ICANN|ARCHIVES

Internet Corporation for Assigned Names and Numbers

File 0114, GNSO

Lisbon, 25 March 2007, Sunday afternoon

>>BRUCE TONKIN: Let's try and sketch out a diagram for the process for string contention.

So we've got a diagram at the moment that sets out the process for the string checks. And that's the diagram that we completed when we were in Marina Del Rey.

So this is trying to map out the process, as I understood the decisions in Marina Del Rey.

And I want to kind of validate it with this group here.

So firstly, we receive applications and we post applications. And then the first thing the staff need to do is review these applications for the same string or applications for strings that are confusingly similar.

Now, I think it's reasonable for the staff to make a judgment of that, but they're still receiving public comment here. Because at this point, the applications have been posted. So even if the staff didn't think two strings were confusingly similar, other parties could obviously say, hey, we reckon, you know, string number five and string number six, those look confusingly similar.

So that then gets it into this process, which is contact the applicants and advise them of this contention. And then applicants have -- and I'm just putting a timeline -- actually, you might want to strip that time out there, because we just want to get a general idea here.

But applicants have some time to conduct meetings amongst themselves to try and resolve that. So if there was, you know, two people applying for dot bank and, you know, it was the banking organization of China and the banking organization of the U.S. or something, they could get together and say, how about we jointly run dot bank, and agree on the rules. And they have a period of time to do that.

If those applicants then reach a resolution, then that would inform the staff of that decision. And that would go into, essentially, the initial evaluation report, which is part of the string check process, essentially.

>>CHUCK GOMES: Bruce, do you want discussion as we go or do you want us to wait?

>>BRUCE TONKIN: No. Go ahead.

>>CHUCK GOMES: The question that just popped into my mind, do we want any evaluation to occur to see if minimal criteria have been met before they -- to see whether we really need contention resolution? Do you follow me?

If, in fact, somebody didn't meet the basic criteria of the application -- certainly want to know the application is complete and so forth. But -- or do we want this to just happen right away?

I don't know what the answer is. But --

>>BRUCE TONKIN: Yeah. Well, I think the issue there, Chuck, maybe we'll come back with it.

Is it worthwhile putting it against the other diagram that you have, just so that we can see it.

I might not be able to find it now.

Yeah, and this is one of the things I was talking to Liz about yesterday. I was -- I don't want to make the other diagrams so incredibly complicated by trying to incorporate this in, but I've got to try to link it into the paths.

>>CHUCK GOMES: There may be cases where contention is eliminated because of -- an application is not adequate.

And at the same time, we're concerned about timely process. So we're kind of balancing the two things.

>>BRUCE TONKIN: Yeah, okay. So Liz has just said it's in the report.

>>LIZ WILLIAMS: If you get your report, the other diagram that Bruce is referring to is in the back sections of the report. Yeah, so refer to that rather than having two things on the screen.

>>BRUCE TONKIN: Can you just grab the report? Sorry, that's something I'll grab off the table here.

So --

>>CHUCK GOMES: (inaudible).

>>BRUCE TONKIN: So the way I see this, Chuck, is this is probably in what I'd call the initial evaluation period that this initial contention resolution is happening . And then so the way I'm very roughly -- and we need to kind of line up the terminology here. What's happened to -- I've unplugged the other end.

So I'm thinking that during the initial evaluation in here, where we're looking at string checks and stuff, a lightweight contention resolution process, being the parties get together and have a chat, I think is okay.

And then what you're saying, basically, is that --

>>CHUCK GOMES: I mean, if I'm an applicant and -- I may not be as willing to compromise and come to resolution with another player if there's a chance that maybe that other player's application was incomplete or -- now, that should have already been covered, based --

>>BRUCE TONKIN: Yes.

>>CHUCK GOMES: But maybe technical evaluation or something like that.

>>BRUCE TONKIN: Yeah.

So I think the concept here, as I understood what the committee was getting at, is at the earliest possible opportunity, we're encouraging these people to get together and talk. So during the period of time when all of this is happening, the idea is that those parties should talk and try to work together.

And if they do come to some agreement, then that then becomes input into the initial evaluation report, which is actually at the bottom of the -- in this first box, basically.

>>CHUCK GOMES: Let me ask another question.

If we go back to the application submission, or it's actually, on here -- yeah, it is the application submission period, if one applicant's application is incomplete and the other one is complete, do we then not have a situation of contention? Or is opportunity given for them to make it complete?

>>LIZ WILLIAMS: This is where I thought that the applications -- and there may well be five in this system. It's not just between two -- but they would be equal in every respect. They've passed their technical, they've passed their financial and business. They've passed everything. And this is getting to the point where we have to make a decision between qualified applicants.

>>BRUCE TONKIN: But I think the distinction I'm making here, Liz, is this is where we've posted thesis applications and we're then notifying these guys they should talk to each other. We're not going into an extended phase, which is the next thing in this diagram.

So I think before we go into the extended phase, they should be validated, anything that goes into the extended contention processes would have to have gotten through the process here.

>>CHUCK GOMES: I'm okay with that. But we still have to know what the answer is to the question, if you have two applications for the same string, one of them's a complete application, one of them's incomplete, --

>>BRUCE TONKIN: Yeah.

>>CHUCK GOMES: -- does that eliminate the contention at that point?

>> Yes.

>>PHILIP SHEPPARD: I think it has to, actually. Sympathetic as I am to the idea of trying to parallel process it so you give valid applicants the opportunity to do that before you go into a full evaluation phase, if you do try and do it early, you're likely to be putting potentially unequal partners negotiating.

>>CHUCK GOMES: That's the place where I was coming from.

>>PHILIP SHEPPARD: Which is a bit bizarre. You tell people, we haven't had an evaluation yet, but you seem to be in contention with somebody else, so go away and talk about it. And they come to some great agreement. And then you suddenly say, "well,

actually," --

>>BRUCE TONKIN: These guys didn't have standing.

>>PHILIP SHEPPARD: -- "party A doesn't qualify, anyway. So you've wasted your time."

>>BRUCE TONKIN: Basically, if I just try and map this with the one that we've got at the moment. We've got three phases, we've got applications posted. Then we've got string and applicant evaluation. And then we have a draft eval report.

So what you're saying is, in parallel -- I've got these other processes looking at contention.

So I guess what you're saying is that you could give notice of contention here, but their time to resolve the contention starts there.

So this is sort of like a give notice, which roughly lines up to that draft initial report. And then at this stage, you're saying, "Okay, I've got three applications that the staff have evaluated and met these requirements." And then we're saying, "Okay, now you guys are in this string contention flow."

Is that what you're suggesting? And I think that helps answer Chuck's question.

>>PHILIP SHEPPARD: Yes. I think it makes sense to give notice, because then they can start to think, "Oh, my gosh, we might have an issue." And therefore you can plan for it.

>>BRUCE TONKIN: Yes.

>>PHILIP SHEPPARD: But you can't actually start anything until you know you are potentially equal partners.

>>BRUCE TONKIN: Okay. So --

>>PHILIP SHEPPARD: Or valid applicants.

>>BRUCE TONKIN: -- so what we need to then break this diagram up is say we're reviewing applications. And I guess this is where the "confusingly similar" comes in this string check here. Because that's one of those checks. So I guess we'll identify the confusingly similar as part of this as well.

So it may be this, then, needs to start. We can contact the applicants, I think, and leastwise to advise contention.

