to figure that out as to how -- what's that? -- reconcile it. I'm just trying to figure out -- >>KURT PRITZ: So I think that -- you know, the ICANN but is commented on by the community every year and passed by the board. And the revenues and particularly the expenses are a fixed amount. And as you know, most of those -- most of the ICANN revenue comes from the gTLD community, by far. And so then the question is whether that gTLD community should be paying for the development costs of the new gTLD process, or if ICANN recovers that money going forward and that the budget is fixed by a bottom-up discussion, then the result would be, in a zero-sum game, that if ICANN recovers the development costs in some way, that fees to the gTLD community will essentially be lowered in the longer term. >>JEFF NEUMAN: But I guess it's not that you're in debt. It's just -- you already have the money. You've already received it. You've spent it. And now you're just trying to go back and get more, which is just going to go into your reserve fund. I guess the part that's kind of hard to swallow is that when it's budgeted, it's collected -- I mean, has there been an outcry from the existing gTLD community that you have to go back and recover those costs with new applications? Is that the need you're trying to address? >>DOUG BRENT: Doug Brent. Hi, Jeff. Thought I'd weigh in on this. I have to leave at 4:00. So let me say one thing before I come back tomorrow on new gTLDs. First of all, on your point about publishing more data, about the -- whatever it is, $12.7 million, we can certainly do that. In the end, it's a set of hours and costs allocated. But it's -- I think getting more detail would allow you at least to inspect that and get some more trust. Second thing, in terms of this argument, so, first of all, this is certainly one of the elements we're looking for feedback from the community around this idea of cost recovery. And it's not only the notion of cost recovery, but from what point in time would you do cost recovery. We certainly did get feedback from registrars, and I was sitting around the table when we got this feedback that said -- the argument was something like, "You should lower the fees, and in fact what you're doing is sort of overcollecting now for this gTLD program, for which ICANN is going to get additional revenue while you're still trying to meet the target for contribution to the reserve fund, so that a valid use of that money would be not to spend it on new gTLDs, but just to collect it and put it in the reserve." So I think we have specifically heard -- I think many registrars would just say, lower fees, lower the reserve fund. But given the strategic goal for the reserve fund, that we've, in fact, taken money that we would have collected for this and put it into new gTLDs, which has been a form of cross subsidization and looking to recover that and put it in the reserve which would lower the cost on current gTLD registrants to top up that reserve fund. So that -- it was directly as a result of one discussion with registrars. >>JEFF NEUMAN: I mean, I hear what you're saying. But if you look at the base agreement, you know, the fees that ICANN is collecting on an ongoing basis, at least from registries, seem like in some cases to be going up or a lot higher than some registries are going to be able to pay. But, I mean, we'll get to that separate. >>DOUG BRENT: Sure. >>JEFF NEUMAN: It just seems like a lot of money. And, you know, you think as an existing gTLD, I'd say, "Hey, jack up the price. I think that would be great." But, no, I think what you're doing here is discouraging a number of people that would like to apply. But, you know, $185,000, $26,000 of which is to cover historical cost and then $60,000 to cover risk, but the risk is to make all of the money back, including the money you're budgeting for historical, you know, it's kind of a little bit of a circle. In other words, your risk is, are you going to collect all of the money to pay for your historical cost and to pay for your costs of implementation going forward. So it's $26,000 historical, plus an element of the risk cost is attributed to recovering money from the history. So it's -- it's more than 26. I don't know exactly what it is. I'm not the best mathematician. But there is -- it's actually more than 26,000. So maybe if there's a mathematician, they can help figure out what the exact cost is. >>DOUG BRENT: I don't want to take up all the time on this. But I don't think the way you characterized it, Jeff, is the way the numbers have been put together. The idea was that the cost components, there is this program development, instead of call it historical cost, it's really a program development cost for the new gTLD, which is being booked back against the new gTLD revenue. The second is sort of this easily identifiable, you know, variable processing cost or processing cost to the applications. The third element is this idea that just like anything else, you wouldn't want ICANN staff larding in every single step of the way, making it the most expensive process it possibly could be to account for any possible exception. If you look back at the most recent STLD round, some were more expensive to process and some less so. So the idea that that risk element has nothing to do with recapturing this idea of program development cost, which is something we should take feedback from the community on. If the community says don't recapture those program development costs, then that's an answer. >>JEFF NEUMAN: I just think something like that should have been known by the community before the whole process started. In other words, you guys hired some pretty expensive consultants in some of your research. I mean, they're the best in the world, some of them. But, you know, did you have to hire Willis, you know, to do your risk assessment. How much did that cost? And did you have to hire all these consultants -- I guess my point is, if the community knew at the beginning of the process that you would be going back and charging all of that cost to the new applicants, we may have had a different discussion as to how long this process would have taken, the type of consultants we may have used, you know. But now we're going after the fact and saying, okay, it costs $12 million, now we've got to pay for that. It may be a comment towards future rounds, because there's nothing we can do about past spending. But it's just a point I wanted to make. >>DOUG BRENT: I will soy to that particular point, there's nothing that has been spent here that wasn't in that initial gTLD budget and in that process. Using Willis, which turned out to be the price performer, seriously, it was the price performer, and anyone doing this analyze analysis was part of that budget. So that wasn't new money that wasn't budgeted as part of the 2008 fiscal budget, 2009 budget process. So, in other words, we didn't just independently go do that. That was built into the plan for this year. >>JEFF NEUMAN: Yeah. I won't take more time. >>AVRI DORIA: Okay. Thanks. Chuck. >>CHUCK GOMES: I have a whole bunch of questions. I'll take a group of them, and then I'll get -- just get added to the end of the queue so I can let some other people jump in. First of all, on page 1.7, next to last paragraph, it says: "If the initial startup requirements are not satisfied so that the gTLD can be delegated into the root zone within the time frame specified in the registry agreement, ICANN may, at its sole and absolute discretion, elect to terminate the registry agreement." The first question there is, is any idea what that time period might be? And you know, should that correspond maybe to the time period by which the TLD has to become active? Or is there any correlation there? Does that make sense? >>KURT PRITZ: Yeah. So -- Yes. So we think that, you know, in our discussions, we think, you know, one year might be too short, two years might be too long. But we'd like to have a dialogue about that. In order to become delegated in the root zone in the process as it's envisioned, there's a predelegation check where the registry actually has to become operational in some aspects. So we think that probably satisfies, as far -- you know, as far as being operational once the delegation takes place, that would serve to satisfy that. >>CHUCK GOMES: So what you're saying is the two would probably correlate. >>KURT PRITZ: Yeah. >>CHUCK GOMES: Okay. Thank you. The next question on the next page, 1.8, end of the second paragraph, the last sentence there says: "ICANN will open a public comment forum at the time the applications are publicly posted on ICANN's Web site, which will remain open through the application round." Would you define "application round" for me, please. >>KURT PRITZ: Karla, that's until the delegation right. So it's until -- you know, there will be an opportunity to comment at any time, you know, in an ICANN-style way on an application through the entire evaluation process. >>CHUCK GOMES: Through the -- So until everybody's delegated? All -- that are eligible? Is that an indefinite comment period until the last application is dealt with? >>KURT PRITZ: Well, as -- you know, I might not be the best to answer this question. But as Karla answered that question before, there could be individual comment for each application. And so that would be open for the length of that evaluation process. >>CHUCK GOMES: So if I've submitted an application and I can assume that public comments will be possible all the way through until it's done? >>KURT PRITZ: Yeah, until the evaluation is done. >>AVRI DORIA: Mike, was your question in relation to this? Or was it another question to add to the queue? >>MIKE RODENBAUGH: Another question just to add to the queue, please. >>AVRI DORIA: Okay. Thank you. >>KURT PRITZ: So we probably need to cut this in the queue, because we need to -- >>AVRI DORIA: Cut this queue after Mike that just got in it. >>CHUCK GOMES: So I should go through all my questions now? I guess not. In your -- On page 1-9, scenario number 9, -- and I may be missing something -- in the approved four subsequent steps, says "no." And I guess that's because they failed the initial evaluation? Is