>>ROBERTO GAETANO: Okay.  Can we start?
 >>DENISE MICHEL: Please take your seats in back.  We're going to start the forum on GNSO improvements.  Thanks.
 >>ROBERTO GAETANO: I know it has been only a very short break from the previous session, but if you look at -- from the other point of view, the sooner we start, the sooner we finish and you can go to the bar.
 So let's start now.
 This is the GNSO improvements working group, and basically what we are -- we are trying to do here is to present a document, the state of the art of the document that the GNSO improvement working group has done.  This is a document that is presented for discussion, to encourage discussion and to encourage input from you, from the community.  It is in no way an established consensus of -- of the current situation, but is only provided as is to provoke and to encourage your input.
 I will keep the introduction very, very short.  I would just introduce the members of the working group.
 I would say that the -- this working group was created at the Lisbon meeting because after several months of working in the BGC and the board and so on, we realized that we were not making the progress that the community was expecting, and so we decided to have a specifically -- a specific working group to deal with this matter in order to improve the process and to come to a conclusion faster.
 So the group is composed by a member of the BCG committee and past board member.
 From my right to my left, Vanda Scartezini, Ramundo Beca, Tricia Drakes, myself, and Denise who is assisting as staff, and Susan Crawford.  Yes, thank you.  And Rita Rodin and Vittorio Bertola.
 So next.
 So basically, this is part of a bigger process of reviews.  I'll -- I will not go through these slides.  You'll have them, in any case, in the documentation.
 I would like only to point out where we are now, and this document has been posted a few days ago, explicitly for the -- as a basis for discussion.
 This includes -- it lists the points where we think we have an area of potential consensus, even if we don't have yet a real consensus.  The -- we still have some devil in the details, and there are also some other points that we want to submit as questions to the community where we would like to have your input in order to complete the picture that is before us in order to come to the conclusion.
 The objectives are -- of the working group are to maximize the ability for all to participate in the process, to ensure that recommendations are developed for the gTLD -- for the consensus policies for board review.  That probably is the PDP process.  Maximizing quality of policy output, ensuring policy work gets support, and support council effort to prioritize and benchmark GNSO objectives, and ensure that PDPs are based on totally researched and well-scoped objectives.
 And you can read yourself the rest of the slide.
 Next?
 So the -- next.
 I would like to have a quick survey of what, in our opinion, are the points where there is an agreement, even if it's not a complete agreement.  Those are the key areas where -- around which we are trying to build a final recommendation.
 The first point is that we have identified the key element in the PDP development in the working group.  So the working group model -- the working group is really the core of the process, and we believe that if working groups work and produce consensus policies, then the process is moot and can work.  Otherwise, there's no point in bringing issues to the council that has to then decide with some kind of voting system and some kind of complicated voting, and based on the balance between the different constituencies.
 So I think that the first -- the first thing on which we have a sort of agreement is in order to make the PDP model inclusive and in order to improve the process is to consider the working group as the key element, and not the council.
 The council now, in our opinion, has -- or at least appears to have a kind of a legislative function.  Everything turns around voting, and as the key issue.  If we think about all the debate on the constituencies, the problem of creating new constituencies, the problem of altering the structure, it all ends up in being a consideration about increasing or decreasing voting power of certain block or certain constituencies.  And so therefore, we were wondering whether this obsession on voting is really necessary for achieving the primary goal of the GNSO, which is to bring forward consensus policies.
 So in our opinion, the GNSO should return to a role that is a strategic role of managing the process, but not really get into legislative mode.  And that is linked to the question about the constituency structure.  There have been, in the London School of Economics, a recommendation -- a lot of impact that we are going to see in detail about the current constituency structure and proposals for improving them, maybe reducing the number of constituencies, changing the voting power, and all this.  This -- we will see this in detail later.  This is a highly controversial area, and -- but is not, in our opinion, the essence of the problem.  The essence of the problem is how to make the working groups work.
 So the draft that we have is a basis for discussion, is thick, but I encourage you, especially those who came here from far away, and therefore, have long flights back, I would encourage you to use the time on your flight to read through, so that when you go back home, you can provide comments and suggestions that are very useful for us.
 We are open, of course, to -- to criticism, but what we would be very happy about is that together with the criticism, you could provide alternative solutions or suggestions for us so that -- to make our work easier.
 So this is where we are now.  This document has been posted for public comment.  We had several teleconferences.  We have had contacts with former and current GNSO chair, just to make sure that things that we're absolutely off, unmanageable, undoable, were taken out, and now we are in the process of gathering input from the community and the idea is to produce a second draft, I would say in -- in a couple of months' time, and that second draft is going to be the basis of the document that will be presented by the working group to the BGC, and for board action, and at that point, of course before taking action, the board will have another period of public comment.
 But our decision was to go to -- to get input early, so that we don't find ourselves at the -- in the last round to have to correct errors that could have been corrected earlier.
 That's basically where we are.
 So those who have seen the structure of the document, there are areas.
 It starts with an area that evaluates the constituencies, the voting of the constituency, the problems related -- identified by the London School of Economics' report, and the second area is the GNSO Council and the -- again, the voting of the council.
 The third area is the working group, in how the working group are working and how they have to be considered the center of -- the core of the process.  And then we have the -- the relationship with the other parts of the -- the PDP process, sorry, how it can be streamlined and optimized, and at the end, a relationship between the GNSO and the council with all the other entities.
 Of these, we recognize that the area that would have raised the most comments is the one that's -- that is more sensible -- more sensitive and is the area of the constituencies structure.
 So we have decided not to start by -- with that area, because otherwise we would spend probably most of our time discussing only that not having time for things that we value even more than the balance of power among constituencies, but that have been the object of less comments so far by the -- by the community.
 So we have a kind of a rapporteur for each area, and I would like to start so with the working groups and I will -- I will give the floor to Rita, for doing this part.
 >>RITA RODIN: Thanks, Roberto.  I just wanted to thank everyone for coming and really emphasize that we're going to look for all of you in the audience to speak today.  This is not going to be about the working group talking to you about our ideas.  Hopefully you all had a chance to read the report and we put an executive summary to that report because it was posted a little bit late.
 So hopefully you can see the key initiatives that we're looking for you all to talk about.
 Very briefly, to set up the issue that I'm going to ask you all to comment on, one of the things that the working group was dealing with is:  How do we balance two very critical elements with respect to GNSO and policy development?  One is eliciting broad international participation in the process -- and we know there's a variety of issues inherent in that -- and balancing that with really quality of output and efficiency.
 One of the things that we're looking for, particularly with the GNSO, is to help formulate consensus policy.  Under the ICANN bylaws, consensus policies are policies that registries and registrars are bound to in their contracts, and the bylaws look to the GNSO and the constituencies of the GNSO to help promulgate that policy.  And Susan's going to talk to you a little more in detail about that.  But that's one of the goals that we are looking at:  How do we encourage development of quality consensus policy?
 Currently, as you all know, the GNSO uses task forces.  The bylaws talk about constituencies appointing their representatives from each constituency to sit on a task force.  Or having the entire GNSO Council sit on a task force for a particular issue.  And we've all been through some of the ups and downs with those task forces.  So one of the things that this group has been considering is:  Should we move to a working group model?
 For those of you that have been following, there's been, I guess, one and a half working groups.  One with respect to IDNs and the other currently with respect to WHOIS, and the general consensus within this group is that those models are working pretty well.
 So we wanted to get your comments on whether you think the working groups should be the default model, if you will, for policy development within the GNSO going forward.
 And we have a couple of recommendations that the group has come up with.  The first is obviously fundamentally there needs to be notice, that the working group is going to be formed, and a discussion of the particular issue.  And this notice needs to be posted very clearly and as broadly as possible, both inside and outside of ICANN, in different languages, in a sufficient amount of time before work begins, to actually have people come and participate.
 Another thing that the working group viewed as very critical -- this working group, sorry, viewed as very critical -- to the success of GNSO working groups is a very strong, experienced, and neutral chair.  In our view, it's very important to have a chair that knows substantively about the issues, but that can also try to facilitate consensus and agreement among the participants.  I just sat through the new gTLD presentation, and I'm always amazed how you can have one person come up to the mic and say, "That is absolutely the worst proposal I've ever heard, it's not going to work, no one's going to accept it" and then the next person at the mic will come up and say, "You guys have done a great job.  I think this is, you know, really a viable option."
 So it's important to have a chair in a position that's going to, in theory, try to bring those two groups together and try to drive the group towards some sort of consensus.  And I think one of the most important recommendations that we're thinking about is having the GNSO as part of this reform develop operating principles for each working group.  I think it's -- it's always helpful to know what you're getting into -- right? -- whether it's agreeing to run for the ICANN board or take a new job or what have you.  And I feel like if people know what's expected to them, that really helps facilitate achieving a goal.
 So one of the things that we're going to look for your comments on, if we can go to the next slide, is whether you think that this working group approach is aligned with the obligation of the ICANN community and, in fact, the GNSO to develop consensus policies.
 So I would love -- I know there's got to be some people in the audience that actually have some thoughts on whether they think this approach of having a working group with an experienced chair, very broad representation not only from GNSO constituencies but members of a working group that are from really just any interested party that could come to the table and work on developing policy.  Hi, Mike.