And so what we'd be saying is that the draft evaluation report essentially publicly identifies that contention at that point. And the need to go into contention resolution, that would be in the draft evaluation report, that says we've done our report, we've done the technical, financial, operational checks. There's no technical issues with the string, et cetera, et cetera.

And now we've got to the draft initial evaluation report. And then it's saying, okay, we need to go into string contention mode, effectively.

Does that make sense?

>>WERNER STAUB: Let me make sure I understand correctly.

Is it that before the applicants negotiate, we have a report? Or is it after that?

>>BRUCE TONKIN: Different reports. I was saying that we now have a draft evaluation report, which is basically the staff evaluation of those applications. And let's say I -- I might have had two applications that were here, and maybe only one gets through here.

In that case, there's no contention, so that's fine.

Alternatively, I have three here. One gets knocked out, but I'm still left with two. So in the draft initial evaluation report, I'd say, okay, I now have two valid applications that have contention. And then I'm saying, right. Now you have -- you have 30 days to conduct your meetings. And if they reach a resolution, they basically report that back to the staff. If they don't reach the resolution, we'll go into that next step in the second.

>>WERNER STAUB: The question is, why would the 30 days be prescribed? I've once been unfortunate enough to be in an arbitration. But the good thing was that the parties could actually agree on a timeline. So, basically, nobody forced the parties to -- if they agree to negotiate longer, nobody forced them to negotiate within so many days if they needed more time.

>>CHUCK GOMES: Yeah, we could probably have an option for them, upon mutual agreement. Because, otherwise, if you don't have a time set --

>>WERNER STAUB: Indeed, yes.

>>CHUCK GOMES: -- one player could play games to delay.

>>WERNER STAUB: Yes.

>>BRUCE TONKIN: So --

>>LIZ WILLIAMS: Just -- how do you want to represent that there? That square box --

>>BRUCE TONKIN: What I want to start with, Liz, is take out the 30 days.

>>LIZ WILLIAMS: Which I have done already.

>>BRUCE TONKIN: Then we can just say applicants have a -- what is it, a minimum time or something? And -- or otherwise agreed amongst the applicants.

>>CHUCK GOMES: Why do you want to take out the 30 days? I think you can accommodate what Werner is saying without doing that.

They have 30 days to come to the agreement. And the agreement could be that they would like more time to work.

>>BRUCE TONKIN: Yeah.

>>CHUCK GOMES: As long as both or all parties agree to that, that's okay. But if you don't put a time limit there, then the game-playing can begin by a party that just wants to delay.

>>BRUCE TONKIN: Let's say applicants have 30 days or as otherwise agreed amongst those applicants.

>>PHILIP SHEPPARD: Unless subject to mutual agreement.

>>BRUCE TONKIN: Yeah.

>>PHILIP SHEPPARD: -- applicants have 30 days.

>>BRUCE TONKIN: Yep. Okay.

So can you make these changes as we go, Liz, and then we'll do another look at the screen.

>>LIZ WILLIAMS: If you could just go a little bit slowly. Philip, just come back. You said unless.

>>PHILIP SHEPPARD: Unless subject to mutual agreement.

>> Do I have the little -- do you have --

>>BRUCE TONKIN: Just for useful --

I want to make sure we capture these changes in the way that we're agreeing.

>>LIZ WILLIAMS: Thanks, Craig.

Could you -- it's.

>>BRUCE TONKIN: Okay. So we're changing this text to say applicants have 30 days to conduct meetings or otherwise by mutual agreement, or words to that effect.

>>LIZ WILLIAMS: It's unless agreed otherwise, applicants have, da de da de da.

>>PHILIP SHEPPARD: No, unless subject to mutual agreement, applicants have.

>>BRUCE TONKIN: Yeah. Can you --

Philip, would you mind sitting next to Liz. You're probably pretty good with that.

>>LIZ WILLIAMS: I don't want him sitting next to me.

>>DAN HALLORAN: Bruce, are we saying we're going to leave it completely open ended? Like, two applicants could just negotiate for two years, if they wanted to, and get back to us later.

>>BRUCE TONKIN: Yeah, because basically you're then locking out other people, I guess.

>>DAN HALLORAN: The round would never close. Would we go into the -- we have, like, these timelines with the final evaluation, initial evaluation.

>>WERNER STAUB: Just maybe probably a cap on length anyway. Like a child is in school and has to repeat, at some point, you basically says no child is going to be forced to do more than nine years of school. And then if it doesn't work out, fine.

>>BRUCE TONKIN: So you want to put a maximum limit on it, essentially.

I think -- I kind of think Dan's got a good point. I think, ultimately, we're trying to bound these processes somehow. Because otherwise the staff are going to have to keep track of this for years, as well.

>>DAN HALLORAN: We've had some experience with, just off the top of my head --

>>BRUCE TONKIN: Yes, you have.

>>DAN HALLORAN: Well, think of pro and --

>>WERNER STAUB: Post.

>>DAN HALLORAN: Yeah, post. Where you get an application and it just kind of festers for a while.

>>WERNER STAUB: We have another application opened from 2000.

>>BRUCE TONKIN: Why don't we cap it, then, at --

>>PHILIP SHEPPARD: There's good reason for that.

>> Not in a bad way. Just takes a long time.

>>BRUCE TONKIN: This might get into implementation areas, so I don't want to complicate the diagram, but I think it's maybe reasonable if we cap it at six months, unless agreed by ICANN. So the applicants might have some flexibility between 30 and, you know, six months. And then other than that, they go to ICANN and say, "Hey, actually, we need nine months," and ICANN might agree to that. But it's a step that ICANN can approve, basically.

>>LIZ WILLIAMS: So capped at six months?

>>BRUCE TONKIN: Yeah.

I'm just trying to (inaudible) dates here, I don't really care what the number is. It just isn't six years.

>>CHUCK GOMES: And one of the factors that would be involved there would be the fact that they're going to be dropped down in the queue in terms of processing order while --

>>BRUCE TONKIN: Yes.

>>CHUCK GOMES: -- this is going on. So the longer they delay, the further down they are in the queue.

>>BRUCE TONKIN: Yeah. I would say it becomes a bit like WHOIS. If they're still negotiating after 18 months, they're probably not going to reach agreement.

Yeah. Olof.

>>OLOF NORDLING: I think we should add some precision that says there could be more than one -- more than two contenders, just to say that unless otherwise mutually agreed to among other contenders.

So it's not just mutually agreed among a couple of or two or three of them. But they all --

>>BRUCE TONKIN: So wouldn't mutual in the context of contention still cover that, though? If I've got four people and I'm having to get agreement, mutual doesn't mean all of them, does it?

>>OLOF NORDLING: Does?

>>BRUCE TONKIN: What do you think, Steve?

>>OLOF NORDLING: Okay.

>>BRUCE TONKIN: Change all sorts of stuff.

I was thinking, that's great. My computer's doing work on its own.

>>DAN HALLORAN: One more, if I could, Bruce, just kind of implementation question.

We went through a lot of detail on the other flowchart about when staff looks at it or when we leave it to other people to jump in and object. And this is one where I'm wondering, is it just staff that can decide if there's contention.

>>BRUCE TONKIN: No.

>>DAN HALLORAN: Or can it also be applicants? Can it be a third party --

>>BRUCE TONKIN: Remember, it's going through -- if you look at this diagram, Dan, this is actually happening after it's got to the draft initial report.