that right? >>KURT PRITZ: Well, let's see. It fails the initial evaluation but passes the extended evaluation. >>CHUCK GOMES: Right. Shouldn't that be a "yes"? Yes. >>KURT PRITZ: Oh, in scenario 9, in string contention, it lost out. >>CHUCK GOMES: Oh, it lost out in string -- >>KURT PRITZ: Right. >>CHUCK GOMES: Oh, is that what the "yes" means under "string contention"? >>KURT PRITZ: Yeah, it's in string contention. >>CHUCK GOMES: Lost. Okay. Very good. This is -- Okay. The next question has to do with the -- to Werner's question earlier with regard to the next round that's coming up. You say on page 1-11 there in that paragraph in between the lines there: "The goal is for the next application round to begin within one year of the close of the application submission period for this round." So that's the 30-day or whatever period? You're pre- -- you're suggesting that it should start one year from -- So if the application round starts on July 1st, it would be July 1st the following year? Is that what's being said there? >>KURT PRITZ: So what's being said there is that when the application round closes, essentially, when we publish the applications, the next round would start within a year of that. And -- >>CHUCK GOMES: No. I'm sorry. Say that again, please. >>KURT PRITZ: At the close of the application round means you can no longer apply. The application period. So it's a different application round than what you were discussing earlier. So you apply for a TLD. The period closes. All of the applications are published. Within a year of the start of the evaluation -- >>CHUCK GOMES: Good. Okay. Thank you. >>KURT PRITZ: Now, that is -- you know, frankly, that's, essentially, a placeholder pending, you know, Mike's comment and others that, you know, we need to hone the evaluation time so we can perhaps improve, but at least be very specific and accurate about that time. >>CHUCK GOMES: But you're not expecting that all of the applications would informal be dealt with from the first round -- >>KURT PRITZ: Correct. >>CHUCK GOMES: -- before you start the other one? On that same page, at the bottom, why is it that applications from individuals or sole proprietorships would not be considered? >>KURT PRITZ: I think 'cause -- well, too, the ability to enter into a contract with ICANN. And also having to do with the longevity of the entity. So it's thought that, you know, an entity rather than an individual has more -- you know, has more infinite life and it will provide better protection for registrants. >>CHUCK GOMES: Well, let me stop this and turn it over to the next person. >>JEFF NEUMAN: Can do I a follow-up on one of the questions? This is Jeff, yeah. >>AVRI DORIA: Oh, okay. I'm sorry. >>JEFF NEUMAN: It was a follow-up on the new round, that there would be follow-up rounds. And is on Werner's, and then Chuck. The question I have is, are we being a little foolhardy at setting that date, because what if this entity kind of collapses in on itself. There's no time for any lessons learned. I mean, I'm not saying that we should delay that long future rounds. But I think ICANN should build in some mechanisms to learn listens in this round if everything falls over or if you get way too many applications and it's going to extend too long. There's just no evaluation period after that. And I'm just worried about it for ICANN to make a definitive statement. >>KURT PRITZ: You know, I think you're right, Jeff. And that one-year period comes from a lot of standing at the white board and mapping out timelines and providing some -- you know, balancing the interests you describe and getting lessons learned with the interests Werner describes and giving applicants who might be willing to wait, you know, a goal line. So all that balancing has kind of taken place. And as I described earlier, I think, you know, we need to nail down timelines with specificity so we can make that call a little more closely. >>AVRI DORIA: Eric. >>ERIC BRUNNER-WILLIAMS: Thank you, Avri. Eric Brunner-Williams. Kurt, I'm going to ask what's going to sound like a policy question, but it's really a process question. And I'm sorry that I can't think of a smarter way of doing this. At the San Juan meeting, the sense of the council was that community would -- the existence of a community application would be dispositive if there was contention for a string between two applications, one of which was community-based and one of which was not. At the February workshop, you suggested that it would be a pebble, and I recall you were holding your hand up, holding the imaginary pebble, to weigh in a scale, so that it would be some kind of evaluation of weight of the claim of the community versus some other claim of goodness by the noncommunity applicant. The subsequent publication of the auctions paper actually gave a number for that kind of weight of .25, but only if the community-based applicant came from a non-OECD country. And I think we're now at the point where there is very little, if anything, left of the weight that you suggested in February. So -- But my question is not about the change in policy or whether or not the policy that came from staff is better than the policy that came from the GNSO Council. My question is, what does the council do, should it decide that its original policy was a correct policy, in order to instruct staff to make community-based applicants or the existence of a community dispositive rather than advisory or evaluated? Thank you. >>KURT PRITZ: Is that a question for you or me? So I -- I'm not -- You know, I saw your e-mail on this a while ago, Eric. And, you know, I'd like to take this offline, because I think that you missed the intention that the pebble -- you know, the pebble still exists. So I thought -- You know, I'm not able to spin it up right now, but I had a good answer for the time when I read your e-mail. Do you want to -- if you can take it briefly, Olof? >>OLOF NORDLING: Very briefly. I mean, the community-based applicants, they have one advantage, and that is to elect a comparative evaluation. They will force that, if they do so, among the other contenders, and they will have a definitive advantage in that comparative evaluation. So if you read the string contention paper carefully, you will see that it's been taken care of. >>AVRI DORIA: And, Eric, one thing I want to add to the answer before you go on is, that is one of the questions -- not specifically this issue -- but one of the questions for our Wednesday discussion is, what, if anything, do we want to do about anything where we believe there is a gap between what was suggested in policy and what is showing up in process. So that what we do about it is an open issue. >>ERIC BRUNNER-WILLIAMS: Thank you, Olof, for your response. But my question wasn't about the difference in policy. I'm not trying to suggest that your answer is incorrect or some other answer is correct. Rather, the council in San Juan made a statement about the nature of contention between two strings, one of which is community-based, one of which is not. Staff has arrived at a different policy. And I'm sure it's a better one, or perhaps it is. But that's not the point. My question is, how does the council inform staff that it actually inform staff that it is dispositive and not as you say, something they may elect to exercise? >>AVRI DORIA: And that's what I meant is how do we do that or how will we go about doing that is, indeed, a discussion that is scheduled for Wednesday, is trying to figure out what actions we need to take. >>ERIC BRUNNER WILLIAMS: Thank you very much. >>KURT PRITZ: And I sort of think the implementation in this case is on all fours with the policy recommendation that says community representation will be a factor in determining who is awarded the string in indications of string contention. So I am sure I mangled the wording a little bit but not the intent. And that's the -- a lot of cases we think we match the intent, and this is one of them. But we know we should discuss it. >>AVRI DORIA: Right. That's one of the things that needs to come out, is as the council and as the GNSO community reads this, do we believe that, as you do, is one of the issues that we'll have to discuss. Okay. I had Zahid next. >>ZAHID JAMIL: Thank you, Avri. My query is regarding page 1-4. 1.1.2.4. It's about objection filing. And I made this point a little earlier. I am just going to try to go through it a little slowly. I am trying to figure out what period for objections. When I read 1.1.2.4, it says it starts after ICANN posts the list of complete applications described in paragraph 1.1.2.2, and it closes following the end of the initial evaluation period. And it refers again to 1.1.2.3. And I look at both of these and they don't give time lines, per se. So what would be the maximum, what would be the minimum? That's my first question. And second, will the applications be published so people would know they have to make objections? And if so, how and where? >>KURT PRITZ: This section is meant to convey that the objection process for economy reasons will occur in parallel will the evaluation process, and as I described in the answer to earlier questions, I think in this document, those time periods need to be nailed down. There's quite an extensive communications plan intended. And the launch of that started with essentially letters to each -- like I said before, each of the governments and territories in the world, and each of the TLD managers. There will be notices placed in publications in each of the regions around the world that this application round is going to be launched, that it has been launched and here are the applications for those who wish to notice it. So it's certainly our intent to take away the ability for somebody to say, "Gee, I didn't know." Well, the ability to take that completely away is just about an impossible. And our intention is to get as close as possible and do it in an economic style. Did you have something, Karla? >>KARLA VALENTE: And during this period of time we also plan on holding Webinars or potential venues where we can explain the draft RFP or the draft applicant guidebook, we can answer questions about it in order to minimize potential questions