 >>MICHAEL PALAGE: How you, doing Rita.  Mike Palage.  I'd like to take that invitation to offer some comments.  You talk about working groups.  I think a couple that you missed were the working groups on the reserved name working group, the work -- PRO working group and, if you will, the working group that has looked at the introduction of new TLDs as a whole, and one of the things that I think -- you know, at this point is to sort of acknowledge some of the work that staff has done, particularly Liz Williams, in trying to move that whole process forward.
 It's been very complex.  But what is important is it goes back to the points that you've talked about, the difference of a much more diverse working base, which very interesting to note, in those working groups that I've participated in over the last year, since stepping down from the board, is we've never taken a vote.
 What happens is, when we produce reports, if there is someone who feels strongly, they will support -- they will submit a minority statement.  So instead of feeling that there is the necessity to win a vote and to horse-trade and do other stuff, what happens is one feels confident in knowing that they will have the ability to submit a minority statement or a statement of interest which is inconsistent.
 So I think what's very important with that approach is, Roberto, what you were talking about is getting away from the legislative model, where someone must win a vote.  These processes that have taken place over the last year, which I've participated in, there has not been that voting mentality.
 Going back to, if you will, the preworking group model, ICANN 10, it seems so long ago, I previously served as the chair of the famous names working group, working group B, and one of the things that we did there which has been duplicated, I think, in connection with the new TLD round, that staff has overseen, is it's been a very broad net.  People have been able to come in, they've been able to leave.  There -- I think it is -- people are -- there is not a necessity that once you join a working group, you're handcuffed into it.  It allows for people to come in, leave, and I think that this is important.
 One of the reservations I had about the bylaw amendments that were put in in 2003 with the evolution reform, with the task force, if you will, generally it was almost always a committee-of-the-whole, and what happens is, we saw a lot of that work focused at the council level, and not being delegated out to other people that have, if you will, interests and have expertise that might be able to contribute.
 So I think with regard to the -- the comments that have been raised by the committee, I think that they're on point.  I think they're in the right direction.  And again, consistent with the comments that I had previously publicly submitted in connection with the LSE report as a whole.
 >>RITA RODIN: Thank you, Mike.
 >>PHILIP SHEPPARD: Philip Sheppard, councilmember, BC member and chairman of an existing GNSO working group.
 I think you're right in terms of wanting to move in the direction of working groups, but I think I ought to make a condition, I think, "where appropriate."  I had experience, indeed, like Mike, of -- of the earlier versions of working groups, which were frankly unproductive, rather noisy and difficult organizations.
 The current concern that you have seen, as did the reviews that we had by professional consultants, with the emphasis on votes, of course, we must all remember is entirely the responsibility of what happened in the 2003 review, with the shift to double voting and the increased emphasis and dichotomy that that particular form produced.  So anything you're doing to go in a different direction to that, I think, is completely right and fully supported.
 Whether or not the working group model should be the default model, I'm not absolutely certain.  I think depending on the nature of -- of the policy question that's there, I would want, as a member of council, the freedom to say, "I think in this case, maybe a task force, short-term, et cetera, is going to be the right way to deal with that.  In this case, maybe a working group.  In this case, perhaps a committee-of-the-whole that really wants to move over to something."  So that flexibility to go ahead and do that in whatever way seems to match best with the policy issue in front of us, in particular given the dynamics, the level of interest in that, the degree of public interest versus specific technical change perhaps that's going to flare out of a policy, and things like that.
 So I think certainly not tying our hands, but giving the flexibility to do that part of that certainly following the proposal that I think has come out of every review we've seen -- and you're also seeing the same -- is get Annex A, most of it, out of the bylaws so that we can -- we can be more flexible in terms of changing rules, time lines, et cetera, as we need to, to respond to the dynamics of the processes in front of us.
 >>RITA RODIN: Thanks, Philip.
 >>JEFF NEUMAN: Hello.  Jeff Neuman with NeuStar.  I, like Philip, and Mike, have participated in a lot of working groups and task forces over the years, although I still -- and maybe I'm the slow one here, but I still don't know the difference between a task force and a working group, so it may help us, if you could actually provide your definition of what the difference is between a task force.
 From reading what you wrote, it seemed to me that a task force was one where people were appointed from different constituencies, whereas a working group was one where a call went out for people and I'll tell you, in practice, it's pretty much -- at least from the constituencies, it's operated the exact same way.
 You know, I'm in the registry constituency meetings and whether it's called a working group or a task force or -- or advisory group or whatever the label you put on it, it's always, you know, who wants to serve on this because we need to put people on it.
 So --
 >>RITA RODIN: If I could just interrupt for a sec, so again I'm going to -- this is from memory.  Maybe, Maria, you can correct me.  Under the bylaws, I believe that the task force has -- it's either the entire GNSO Council or one member from each constituency of the council, and then each of the constituencies has the right to appoint, I think, three other nonaffiliates, so the constituencies can appoint anybody from the community when there's a task force.
 I think what the working group is looking at is really anybody can join from wherever in the world, whatever constituency they might see that they're a part of, or maybe they've never been to ICANN before.  So it's really looking to increase in a marked way the representation and participation of a working group.
 >>DENISE MICHEL: And also a key difference, of course, would be the -- currently the constituencies have voting rights on task forces.  The working group model that this working group envisions is one of building consensus, noting majority and minority opinions, and as Rita said, doing outreach to stakeholders that may not be participating in ICANN, but have a vested interest in the issue.
 >>JEFF NEUMAN: Right.  And I think on that point, you know, I -- there's been some analogies to the IETF and their formation of working groups around certain issues, but -- and I think that's a good model for that group.  I think, as you recognize, that the difference here is that here, the outcome of a consensus policy is -- you know, contractually binds certain parties.
 Unlike the IETF where you would come up with a standard and while there's no contract -- you know, there's nothing contractually binding anyone to follow it, it's, you know, just generally accepted and done.
 So, you know, that makes the working group model a lot -- the implications of a working group model a lot more difficult, and in the working groups that I've participated on, I guess there were some other working groups other than the ones you mentioned.  I think the new gTLD task force had created working groups, you know, the protection of the rights of others, I think is a working group, I think, under that definition.  And again, the problem that we've had with working groups to date, although some have been very good, is that if you feel particularly passionate about that subject, you are going to have more and more people from that particular interest join, and others who may not be so interested in that subject not join, and so there have been some working groups that have been a little biased towards the position that -- where more people participate.  And in certain circumstances, that may be a good thing.  In others, it may not.  And I'd be happy to go into detail!
s, but there's so many other people here, so I will leave.
 >>RITA RODIN: Yeah.  We're going to have to close the queue I've just been told by the moderator, but the other thing, too -- and sorry, we're just trying out this format -- there's a number of questions on here.  I'd be interested to know Wendy, your thoughts on the working group but also one of the things that we think is going to be necessary to make this work, to address things that Mike and Philip and Jeff have addressed, is making sure that there are some principles that are developed up front.  I know that there's been some that have been developed for the working group for the WHOIS.  Having operation principles, having an articulation of how the committee is going to meet, having a clear articulation up front of what are the goals of the working group are, I think all those kinds of things that are going to be done up front are really create some efficiencies here.  And what we're looking for is also, to the extent any of you on-line have comments on that, it's the!
 third bullet:  What kind of operating principles should the GNSO develop for working groups to promote both the development of sound policies and then in that next bullet, the development of consensus.  Wendy, thanks.
 >>WENDY SELTZER: Thanks.  So Wendy Seltzer, joining the parade of veterans of both task forces and working groups.  As an unaffiliated individual or participant in the At-Large Advisory Committee, which is not a constituency and, thus, has been denied voting rights in any of the previous task forces, I have to say that I've strongly preferred the working groups and I thank Maria for helping to make the WHOIS working group work much more effectively as a consensus body than I had initially anticipated.
 I think that the -- the contrast between the WHOIS task force and the WHOIS working group was that the task force force tended to stalemate around the likely constellation of votes and led to holdout problems and lack of progress, whereas at least the working group may not have reached a consensus in many places, but at least provides a sheet of paper on which everyone can add suggestions whether those will end up as majority or minority positions.
 So in terms of operating principles, I think one of the useful -- one of the ways that working groups can be useful is providing a slate on which people can argue for positions and -- rather than the outcome being that the consensus of the working group will be adopted as consensus policy.  The better-argued positions from the working group will be those that have more force going forward to council for adoption, as consensus policy.  So that it doesn't wind up being a, you know, beauty contest of who can rally the most supporters to stand up and say, "Yes, I support that position" in a working group, but, rather, if it's a convincingly argued position, even if only one or two of the -- of its supporters happen to have joined the working group, it may still be something that can gain consensus on council.
 So I guess would recommend seeing these working groups as -- as inputs and not as votes, even in terms of support or disagreement.
 >>RITA RODIN: Thank you, Wendy.
 >>MARK McFADDEN: Mark McFadden.  The first thing that I want to make an observation on is that it's very difficult to peer inside of another organization and say, "Well, there's a working model that we could adopt for ourselves in terms of making decisions and coming to consensus."  And if people sort of look into the IETF and say, "Well, look at the way those working groups work and the way that they engineer consensus in those organizations, and could we adopt a model like that that would work for ourselves," I -- I want to express a real concern there.