>>DAN HALLORAN: Right.

>>BRUCE TONKIN: So people would be in here, under these sort of string check processes under here. That's, you know, the public comment. And other (inaudible). So --

It's as -- when you've done the draft evaluation report, that's where you're identifying the confusingly similar. And that would presumably, because you've got some external advice or someone's raised an issue and you've got advice. So it's already gone through that string check process.

>>DAN HALLORAN: But, ultimately, withr and who makes the call? Is color similarly confusing to colour with a "U"?

>>BRUCE TONKIN: Yep. So that would be --

>>DAN HALLORAN: Someone has to decide.

>>BRUCE TONKIN: Yep. That's the dispute resolution panel for confusingly similar.

>>DAN HALLORAN: We'd have to convene the panel to decide?

>>BRUCE TONKIN: Yeah, that's correct.

Well, look, I don't know. I'm just trying to be practical here. But I think if you're doing "color" and "colour," it's a bit obvious. But I'm trying to sort of encapsulate some of the IDN-related things that, you know, people could make claims. And I'm not saying that's staff expertise. You're going to go out and get expertise to go through a process and then come back with an answer, I think.

I think, by and large, you'll be dealing with exact matches, I think, plus stuff that's pretty obviously confusingly similar, which would be enough for the staff to sort of kick this process off. Then there might be areas where it's pretty gray. And then you're going out to, you know, an appropriate body, just as you would with a UDRP dispute or something like that. And I'm not quite sure how long those things take. If that's over 60 days or something like that, they would come back with an answer saying, yes or no, we think these are confusing.

>>LIZ WILLIAMS: Bruce, I just had a question from Eun-Joo, who's not here. Eun-Joo Min is from WIPO. And she and I were just talking about this thing just before the meeting with Michael Palage.

>>BRUCE TONKIN: Yep.

>>LIZ WILLIAMS: And I was giving her examples of where I thought certain kinds of strings would fit into this, like Web and wed, W-E-B, and W-E-D. And she didn't think that was confusingly similar.

>>BRUCE TONKIN: Fine. I mean, that's -- I'm not --

>>LIZ WILLIAMS: But you're wanting -- you don't want to put any limit -- this is a question -- you don't want to put any limitations on what could fall into this bucket, do you, because it's a complainant-driven process?

>>BRUCE TONKIN: No. It has to go through this process, just as though it was an existing TLD. Let's say dot web existed, and what was the other one, dot wedding or something, dot wed? Someone applies for, it I'm pretty sure the dot web guy would complain about that. And then it would go through that process. And it's that process that determines the outcome. It's not a staff arbitrary decision.

>>WERNER STAUB: Question.

>>BRUCE TONKIN: Yes.

>>WERNER STAUB: If somebody objects to the decision of the initial panel, --

>>BRUCE TONKIN: Yes.

>>WERNER STAUB: -- say, you know, I was thrown out, there were three contenders. I was thrown out because, you know -- or I'm actually not confusingly similar, that kind of thing, that would also be a process problem, because it would have to delay the subsequent --

>>BRUCE TONKIN: Just give me an idea of what you're appealing against. There's a draft initial eval- -- you mean you filed on a technical, like it turns out you don't know how to run a name server?

>>WERNER STAUB: Or even before that. Because you basically say, I don't think that we should actually be considered to be contenders, because the decision by the panel that decided we are confusingly similar is wrong. I have nothing to do with that guy.

>>BRUCE TONKIN: You're bound by that panel.

>>WERNER STAUB: You're bound by that panel? There's no appealing against that?

>>BRUCE TONKIN: Hmm.

>>WERNER STAUB: So the whole thing is that I was mistakenly put into contention mode by someone who was unrelated. But now I have to fight it off with him.

>>BRUCE TONKIN: I think we're getting into corner cases here.

But the general concept is, you'll be bound by the decision of that dispute panel as part of the process I think is where we're talking about.

I guess, ultimately, you can go to a court of law. But I think that's going to get -- the whole process is going to get extremely messy. I think, basically, we would be requiring the applicant to agree to abide by the outcome of that dispute process. Because, otherwise, it's just going to get immensely complicated if you can --

>>PHILIP SHEPPARD: So, Werner, how can you mistakenly go into a contention panel?

>>WERNER STAUB: (inaudible) I understood, somebody not a complaint, but somebody's going to decide that these three strings are confusingly similar.

>>BRUCE TONKIN: Yes. Not somebody. Some panel of experts.

>>WERNER STAUB: Oh, some panel of experts.

>>BRUCE TONKIN: Yes, it's going through -- it's the same as the dispute process for string contention.

>>WERNER STAUB: But that would, of course, have to be a single process for all TLDs at once. So you basically have 50 TLDs that come in, and then they have to be grouped by one panel that says, five of them --

>>BRUCE TONKIN: But they do not need to be grouped by one panel.

UDRP is not a single panel.

>>WERNER STAUB: I know. But how do you know that out of 50 applications, five are confusingly similar against one another and the others, you know, have no problem?

>>BRUCE TONKIN: Because the others didn't -- nobody raised an issue with those other applications.

Is that what you're getting at?

>>WERNER STAUB: Okay. So, basically, one would only become lumped into a confusingly similar discussion if somebody complained?

>>BRUCE TONKIN: Or the staff just looked at it and worked that out for themselves.

So we're trying to sort of look at the scenario. So I've got a set of names, Werner. So I've got A, B, C, and --

>>WERNER STAUB: Wed and web.

>>BRUCE TONKIN: Actually, let's go AB1 and I've got ABL. So I've got these two. I think the staff could sort of look at these two and say, yep, that looks like it's potential for that.

And that could then decide, okay, we've got to confirm this with this dispute body.

Then there's other scenarios. So there's a bunch of Chinese characters that, as far as the staff are concerned, they're not really sure. Two things, one is the staff might be conservative and just say, hey, they vaguely look like they might be the same. We'll go to a dispute panel. Or there's a complaint that comes in here and says -- which could be one of these guys, incidentally -- so this applicant one and this applicant two, this applicant one goes, "What -- you know, that looks way too similar to what I'm asking for." So that applicant itself would initiate that process, because they want to knock that out.

>>WERNER STAUB: Mm-hmm.

>>BRUCE TONKIN: So I think the process is essentially that -- yeah, it's going through a dispute resolution process to determine that confusingly similar.

If it is confusingly similar, then we have contention.

>>CHUCK GOMES: Wait a second, I need to clarify something. You said that applicant one could submit a complaint regarding applicant two when they're in the same round?

>>BRUCE TONKIN: Yes.

>>CHUCK GOMES: Why does -- Oh, so applicant two could do the same thing regarding applicant one.

>>BRUCE TONKIN: Of course. Yeah.

So, basically, that dispute gets filed, and then it's decided, yep, they are confusingly similar. And then they get bound into the -- so, basically, if they're not, they can both go through.

>>PHILIP SHEPPARD: Yes.

>>BRUCE TONKIN: Right?

And so if you were the applicant of that and then you saw another applicant that looked really close to you, you'd probably say, "That's not good," just as if you are an existing TLD, like com, and, I don't know, there's a Cyrillic com or something, I don't see how that's any different to that scenario, just because you happen to have it already in the root.

>>PHILIP SHEPPARD: So there are -- let me understand it.

So there are three scenarios, I guess. Either a bunch of names are similar to the themselves and/or identical, but also confusingly similar to an existing TLD. The existing TLD complains, knocks them out of the process, we don't hear about them anymore. Scenario one.