during the application process or issues during the application process itself. >>AVRI DORIA: Thanks, and as I am listening to the questions coming out, I am realizing that as it's 4:18 at the moment, we may not get through our questions. So we may actually have to have special phone calls for questions. At the moment, the list had Mike on it, and then we had tried to draw a line, but since then I have gotten Ron, Tony Harris, Jeff and Marilyn with questions. >>JEFF NEUMAN: I was on a previous list, and you may have erased me because I asked a follow-up to something, but that wasn't my question that I had. Sorry. >>AVRI DORIA: Okay. So I will go with Mike, and then -- yeah, I did cross you out, but that's probably because you had.... >>CHUCK GOMES: I am going to ask a process question here to get feedback from the council. There's no doubt we are going to have to ask a lot of questions after this week's over, because we are not going to get through them all. I don't think there's any chance of that. Would it be better to get through -- in other words, not cut off the queue and get through as many questions as we have today on Module 1, or to cut off Module 1 and go to Module 2 and we will have leftover questions on both modules? My own inclination, and this is my opinion, that it would be better to take all the questions for Module 1 that we have today and then go to Module 2. What's the feeling of the council? >>JEFF NEUMAN: I am not on the council so -- >>CHUCK GOMES: Anybody here. It doesn't just have to be the council. >>JEFF NEUMAN: My feeling on that is I think you're right because Module 1 is an overview which touches on every single other module and will probably reduce the questions when you get down to the specifics of the modules. >>AVRI DORIA: Does anyone object to continuing on Module 1? I don't want to spend talking about; otherwise, we don't even get to the questions. So no one objects. Fine. I had Mike, then I will come to Jeff who I had cut because Mike had started already, and then I have Ron, Tony Harris, Jeff and Marilyn. >>MIKE RODENBAUGH: It's -- >>TIM RUIZ: I do have a comment I would like to make about what we just decided. It's not really an objection, per se. Is that all right? >>AVRI DORIA: Okay. >>TIM RUIZ: Tim with the registrars constituency. An additional suggestion would be -- and I think Adrian was kind of hinting at it -- was on modules going forward that we just get right to the questions. Do we really need to go through a detailed step-by-step review of each of the modules? >>AVRI DORIA: Okay. Thanks. >>MIKE RODENBAUGH: I agree with you, Tim. Actually, I think we could just go to questions. I assume everyone has read this stuff, but I may be very wrong on that. Anyway, I have quite a specific question on the community-based gTLD requirements, which is -- they are outlined in section 112, but my concern is actually on page A-7, the questions that are intended to be asked of every applicant. And you state that -- or I'm sorry, the document states that regarding nexus between a proposed string and community. String name is name or well-known abbreviation of community institution. So, okay. I assume that means INTA, dot INTA, no problem. But what about, let's say, the world health organization wanted dot health, or the motion picture association wanted dot movie. Do you consider those to be potential community-based applications? And I please hope you say yes. >>KURT PRITZ: They are potential community-based applications. When we get to module that describes that, the scoring would be higher for an application for -- what is it? MPAA? Than for dot movie, because the nexus between the string and the organization is greater. So that's Kurt's opinion if he was a scorer, in Module 4. >>MIKE RODENBAUGH: Okay. That's helpful. Thanks. One other question. It goes back to what Chuck was asking about sole proprietors and so on. I think you gave a good answer on that. That makes perfect sense. I am concerned -- because I think there are going to be a lot of new ventures applying for TLDs, and it seems to me that they are at a disadvantage out of the gate. Not only do they have to provide one-year financial statements, which may actually not be possible. I am just wondering, what's your view on that? Are they disadvantaged? And how? >>KURT PRITZ: Well, there's actually an alternative for those organizations that don't have that documentation available that's described here, or it's described in the questions themselves. And two is, disadvantage is a relative thing. So if we get an application from a major multi-national corporation and an application from a company that was started yesterday, which one will probably get a higher score in the financial evaluation? Probably the existing multi-national organization. But that doesn't mean the smaller one cannot pass. As a matter of fact, if you read the scoring criteria, I think it's reasonable for a startup to pass that criteria, but not get 2's on every scores but 1's. >>MIKE RODENBAUGH: So you are saying essentially they don't have a disadvantage at all in your view? >>KURT PRITZ: You know, they will get a lower score, but I think they can still get a passing score. >>MIKE RODENBAUGH: What about in the event of string contention between a startup and a multi-national? >>KURT PRITZ: Well, that's -- again, that's Module 4, and that's why we are running behind schedule. But in string contention, if you are a community-based applicant, you can elect for comparative evaluation. And if you are not a community-based applicant, you can't. Then you would go on to the next type. >>AVRI DORIA: Okay. Jeff. >>JEFF NEUMAN: Okay. My question is on dispute resolution, and the filing -- not on the filing fee itself, but more the timing. >>KURT PRITZ: Module 3. >>JEFF NEUMAN: It seems to me, according to this, you have to file your dispute during the objection period which is during the initial evaluation, and you have to pay your up to $5,000 nonrefundable fee.. It would seem too me you should maybe think about the timing a little bit because if it doesn't pass the evaluation, there really is no necessity to have a dispute resolution. But you filed your objection and paid $5,000, which is nonrefundable, for which there was never anything that passed initial evaluation. So when you work on your next draft, think about the timing or the payment. >>AVRI DORIA: Ron. >>MIKE RODENBAUGH: Avri, I forgot to ask two questions, so put me back on. >>AVRI DORIA: I will put you back on the queue. Ron. >>RON ANDRUFF: Thank you very much. Ron Andruff. I want to say that the work you have done has been very good. [ Applause ] >>RON ANDRUFF: I just arrived. It's very good work. It's clear, it's concise, it's modules, it's very well done, and hats off to all of you for having done it. I hope you get a good bonus, Kurt. But all joking aside, I want to talk about two things. One has to do with what was just being discussed about the idea of Module 4. It's 4.2.3. It's the comparative evaluation criteria. >>KURT PRITZ: So, Ron, actually we have cast this meeting to have experts here for modules 1, 2 and maybe 3, and those experts are going to be gone tomorrow. So can we hold the Module 4 questions until Module 4 time? >>RON ANDRUFF: Certainly. No problem. We can back off that one. The second question had to do with something that's missing from the plan as I saw did and it has to do with IDN transliteration -- what I would call IDN transliteration. If an applicant files for dot finance, goes through all the technical evaluation, financial evaluation, all the evaluations, scores high, wins the right to manage that on behalf of that community but would like to also have that in Arabic or in Cyrillic characters, Russian Cyrillic characters, why would an applicant have to pay another $185,000 when it's really just a question of looking at the transliteration of that word in another language? Maybe there's a fee of 25,000, maybe there's a fee of 50,000 for someone to look at that and make sure it's correct, but I think it's unfair to the community that if someone wants to manage a space on behalf of a community, they have to go out and maybe buy five, six, seven names, costing them one and a half million dollars when they could have spent 185,000, won the award, and then manage that space on behalf of that community by paying a smaller fee. Now, what that would mean, then, is if you manage that space on behalf of a community, you would then offer that registrant when they bought dot finance, they would get it in four or five languages, use which ones they want or not. But you would also be very sure you don't have collision. My biggest fear is when you sell dot finance to a Russian in Cyrillic character and then dot finance in English and dot finance in Arabic, we are going to see collisions and lots of them. That's my fear. >>KURT PRITZ: And that's a very interesting point. And in thinking through this, there's a balancing between that and the gray areas that one gets into as to what's a translation or a transliteration of a string. And when that's very clear and when that becomes gray and how that determination is made. So the way -- I think you should make that comment. The way this process is written is if it's a different label in the root zone, then it's a different application. But I think there should be discussion about that. >>RON ANDRUFF: You would depreciation though, that once one -- once a company or organization has gone through the whole process, they have gone through the process of being able to run the TLD, et cetera, et cetera, all of the criteria have been met, you would agree when you go through the first one, you have gone through it; right? >>KURT PRITZ: Volume discounts. So again, what are applications from the same entity and how can that be gamed? There's a whole host of issues that add to the complexity of trying to administrate the process that, in the first round, we wanted to make as straightforward as possible. But I think that's a valid discussion, if it's exactly the same entity, you know, the evaluation -- the evaluation panel is probably going to spend less time on it. >>CHUCK GOMES: A follow-up on that particular issue. I