 Because in the engineering community, what happens is, the protocol or the engineering work either works or it doesn't.  It's very binary.  And in the policy development world, it's a very, very different situation.  To adopt the model of the IETF working groups, I think, is a poor, poor decision.
 Now, we could call the working groups something else, but to look at the way that consensus is actually managed in the IETF is not something that we would like, I think, in ICANN.  And for instance if -- if a working group chair in the IETF decides that there's no consensus around a particular proposal, or a particular Internet draft, the draft is effectively dropped.  The author has a way to move forward on their own, but I think that it would be difficult for us, in terms of policy development, simply to drop things on the floor, in the event that consensus was not reached.
 I think that's a real danger in this model.
 As a veteran, I guess, is what Wendy said -- as a veteran of the IDN working group, one of the things about this model that's very important to understand is that you are completely dependent on the quality of that working group chair.  You put a lot of hope that that working group chair can engineer the kinds of consensus that you are really hoping for here.
 What I would say to you is while in the IDN working group we were really blessed with an outstanding working group chair who worked very hard at preserving minority positions, I think a number of people in our organization who are in a position to do that is very small.  We do nothing to develop those people.  We have no leadership development as part of our organization.  We are all a bunch of volunteers muddling about the best we can.  So one of the budget implications of this is that we would have to do some internal leadership development before we would ever have the people who could be these working group chairs.
 I think if the committee is happy with the 1 1/2 -- as Roberto said, with the 1 1/2 working groups that have come along so far, you should just be thinking you are accidentally happy because it could have been much worse.
 Another thing I will say about the budget implications is this.  If the working group model is something that you like, think about what you're setting yourself up for.  You have a working group on a particular issue, another issue, another issue and another issue. 
 And I ask you, how do you get global policy done?  How do you get a sense that the policies are coordinated when you have individual tactical working groups addressing individual policy areas?
 Someone has to do that, and that's not just an executive activity for the council.
 In terms of budget implications, another thing I have come to, if you have many more working groups that are addressing policy issues, what are the staff support requirements for the multiple working groups?  It is a question the committee has to ask itself.  But I think will end up with more working groups than you have in terms of staff support for the existing council.
 I support the idea of working groups, but I think there are some really serious things that have to be thought through before you can make it a workable plan and think through the implications it might have not only on the budget but the way the policy gets done.
 >>RITA RODIN:  Thanks, Mark.  Not to throw this back to you, those are great questions, some of which are up on this bullet slide and in the report.  It would be great if -- I welcome everyone to send any written suggestions.  I think we want to move this sort of raising questions phase and actually talk about suggestions about how we might augment staff  or what what we need to do would be great.
 >>MARIA FARRELL:  Hi, Maria Farrell.  I'm GNSO policy officer, and I have been working with these working groups for several months now.  My colleague Olof Nordling as well.  So we have been working with this working group model.  We have been, I guess, developing -- we have been bootstrapping it on the fly.  In that sense we have been trying to almost channel things that are coming to us in the ether that we think are possible to develop,  I suppose within the bylaws but yet not strictly following the bylaws.  I think it might be helpful at some point, not at this moment, for me to make more explicit what we do think this working group model currently means.  It's not voting.  It's taking some of the IETF models but not bootstrapping the whole thing, I have to say because engineers are other exotic birds and you can't transplant their culture either.
 But several things we have learned with this have been reiterated by other people and importance of chairing.  Denise, in fact, has worked very hard to make the funding and project available for us to develop people more internally.  I am responsible for that and haven't yet had to have quite the opportunity and the time to make that work.
 But working groups, I think are very much part of the way forward around here and I think it would be helpful for us as staff to work with some of the people who have been talking here today and make explicit about some of our assumptions about the working group model.
 It is not a panacea.  It is something that must exist within the bylaws and I think possibly I was talking to Milton yesterday about how wonderful I think the non-voting thing is and it makes -- it removes a lot of the incentives against cooperation.  But it also exists in a political context, and so I think we should look at how this moment in WHOIS working groups do work very well but they may not work in every moment in every PDP.
 Anyway, as I say, I will be happy to give more explicit maybe written comments or something like that to the group about what the working group model is, perhaps, we can work on it together to develop that and see when it is appropriate and look at answering some of these very pertinent questions.  Thanks.
 >>RITA RODIN:  Thanks, Maria.
 >>THOMAS NARTEN:  I'm Thomas Narten and a lot of people have already said a lot of things that I could say as well.  Let me just sort of say generally -- this comes from my perspective of having worked in the IETF for many, many years and having seen policy development processes in the RIRs, there is a lot to be said about moving towards a working group model and sort of two big benefits, I think, is moving away from voting and being more inclusive and allowing the people that really care about the issue to participate.
 That said, I don't think a working group is a panacea.  Even if you look at the IETF, there are plenty of examples of working group efforts that have flat out failed and didn't deliver.  You have to look at individual cases to why that is sometimes, because fundamentally there wasn't really agreement among the people participating and there was never going to be agreement so there is no way they would ever come to agreement.
 In other cases, the way the group was orchestrated or the way the process went and the way it was chaired, there were failings there that caused things to sort of fall apart and so on and so on and so on.
 There is a lot of lessons that can be learned from what happened national IETF and other places.  If you apply those lessons carefully, I think there is a lot of benefit here, but it is not a panacea at the same time.
 >>TONY HOLMES:  Tony Holmes from the ISPs.  Like the last speaker, a lot of things I was going to say when I got in the line they have already been said.  I would like to express support about to think about moving toward a working group model, and I think there are key issues around that that need careful consideration as we do that.
 I've heard people here say that the balance of the working group is critical, and I certainly support that.  Getting that right balance is the difference between success and failure.
 But it has also been an issue under the current situation.  The voting arrangements are very confrontational and coming from a constituency that clearly feels they were disenfranchised under the last reform process, it has been no incentive to get involved in some of the key elements when you know at the end of the day it comes to a vote and when you don't side with those parties and share the views with those who have double votes, you are always going to lose.
 Just look back at the records if you don't believe that.  Those parties that were in that position never won a single vote against those with double votes on their own.
 So we need to move away from that and I think going into the working group format where we can strive for consensus is absolutely key.  It is a way forward.
 How we tackle areas where we can't get consensus demands a little bit more thought around that.
 The other issue that I'd like to raise here is the actual setup and executive control of such a practice.  I think that that's as critical to the success as that model as anything else.  Maybe having a form of council as an executive body might offer a way forward.
 But we need to work with staff and determine how we handle that and recognize that we have to have priorities.  We won't be able to do everything at once.  There isn't resource to do that.  If you try to tackle everything at once, I think you are heading for disaster again.
 Having a real program that's mapped out with the key priorities, the right approach, getting the balance in those working groups offers a way forward.  I think that's where we should focus in the new term.
 >>RITA RODIN:  Thanks, Tony.
 >>MILTON MUELLER:  Don't look so sad, Vittorio.  Wake up.  Perk up.  I will try to say something that ticks you off.
 [ Laughter ]
 I have a philosophical problem as a political economist with consensus as a touchstone of policy making.  We need to recognize that what ICANN does is economic regulation of an industry fundamentally.  When that happens, there are conflicts of interest.  Decisions allocate costs and benefits.
 And when there are these economic conflicts of interest, there may not be consensus.  I mean, you just have to recognize that as a fact but, nevertheless, decisions will need to be made.  To give you two simple examples, it doesn't, I don't think -- I hope he doesn't contradict me afterwards, but I cannot see that VeriSign would have sat down with a group of its peers and consensually agreed to give up dot org and dot net, that that was essentially a bargaining game that entered into with the Department of Commerce with a certain contractual sword hanging over its head.  That's the way it goes.
 Another example is the WHOIS situation in which one party that wants open access has exactly what it wants now.  There is absolutely no reason for it to agree to give that up unless there is some way of and effectively outvoting it or requiring it to give that up by law.
 That being said, the working group model as a way of developing policy in an initial stage is something that I like, at least as long as there is a council structure sitting on top of it that determines how working groups get created, what their charter is and this council group on the top is balanced and representative.  I see we are only talking about working groups now.  We are not necessarily talking about the other recommendations, so I won't go into that.
 But the good thing about working groups are, indeed, their openness and their ability to incorporate multiple points of view that simply would be and have been excluded by the normal council task force process.
 And in some respects, it is good to have a flexible thing with no voting at that stage of the process where you are finding out what people can actually agree on.  I would think their recommendations would have to go up to a council-type structure that does, in fact, vote.  Again, you can require supermajorities and all kinds of those things.
 The problems with working groups have already been noted very effectively by Mark McFadden which there is a selection bias.  People jump into working groups based on whether they're interested in them or not and you can get a special interest group jumping into them, 20 people jumping in and getting an unbalanced view.
 And they're heavily dependent on the chair for management.  However, both of these problems can be adequately resolved by properly managing the input and output of the working groups.
 Thank you.
 >>CHUCK GOMES:  Chuck Gomes.  Let me thank each of you for your work on this because I think the report demonstrates that you've put a lot of thought into this and I appreciate that and I think you're on the right track.
 As you know by your questions, we have a lot more work to do.  I want to comment on a few comments I've heard preceding me here.  I know this isn't the time to talk about weighted voting, nor am I going to say very much but I think there are some misconceptions there.