>>BRUCE TONKIN: Yes.

>>PHILIP SHEPPARD: Scenario two, they're confusingly similar to each other, but to nothing else that's gone before.

>>BRUCE TONKIN: Yeah, to nothing that exists, yes.

>>PHILIP SHEPPARD: And if they're judged not confusingly similar --

>>BRUCE TONKIN: Then they're just treated as independent applications.

>>PHILIP SHEPPARD: -- they just go ahead as two separate applications.

>>BRUCE TONKIN: Yes.

>>PHILIP SHEPPARD: If they're judged as confusingly similar, then they go into the same process as they would if they're identical.

>>BRUCE TONKIN: Yeah. And bear in mind that a number of these things can be kicked on and off parallel.

>>PHILIP SHEPPARD: Yeah.

>>BRUCE TONKIN: So you might say that one of those is a technical problem, you know, some -- it sort of scales across, because there are different things that are happening there.

>>WERNER STAUB: Just one thing F there's going to be a complaint-based process, there must be a deadline as until when you can complain about it's being.

>>BRUCE TONKIN: That's part of the previous diagram. Yeah.

>>WERNER STAUB: Okay.

>>BRUCE TONKIN: You were involved in that discussion in Marina Del Rey. We went through it in detail.

>>WERNER STAUB: Okay.

>>BRUCE TONKIN: But there's a whole process around complaints for dispute resolutions around string contentions. We're not saying that. So we're saying that that whole process has come up, instead, there is string contention. And now we're in the string contention.

So I'm not going to reinvent the other diagram. We've already got that.

>>WERNER STAUB: It may be because of the symbol.

If you use that symbol, that means it's a single decision. If you -- there's another symbol that you would use in a graphic to say there's a process.

>>BRUCE TONKIN: We have --

>>WERNER STAUB: For, you know, review applications.

>>BRUCE TONKIN: This one?

>>WERNER STAUB: Yes.

That's a decision at one point in time. It's -- if you want to say it's a process, would you use a different symbol.

>>BRUCE TONKIN: Okay. You can advise Liz on this afterwards.

But the concept of this is USC got to read the other diagram in tandem, yeah.

And the concept, I think, that we were coming up with here, which I think is the new element, is the fact that we're talking about something that's come out of a draft evaluation report, essentially.

So there's a notice period. And, look, maybe we're just making it too complicated by the notice period, maybe we just get rid of that and therefore we start from the draft evaluation report. Maybe that simplifies the understanding.

>>CHUCK GOMES: They will essentially get notice when we post the string.

>>BRUCE TONKIN: When this is posted.

What they won't know is whether we consider them to be confusingly similar.

>>CHUCK GOMES: That's right.

>>BRUCE TONKIN: But I guess what we're saying is they can make their own judgment on that. And then when the draft evaluation report comes out, they'll see what the view is. So that might clarify a bit.

So what you would do, then, Liz, is, I think, this becomes -- if the draft evaluation report finds -- if the draft evaluation report reports that they are the same or confusingly similar, so that -- that point is coming from this diagram, I'm going from the draft evaluation report as input into here, so the decision is "if" on that report, I think.

Yeah, so just put an "if" there, "If input from draft evaluation report."

If input from draft evaluation report identifies that the applications are the same or confusingly similar," something like that.

"If input -- if the draft evaluation report" -- so just delete "input and from."

If the draft evaluation report identifies that the applications are the same or confusingly similar" -- strings, sorry. Yeah.

And then you can (inaudible).

>>CHUCK GOMES: (inaudible).

>>BRUCE TONKIN: Yeah.

And add some rubies and emeralds, and we're done.

(inaudible).

>>BRUCE TONKIN: By the end of this, you'll be an expert at -- that's right.

(inaudible).

>>BRUCE TONKIN: Just keep saving this as you go, by the way.

Exactly.

(inaudible).

>>BRUCE TONKIN: Okay. So just say "If the draft evaluation report identifies that the application is the same or confusingly similar," just full stop. And then delete the -- that stuff there. Just here. From there on was just delete.

Yep. That'll do. Okay.

Then we'll contact the applicants to advise regarding contention.

So just delete this bit here. Let's put in "draft report," that text, after the double hyphen. Yeah, just from there to there, delete that.

>>DAN HALLORAN: Bruce, can I please ask about this contact, the competing applicants? Is this the only place where we're going to contact the applicants, like, we're publishing this report, and it might identify a whole boat load of different kinds of issues, technical issues. And this would just be, in my mind, one more issue with the application.

>>BRUCE TONKIN: Yeah, so what -- the answer to that is "no." But what we're trying to do is we're just trying to map out the string contention process.

But I guess publishing the report is effectively getting in contact with them, anyway. But I think what you're trying to do is to say, hey, we're giving formal notice that you guys are similar and should try and meet to resolve.

>>CHUCK GOMES: Now, is that draft report publicly --

>>BRUCE TONKIN: Yes.

>>CHUCK GOMES: -- posted?

>>BRUCE TONKIN: Yes, it is.

>>CHUCK GOMES: Okay.

>>BRUCE TONKIN: Actually, there's two things that happen here. There's a draft and then there's public comments, and then there's final. I don't know whether we want to talk about final or draft here. I think draft is probably enough. I think that's basically saying it's gone through a fair bit of analysis. Obviously, comments can still come in. Yeah.

>>CHUCK GOMES: Preliminary check (inaudible).

>>BRUCE TONKIN: Yeah, yeah. So, basically, unless the applicants agree differently -- yeah, that'll do. I think that's fine for the moment, Liz.

Okay. And this is only the easy part of the diagram. The next part of the diagram is a lot more complicated.

But I guess we're improving it as we go.

Yes.

Okay. So now we're basically looking at the scenario where they couldn't deal with this via mutual agreement. And so these are the options, as I understand them. One is, the parties could decide to mutually agree to use external arbitration. And this would be binding external arbitration. Maybe just put that in there just for clarity, just put "binding external arbitration."

>>CHUCK GOMES: That's a good idea. Because this some cases, they talk about -- at least in the U.S., they talk about binding arbitration or nonbinding arbitration.

>>BRUCE TONKIN: Yeah.

So the parties may agree to binding external arbitration, and the concept we're trying to get across here is, if you do that, you know, it's obviously a cost. You've got to pay that arbitration body. So there's an incentive to try and resolve it yourself to not incur that expense, essentially.

So the idea in this diagram is saying, okay, you agree to this. You have some period of time. You pay the fees. And, you know, once the answer of that arbitration, that, you know, forms part of the final evaluation report.

The other scenario is that they don't agree to that. And so, essentially, these parties are in dispute. Then what we're saying here is that the decision that -- as I understood it from the meeting in Marina Del Rey, was we're saying is the name related to some sort of existing institution? And that's kind of the wording that's currently in recommendation 14C, which is, you know, the institution could be related to, you know, cultural or whatever sort of list of the varieties. But the examples we talked about were bank. Another one would be talking about would be, like, a country name or a geographic location or something like that were examples that would fit into that path.

If yes, then we're moving into a more detailed comparative evaluation process. And picking up the IDN discussion, you know, part of that might well be, you know, do they have expertise in IDNs, for example, if they happen to be IDN names.