understand the fact that the string itself is going to require some costs for the IDN string that the applicants doing that. But presumably all the rest of the application would be identical and require no additional costs at all. >>RON ANDRUFF: Exactly. That's my point, Chuck. Thank you. >>AVRI DORIA: Thank you. >>TIM RUIZ: Avri, I have a question on that. >>AVRI DORIA: Sure. >>TIM RUIZ: But there will be additional associated costs -- some minimal associated costs with evaluating the technical aspects of actually supporting an IDN as a registry; correct? So that wouldn't -- So you wouldn't be able to roll that up completely into one $185,000 fee. >>RON ANDRUFF: And that was my point, too. Exactly that. There should be a fee, whether it's 25 or 30. Whatever it is, there is a number for that, but it shouldn't be 185. Thank you. >>AVRI DORIA: Paul, you wanted to follow-up on that. Also, I think when we were making policy, we had made a policy recommendation that each application was an individual application and was to be considered that way. So I think partly, the staff is going from that basis, that each individual is an atomic unit and should be considered as such. So that should be a new policy recommendation to them. Paul. >>PAUL STAHURA: I agree with that recommendation. Otherwise I might apply for dot shoe, have the exact same application for dot sock, and say, well, I already put in the application for dot shoe, and it's exactly the same as dot sock so I should pay less because I went through the whole analysis for dot shoe, and, you know, I want that sock for dollar extra. And also, if I put in dot shoe, maybe I can get dot shoe in Chinese and Cyrillic and Arabic and every other language. So I agree with the policy as it is. >>AVRI DORIA: Okay. Thanks. You have yet another -- okay. >>MICHAEL PALAGE: If I can, Kurt. As we're talking about right now, we are talking about providing an applicant a benefit of getting multiple -- an IDN equivalent or transliteration of the string name they applied for. Although ICANN in the draft contract proposes removing the equitable treatment language that exists in the current contract, ICANN does have contractual obligations to existing registry operators to treat them equitably. So if, in fact, ICANN was to give an applicant some preference for a transliteration, how, under the requirement of treating existing registry applicants equitably, would you propose to handle that for existing TLD operators? >>KURT PRITZ: You know, I don't know. It's not envisioned in this implementation plan that that occur. >>RON ANDRUFF: Mike, -- I'm sorry to jump in. My thinking simply was you would pay that extra fee. So in this case of -- let's say it's dot travel, and dot travel wanted to take it in dot voyage. They would pay a fee for that transliteration investigation. So whether it be 25 or 30, whatever that number is, that was it. >>AVRI DORIA: I want to not get into redoing the policy discussion that we had on it. It may, indeed, be something that the GNSO will want to review the policy on, but certainly for this round, the policy is that applications are atomic units, and they are individuals and they stand alone and such. And I don't think we have any drive at the moment to impose a new policy decision when we actually were quite specific in making that one. >>JEFF NEUMAN: I'm not sure that is necessarily a policy. This is Jeff Neuman again. You can have each application be separate. We are only talking about a fee here. And the fee was not a part of the GNSO policy discussion. >>AVRI DORIA: That's true. >>JEFF NEUMAN: So you can have a separate application for it but since it has already gone through all of the evaluations or most of them, ICANN is not going to spend -- they shouldn't spend $185,000 again. So I disagree with you -- I agree with you, that's a separate application. I disagree that the separate application by policy has to be 185,000. >>AVRI DORIA: Okay. >>TIM RUIZ: Just a follow-up on that, then, because now we are expanding the whole fee thing to, well, if an existing registry applies for a brand-new gTLD, but by virtue of the fact that they are already an operating registry and have already gone through a lot of the evaluation processes and proven themselves, blah, blah, blah, do they get a reduced fee even on a brand-new gTLD. So at some point this turns into a snowball effect on the fees. >>CHUCK GOMES: Yeah. Good point, Tim. And I think in that regard, even an existing gTLD I think would have to have one -- at least one application that was the full fee, assuming that we -- they did something on this other fee. Because you can't just assume, because you are an existing gTLD, that you don't have to be evaluated at all, I don't think. So it wouldn't -- There would need to be one full evaluation. If there were subsequent ones where it was the same application, separate applications like Jeff said but a lesser need -- cost of evaluating the string or whatever, then that's where it would come in. That's my thoughts on that. >>KURT PRITZ: So you guys could get all together and funnel all your applications through one entity and get a barring begin. >>AVRI DORIA: You wanted to add a comment to this particular thread, Marilyn? Okay. >>EDMON CHUNG: I don't think that's what we are talking about. >>MARILYN CADE: I think I am in the queue. >>AVRI DORIA: You were in the queue for later, but you wanted to make -- >>MARILYN CADE: I do. >>AVRI DORIA: Because I don't want to let this one become a free for all discussion. >>MARILYN CADE: My name is Marilyn Cade. I am a member of the business constituency. I consider myself a member of the broader user business community, and I'm going to make a comment about my observation about what we are doing right now. I thought we were, as the council and interested observers, commenting on, to the staff, the implementability and the feedback about the excellent resources they have given us. And I, too, am very impressed and want to lend my appreciation. It's easier for me to read and evaluate in the way that it's organized. But I want to say right now, so far what I have heard since I have come back in, is comments from people who want to run registries. And I think that's a very important set of comments and feedback, and I really want to hear more about it, but I would like to hear more about the broader concern and views of the council and those here about the implementability of the things that are being presented. I have a lot of questions as a member of the business constituency about those things. And I know many of you have specific business interests and I respect that. I think you ought to declare it when you speak, if that's the case. And then we ought to try to make sure that all of your concerns are identified so they can be addressed, but really focus on the policy. >>JEFF NEUMAN: Marilyn, this is Jeff -- >>AVRI DORIA: Hold on. I am afraid we are getting closer and closer to a rat hole every minute on this one. >>JEFF NEUMAN: I am wondering, is there a question there? >>AVRI DORIA: Okay. I would like to continue with the rest of the list of people that are waiting to bring up issues. This issue has certainly been noted, and its variety of opinions has probably been noted and I suggest we continue talking about it in comment space. But I would like to go to Tony Harris now, who has got the next question on the linear list, as opposed to the side list. >>TONY HARRIS: I'm sorry. I have to move you off this topic for a few minutes. I hope you won't hold it against me. Tony Harris from the ISP constituency. Sorry. And getting to Marilyn's point, I am a future applicant, hopefully, for a TLD so I am an interested party. I would like to refer to page 119, breakdown of fees and demands. This particular item ends on page 121, and it says, "This list does not include fees; that is, registry fees" -- Excuse me -- "that will be payable to ICANN following execution of a registry agreement. I found the information on that fee in another document which was issued, the new gTLD program explanatory memorandum, cost considerations of the new gTLD program, which is a separate document. On page 10 of that document, there is paragraph which refers to this, and it says building on existing registry fee models, new gTLD registry fees are projected to be the greater of $75,000 or approximately 5% of registry transaction revenue. Using this approach, a new TLD oriented towards high volumes of registration would have a fee based on transactions, obviously, and a new TLD that might be a for community or of a limited registration uses would pay the fixed fee. My only question is, is not 75,000 perhaps a little high for some applicants? Let's say at least in the first two years, until they have time to actually implement their TLD, obtain customers and come through -- you know, give their marketing a chance to get underway. It's just a question. I'm not strenuously objecting. >>KURT PRITZ: I think the proposed fee model comes from a discussion about trying to accommodate various business models. And without knowing, you know, the full extent of the business models that are going to be described. So I'm sure it's going to be inadequate in some way, but it's intended to have those entities that won't be registering names, but maybe use the TLD for corporate purposes, you know, a reasonable fee. And those that go for volume registrations, a reasonable fee. I would like discussion on that because I would like to better anticipate what the business models are going to be and try to fashion a sort of one-size-fits-all fee model so the contracts can be the same. That's fair and equitable. So I understand your comment. >>AVRI DORIA: Adrian, do you have a follow-up on that? Is that what you are saying to me? >>ADRIAN KINDERIS: Follow up. >>AVRI DORIA: I thought that's what you were trying to say. Go ahead. >>ADRIAN KINDERIS: I can't hear with this thing on. Kurt, following up on that question -- Adrian Kinderis, and hoping to take over the world with many, many gTLDs, just declaring an interest. Can you tell us how you arrived at 75,000? Was it a budgetary thing? Is there an administrative cost per TLD that you have allocated and you have