 If you want to look at the record, look at the record -- the voting record before that was put into place and you will see that there definitely was a problem.  The emphasis on voting didn't start after weighted voting came.  That happened before weighted voting came and it was not -- so it was not because of weighted voting that there has been that emphasis on voting.  Not at all.
 It was because people were frustrated that it takes a lot of time to reach consensus.  It takes too long.  It's easier to vote.  But that all occurred before weighted voting was ever established.  I would be glad to talk to anybody about that separately.  I won't belabor that point.
 At the same time, I totally agree that we need to get away from voting and we've had some examples in that.  Mike's shared some of those, the new TLD working group hasn't taken a vote.  It's called a committee.  It is kind of like a working group because anybody could participate.
 And the reserved name working group, we didn't take a vote.  So we've had some good successes in that and so we need to move away from voting.  You're absolutely right.
 Leadership is a key but you need to understand that leadership is very time consuming, and when you consider you're working with volunteers, that's tough.  Certainly -- and I've communicated this to Denise, she knows this.  Chairs need administrative support.  I think it is totally a poor use of resources to use our excellent policy staff -- and I really mean that sincerely that they're excellent -- but they shouldn't be doing admin work and there is a lot of that with regard to the work of whether you call it a task force, a working group, a committee, whatever.
 You know, we may have to talk more about the idea of maybe paid facilitators to do these things.  I don't know.  I haven't really thought that through very more.
 We have to also accept -- and someone else said this before me.  You can't always reach consensus.  That's not always bad.  We need to accept that.
 The goal of a working group should be to motivate members to work towards a common position.  That is really critical for working groups.
 If you just have working groups where the people involved want to make sure their position is the one that wins, you need commitment from individual members to really work towards that.  We have seen that happen in several instances already where people are willing to do that.
 I heard the term "balanced" mentioned by a couple of people.  The balance that's needed is make sure you have good balance of competing points of view.  It is less important to have good balance of constituencies than it is to make sure you get as many different viewpoints as possible and then a commitment to work together to try and see how we can reach positions that the majority can support.
 >>ROBERTO GAETANO:  Excuse me.  The queue is closed.  We said that already before -- we said that -- to be honest, we said that when Bertrand was the last speaker.  I'm willing to reopen this -- the issue on this point at the end if we have time.  But, otherwise, we are going to debate -- consume all the time for this point.  I'm really sorry, but I think that we have to keep a balance on the different parts.
 And then if we can stay after 6:30, I will reopen the debate on this.  I'm sorry.  Please be brief.
 >>BERTRAND DE LA CHAPELLE:  I will try to be as quick as possible.  First of all, I want to endorse what Chuck has said as the working group -- or the group is moving in the right direction and addressing exactly the right issues.  The first element is the importance of the working group format to enable the participation of all the different actors and all the different categories of actors.
 I have been a pioneer of sorts in participating as a GAC member into the WHOIS working group and although it is a difficult exercise, it is very beneficial that the technical, the economic, the social and the political dimensions are taken into account from the very beginning of an issue when it is addressed so that the full pictures there.
 The second element is that the policy development process is a process that goes through a certain number of stages.  In the working group -- the joint working group between the ccNSO and the GAC on the IDNs and ccTLD space, we have moved in the direction of putting an issues paper that is a very early document -- it is a joint document -- very early on just listing the questions.
 It is very important to allow a broad understanding of all the different actors.  And the next stages is something that Avri has mentioned in previous sessions.  It is the importance of the terms of reference for a working group.
 Between the issues paper and the moment that the working group is actually formed, some discussion is beneficial on the terms of reference and the objective of the policy development process should be.  My feeling about the work of the new gTLD and the WHOIS working group is that there was not a preliminary complete agreement among all the different constituencies on what the real objective was on that.
 And the last element is that in the process of developing policies, the drafting of the policy is something that is the responsibility of course, in other words, responsibility of the chair.  It is not only responsibility that should rely completely on the staff, it is usually useful to have a small subset of the working group that is, as Chuck said before, representing the diversity of viewpoints, not necessarily the constituencies, but that can work on behalf of the working group to prepare something that is acceptable and represents different positions.
 This is what we've tried to do in the GAC in the drafting of the new gTLD and WHOIS principles, and it worked pretty well.  Thank you, thank you.
 >>RITA RODIN:  Thanks, everyone.  Now we've learned a little bit -- I didn't specifically mention voting but everybody had something to say on that and Roberto will talk about that.
 If we can limit the next bunch of questions to issues that Susan will talk about and try to be brief and we will think about a better format if we do this in the future.
 >>ROBERTO GAETANO:  Susan is reporting on the policy development process part.
 >>SUSAN CRAWFORD:  Thank you very much, Roberto.  And thank you to all of you for your comments and energy here as we drift into the late afternoon.
 The consensus policy regime is at the heart of ICANN's legitimacy.  We have no power delegated by government.  No special expertise like an administrative agency, and we can't claim to represent everybody so we're not a democracy.  So we have to ask ourselves, what's the theory that allows ICANN to tell actors in the domain name system, registries and registrars and registrants, what to do and what's the theory that persuades governments to defer to its decisions?
 ICANN, I think, can best be understood as an organization whose authority to make rules has been based on contracts that require compliance with future rules only when those rules are based on the demonstrated consensus, those who have chosen to participate its forums and affected by its policies.
 The premise of this regime is pretty simple.  If most of those affected by a rule agree that it will improve things and intense opposition is absent or irrational or limited to those who don't bear the costs of the policy in question or -- and this is very important -- or if opposition is limited to those whose objection is based solely on a desire to continue to engage in unjustifiable wrongdoing, everybody else decides some direction is the right way to go and one group just wants to continue to engage in unjustifiable wrongdoing, then those who are supposed to implement this rule have agreed by contract to do so.
 We have two problems.  One is that the bylaws speak generally about the GNSO creating policy but that the contracts that ICANN has with gTLD registries and registrars talk about consensus policies and limit the topics those consensus policies can concern to those within what we call the picket fence.  It is a very big fence, but it is a list of topics about which those policy cans make.
 That's our first problem.  There is a mismatch between the bylaws and the contracts.  The second problem is that the current PDP as people involved in the GNSO well know is too constrained, too prescriptive and really only honored in the breach.  It has short time scales and lots and lots of rules about how it proceeds and it hasn't proved to be workable, I think it is fair to say.
 So the draft produced by this working group suggests a way to keep voting on consensus policies at the board level only, tries to suggest a way to fix the PDP prescriptiveness by taking it out of the bylaws and letting the GNSO create its own policies and tries to draw a distinction between those consensus policies that are binding in the future on contractual partners and, on the other hand, other advice like, for example, the new gTLD policy that's being worked on so hard right now.  That's advice to the board.  It is not a consensus policy finding on existing registries and registrars.
 Okay.  So we've made some recommendations to make this happen.  And we've got a bunch of questions, just as we did for working groups.  So you can see them, perhaps, on the screen.  How could we align the PDP process better with ICANN's strategic plan?  How can we do this in the bylaws or otherwise?  And then most importantly for this session, how might the GNSO improve PDP rules?  What can we do to make the PDP work better?
 I know there are lots of people out there who have deep experience with these PDP rules and we're looking for your input on this one.  Come on!  All right.  Come to the mike.
 >>RITA RODIN:  Everyone we told to sit down, come back.
 >>PETER DENGATE THRUSH:  I'm going to talk about this topic, not the one I wasn't allowed to talk about last time.  Peter Dengate Thrush from the board but speaking personally.  I'm alarmed, I think, at the suggestion which I don't quite understand, Susan, about shifting voting on the narrow range of policy issues to the board.  And that's because a number of us fought very hard -- and I think you were one of them at the time of the reform of ICANN -- to make sure the board was never allowed to make votes on policy decisions.
 The whole point of the restructuring and we managed to achieve this particularly with the bylaws governing the ccNSO where I fought very hard under a remit of the board being allowed to remit or reject but not to remake.  We had a situation before the reform where the ASOs would go through their process, come to a conclusion, send it to the board and the board would feel entirely free to disregard the work done below and substitute its own judgment for the work of the community.
 And my fear is that you may be reintroducing that kind of a situation which we fought very, very hard to repair.
 Let me just clarify, why should the board resume the power to be voting on important policy decision when is we fought so hard to make it clear that would be the per view of the GNSO?
 >>SUSAN CRAWFORD: Just to respond very briefly, before the next questioner comes up, I'm speaking here for myself.  This is not a consist view of the working group.  I would suggest that we would amend the bylaws to note that what we need for a consensus policy is a process for consultation and expression of views, rather than a voting by representatives at the council level.
 So the working group, as Rita has described, does a lot of outreach, gets statements on all sides of an issue, tries to come to a majority view, a minority view, forwards that to the council, we'll have to talk about what the council's role is then, but we have envisioned as a group that the council would not necessarily be voting at that moment, but, rather, commenting on the report and passing it up to the board.
 >>PETER DENGATE THRUSH: Why that last step.
 >>SUSAN CRAWFORD: Why that last step.  Because the board ultimately has to adopt a consensus policy.  That's how it has always work.  So Philip.
 >>PHILIP SHEPPARD: Philip Sheppard.  
 Peter addressed one of my questions, which was:  How does the board feel with consensus policy moving up to them.  He's expressed one view.  I'm certainly aware of other board members who have the same view.