So here there's a fee for this. This is an additional cost from the normal application fee. And then they need to submit additional material for those external reviews. And that material would be posted and be subject to a public comment period. And then there's a report, basically, recommending -- a recommendation saying which one of those two parties get the string.

So that would be a fairly similar process to what was done with, say, dot net or dot org, where, essentially, we're saying, okay, we're now moving into this extended process. Here's the criteria or the extra information we require. They supply that information. That information's posted. The public get to comment on it. And then there's a final outcome of that, which is a recommendation to the board basically saying, hey, we recommend A or B.

The other process, as I understood, was if it's not relating to one of these and it's basically just a generic word, it's "red" or "green" or it's "house" or "tree" or whatever, then we're sort of moving into saying we want to then just have a -- just a straightforward process. And the option here, I think, that seemed to have -- seemed to be most implementable, I guess, as an efficient process here was to use an auction.

Can you just strip the square bracket off -- actually, you need a closing bracket, I think, is why I'm getting confused.

And what I've put in here is just to indicate this is not binding, in a way, but I just want to be clear that the auction is not suddenly the money is going to the winning applicant or whatever. It's that if ICANN has any surplus funds, that would go into something like security and stability. There might be a trust fund that's established. It doesn't go into the operational expenditure fly to upgrade everyone from business class to first class for the next ICANN meeting or something.

>>CHUCK GOMES: And another idea that would fit in nicely here is some funds for the undeveloped world applications.

>>BRUCE TONKIN: That's a good -- can we stick that in, Liz. I think that's a good one.

First is security and stability. Or -- yeah, don't put "or NCUC."

Or --

>> (inaudible).

>>BRUCE TONKIN: Love that. That's suitably meaning anything.

Or for the digital divide. There you go, Dan.

>>LIZ WILLIAMS: Bruce, I just want to put up some questions again that Eun-Joo posed to me whilst we were having the discussion. Just go back to the diamond which says parties may agree to binding external arbitration.

>>BRUCE TONKIN: Yep.

>>LIZ WILLIAMS: And I'm paraphrasing what she said, so these are not my words.

She said it's very expensive and very long.

>>CHUCK GOMES: What is?

>>LIZ WILLIAMS: Binding external arbitration.

Could be expensive. I think it could be expensive and long.

>>BRUCE TONKIN: Someone had experience with it. How much did it cost and how long did it take?

>>CHUCK GOMES: I imagine it varies a lot.

>> Varies a lot.

>>WERNER STAUB: We had one that was two years. And it cost about -- well, more than half a million dollars. Probably both parties paid, including lawyers' fees. The total was probably a million dollars.

>>BRUCE TONKIN: Yeah.

>>LIZ WILLIAMS: Okay. What I'm trying to get to, Bruce.

>>BRUCE TONKIN: Yeah, what I'm trying to convey here, and I think that would be worthwhile in the reports, is trying to get an idea of costs.

But one of the things that I'm trying to get across here is there are some different paths. To some degree, you have choice. you have a choice to go down here or perhaps down that route. You don't have a choice if it's relating to existing institution. But I think even there, I think in Marina Del Rey, we talked about that those parties may mutually agree to do this, as opposed to go down that comparative evaluation path. Because that may well be the cheapest option.

>>CHUCK GOMES: Now --

>>LIZ WILLIAMS: What I think I'm trying to bring out here -- sorry, Chuck -- is that this piece of the puzzle is expensive whichever way you look at it.

So there's a lot of -- what I'm trying to get to is the text in the report that says "for applicants, be prepared that this might happen to you, and there are incentives for to you reach agreement much earlier in the process." So I'm trying to develop things around incentives as opposed to identifying problems, and then working out exactly -- we need to provide some instructions about, for example, binding external arbitration and then I need to write in the report, for example, the International Chamber of Commerce, the London court of international arbitration and the stuff that I released to the group the other day, that identifies a rough estimate of cost and the choices that we have about these kinds of things, because applicants will have to warrant that they would accept that line, and these would be their choices in this case.

>>CHUCK GOMES: Now, we have binding arbitration there. But -- and I know this complicates it more. But couldn't they maybe first select mediation?

>>BRUCE TONKIN: Yeah. I debated those two.

Just go back up. I wondered whether to stick that in explicitly or not. That is true, Chuck. And the question is, maybe that's just up to them to decide that internally here, whether you want to spell that out. But, yes. And maybe you want to say "applicants have 30 days to conduct meetings or use mediation," or something, maybe put that in there. But you're right, I think what we're trying to do is to say in this flow, this is a hell of a lot better than any of the options down here.

>>CHUCK GOMES: Yeah.

>>BRUCE TONKIN:In terms of cost and time. Because, obviously, all of those other options are time, too. It's not just the expense; it's the time it's going to take.

>> (inaudible).

>>BRUCE TONKIN: Just hold on a second, Amadeu.

Put it in a separate box --

(inaudible).

>>BRUCE TONKIN: Okay. Just say, unless the applicants agree differently, applicants have 30 days to conduct meetings or use mediation. Trying to capture that concept. Because I think Werner used that word as well when we discussed this a few times.

>>CHUCK GOMES: And it could be conduct meetings and/or use mediation. So they could try meetings first and then go to mediation.

>>BRUCE TONKIN: Yeah. Put an "and/or."

And -- yep.

>>LIZ WILLIAMS: Do you still want (inaudible).

>>BRUCE TONKIN: Yeah. I think it's just sort of picking up Dan's point there.

Of course, I suppose this arbitration phase has then got the same issue. We might have to play around when the -- with some experience on this. But -- yeah.

Amadeu.

>>AMADEU ABRIL i ABRIL: When we talk here about negotiations between the parties that have submitted the same string or confusingly similar strings, the most usual agreement would be that one of the parties use another string.

>>BRUCE TONKIN: Yes.

>>AMADEU ABRIL i ABRIL: The problem is that I'm not sure this is feasible in the process.

>>LIZ WILLIAMS: Yes (inaudible) change the application.

>>BRUCE TONKIN: So just -- that is something that I think is a feasible outcome. And I think Dan raised this as well. It's worth talking through just what sort of agreement is possible here.

One agreement is they combine. Or the second area of agreement is they agree to use different strings.

If they use different strings, I think what we were talking about -- because we did have this discussion, Amadeu -- I think, essentially, those strings then get posted and that starts again with respect to those strings.

But I think that's okay. I think that's is a reasonable outcome.

>>LIZ WILLIAMS: That could allow two applicants to be successful.

>>BRUCE TONKIN: Yes.

>>LIZ WILLIAMS: Which is what you want. You don't want one aggrieved party losing. You want an option for a possibility -- we just need to be careful that we're not opening the application process again.

>>PHILIP SHEPPARD: And I think the other -- we did discuss it before, didn't we? And the other refine that we made was, that have, in those circumstances, the parties who agrees to use the different string would not use the priority date for the original application in terms of the processing first come, first served order that ICANN would be using. Yeah? The evaluations --

>>CHUCK GOMES: Let me make sure I understood what you said.

They would not use that prior --

>>PHILIP SHEPPARD: They would not lose --

>>CHUCK GOMES: Oh, lose.

>>BRUCE TONKIN: Oh, yeah --

>>CHUCK GOMES: That makes more sense to me.

>>BRUCE TONKIN: I got confused, too. I must admit, yes.

>>PHILIP SHEPPARD: Sorry. I did say "use." They would not lose that priority, yes.