done your math and worked out that it's 75,000? Or what sort of parameters did you use to arrive at 75? >>KURT PRITZ: Again, it was looking at, you know, the current revenue models and fees currently paid by registries, and balancing the foreseen business models of those registries. So some new registries will be focused on large volumes of transactions, some won't. So should necessarily those registries focused on large volumes of transactions, you know, pay a fee proportional just to those transactions? Or should there be some fixed fee associated with the registration? >>ADRIAN KINDERIS: I understand that notion of it, but you have arrived at $75,000. I would have thought in your business plan, you have said, well, we need to generate, if there is X amount of registries, there is an administrative cost to that, we have to make a profit, so therefore that can be $10,000 if there's -- the more registries you have, the less administrative -- there might be -- I will take that back, actually. But you understand where I am going. You arrived at 75,000. I understand why you have got a set fee. >>KURT PRITZ: Got it. >>ADRIAN KINDERIS: That's fine. But why did you do 75,000? >>KURT PRITZ: ICANN is developing a three-year budget that's a separable budget associated with new TLDs, new gTLDs that forecast revenues and expenses associated with the larger volumes of new TLDs. And you will see, we are reviewing a draft of that, a rough draft of that with the ICANN board finance committee at this meeting. And I'm not sure of the exact timing of the publication for that. There's a balancing in there, as ICANN gets to some steady state -- you know, approaches a steady state model of revenue and expenses. And so you will see in that budget that there's registry fees where there's an increase in registry fees based on the new volume of gTLD registries. And there's a line item in there for potential fee reductions across the board so it brings the revenue model back down to what the expense needs are. >>ADRIAN KINDERIS: So that fee can actually be variable? >>KURT PRITZ: Right. >>AVRI DORIA: Paul, you had a follow-up on that, and just before Paul -- >>PAUL STAHURA: >>TONY HARRIS: Just to make sure I understand this correct, 75,000 is a minimum charge. >>KURT PRITZ: Sure. >>TONY HARRIS: I wanted to make sure I had that right. >>PAUL STAHURA: 75,000 or 5%, that means when the 5% gets over the 75 minimum, you have to do about $1.5 million in revenue for that TLD. That seems really high to me, especially considering that most current gTLDs are around the two and a half to three percent range. So that was one point. >>AVRI DORIA: Okay. That was the follow-up? Okay. Was there an answer? I don't know if you wanted to comment on that comment. No. Okay. Okay. So that was the follow-ups on that. I think we've hit that. At the moment, I have a list of people that still wanted to ask questions about Module 1 before we move on to Module 2 and 3. And we hope to get to 2 and 3 today. And we have 45 minutes left on this session today. But I had Marilyn, Chuck, Werner, and Mike. And then also Claudio. These are all on Module 1. So, Marilyn, question on Module 1. >>MARILYN CADE: I'm going to ask my question, I think it's about Module 1. And the question goes to Kurt. Some time ago, I think it may have been as long as three years ago, the board instructed Paul Twomey to retain economic advice on the impact of introducing new gTLDs. And that work was to inform the whole process. And it was a broad study. I've seen only one element addressed with a report from Charles River Associates. But the broader underlying implications of the resolution from the board, I believe, was to also examine the implication of the introduction of new gTLDs on the rest of the community. So, for instance, the last bullet point there, on -- that is related. >>JIM REID: Additional fees paid directly to outside panels and providers." That is clearly a cost which will be incurred by both applicants. And I will have a similar question under Module 2 when it has to do with disputes. Can you give us the status on where that robust, well-documented, thorough, long-awaited economic study is? >>KURT PRITZ: It's going to be published. I don't have a precise date for you. But it's in the near future. >>MARILYN CADE: I have a follow-up. Will it be published in order to be taken into account before we complete this work? >>KURT PRITZ: Oh, I think so, yeah. >>AVRI DORIA: Okay. Chuck. >>CHUCK GOMES: Thank you. Some of these I'll just try to leave as comments. >>AVRI DORIA: If you finish before I'm back, Werner's next. >>CHUCK GOMES: Got it. Excuse me. With regard to the startup issue that was discussed earlier, there's a requirement for proof of good standing. It seems like that would be difficult for a startup to do. But I'll just throw that out as a comment. And I don't know that we necessarily need to discuss it. Okay? On page 1-14, the two requirements 4 and 5, financial statements, and documentary evidence of ability to fund an ongoing registry operations, how will those be evaluated? What objective criteria will those be measured against? And I have other questions, but I don't know if -- Do you want me to just throw that out and -- >>KURT PRITZ: No. So these are documentary requirements. So financial statements we think are well-defined, accepted standard. There's four documents. Kevin's not here, and I'm not a genius on this. But, you know, it's a balance sheet, it's a cash flow, it's something else and something else, I forget. >>CHUCK GOMES: No, I understand the definition of a financial statement. >>KURT PRITZ: So -- Yeah. So they have to be furnished in order for the application to go forward. How information will be used is described in Module 2, where the applicant answers questions and the evaluator will use the criteria associated with those questions to measure the application. So in some cases, the information contained in those statements will be used by the evaluators to score the criteria in number 2. >>CHUCK GOMES: And they would use that relative to the proposal that they've made? >>KURT PRITZ: That's right. >>CHUCK GOMES: Yeah. >>KURT PRITZ: So it's intended to scale. >>CHUCK GOMES: The next question, at the bottom of page 1-18, it says: "Although ICANN intends to follow the security precautions outlined here, it offers no assurance that these procedures will keep an applicant's data confidential and secure from access by unauthorized third parties." That seems pretty wide open with regard to confidential information. I doubt if any of us's businesses could get away with privacy statements or confidentiality statements that say that. And I'm not necessarily looking for a comment on that one. You can take that as just input. >>KURT PRITZ: Thank you. >>CHUCK GOMES: Then going to page 1-19, second to last paragraph, -- or, actually, the end of the second bullet there. "Details will be made available when the application process is launched on the gTLD evaluation fee." When the application process is launched. What does that mean? Does that mean at the beginning of the four-month period or at the beginning of the actual application period? >>KURT PRITZ: Yes. So, at the worst, it would be when the final applicant guidebook is published, which is four months -- >>CHUCK GOMES: The four-month period. >>KURT PRITZ: -- before. >>CHUCK GOMES: That's what I was hoping. That's very helpful, I think. It would be the shame if it was the going on of the formal application period. That's really late to get that information. Thank you for that clarification. And then on page 1-20, the bullet there, "dispute resolution adjudication fee," it says: "Both parties -- I think this is the second sentence -- both parties in the dispute resolution proceeding will be required to submit an advanced payment of cost in an estimated amount to cover the entire cost of the proceeding." I don't know if it's possible or not, but when I read that, I wondered, is it possible that there maybe even could be three parties in that that have to pay that fee and would they all have to pay the full fee? I don't know. You can take that offline as far as just maybe it's not even possible. >>KURT PRITZ: It's going to be operated by the dispute resolution provider, who has a standard practice for how they operate. And so there's -- in subsequent modules, there's, you know, fairly detailed procedures. And then the dispute resolution provider will have their own set of rules that they publish that govern this. So how those fees are paid and how they're shared if there's multiple antagonists in the plot will be published by them, according to their standard operating procedure. >>CHUCK GOMES: And then last of all -- and this will just be a comment -- page 1-22, middle of the page there, it says: "Further details on refund amounts will be available in the final version of the RFP," talking about refunds for applicants who choose to withdraw at certain stages. I think that's helpful to get it at the beginning of the four-month period. I think what would be even more helpful, if some guidance on that could be provided even before then. But that's just input. >>KURT PRITZ: Thank you. >>AVRI DORIA: Okay. Thank you. Werner. >>WERNER STAUB: Werner Staub. I have a question on the statement you have on the top of page 1-12, where it says there are two kinds of communities -- sorry, two kinds of applications, open applications and community-based applications. So this, I believe, is an unfortunate kind of a misnomer. Community is not defined by being closed. The closed -- By locking others out, I do not create the community. And I think many of the communities who want to go for new TLDs would consider themselves to be open, and they would not have policies that would say, this person is in and this person is out as a registrant. They would have policies as to what the domains should be used for. >>KURT PRITZ: That's interesting. >>WERNER STAUB: The use is the term, in fact, and not who is in and who is out. Now, if you look at the word, it comes later in the evaluation criteria, and I think, really, there it was misunderstood, because people are getting points for locking out. And that should be