 My question, though, from my perspective is:  If you're going to do that, if consensus policy is moving up to the board, if it's still rather weak on contract compliance, as we currently are, what on earth is the incentive of me, as a volunteer user, in coming and doing anything here anymore?
 [Applause]
 >>SUSAN CRAWFORD: Thank you.
 >>JEFF NEUMAN: So Jeff Neuman.  I don't know how I follow that, but I guess my answer to that question is:  The board has a fiduciary duty to the corporation as opposed to the volunteers that participate, and while, you know, we certainly appreciate the volunteers that participate, the -- I don't know if I should get closer here.  Their -- wow.  Hopefully.  Is that on the record.
 >>SUSAN CRAWFORD: While we have a pause now.  We're going to stop the queue after Thomas.  I just want to make that clear.
 >>JEFF NEUMAN: All right.  So I just wanted to clarify that question and then the other point I want to make is I just want to go back in history a little bit as to the reason why the 90 days for the PDP was put in there.  It wasn't because -- and I was on the advisory committee that -- to help write that.  It was not because we were overly optimistic that we hoped policy could be done in 90 days, contrary to what, you know, I hear.
 At that time, you know, it was input from the registries as to why we got the 90 days because at that time the policy for approving registry services was grouped in with PDP.
 So if a registry wanted to propose a service, it was the fear of the registries that it had to go through this long, drawn-out PDP before that service could be rolled out, and that could be extremely time-consuming.
 Now that the registry services is part of a separate funnel, I agree that the reason for the 90 days, I don't believe -- I don't believe that reason exists any longer, and I would certainly recommend, obviously, the -- the extending of the time frames to actually reflect a reasonable process rather than keeping it at that really -- at the tight constraints.  But I wanted to provide that background, just so that you understood it wasn't something that was just optimism or those -- those -- you know, people who didn't know anything wrote that policy.
 >>SUSAN CRAWFORD: Thank you, Jeff.  Appreciate that.
 >>STEVE METALITZ: Steve Metalitz.  I just want to comment on two small -- relatively small points in these questions.
 One asks whether some change to the PDP process should be reflected in the bylaws.  I think that almost nothing about the PDP should be in the bylaws.  Because it's too hard to change it, and I'm sure perhaps like the 90-day period that Jeff referred to, I guess it seemed like a good idea at the time, but I think it's been one of the real shortcomings of the PDP is it's etched in stone and so now we have to go through a very involved process to change it.
 The other point I wanted to mention was the question about public discussion, fact-finding, and expert research before launch of a working group or other activity.  I think that's an excellent idea, and I think we have two examples that we might learn from.  One most recently is we have a proposal to start a PDP on domain tasting, and an issues report that says, "Maybe we don't know enough about domain tasting -- I mean we have a lot of unanswered questions about domain tasting."  Maybe if we had spent some time gathering facts about domain tasting before we had to initiate this process of an issues report, which in itself was kind of a difficult birth -- let's put it that way -- maybe we would be farther along now.  And of course the other example is the WHOIS process, because I think rarely have so many spent so much time to so little avail because they had so little data, and I think if we had data to begin with, or earlier in the process, we -- we would be farther along!
 on that, too.  Thank you.
 >>SUSAN CRAWFORD: Thank you.
 >>RITA RODIN: Hey, Steve?  Steve?  Could I just ask you a question?  
 So I've heard this a lot about the data and there wasn't enough data with WHOIS.  Who could that be fixed?
 >>STEVE METALITZ: Well, there are a couple of ways.  One might be that ICANN could actually commission a study and we would have to figure out what were the important pieces of data to try to collect.  But some of this data is held by registries and registrars.  And if we asked the questions, presumably they would answer -- at least they should be encouraged to answer -- but as it stands now, we've asked those questions maybe informally in the discussion within the WHOIS task force and WHOIS working group but we don't often get answers.
 I think it's -- first, it would be important to get them -- try to get some of that information out first.  And second, I think if there were a process whereby ICANN decided -- or there was a process whereby it was decided that we need facts on certain topics, facts that may be within the -- within the purview of registrars and registries, there would be a format for getting that information if there needed to be some type of confidentiality or anonymization or something like that.  It could be done too.
 >>RITA RODIN: Thank you.
 >>MICHAEL PALAGE: Mike Palage.  This is a tough issue.  And again, I'd like to start off by supporting the comments of Peter Dengate Thrush and Philip Sheppard.  And for those of you that know me, Philip Sheppard and I tend to not agree on a lot of things, so I think that's noteworthy that I do think there's an incredible danger here that if you somehow -- if you allow the board to just impose or vote on consensus policy, if -- if it doesn't come from the bottom, that there's really a danger of alienating the volunteers that have really contributed to this process over the late -- last eight years.
 >>SUSAN CRAWFORD: If I could just one very briefly, the question before the board would be:  Does this policy maximize consensus.  That's the question for the board.  Just to say -- say is there a consensus based on the outreach that's been done.
 >>MICHAEL PALAGE: And so that's the one side.
 >>SUSAN CRAWFORD: Yeah.
 >>MICHAEL PALAGE: But now as someone who has served on a board for three years, while looking at the, if you will, inefficiencies or imperfections of the old DNSO and the current GNSO, there is a fiduciary duty that each director owes to the corporation, to the organization, to do the right thing.  And therein is a balance that I struggle with.  So I guess the -- the statement I would say is:  If one goes down the direction of letting the board make that decision of trying to gauge whether there was consensus -- because as Milton has recognized, we're dealing with people that have financial interests.  The ability to make noise and throw sand in the wheel and prevent anything from happening is a reality in this process, and we can't let that impede the board from doing the right thing for the corporation.
 The only thing I want to say is if the board is going to go down that path, there really has to be some type of supermajority for them to take that action if there is not clear consensus within the community or if that didn't happen.  There really needs to be a safeguard from the board just usurping that consensus process or the consensus philosophy that ICANN was conceived upon and built upon, and again, if we -- if we lose that, the dangers that Philip said about people just walking away is going to happen and perhaps there's some people that want that to happen.  Maybe they want the volunteers to just go away and, you know, ICANN with its $50 million budget could just pay for those people that it wants to attend to come and speak at the mic instead of the volunteers.  That could be a cynical view of what could happen, but I -- again, there's a real danger in going down that path.
 >>ELLIOT NOSS: So -- Elliot  Noss.  Before I reached the microphone, I wasn't sure exactly what I wanted to pose as suggested answers, but I knew what the questions were for me.
 I think I had the good fortune of listening to previous comments and things have crystallized a little bit.
 First, this will be an extremely difficult and contentious area to effect a structure that's going to arise in clean consensus being presented to the board.  I'd like to highlight some of the structural difficulties.
 The first would be that the subject matter is incredibly arcane.  It's very, very narrow.  There are very few people in the world who understand it in detail.  That's just the nature of it.  I've had discussions with board members who have been doing this for years now, and are still amazed at how the substantive issues they're only able to scratch the surface on.  So we have very difficult subject matter.
 Second, we have a very diverse -- which is a positive thing -- diverse group of participants representing a diverse group of interests.
 There are "coming at" questions from all angles and points, and again, I don't think that's a bad thing but I think it's an environmental factor that we need to keep in mind.
 And third, and certainly for me the most important as it relates to looking at governance models like the IETF, W3C, this is overwhelmingly, or to a great extent, an organization or a room that's populated by lawyers.  Not by engineers.
 [Laughter]
 >>ELLIOT NOSS: There are certainly some engineers in the room, but there is a large number of lawyers.  And we need to recognize -- that's not -- that's an observational statement, not a judgmental one.  We need to recognize --
 >>RITA RODIN: You're a lawyer too.
 >>SUSAN CRAWFORD: He's a lawyer too.
 >>ELLIOT NOSS: Yeah, well, I fled the possession, as I like to say.
 Engineers are trained to solve problems.  Generally lawyers are trained to advocate interests.  And again, I say that observationally, not judgmentally.
 So we have those three big structural, environmental challenges in crafting any policy process.  So when I hear, you know, people talk about volunteers in this room, I think that's a gross misuse of the term, first of all.
 Second of all, when we talk about consensus being presented to the board -- and I understand that the pointed of the PDP is to effect consensus.  At the end of the day, the fiduciary role of the board and the goal of the organization is to make good, smart decisions.  It's not to hold an election as to what those decisions should be.  And it's not to effect the decisions that are, by structure, the most popular.  I think that as you think through these difficult problems -- and, you know, boy I don't think we're at the place where we have rock solid answers, you know, we're deep mired in the questions, I think you really need to try and find some way to allow decision takers to emerge.  To create a process that allows for well-informed individuals to take hold of issues and to put forward strong solutions and to find a way to effect them.  Sometimes in the face of opposition, sometimes in the face of differing interests, and I fear that if we look at the last few years as a!
n existence proof, they may be unfortunately an existence proof of one thing:  That trying to do that with this concept of consensus as we try to employ it might be an impossibility.
 >>SUSAN CRAWFORD: Thanks, Elliot.
 >>ELLIOT NOSS: Thanks.
 >>THOMAS NARTEN: Okay.  So Thomas Narten here.  I would say something fairly narrow, based on some things that were said earlier.  I think on the question about whether the board should be, you know, sort of signing off on the policy at the end of the day, I think that's a little bit of a tricky thing.