>>BRUCE TONKIN: So, basically, the strings would get posted again -- how are we going to capture that? -- combine -- agree to use different strings --

>>CHUCK GOMES: (inaudible) they can stay the same.

>>BRUCE TONKIN: Yeah, that's right, yeah.

>>AMADEU ABRIL i ABRIL: Then I have a second comment regarding the word "arbitration." But it's a fairly legal and technical point. It doesn't change the process. So perhaps we can discuss that in another point in time.

>>BRUCE TONKIN: Can we just be specific, string evaluation. None of the rest of the stuff gets evaluated again.

>>AMADEU ABRIL i ABRIL: No, arbitration. The word --

>>BRUCE TONKIN: Sorry, different topic. Sorry, Amadeu.

I just wanted to make clear for Liz to just make clear that we're not going through a whole huge process here.

Sorry, Amadeu, separate point. What were you raising?

>>AMADEU ABRIL i ABRIL: I also had some comments regarding the use of binding arbitration.

I agree with binding but I have some reservations about arbitration that something quite particular, you may have binding decisions that are then implemented by ICANN and create, you know -- in just creating the TLD and the conditions of that decision. But probably it's not arbitration as we lawyers know that.

So it's a minor point, so we can discuss the technicalities of putting "arbitration" or not. The question is external binding decision. That's what's important.

>>BRUCE TONKIN: Yeah. I -- let's just leave these words as they are. I think that might be something that comes up. I think Jon, being -- I don't know whether he's available, was going to have a little bit more discussion about what those different types of arbitration are. But I think we're getting into different legal terms here, the principle.

Yeah, Dan.

>>DAN HALLORAN: Can I just flag again, or re-raise a strong concern about the idea of letting people start over again at the top?

>>BRUCE TONKIN: Yes.

>>DAN HALLORAN: For one thing, you're changing the whole process from a defined set of phases into a continually recurring loop, which I'm not sure -- you know, you get down here and then start over. Then we have to reconvene the panels and post all the applications again.

>>BRUCE TONKIN: Yes.

>>DAN HALLORAN: And start the whole process over.

>>BRUCE TONKIN: It is an issue. But don't you agree that -- I agree that you could potentially get into an infinite loop. Because you then propose some more confusingly similar strings.

But I think -- you've also got to kind of understand what the incentives are for the applicants here. So why would they do that? You know, surely they would be trying to do their best to get this thing through as fast as possible. So why would they then choose something else that looks like it's going to be confusing and start the whole thing again?

>>DAN HALLORAN: My other big concern about it is that I think it's really susceptible to gaming. Because if we set it up like this and I was going to apply for a TLD, I would put in two applications for that TLD and you can wait and see how -- let's say you don't do so well on the initial evaluation report, you get a do-over card. You can pick two new strings and file two new applications, tighten up your technical and improve your answers on the financial.

>>CHUCK GOMES: They don't get to submit a new application, do they? Just a new string. The same application would apply, would it not?

>>BRUCE TONKIN: That's correct, yes.

>>DAN HALLORAN: The same business plan and the same --

>>BRUCE TONKIN: Yeah, exactly. That all stays the same. So all you're doing is have a choice of the string. You're quite right, Chuck. So if you submit a weak applications that fails on other grounds, it fails.

(inaudible).

>>BRUCE TONKIN: You don't even get into string contention, actually.

Yeah, Steve.

>>DAN HALLORAN: But if could get you out of hot water about the string itself.

>>BRUCE TONKIN: Sorry, what was that?

>>DAN HALLORAN: So if I want to get -- let's say I want to get dot, you know, controversial, something that might run into trouble, I might catch a lot of objections to that. I could put in two applications for dot controversial. And then instead of getting stuck in a morass and not getting it, I could just bail out and pick two new strings when I kind of got in hot water halfway through the process.

I'm building in contention on myself to give myself a do-over on the string, which no one else gets.

>>BRUCE TONKIN: To give you an option to change a string.

>>LIZ WILLIAMS: But doesn't it mean you'd pay two lots of fees? You'd have to be willing to pay two lots of fees.

>>DAN HALLORAN: Well, you're going to get two strings no matter what. Because I can resolve the contention among my affiliated companies, you know, keep the string for one applicant.

>>LIZ WILLIAMS: I see what you're saying.

>>BRUCE TONKIN: It's the registrar accreditation game, basically. You game it by creating accreditations that you can then trade, essentially.

But, I don't know, I --

>>DAN HALLORAN: It's just a reservation is all. I'm still concerned about it.

>>BRUCE TONKIN: Yeah. And I'm just trying to think it through a little bit.

The alternative is you're basically forcing them to try -- it makes it very hard to reach agreement, I think, if you're not giving them that option. Because you're forcing them to sort of combine in some way or to just stop altogether.

>>CHUCK GOMES: Of course, you could manage that by requirements that organizations that are submitting applications identify themselves if they're related in any legal way to any other applicants. You attorneys can deal with that more than me, but --

>>DAN HALLORAN: We need an Army of investigators out there investigating shell companies in India and Taiwan, whatever.

>>BRUCE TONKIN: Ken?

>>KEN STUBBS: Yeah. You could also have the situation where you may have executory agreements. Even though there's no relationship now, you could have an agreement that going in, these people would agree up-front to doing something.

So --

>>BRUCE TONKIN: I agree you can game it. So the question is, do you want to allow it to be gamed or basically just shut it off altogether and say, "You can't do it. You either agree on the string or that's it. If you don't agree on the string, you're into this expensive process."

>>DAN HALLORAN: Or the evaluation or the auction.

>>BRUCE TONKIN: Yeah, that's the expensive process, yeah. Any one of those three expensive processes. They're all expensive.

So I think what we were trying to do here -- and this is, I think, Werner's original idea -- was the cheapest way for all of this is that they can reach an agreement without having to go into these more involved processes, and we should create an environment to allow them to do that. But if you're saying they can't choose a different string -- and I understand, because you're saying the

reason you're not allowing them to do that is because it can be gamed -- this step may become hardly ever successful. I don't know. Maybe it is. Maybe you just say that the incentive's strong enough that they combine their resources.

>>CHUCK GOMES: You know, I think we should obviously -- anywhere where we can minimize the chance for gaming is good. But we also have to be realistic. I don't think we'll ever create anything that can't be gamed, with the creativity of the people in this industry.

>>DAN HALLORAN: I think -- what would happen there? You're talking about can they combine and reach an agreement. I think in a lot of cases, you'll get a settlement. One party will buy the other two out or something. "You go away, and here, here's $300,000" or something, "let me have the TLD."

>>PHILIP SHEPPARD: I understand the theory. I just doubt the practice, I think. Because I don't see the point. I don't seat incentive in -- if -- in putting in two similar applications that are going to get into that trap and have the expense of that rather than putting in two good applications in the first place.

I just don't -- we're not offering anything extra. We're simply saying, if you're good guys in the process that you've already gone through and it's taken a little bit longer, then we're not going to penalize you even further if one of you is changing track.

I don't see the incentive for that being gamed any more than just putting in two good applications, if that's your objective.

>>DAN HALLORAN: I'm not sure, either.

>>KEN STUBBS: Well, if there's a perception out that there's going to be a limited number of, let's say, generic TLDs, then people will do this principally for one reason, and that is some sort of assurance that if they get bounced out -- if they qualify technically but they get bounced out on the string that they have a backup position that will assure them, you know, quote, the lucky loser type thing. And I would see that it would make sense for companies to do that, put two applicants in for the same string, have a higher probability of getting the string number one, and number two is they have a fallback position.