corrected, first in the word, and then in the conclusion that we derive from it. >>KURT PRITZ: Thank you. >>AVRI DORIA: Paul, was your hand up as a follow-up to that? >>PAUL STAHURA: No. >>AVRI DORIA: No? Don't let me force you to talk. Okay. I'll move on with the list, then. Mike. >>MIKE RODENBAUGH: Actually, my questions are for Module 2. >>AVRI DORIA: Okay. Hopefully, we'll get there real soon now. And the last one I had on my list was Claudio. >>CLAUDIO DE GANGI: Thanks, Avri, Claudio from the IPC. Page 1-13, it makes reference to post delegation contractual restrictions. Are those reflected in the base agreement? >>KURT PRITZ: They are. There's a placeholder for them. So in two or three cases, I think (inaudible) there's -- the base agreement calls out that these post delegation procedures can be developed and included in the contract. >>AVRI DORIA: Okay. Having ended the list for Module 1, I'd like to start with Module 2. I've already got Mike's name on it, because he's -- Now, what we said is he's just going to flash the slides, not going to talk. And I'll start getting names down on the list so that -- so I've got Mike, I've got Mike R., Mike P., I've got Jeff. I've got Tim. I've got Chuck. >>KURT PRITZ: So there's -- as alluded to earlier, there's a six-step review in the evaluation process. There's -- the string is reviewed for certain aspects, and then the applicant himself or herself or itself -- itself, right -- is reviewed for technical capability, financial capability, and registry services offered. And this slide also shows the policy recommendation numbers. There is several specific criteria and procedures developed just for this. And these are all described, all these -- all these criteria and procedures are described. And then there are some really good slides here that I spent some time working on. >>AVRI DORIA: But we'll get to see them on Monday. [ Laughter ] >>KURT PRITZ: That -- Well, you know, actually, these are the ones you don't get to see. >>AVRI DORIA: Oh. >>KURT PRITZ: Because it's really quick. So it describes that DNS stability looks at IDNs, too, and there might be a follow-on. String confusion describes the fact that we've got this preproduction algorithm that will inform the work of a string examination panel. Geographic names that I'm sure we're going to talk about in just a minute. There's a fairly complete procedure and a complete memoranda on the thought that went into geographical names and the balancing there. Then there is the technical and operation capability set of questions that you can read. There's 20 questions, each one scored. You've got to get 20 points. A couple of the questions are optional. The questions are scored 0, 1, and 2. And that will have some predelegation testing. And financial capability, there's 11 questions that are sort of remixed to get seven scored criteria. So you have to get nine points there. And there's -- And we really are emphasizing registrant protection here. So there's a lot of questions about contingency planning, failure scenarios, creating some sort of financial instrument so that the registry operation continues on in the event of failure. That's a way to win, right, if you read the criteria, if you provide that financial instrument, then that's a path to -- that's a path to passing. Because what we care about is protection of the registrant. And then a registry services review, which is very similar to the existing process for reviewing registry services. So those are the aspects of Module 2, and the six-step evaluation that occurs there. >>AVRI DORIA: Thank you. Okay. Mike R. >>MIKE RODENBAUGH: To the surprise of nobody, I'll focus on the confusing similarity aspects here. The algorithm, the magical algorithm. What percentage will trigger rejection in the initial evaluation fees. And also just a subquestion on that is proximity on the keyboard, at least on the QWERTY keyboard, which is, obviously, the dominant one, is that a factor in the algorithm. >>KURT PRITZ: So confusion is a human trait. And the score will serve to inform panel members, provide some objective measure of how close strings are in -- visually. So the algorithm actually does an alphabetic similarity check and a visual similarity check. Craig, do you want to talk about this for two minutes, how the visual similarity check is done? But it has nothing to do with how close the keys are on the keyboard. It has to do with how similar the characters look. >>CRAIG SCHWARTZ: Sure. The way the algorithm works is that the string is compared based on the number of characters within the string, based on the number of similar letters within the string. It also looks at the number of letters that are in the same sequence within the string. It looks at common prefixes. It looks at common suffixes and the lengths of those prefixes and suffixes, and then assigns a score based upon those variables. It also looks at, for example, how similar a letter "B" is to a letter "D" and assigns a score to that, same as a letter "O" and the number zero, and assigns a score to that. So there are a lot of letters, both intrascript and intra-- I thought you had a question. The scoring is both intrascript and interscript. So letters within the Cyrillic alphabet that look similar to letters within the Latin alphabet will also have scores, comparison scores. So between the variables within the string itself, all those -- all those items that I first mentioned when we started talking, and then a combination of the scores of individual characters compared to other individual characters, the algorithm produces this measurement, and, as Kurt said, will be used as a guide to the string examination panel. >>MIKE RODENBAUGH: Okay. Why is this limited to visual similarity only at this stage? Our recommendation was, you know, much debated, but very clear in the end that it should include all sorts of confusing similarity. Frankly, I think the QWERTY keyboard proximity issue is a very, very important one also, so I'd urge you to think about figuring that into the algorithm. A lot of money can be made off dot com typos, as we know. But, really, why is it visual only? And the only other question I will have and then I'm done, is why is there no appeal from rejection at this phase? >>KURT PRITZ: So I -- just to cut to it, in the development of the -- this examination and the objection process for confusingly similar, the discussion with the council here and others publicly has always been that the initial evaluation would look for visual similarity, but objections could be made on any kind of similarity. So that's been the model that's advertised all along. And so the objection-based process accommodates an objection for any type of similarity, you know, or appearance or meaning. And follows -- you know, it's not a trademark law kind of inquiry, but when we did our examination of what kind of similarities there were and how those comparisons will be made and the guidance to be furnished to examiners, all those rules are fairly well established and will be followed. >>MIKE RODENBAUGH: But take, for example, meaning, similarity of meaning. That's something that you could very easily determine at the outset and save, for example, dot travel registry from having to file and pay for an objection on dot viajes. So I don't understand why you wouldn't just knock it out early, give a refund to the person who applied, you know, obviously, not a full refund. And then it's over. Instead, you're causing delay and basically waste. And then what about the appeal issue also? Why would this not be subject to some sort of appeal mechanism? >>KURT PRITZ: So that's a very good question. And the development envisioned how an appeal might work. And if you get a different answer than to the initial query, then what's the effect of that different answer and the chance, though -- you know, in designing the process, how that might sort of auger into circles of appeals that aren't resolve. And so an alternative to that is, you know, because this is at a very early stage, if, in fact, two strings are so visually similar that it's probable that user confusion would result, you know, the application could be withdrawn and the applicant could get a refund for, you know, a significant portion of the application fee. So that would be a very good comment to make, because that's a -- that was a lot of discussion about how to go forward with a proposed model. And that's exactly what -- this is exactly the kind of discussion. So what arguments you have in favor of an appeal and how that would work might be really helpful. >>AVRI DORIA: Okay. We had two follow-up requests on that issue, one from Chuck and one from Edmon. >>CHUCK GOMES: Yeah. One -- Several points with regard to the confusing similarity. Number one, I think it needs to be very -- made very clear to applicants that the definition of confusing similarity is not just visual. I understand that the algorithm can't do much more than that. And that's fine. But it's very important to communicate that the definition of confusing similarity goes well beyond visual. So just make it clear. 'Cause it could be easily concluded, because the algorithm only covers visual confusion, that that's all that matters, which is not the case from the GNSO recommendation, as Mike pointed out. Secondly, I think he makes a very good suggestion with regard to if some of the other forms of confusion could be weeded out earlier rather than using the dispute process, it would make a lot of sense. And that isn't necessarily all that hard, I don't believe. So that would be, I think, a very good recommendation. >>AVRI DORIA: Okay. I have Edmon, and then I have Paul still on follow-up. >>EDMON CHUNG: I actually have a question that's following from him and then also another question on this module, so could you put me -- >>AVRI DORIA: Okay. I'll put you in the queue for the other question. >>EDMON CHUNG: Okay. Mike mentioned about the -- how the percentages will be used. I'm not sure I got a clear answer for that, you know, what percentage as a result of that algorithm would trigger further discussion. Or is there any at least a range of that? 