 And I'd sort of step back and say, what I think actually works best, if your goal is to have good clean, clear policy or engineering documents or anything that's a consensus or as close to consensus as you can have document, you have to iterate, you have to have multiple levels of serious review, and, you know, one way of achieving that is you have a lot of the -- the initial work done at a working group level, where the rules are maybe a little bit looser and the people that have the most interest get together and do things and then you have a broader call for comments, you might have another review at the GNSO level where there may be voting, there may not be voting, the rules may be different.  But as you go sort of higher up in the system in getting approvals, the bar for blocking something also gets higher.  I mean you can't get to the endpoint and have somebody say well, I don't like that paragraph because it just rubs me wrong and then just pull the rug out from the !
entire process.  And so, you know, at the board level, in some sense, you want to be able to look at it very broadly and say, you know, is this reasonable policy, did this go through the process properly, is this something we can really live with at the end of the day, and if so, you know, push it forward.  But not try to second-guess and redo all the discussions that happened earlier on in the process.
 >>SUSAN CRAWFORD: Thank you so much.
 >>CHUCK GOMES: Chuck Gomes again.  The board already decides whether consensus policies happen.  I mean that's nothing new and that's not going to change in the future.  What we need to look at is what does -- what information does the board need to make that decision.  And that should be clearly defined and outlined for the GNSO in what they put forward to make sure that they give you the information you need.
 Now, one of your recommendations -- and this -- I'm sorry if this goes into another one of your categories, but you talk about the council being a management of the process instead of being the policy development body, or the legislative body.  That's consistent with the bylaws, the way they're worded, and I believe that's the way it should be for a long time, my own personal belief.
 In light of that, the council, if they have to take a -- if they need to take a vote to determine whether they have something that they believe is -- is consensus enough to put forward, then I don't think that's -- I don't think that's a problem.
 So I don't think you have to go so extreme that there can be no vote, even though I do support moving away from voting in working groups and so forth.  So to have the council make a vote, yes, we think we've met the criteria that the board wants from us and we have enough information to give them that they can make a valid decision, I think a vote there is -- is okay.
 >>SUSAN CRAWFORD: Thanks so much.
 >>ROBERTO GAETANO: Okay.  Then we move now to the third item, which we thought was the -- the one that was going to create more debate, but --
 [Laughter]
 >>ROBERTO GAETANO: So I fear what's going to happen now.
 [Laughter]
 >>ROBERTO GAETANO: It's about the constituency structure.  I would make a proposal, and actually I will use the privilege of the chair to make -- to enforce the proposal.
 I would try to group together the constituency structure and the GNSO Council for the sake of time, since we are really approaching fast the deadline.  So the constituency structure.
 The key issue is obviously the voting -- the voting power, and the balance of voting.  However, what we have seen as problems is that in the initial conception of ICANN and of the then-DNSO, we were having this concept of creation of new constituencies by a certain process.
 That has never happened, and we have wondered why it has happened -- it has never happened.  And the answer that we have given is that most probably, this is because it was going to alter the status quo of the voting power.
 So I think that we tried to start attacking the problem from another side, taking the idea of the London School of Economics of the three large groups.  We are thinking of three or four large groups.  You'll see in the document why three or four.  And -- but basically, the idea we -- we don't want to debate the details here, but to go to the concept, and the concept is to have -- establish the balance of voting with large groups of what are the current constituencies.  And then let inside the group to form and dissolve different groups, like for instance if there is any synergy between the business constituency and the IP, they can merge together or remain split, but the important thing is that overall, the group in which they are in will keep the same voting power.
 The other -- the other thing -- so that is mainly to address the dynamics.  It will be easier in this -- in this context to have new groups that are formed, new constituencies.  We're thinking about a constituency for registrants, for instance, that also the LSE report was suggesting could work in the same group as the noncommercial constituency.
 But also, for instance, to have a specific role for domains, so that they could have a place where to get together and discuss their -- the issues related to their business, and to have a place in the -- in the ICANN structure.
 The other issue about the constituencies is participation.  There's a wide difference among constituencies about how much they cover their potential membership, how representative they are, how transparent the procedures are in order to join the constituency and we can go also to the next slide.  And -- and whether we should have, at a certain point, a centralized registry of participants of all constituencies and move to a concept that is -- that makes belonging to a constituency not really the major issue, but is a participant to the ICANN process and then you can choose the constituency in which you operate.
 Of course we believe that ICANN should provide increased support in terms of staff, in terms of resources to support the constituencies, in order to allow the constituencies to also do outreach and support their internal work.
 The questions that we have are here.
 Which of the groups that could better balance?  We have an initial shot, as registries and registrars with a question of whether registrars should -- whether this is a homogeneous group, whether there are some other constituencies that can form in that group, and then the business commercial users and noncommercial and individual users.
 So those are the -- basically the four groups that we're thinking of, but that is open to debate.  Are these or others the ones that have to be -- that have to be defined?
 Then the question, the well-known question, whether the noncommercial users should be grouped with the individual constituency, what is going to happen with the at-large in this case, is there an overlap or not, and basically those are the questions that are related to the composition of the constituency.
 Since you don't have to stick to those specific questions, I think I move to the next slide, and then I gather the -- your opinion on what -- on what -- your comment that you have.  Related to the council, the issue was to move the council from I would say a legislative body to some kind of management of the PDP and management of the working groups, this, incidentally, in the paper we have addressed the question on the creation of the working group and the role of the council in the creation of the working group in the assignment of the chairperson and so on.
 So the key function will be this one:  Will be the management of the working group and of the PDP, and we assume that voting should be less important.  However, we leave open, the open questions on this.  Next slide.
 And the questions are related on the voting system and related to the first three bullet points relate to voting, and then is there a threshold and how many councillors should each stakeholder group elect, which is optimal composition in terms of numbers of the council, and then the relationship with the NomCom.  Is the mechanism by which the NomCom appoints a councillor who has been effective in the view of -- of the council.  I think that we can take this question out, because the council itself has -- with the appointment of Avri Doria, NomCom nominated to the council as chairperson of the council has -- has proven that there is a usefulness in appointing noncom members.  So those are basically the questions.
 I would like to limit the -- in order to give the opportunity to everybody to speak, to limit the duration of the -- of the intervention.  I'm sorry that I have to do this, but we have time limits.
 So I see already people strategically positioned in the --
 [Laughter]
 >>ROBERTO GAETANO: So you can already line up.  You don't have to wait until I -- because I'm actually continuing talking until I see the first -- okay.  I'll stop it here.
 [Laughter]
 >>MILTON MUELLER: When does the time limit -- what is the time limit?
 >>ROBERTO GAETANO: Well, three minutes.
 >>MILTON MUELLER: Three minutes.  Okay.  All right.  This part of your report, we've actually given some attention to, although you cast a pall over it by suggesting that the council would essentially dissolve into the process of filtering un-aggregated opinions to the board, which would then do the real voting.
 So number one, there's no support here for the idea that you should eliminate or completely move away from voting.  Voting can encourage people to go out and seek the support of other constituencies to get their votes.  And there's no necessary connection between eliminating voting and gathering consensus.  In fact, what you can end up doing is simply making the whole exercise meaningless because at the end of the day you don't have an agreed position, you just have a collection of opinions.
 Now, with respect to the stakeholder groups, we really are pleased with the way the group has approached this problem.  We think that the fourth stakeholder groups, or three stakeholder groups, are a good idea.  The question of noncommercial versus individual is probably too complicated to go into in a 3-minute space so we'll be giving you a report on that.
 We think that we're concerned -- we just think that those three or four groups should be constituencies and you should not have a double structure of constituencies, then forming another layer of bureaucracy of stakeholder groups.
 If you have stakeholder groups as a formal structure, then people within them can ally and make groupings as much as they like and you don't need to formalize it, you don't need to stick your noses into it at all.  They'll do that anyway and the broader the stakeholder groups are, the more likely that activity will happen and it won't need any help from you.  What you do need is basically a firm voting structure at the council level that decides, you know, what work -- working groups gets formed, who manages that process, that does, in fact, ultimately take the input of those working groups and votes on them, whether to pass them on to the board as consensus policies.  You need that.
 What question have I not addressed?  Oh, yes, the weighted voting issue.  On that issue we think that if you do only have three shake holder groups, we think that you should not have weighted voting and probably not even with four.  We do think the broader groups are good, and on the issue of whether registries and registrars should be confined.  Those of us who enjoy seeing other people suffer think that this is a great idea.
 [Laughter]
 >>MILTON MUELLER: No.  I'm just kidding.  There is something to be said for that, because if it's a broader constituency group, they do have have to form alliances among themselves and as the market evolves, the relationship -- the rigid distinction between registries and registrars starts to dissolve also.  So you could think in terms of think, rather than four.  But either one would be an improvement.
 >>ROBERTO GAETANO: Thank you, Milton.  Who is the last person in the queue?
 [Speaker is off microphone]
 >>ROBERTO GAETANO: Thank you.
 [Laughter]
 >>ROBERTO GAETANO: Okay.  Michael, you're the last one.  You have the last word, as often.
 [Laughter]
 >>ROBERTO GAETANO: There's no obligation for using completely your three minutes.