Hypothetically, if Chuck and Bruce were both separate companies but they had a loose affiliation, Chuck might end up with the string and Bruce gets to be a lucky loser in a case like that.

So....

I don't like the idea that you don't know up-front what a lucky loser's position is because, hypothetically, you could all apply for dot notebook, and all of a sudden you get into the process and discover that nobody applied for dot bank or something else, you know. And I am concerned about that, that there has to be --

>>BRUCE TONKIN: Yeah. I think the thing, too, probably to think about it is to think about this in rounds and what happens in the first round versus subsequent rounds. And I think the gaming you're talking about is probably going to be especially strong in the first round. And so maybe it's a bit like the discussion we're having on reserved names. Maybe we need to be more conservative in this first round than we might otherwise be later on.

>>KEN STUBBS: Look at 2000, with Afilias and NeuStar and stuff like this. It was just a mess. People were applying for three or four strings, you know, at the last minute.

>>LIZ WILLIAMS: Bruce, the other question -- sorry, Ken, were you finished?

>>KEN STUBBS: That's fine.

>>LIZ WILLIAMS: The other question that Eun-Joo and Michael raised was that if a string ends up out of the process, is it out of the process forever? Or could it come back into a subsequent cycle?

So are we creating a quasi-reserved list of strings that could no longer be applied for and that list would get exponentially bigger the longer the rounds go on?

>>BRUCE TONKIN: That's correct. Yeah, that's the discussion we had this morning on Chuck's report around reserved names. So we're saying let's just take string contention out of here for a second.

If you get a string kicked out because it's confusingly similar, it then goes on to, effectively, a disputed names list.

>>LIZ WILLIAMS: Meaning it can never be applied for?

>>BRUCE TONKIN: No, no, no. Meaning it's on a disputed names list. And then there would obviously need to be a process to get it off.

>>CHUCK GOMES: Which means that you wouldn't be very smart if you applied with one of those.

>>BRUCE TONKIN: That's the intent, yes.

So, in other words, it's a public list. It's basically saying, "Here's the strings that, you know -- the fish that were rejected," and, if you want one of the rotten fish, you can have it, but --

>>LIZ WILLIAMS: I think we just did a little bit more thinking about that.

The examples that were brought up were "bank" and "banc" and "bank" and "banque."

>>BRUCE TONKIN: Yeah. So let's take the scenario there. So I think this is a separate -- I think this discussion isn't related to string contention, okay? So just keep it simple.

You gave -- what we're saying is that in the first round, bank was accepted, so that exists. And then in the second round, someone says they want bank. And these guys complain about that. And that one gets rejected. That, to me, then goes on a dispute table, effectively. And as we talked about before, there might be a process which I don't think we need to design right now, but there might be a process to sort of say, okay, there is a way of getting something off that table. And that would probably be along the lines that whatever you're using dot BANC for was sufficiently differentiated, it really tight eligibility rules, you know, there might be a whole separate process for deciding whether that would be allowed, which is much tighter rules than if I was going to apply for something like dot loan, which is, you know, not confusingly similar.

So I think, you know, that's sort of the idea that we were talking about with the dispute processes, that the outcome of those dispute processes, the name then gets parked and it's put there and you know about it, so you're not going to keep running through the same process.

In fact, I used some examples in Australia. We have some rules about misspelling. And what was happening is that somebody -- taking Qantas as an example, and this is just sort of giving you an example of how this works in practice -- Qantas has com.au, and someone was registering Qantas with a "U." And that was grounds for deletion. But then a week later, it just popped up again. Got deleted. A week later it popped up again, got deleted, each with different parties playing the same game. So eventually, they just took it out of circulation and actually put it on a reserved list.

So in Australia, we actually have a reserved list of names that are cancelled effectively as an outcome of that dispute process.

Now, you can always go and get one of those taken off. But, you know, you'd have to have strong reasons to say, that is my name, and I've been using it forever, and nothing to do with the airline industry, but it's just not subject to the normal process of first come, first served.

Ken.

>>KEN STUBBS: Well, I'll defer to Dan first, and then I've got something.

>>DAN HALLORAN: Go ahead, go ahead.

>>KEN STUBBS: Okay.

Please confuse me -- I mean, please --

>>BRUCE TONKIN: Yeah, I'll gladly confuse you, Ken.

>>KEN STUBBS: No, that's all right. I'll be with you in a few minutes here.

Please correct me if I'm wrong, but we are, in effect, developing a process that's supposed to work for all new TLDs. That includes IDNs.

This gaming issue could be very, very serious in the IDN area. And I think we need to acknowledge or we need to come out front and acknowledge that there may very well be a different methodology that's needed to deal with string contentions in IDNs.

The reason, very simply, is I don't understand it to begin with. So I -- I have enough trouble getting along in English with bank, banc/bank and banque. But, boy, when you get into IDNs, you really start running into potential issues. And I think we need to have a methodology to ensure that we don't box ICANN in in a situation that -- if we don't acknowledge it up-front and then we start applying a different treatment to the IDN contention process, we're going to get criticized for deferring to higher powers, and so forth.

So I think it's a good idea to deal with it that way.

>>BRUCE TONKIN: Yeah. So I think where I'm coming to after this discussion is I think we'll take out that option of posting different strings. I think particularly in the early rounds, I just think -- I think you're right, I think we're adding more complexity there than -- and we've got enough already. So I think that's probably a fair decision to take at this point.

>>LIZ WILLIAMS: What do you want to delete?

>>BRUCE TONKIN: I'm just suggesting you delete this box for the minute.

>>PHILIP SHEPPARD: Sorry, but can you talk me through that rationale once again.

I don't see where the harm lies. I mean, even if -- I know, your strategy is to have dot web at the end of the process, you put in --

>>BRUCE TONKIN: What you could do, Philip.

>>PHILIP SHEPPARD: You put in bank and bank in your application.

>>BRUCE TONKIN: But do you it deliberately so that --

>>PHILIP SHEPPARD: Deliberately.

>>BRUCE TONKIN: You then benefit from the knowledge of what the other strings are that have been applied for.

>>KEN STUBBS: That's right.

>>BRUCE TONKIN: So, basically, in a particular round --

>>PHILIP SHEPPARD: But why is that gaming? I mean -- why is that a problem?

>>BRUCE TONKIN: I could tell you offline. But I'll tell you how you game it from a financial point.

Because you're basically saying, let's say there were 50 strings that get accepted in the first round, and you know that the next round is going to be six months, maybe a year away. So you have a first-mover advantage. Then what you do is put in multiple applications for dot web deliberately, because you know there will be others in there. And then you basically survey the game. And you go, oh, okay, I've got two applications in banking, I've got no applications in this other sector, real estate. So now I'm going to apply for real estate.

>>KEN STUBBS: That's exactly --

>>BRUCE TONKIN: Effectively, I cherry pick, because I'm in the round and I can cherry pick. And I know what else is in that round. And I've got a first-to-market advantage until someone else can apply in the second round.

>>DAN HALLORAN: You can have ten other companies trying to do the same thing at the same time, all trying to guess which one to jump to first.

>>BRUCE TONKIN:Yeah.

>>DAN HALLORAN: It's really a mess.

>>BRUCE TONKIN: Exactly the sort of thing that was going on with dot EU, actually, and not for the strings that way, but, basically, you increase your chances by having as many accreditations as you have and you get slots. And then you grab as much as you can. And then what you grab you then sell on.