'Cause when I'm trying it out, I guess when applicants are trying out, they want to see, "Okay, it's like 50%." Is that something that would trigger some further -- that's pretty much the.... >>KURT PRITZ: So again, and particularly in discussions with this group, it's been conveyed that confusion is a very human reaction. And also -- and maybe Amy or somebody else can speak more clearly on this -- our investigation indicates that while this isn't a trademark kind of examination, trademark examinations very similar to this take place where the examiner uses the result of an algorithm. We're using an algorithm that was designed for this purpose by experts in this area that advise national trademark offices, use this information, and then make -- you know, examiners sit in their offices and make judgments regarding these. >>EDMON CHUNG: Sure. I understand. And exactly because they are experts and they probably have some statistics of, you know, around the range where it would be more likely that there would be a problem. So I think it would be useful for applicants to know that. >>KURT PRITZ: I don't know. My fond dream would be that someday the algorithm would be so reliable that it becomes a very objective test. But that's far off. >>EDMON CHUNG: No, I understand. It's not aimed to be that I would have. But I guess my guess is -- this is just an idea, like, if you tried it out and it's, like, between 50 and 70%, it's 90% chance that you -- you know, it would be like this. 'Cause they are -- as you mentioned, they are the experts in this area. They should have some -- you know, some idea of what the number means. >>AVRI DORIA: Thanks. Paul, you had a follow-up on this one? >>PAUL STAHURA: Yeah. Back to string confusion, or sticking with string confusion. I'm making sure I have it right. It sounds like that there are two definitions of string confusion, one that's used for objections and one that's used for string contention, with the string contention one being visually similar and the objection one being more than visually similar. Is that right? >>KURT PRITZ: Yes, that's what -- how it's designed right now. I withdraw my -- wait. I -- can you ask your question again. >>AVRI DORIA: I think Olof might have a -- I guess ask the question again. >>OLOF NORDLING: Well, let's put it like this, that if an objection leads to a conclusion, that, yes, they're confusingly similarity, it adds on to the already established contention sets. So it will be part and parcel of the contention set handling. So, from that perspective, they may come from two different sources, one from the initial evaluation and the feignings there, and the other from the objection process, but they'll converge into a single handling of the contention set. >>PAUL STAHURA: Yeah, that's my understanding. But if there is no objection, then it's visually similar is the bar level, not more stringent than that? >>AVRI DORIA: Okay. Moving down the queue, I've got -- yes, Mike P. >>MICHAEL PALAGE: Thank you. Mike Palage speaking as an individual who will be assisting clients in navigating this process, as if anyone was in doubt about that at the beginning. My first question is with regard to section 2.1.1.2, which appears on page 2-4, "Reserved Names." And, Kurt, this question is directed to you. I was looking at this reserved table, and I was comparing it to the original table that was produced from the GNSO Reserved Name Working Group. And I did not see -- and I could be wrong on this -- but I did not see our original recommendations, including ALAC, GAC, NRO, RSSAC, and SSAC, as being in the original recommendation from the GNSO Council when we produced that working group document. I was wondering how those strings got added to the list -- >>KURT PRITZ: How? >>MICHAEL PALAGE: How GAC, ALAC, and -- >>KURT PRITZ: Oh, how they were added to the list without -- >>MICHAEL PALAGE: Yes. >>KURT PRITZ: -- without being part of the Reserved Names Working Group. They were viewed as names very similar to those already on the existing list. >>MICHAEL PALAGE: Okay. So staff made that determination? >>KURT PRITZ: Right. For public comment. And there's -- there was a lot of staff discussion about reserved names list at the first and second level. And I remember we had this discussion during the Reserved Names Working Group on these specific names and didn't quite resolve it. So I would like to have that discussion, because there was, you know, quite a discussion at the ICANN staff level about, you know, whether the reserved names list should be more extensive or way skinnied down, and arguments on both sides. >>MICHAEL PALAGE: So, again, as someone who kind of went through line by line and noticed this, it would be helpful, if staff did deviate from the original GNSO recommendations, if they could call that out. The one follow-up point on this first question I'd like to raise is, in this document, ICANN talks about preserving and respecting the rights of intellectual property owners. Just for the record, there are at least two U.S. trademark applications for the string "GAC." There also is a trademark application in the United States Patent and Trademark Office for "NRO." So perhaps staff might be able to reconcile how, on one hand, they are saying they want to protect the rights of intellectual property owners, when they are, if you will, adding, without -- I appreciate this notice right now, but they did take steps and recommendations to add to a list which was not originally proposed by the community. That's question one. Next is question regarding section 2.1.1.1 appearing on page two three. In this -- it says an application that failings the string contention review, and it's too familiar, it says there will be new further reviews available. This follows up, I think, on Mike Rodenbaugh's original question, not so much with regard to the applicant but also the potential challenger. Will ICANN impose any limitation on a challenger's right to seek redress through a court of proper jurisdiction? So in the UDRP, if you lose, you have legal recourse to go to a court of competent jurisdiction to enforce your rights. Does ICANN envision imposing any limitations on redress for a challenger that does not prevail? Marilyn, since you seem confused, let me explain it for you. Sometimes what happens is if you participate, say in binding mediation or something like that, that's it. That's your one bite of the apple. You don't get any additional recourse. My question again, just for purposes of clarity, I do not want an intellectual property holder trying to participate in an administrative dispute process, because it might be cheaper, to somehow limit their rights to go to a court. It's not clearly addressed. I want to make sure that that is not preventing a third party from seeking redress through a court. >>AVRI DORIA: Any comment? So I didn't hear, and I'm sure other people didn't hear whatever it was. >>DAN HALLORAN: I was explaining to Kurt that we will look into that. >>AVRI DORIA: Oh, you will look into that. >>MARILYN CADE: That's a question I had, Mike, to raise I thought tomorrow, but it's worth just making a comment. When we set up the UDRP, there was -- there was some discussion about limiting contracting away your right to go after the UDRP to a court of competent jurisdiction. And that idea was rejected. So in the UDRP -- and I would just point this out to the ICANN staff -- in the UDRP, even though you go through the UDRP, you do have the right to appeal that decision to a court of competent jurisdiction. >>MICHAEL PALAGE: There we go. Marilyn and I are in agreement. >>AVRI DORIA: And Jon had a follow-up on this. >>JON BING: Only to point out that the UDRP is between people or actors which has a contract with ICANN. And indirectly with each other. But your question is, of course, addressed in a situation of where a third party has no contract apart from being a party of the dispute resolution or the confusingly similar discussion, and obviously we cannot limit the jurisdiction of court and the possibilities through this instrument. >>MICHAEL PALAGE: And we can talk about this off-line, we have limited time, but we can talk about the nuances of the UDRP. My next question is with regard to 2.1.1.4.2 that appears on page 2-10. And this is with regard to the geographical names panel, the GNP. Could ICANN at this time elaborate on whether they view this as an internal or an external body? Is there -- Could you provide any elaboration at this time on that panel, whether it's going to be one person? Three? Five? >>KURT PRITZ: Certainly it would be external. It would have to have members with -- I'm going to say this the wrong way, but with multilingual and linguistic expertise. So it's an expansive sort of expertise that needs to be retained. >>MICHAEL PALAGE: Okay. My final question is, with regard to section 2.1.1.1 appearing on page 2-2 and this again, if you will, revisits one of the points that Mike Rodenbaugh had raised. With regard to the current -- The staff document talks about the objection of this review is to prevent user confusion and loss of confidence in the DNS. Now, as we have heard, confidence in the DNS is just not visual. We heard earlier from Ron Andruff about the potential need of transliterations protecting them. There is under article 10 BIS of the Paris convention regarding unfair competition. Also under U.S. law there is a recognition of the doctrine of equivalence. Therefore, getting back to this intersection of IDN equivalence and transliterations, I do think the current definition is narrowly construed, and I think has been articulated before but I just wanted to get those legal points on the record from the prior discussion. Thank you. >>KURT PRITZ: So to go back to your first comment, I think -- I think -- about the rights of a complainer or a challenger to pursue other remedies, I think it's -- the balancing to be done is we want to encourage people to use the challenge process; right? So the balancing to be done, then, is if we were to preclude their rights to -- if they have to sign away their rights somehow in order to enter the challenge process, they would be more reticent to do that. So we certainly want to encourage use of the challenge process in that way. You