 [Laughter]
 >>STEVE METALITZ: And I will not.  Thank you.  Steve Metalitz.  I really have a very small point to make.  To the extent that the recommendation here or the proposal here is premised on the idea that we don't -- we haven't added constituencies because they would dilute the voting power of the existing constituencies, I would ask the working group to reexamine that premise.  That premise makes sense if, in fact, the constituencies were the gatekeepers and made the decision on whether there would be new constituencies.  If that were the case, then that would be the most logical explanation of what has happened.
 But, in fact, under the bylaws, constituencies have no role -- no role whatsoever -- in the recognition of new constituencies.
 That decision is made by the board.  It can be made by the board on the petition of somebody who wants to organize a constituency.  It can be made by the board on its own motion.  That's specified in the bylaws.
 So if the board thinks that there should have been other constituencies, it certainly can't look to the existing constituencies and blame them for this not having happened.
 If that should have happened and it hasn't happened, it's because the board decided not to make it happen.  Thank you.
 [Applause]
 >>KRISTINA ROSETTE: Kristina Rosette, IPC.  I am not actually going to be answering any of the questions you've posed but I do have another one.
 This report could easily and particularly the section about constituencies and the council, could easily be interpreted as granting great deference to both the registrars and the registries.  And instead, effectively grouping the other constituencies together and then pitting them against each other internally.
 And that -- if you -- in light of this deference, it's entirely possible that you are going to have a public perception that as far as ICANN is concerned, these two revenue-generating constituencies are the most important ones.  And my questions would be:  Is this the intent?  And if it's not, I would just ask that you recognize that that perception is not only really starting to take root, but it's really growing exponentially.
 >>ROBERTO GAETANO: Just one -- one very short comment.  There's nowhere written in the report that the four groups have an equal number of representatives, and that have an equal voting weight.
 >>PHILIP SHEPPARD: Philip Sheppard.  Roberto, you asked me earlier, when we were discussing this, and also on an earlier phone call, that part of the reason for having an earlier discussion about this was that you were very keen to learn when you got to a point, when the community may be saying, "No, this is the wrong way to go."  We have reached that point. the.
 Voting does have, finally, merit because that produces closure in terms of something so I think you can't completely get rid of it.  We talked about working groups in terms of getting there.  I think we need to factor the dynamics.  
 In terms of structural change in the next set of your report, first of all, is a definition of public interest.  What is it that you are creating as you move on in terms of structural change that is fulfilling the need for the public interest there.  You may talk about these people may -- we think you might want this and that, let's see it flowing from the perspective of public interest first so we can then see where it's going into the wider dynamics.
 I mentioned the word "dynamics" as well.  Now highly concerned about the change in terms of the current dynamic structures where you do have different groups of users, those who are affected by the decisions of those who otherwise are making money by -- because of the contract parties to ICANN, somehow being put together because occasionally those users have similar interests vis-a-vis what is happening to the contract parties.
 And the fact that you sometimes see common interests and say let's put them all in one group together, I think, miss as point not necessarily about the diversities of those interests but the legitimacy of those interests out to be around and ought to be here.  Again, the wrong direction.
 If we're talking about structural change, you mentioned Nominating Committee or so.  I think you must separate the concept of the fact that council has recently recognized the merits of one particular member of the Nominating Committee to be our current chairman from the structural organization that bought into concept which is all based on the double voting system and supposed balancing that fulfills that.  That's how we got to that particular point.
 And if you are going to unravel anything, you need to look at how that works again.
 My final point is structural change.  If you are looking at merging constituencies and changing them, I find it very difficult to do without fully understanding the nature of the review we will have to do with the ALAC where we are just reaching the point where a very significant grouping is coming to, finally, be something that is meaningful and contribute to policy development. 
 And we've just seen from them a very good public interest-based document that has come to council recently.  A structure is beginning to look there.   We will suddenly do some reform that is all going to leave in an advisory capacity.  I just find that very strange.  I think if you are looking at dramatically structural change, I would advise against it.  We can't really do that seriously until we see where we are with ALAC and how that might be factored into the same dynamics.
 >>WENDY SELTZER:  Wendy Seltzer.  Coming from ALAC, I will , I think, invert much of what Philip has just said.  I think that ALAC -- the individual Internet users whom ALAC is chartered with representing would be much better served were they part of a stakeholder group in GNSO working along with NCUC where issues often align working along with registries and registrars at times when interests align there.
 But having the opportunity to work within that group rather than simply providing advice as currently is the structure.
 I think fewer groups, as three or four as outlined here, would drive toward better consensus at the same time I agree with what Milton said earlier that given the divergent interests represented here, expecting them to reach consensus is really often wishful thinking and consensus ends up being a synonym for who can hold out longest in favor of a status quo and block consensus and, therefore, block the ability to move from the status quo.
 Along with the recommendations up here, a recommendation that I haven't gotten a chance to throw in elsewhere so I will throw it in here is a recommendation that provisions that are status quo -- or that all policy have a sunset period after which it is reevaluated so old policies don't simply stick around because there cannot be consensus to change them.  I'm thinking back again to WHOIS where a strong poke that nobody would be satisfied might progress better than the number of groups that are satisfied with the status quo and block change.
 >>ELLIOT NOSS:  Those are great words to end on, "block change."  I think that I'm going to repeat two comments I made in Lisbon, expand on them a little bit with a third, hopefully all within three minutes.
 First, it has been again at an existence-proof level the case that the GNSO has been an exercise to date in blocking change and, I think, WHOIS, new gTLDs are the two best examples of that.
 The first substantive course that I recommended in Lisbon that I will repeat here was a combining of the ISPC and the BC because those two constituencies, in my view, are effectively one, both representing intellectual property interests.  If you look at the membership and the composition of those groups, who attends, who participates, I think that's just on its face the case.
 Sadly, that infiltration expanded to the ISPC where, for me, an ISP guy for many years, somebody who has been in the service provider industry for a dozen years now, really since its inception is saddened to see ISPs in some cases, not all, represented by intellectual property lawyers from inside large telecos.  That is not what the ISP constituency was intended to represent in terms of voice at the table.  So the specific recommendation as it relates to that grouping is twofold.  One, to combine the ISPC and the BC along with other commercial users who deal with suppliers of domain names into one constituency or one group, call it what you will.
 The second is to expand the concept of ISPC and instead have that be a service provider's constituency.  Those would be the thousands -- tens of thousands of companies who are between the registrants of domain names and the contracted parties, registries and registrars.
 It is a very easily defined class and one that is a huge contributor to the distribution chain that deals with domain names and is effectively without a voice here whatsoever.
 Last point I'll make is that while registries and registrars have combined to agree on issues, their issues are not issues in common and however you deal with the totality of the constituencies, I would strongly urge not lumping those two groups together.  They're in very, very different positions in the distribution chain.  And the technical way to tell them apart is that the registrars contract with ICANN -- sorry, the registries contract with ICANN.  The registrars contract with the registries.  But the more informal way is it's the registries are the one that make money and the registrars are the one that do the work.  Thank you.
 >>MARK McFADDEN:  Being with the ISP constituency, it is great to be able to follow that.
 >> ELLIOT NOSS:  You are the exception.
 >>MARK McFADDEN:  I appreciate that.  Since you were so progressive as to suggest who should be in our constituency, perhaps we should suggest who should represent your constituency.  But to the point, one of the things that has been a fundamental flaw in this process is something that I think has been misunderstand a lot in the last few years and that is trying to treat all the constituencies as if they were the same.  They should have the same sort of bylaws, the same sort of representational requirements, and the constituencies are not like that.
 Some of the constituencies have, for instance, a direct monetary, as was so clearly pointed out -- a direct monetary interest in the policy work that's being done.  Other people are participants, not because they have a dark monetary interest but because, for instance, they have an operational interest, they have a user interest and so forth.
 And so there are real differences in the way that the constituencies -- and if we believe in the bottom-up process -- the constituencies are built up from the bottom and, thus, they will have different membership and different self-organization.
 That's one of the most important things for me here is that I think we should focus on the fact that the constituency should self-organ that's one of the things I like about this proposal.  Frankly, Roberto, it is hard for me to stand up here because there is so little that I like about the rest of it so I thought I would stay on the positive.
 One of the things -- he's not here to hear this.  I wish he were really here.  I was going to add three words in that Milton is right.  I just wanted him to hear those words.
 [ Laughter ]
 But he is not here so we have to probably say it to his face.  Number one, I think that what you've done with the idea of these stakeholder groups is insert a level of indirection that does not solve the problem you are trying to solve.
 You know and you've heard from other people here that voting is going to be a part of the work we do.  As much as you would like to get around that fact, at some point, decisions have to be made.  We have to get some closure.  So now what you've done is you have put another layer of management in between the working groups which I do support and the council which has not been adequately defined for its new role.  And now you have the stakeholder group that you presumably want to have doing voting.
 What needs to be cleared up here is, I propose to your working group that you remove this concept of stakeholder groups and, instead, in its place solve the problem of the council which you've done a very good job of identifying what the problems are and go about the work of making it easier to come up with these new constituencies you identify.  Solve that problem and don't layer on a new layer of management inside the organization where you're frankly going to have the same problems.
 >>TONY HOLMES:  Tony Holmes.  Some of my comments are many miles away from Mark's.  I think as a group, it is your responsibility to focus on what are the key issues you are trying to solve and if you do that, I think some of the discussion we had earlier takes us a long way down that path, making things actually happen for the policy development process, the discussion we had around the working groups takes us a long way.