So the whole objective is to grab as many of the good names as you can.

>>LIZ WILLIAMS: So are you happy to delete that box?

>>BRUCE TONKIN: So I think I'd remove that.

The more I think about what Dan's raised, the more I can think of how I'd easily game it. And I think we've just got to be careful in this first round.

It's not like we wouldn't ease up after that. Because I think it's the first round when it's subject to -- there's still a fairly strong first to market advantage in the first round.

>>CHUCK GOMES: I'm not sure that we should necessarily assume that it would ease up over time. Because, actually, what has happened over time in the industry so far is the gaming grows as people get more sophisticated.

>>BRUCE TONKIN: They get better at it. That's right.

Well, I guess, again, starting conservative, you're better off being able to manage that --

>>CHUCK GOMES: I'm not arguing with your point there. I'm just saying I think we ought to be careful about assuming that -- because they actually seem to get better about gaming as time goes on. They learn the processes better, develop their software, et cetera.

>>PHILIP SHEPPARD: So the only option for mutual agreement is to combine applications for the same --

>>BRUCE TONKIN: Essentially. And then you're going into some other process.

>>PHILIP SHEPPARD: Otherwise, you're going into an external decision process.

>>BRUCE TONKIN: I think that's essentially right.

>>PHILIP SHEPPARD: Okay.

>>BRUCE TONKIN: Any other comments, then? Any further discussion on this?

Yeah, Olof.

>>OLOF NORDLING: Shouldn't we state somewhere that we actually do have to maintain a disputed names list and have some kind of repository for all --

>>BRUCE TONKIN: That's right. That'll be part of the reserved names part of the reports. It's nothing to do with this, actually. It's just what we're doing in terms of creating a list of names that came out of the dispute resolution processes. And that -- I'd rather not call it a reserved names list. I just think we want to separate --

>>CHUCK GOMES: So you probably don't want it as part of our report.

>>BRUCE TONKIN: I think that's correct, yes.

Part of the new gTLD report, but we call it something else that the -- what did I call it before, the dispute table or something, whatever. Yeah.

I don't want to confuse it with something that's we're mandating across all registries.

>>DAN HALLORAN: Is it possible somewhere in the report to say, in future rounds, ICANN should consider this sort of negotiated --

>>BRUCE TONKIN: Yeah, I think we put that in the text.

>>DAN HALLORAN: Because one thing.

>>BRUCE TONKIN: We can put it as an implementation guideline, for example,.

>>DAN HALLORAN: So we don't have to change the policy if we want to implement that two rounds from now.

>>BRUCE TONKIN: That's right.

>>LIZ WILLIAMS: I can't go back to the report. Because what I have been doing with everyone's good ideas is inserting them into a version of the report, to capture. So I've got notes here that will go into the report on the disputed names list and on the evaluation implementation guidelines.

>>BRUCE TONKIN: Yeah, yep.

So -- yeah. I'll have to -- I think now na we've got this diagram, we can go back and look at the recommendations and get the wording a bit cleaner. Because the current wording of between sort of 14A, 14B, and 14C is all a bit confused. Because 14C actually applies somewhere else in the process. And I think our 14A and B needs to combine. And I think it's more about more clearly documenting this process, if you like, once we've got a diagram that maps this out, I think this does.

>>LIZ WILLIAMS: Bruce, in terms of the reporting, --

>>BRUCE TONKIN: Yep.

>>LIZ WILLIAMS: -- is it -- what I would like to do, just for clarity, is to name these boxes in the diagram, like, letter them, and then have a portion of the text that refers to that box. Because it's actually quite difficult in the reporting when you're writing a narrative to refer to the table. It'll be in the document. But I'm going to do it that way, if you don't mind.

>>BRUCE TONKIN: Yeah, I think that's a good idea. You number the documents in the table so you have a reference point.

>>LIZ WILLIAMS: Yeah, say box A equals this, and box A relates to or diamond A relates to recommendation 1, et cetera, et cetera, et cetera. And then 2 also relates to applicant guideline. It also relates to contractual condition, it also relates to dot, dot, dot.

>>BRUCE TONKIN: Yeah, I think that's fair. I'd like to have a look at how you're doing that. But, in principle, as much of the cross-referencing as you can do, the better.

All right, I think we've done this topic to death.

The other topic I did want to talk a bit about was disputes. But I don't think we have the right people for that, because they're all in the PRO group probably.

(Inaudible.)

>>WERNER STAUB: What kind of disputes?

>>BRUCE TONKIN: Many, many types of disputes.

[ Laughter ]

>>WERNER STAUB: They are the ones that are too easy to solve.

>>BRUCE TONKIN: If you could find Jon, yeah, see if he's available.

>>DAN HALLORAN: Can I suggest maybe one thing to simplify this is, really, the 30 days, the meetings, mediation, negotiation, or mutual agreement to go to arbitration is all sort of the same thing. It's all the parties getting together and either with an arbitrator, a mediator, drawing straws, somehow they're agreeing to settle the contention themselves.

>>BRUCE TONKIN: Yeah. I think the -- I think that's right, Dan. I think what we're -- I was just trying to do is map it out in terms of a timeline. Because I see that there's an order of magnitude and expense and time that's different from going to that formal binding arbitration versus saying, you know, let's get together in the bar for a couple of hours and see if we can agree.

So that's really the only reason I've separated those. But, conceptually, yes, they are, they're part of a mutual agreement as opposed to using ICANN to solve it.

Ken.

>>KEN STUBBS: I apologize for being redundant, and I'm sure I probably am.

I have a very simple philosophy when it comes to timelines, and I'm hoping that we keep to this. We need to ensure that we have a process that has timelines that in fact represent the worst case that we could expect, and anything else would be good news. I don't like the idea of setting timelines which are going to get broken. We've lived through that with the PDP process for years. There's no need to have distractions in the form of criticism or complaints about a process that rewards super efficiency and yet, you know, I think -- so all I'm really saying is just make very, very certain that the timelines can be met by the parties who have those obligations, be it ICANN, whoever.

>>BRUCE TONKIN: Yeah. It's meeting expectations, isn't it?

>>KEN STUBBS: That's right.

>>BRUCE TONKIN: And I think this is -- yeah, in the public forum, I'll get Kurt to address this specific issue. But I think that's very important that whatever we come up with here, the staff set the right expectations in terms of how long it takes them to do their work, and we're also clear on how long it takes these dispute bodies, whatever they are --

>>KEN STUBBS: That's right.

>>BRUCE TONKIN: -- to do their work as well, and we're clear on that. So we're saying, this is a 90-day, or whatever process it is, roughly, but we're not talking about a nine-year process.

>>KEN STUBBS: Right. That way, the worst-case scenario is what you expect. Anything else is good news.

And I think that's the important thing to work with. It's like, you do project management, you -- it -- from a morale standpoint --

>>BRUCE TONKIN: Yeah, I was trying to put a buffer in there so I can meet it. You're quite right.

>>KEN STUBBS: That's right.

>>BRUCE TONKIN: Okay.

What's happening with -- oh, good.

(Inaudible.)

>>BRUCE TONKIN: How are we doing timewise?

(Inaudible.)

>>BRUCE TONKIN: What time did we say we'd finish?

So we have half an hour perhaps just to get Jon to run through some of the concepts, yeah.

(Concluded.)