know, the other side of it, though, is, you know, I want to limit litigation if we can. >>MICHAEL PALAGE: If I can, let me just give you a real-life experience on that. I testified as an expert on behalf of the NCAA in marchmadness.com. When I testified, that was -- this was back when NSI, Phil Sbarbaro, after the UDRP was implemented, he sent out notices saying I think you have third 30 to 60 days after the passage of the UDRP to make a determination do you want to use the UDRP or else release it. And at that particular time, the NCAA said there's no precedent to go on, we haven't seen any decisions. Given that begun to the head scenario, they elected to go to federal court in Texas to resolve their dispute. So I just want to point out the realities of why some clients may choose traditional courts as opposed to an administrative process. >>AVRI DORIA: Okay, thank you. At the moment I have got Jeff, Tim, Chuck, Philip, and Edmon. And I've gotten minutes. So Jeff. >>JEFF NEUMAN: All right. I will try to be brief, although I wanted to start with a little history. But -- sorry. Actually, there is a little bit. Have you guys started the process to look for evaluators? Who are they going to be? How are you going to find them? As a company that was let's just say affected by the evaluators that you had selected for dot net, when you selected a competitor of NeuStar, what are you doing to find evaluators to run that through the public and to let people know who the evaluators are going to be prior to the application period? >>KARLA VALENTE: We plan on starting procurement activities sometime in the next quarter. And it's going to be an open, transparent process and we are going to define exactly what evaluators are needed, what we expect them to do, et cetera >>JEFF NEUMAN: I guess more basic than that, are you looking for an organization? Are you looking for individual people? >>KARLA VALENTE: So you will see, when we initiate the procurement activities, we will have specific criteria what we are looking for. In some situations, I think it might be an organization. In some situations, it might be individuals with a specific expertise, either gathering for an organization or individually. >>JEFF NEUMAN: Will you let people know prior to the application period who those evaluators are? Is there a plan to announce who those evaluators are prior to the application period? >>KARLA VALENTE: Yes. It's going to be an open -- So procurement activities will be -- we initiative sometime in the next coming months, and once the selection is done, we will announce who. >>JEFF NEUMAN: But once the selection -- sorry to harp on this, but just to review that what happened with dot net, is everyone applied, paid their fees, and then it wasn't announced who the evaluators are until after everyone had applied and they were in their evaluation period. I would like to avoid that because it was difficult for some of the parties involved. So what you are saying is you will have the process, you will select the evaluators, you will announce the evaluators prior to the submission of applications. >>KARLA VALENTE: That's right. >>AVRI DORIA: Marilyn, you had a quick follow-up to this? >>MARILYN CADE: I do and it directly relates to the points Jeff is making. So when you announce who the evaluators are, if the community comes forward with evidence that some of the evaluators have a relationship of some kind, they would be then removed from being involved in that evaluation. Is that right? >>KARLA VALENTE: Conflict of interest? >>MARILYN CADE: Yeah. And that could be that someone on their board is on a related board. I mean, there's lots of ways to have a conflict. >>AVRI DORIA: Are you looking for automatic removal or just that you would review any of that other evidence that came up? >>MARILYN CADE: No, I think that given how important these decisions are, and given that we are outsourcing, we are outsourcing a number of these decisions, we are removing ICANN from any accountability by doing that, because ICANN has put together a process that needs to be trusted. So my question is, if you publish the list of dispute resolvers and the community comes forward and says in the approved list of evaluators, there's a conflict of this nature, is there a process to change that so that the party who is being evaluated does not feel they are being disadvantaged, and is not disadvantaged? >>KARLA VALENTE: So my recommendation is that you put that into the public comments. I think it's an excellent point. >>AVRI DORIA: Thank you. Moving on, Tim. Oh, you had a follow-up on that one? Sorry. Eric. >>ERIC BRUNNER WILLIAMS: Thank you. Eric Brunner Williams. Following up on Jeff, it wasn't just the dot net bid where this problem existed. It existed also with the dot org bid. And what we haven't done as a body is we haven't attempted to draw lessons learned from our two previous failures to do technical evaluations. So when you say you are going to have an open, transparent process and release an RFP, I'm not sure that you are going to be asking the right questions yet. If you say you would like a company of some particular size with some particular set of tail fins and chrome, I'm not sure that there's a right criteria for selecting an evaluator. So until we have learned our lessons from net and org, I'm concerned that the lease of an RFP, as transparent and as well meaning as it may be, may still result in a very -- in a process that we aren't proud of after it's over and will lead to complications. >>AVRI DORIA: Thank you. Any comment? No. Okay. Tim. >>TIM RUIZ: (Inaudible.) >>AVRI DORIA: Okay. Thank you. Chuck. >>CHUCK GOMES: I have several questions, so it may take us past our deadline. Is that okay, Madam Chair? >>AVRI DORIA: Yeah, I don't know in terms of the deadline. It affects a lot of people other than us who have been transcribing all of this time. So we really do have to finish, and we will have to continue. So we'll take your questions up until 5:30 and then -- >>CHUCK GOMES: Okay. Thank you. The first -- there are actually three questions that are part of the first category. On page 2-4, the similarity to TLD strings applied for as ccTLDs. The first question is, I assume that it's -- that means fast-track IDN ccTLDs, for this round anyway. And a second part of the question has to do with timing of the fast-track and the gTLD IDNs, because this part of the evaluation actually presumes that you know what the fast-track applications are. Now, there's two ways that could happen. Either the ccTLD IDNs go first, which as you know, the GNSO strongly opposes, or they go in parallel. So is it fair to assume that the plan goes in parallel and are timed in such a way so that you will know what the applications are. And then the third part of that question is will ccTLDs be compared to gTLD strings to see if there is any confusion in that regard? >>KURT PRITZ: What was your very first question? >>CHUCK GOMES: Sorry about that, Kurt. The -- Well, you were talking about fast-track. >>KURT PRITZ: So, yes, we are talking about fast-track. >>CHUCK GOMES: That was the easiest question. >>KURT PRITZ: Yes. And it's -- There are two independent processes that are coming to a conclusion at essentially the same time if both time lines that are envisioned are accomplished, the fast-track process and the new gTLD process. So there would be applications for both, I think, at the same time, although the application periods of the fast-track process and the gTLD process I am sure will be different. And so there will be complexities introduced by the extent to which the processes run at the same time and to the extent that they differ. It's envisioned that the applications from both processes be examined for string confusion. >>CHUCK GOMES: And one follow-up there. Because in the gTLD recommendations, we reserved, or we're -- we reserved the two-country -- >>KURT PRITZ: Two letter. >>CHUCK GOMES: Two-letter codes, so we are covered there, so there doesn't need to be any check on the -- so this would only apply to IDNs; is that right? Yeah. >>KURT PRITZ: I think that's correct. >>AVRI DORIA: Okay. I realize you probably have more questions, Chuck. >>CHUCK GOMES: I do. >>AVRI DORIA: But at this point -- >>KURT PRITZ: Can we do some geographical names questions before we wrap up? >>AVRI DORIA: Okay. We were going to wrap up now because it was 29, but what did you want to -- >>KURT PRITZ: Chuck probably had some. >>CHUCK GOMES: I was going to actually leave that to others. I had lots of notes on that part because obviously, as I think everyone knows, the implementation plan with regard to geographic names is not consistent with the GNSO recommendation. So -- >>AVRI DORIA: There wasn't a lot to ask. >>CHUCK GOMES: Yeah. >>AVRI DORIA: Basically there's probably a lot of discussions. I think that's one of the things that will come out Wednesday. It's certainly one of the things that will come out in the discussion with the GAC because we know their principles and their questions to the board differ from our recommendations. So the fact that there's variance there has, I think, definitely been noticed. I don't know if anybody had questions on it. But since we did have to break at 5:30, I have got a list that -- I don't know if Chuck was finished. >>CHUCK GOMES: No. >>AVRI DORIA: So Chuck starts out when we meet again tomorrow morning, and I think at this point we should just continue walking through this way and asking questions. And I have Chuck, Philip, and Edmon as the head-off on the list. Yes? >>ADRIAN KINDERIS: Can you add me to the list? (inaudible). [ Laughter ] >>AVRI DORIA: Okay. No one is asking you to prove you have a question ready at the moment. So was your name going on the list? >>ERIC BRUNNER WILLIAMS: I have two questions ready. >>AVRI DORIA: You have two questions ready. I am still going to let him go first because I saw his hand first. >>JEFF NEUMAN: Add me to the list. >>AVRI DORIA: Okay. And the list will still be openable tomorrow morning. There is not going to be a line drawn under it. So I thank everybody for their questions and patience and good humor, and see you all tomorrow morning. Thank you. Thank you. See you tomorrow. [ Applause ]