 I must admit, I find it amusing when people come along and tell you how to merge constituencies and then give you the reason why they're different and separate must be kept outside of that.
 Going down that path, the only comment I want to add to that is that there was a reference earlier to what I think is a very unhealthy linkage that's occurring within ICANN, and that is between ICANN as a body and those at the top of the food chain.
 We heard earlier today a remark from Paul about what is happening in IGF, the whole issue of ICANN and critical Internet resources is back on their agenda.  If you want that to be a successful dialogue, then I think you need to take note that that unhealthy linkage between those at the top of the financial food chain and ICANN is not standing this organization in good stead.  If you are looking to well together other parts of the community that bring differing views, differing perspectives, you put them all in one bucket and so they're all together and we've got this big lump that we need to take care of.  I don't think that does the organization any good work whatsoever.
 The other point I want to make is we need to fix this problem.  I do not believe -- and I think previous experience in ICANN underlines this.  I do not believe a big-bang approach is the way to do it. I think you need to slice this up and you need to make some immediate improvements in the working methods in terms of working groups and you need a new revolutionary path through within a framework to get to a set of principles that really make this thing work in the manner we're all seeking to.  And playing around with structures and constituencies isn't going to help that in the near term.  I'm not suggesting there shouldn't be changes in the longer term.  I think that's probably right.
 But some of those changes will only become apparent on the way this thing is reshaped in the early days.  So let's tackle the things that we can do, start making the process work and I think in terms of putting some thought into how we organize things, how we organize the working groups, the working methods, we've probably got a structure within reasonable do that now.  And it may need changing later.  In terms of the constituencies, I don't see any need to change that at all.
 I would welcome new constituencies and I think helping that along would be to put much more focus on having a clearly defined remit for each of those constituencies.  So we know who is eligible to join in a much better way, get some clarity around that.
 And that may generate new constituencies in themselves as well.
 >>ROBERTO GAETANO:  Tony, the point is taken.  You are over four minutes.
 >>TONY HOLMES:  Thank you.
 >>AMADEU ABRIL i ABRIL: Good afternoon.  I'm Amadeu Abril and I plead guilty of the current system of constituencies creation, among many others.  But, you know, the problem with constituencies, if there is one, is that they have been used as power brokerage centers, as political parties.  
 The original idea behind the constituencies were that they represent functions, not interests.  VeriSign and MuseDoma probably have different interests regarding how registries should work or what are the priorities for ICANN or whether to use registrars or not.  Large registrars and small registrars, registrars that act through resellers and low-margin direct customer registrars have probably different interests.  This is not the question.
 The question is:  They represent for the system the same function, and therefore, they are asked to give their collective opinion or agreement because of the bylaws structure in a certain number of things.  Now, for the rest, we have these other functions, and we come -- you know, individuals may have a different interest or you may have a different interest than that other one.  Okay.  Here is the place.  The general assembly, the public forum, the GNSO, wherever you call that.
 So my personal original idea for the constituencies was that they only existed as an electoral body for electing people to the council.  That is, I joined the GNSO and I declare whether my function within the -- you know, the system is being more a registry or registrar or a commercial user or not, and I will vote for, you know, having these interests represented there.  But for the rest, we all work together in the working groups in, you know, the general assembly, public forum, call it that the way you want.
 I am absolutely against the fact of multiplying political parties within the GNSO, because then the situation you have is that, I mean, interest parties giving them structure and structural power.  What you have then is a constituency voting for putting someone in the -- you know, just deciding what's the position, sending someone to a task force and the task force, you know, fighting for that position, making a second level of mediation for, you know, the consensus, then sending that to exactly the same structure in the GNSO Council, and at the end, you have three levels of power intermediation before, you know, this goes to the board.  It's too many levels of intermediation.  This should only be done once at the council level, and then the rest should be work.  And the current system of interest-based constituencies has led to, you know, republic -- Banana Republic-like dictatorships, phantom constituencies, the Marilyn Cade commonwealth of constituencies, and all sort of!
 strange bids that have no other stands than power brokerage, and this is not what we need.  We need to work.
 >>ROBERTO GAETANO: Thank you.
 >>AMADEU ABRIL i ABRIL: Oh.  So --
 [Laughter]
 >>AMADEU ABRIL i ABRIL: So no constituencies, yes to stakeholders that is function representation, and then interests can work and can coalesce in those structures, but not formally, with granted representativity at each stage.
 >>ROBERTO GAETANO: Yeah.
 >>IZUMI AIZU: Izumi Aizu, a member of the at-large from Asia-Pacific, and I think there are very few members from at least Asian region in this room or those who have spoken or are in the working group.
 One of the -- ALAC's sort of characteristics, as you may all know, is the geographic diversity, or we have -- we come from all five different regions.  It's vast.  And I think the GNSO doesn't have that.  Which I'm not saying it's bad, but it's different, right?
 We haven't -- at ALAC, we haven't really discussed about the report yet, and so I cannot really speak of any of the committee, but just as an individual from Asia-Pacific.
 I generally appreciate the inclusion of the individual users into this GNSO's stakeholders' group.  
 However, there still remains some tough questions, perhaps, where ALAC at least is tasked not only to deal with the GNSO areas, but the whole area of ICANN, including IP addresses, perhaps for the expiration of the V4 may come or migration to the V6.  
 So that means that, you know, putting individual users into the GNSO doesn't really, I hope, mean that -- folding them all into the GNSO basket and just forget about the rest of the world, which is not what you or we intended to.  But then do we represent both sides in sort of policy development?  It's a little bit complex.  So from the, you know, me advisory to post-advisory and have the individual users having, you know, voting power inside of GNSO is fairly tricky.  I have no solution, but I just have more questions.
 For the consensus or voting, in relation to the grouping of the constituency, I think sometimes we could think a kind of hybrid model.  I don't think the report suggests that everything should be by consensus or everything should be by voting.  It depends on the nature of the businesses sometimes.  
 Some of the RALOs discussing their own internal procedures agreed that, okay, we'll try to reach as much as possible consensus, and when we cannot reach or when it's better to vote, then we -- by consensus, we take a vote.
 So it is not an either/or situation that we think we can come up to some hybrid model that might be considered.  Thank you.
 >>ROBERTO GAETANO: Thank you.
 >>MIKE RODENBAUGH: Hello.  Mike Rodenbaugh, with Yahoo with the business constituency, GNSO Councillor, obviously speaking personally, since we haven't had anywhere near enough time to develop any sort of official position on this proposal at this point.
 But it's going to be pretty clear that we're going to be strongly opposed to the notion of merging together the BC, the IPC, and the ISPC for a number of reasons.  
 I think, first of all, there doesn't appear to be a strong reason to have such a, I guess, nuclear option at this point even on the table without trying all of the other options first.
 You know, focusing on working groups, modifying the PDP process, trying out all the different ideas, professional facilitators and that sort of thing, and then see how things are -- are looking, and whether there's still a lot of public pressure or -- or perception that the GNSO continue -- needs to have some sort of structural and major reform.
 Also, that should come in conjunction with the reviews of the other SOs and the ALAC.  So that everything is consistent and at the end we have one sort of cohesive, beautiful ICANN community.
 You know, generally adding new layers of hierarchy, as is suggested by this proposal, with the sub-constituencies underneath the stakeholder groups, generally just not a good idea, in my experience.  And of course, you know, restructuring in the real world is generally looked at as a last-ditch resort.
 So -- and then there's just the perception amongst those of us in these groups, at least the initial perception, that, you know, combining these constituencies together seems to be designed to further minimize our influence on the process.
 You know, despite the fact that business constituency, the IPCs, you know, they pay the ISPs, essentially are paying the bulk of the registration fees that drives this whole process, and, you know, have done a lot to build out the Internet.  Probably more so than -- than the other two registries and registrar constituencies.
 Also, it seems as if it is designed, perhaps unintentionally or will have the effect of turning those constituencies against each other, just with, you know, the -- when you put three people underneath one and they're all competing to have a certain number of councillors that are elected, it just seems more likely that there will be more fights between us than what exists today, and, in fact, that's the sole and only reason that we've heard from the registrars and registries for not combining those two constituencies.
 Despite the fact that to 99% of the public, they're one and the same.  They clearly have the same primary business model of trying to maximize registration revenue.  And then, you know, moreover, there -- there are -- many of them -- or most of them are much broader businesses than -- than registration providers.  So I think that also needs to be considered.  Whether there really is a problem once we end up trying a whole bunch of other options.
 >>MICHAEL PALAGE: Hi.  I always love the last word.  Mike Palage.  I'll try to keep this short.
 I supported in the LSE report the proposal for a 15 council member.  I think some of the terminology we used here has led to some -- some confusion.  I do believe that there are three -- three stakeholder groups, if you will:  Registration authorities, which are the registries and the registrars; the business interests; and the, if you will, noncommercial interests.  And the reason that I support -- or I believe in the concept of a registration authority community is, there are certain registries that have registrar shareholders, there are certain registrars that in the next round will be coming forward and proposing registries.  So the blurring of the lines, although as Elliot has said there are at times interests where their vehemently opposed, I think over time we'll see a blurring.  So my proposal, if you will, from an implementation plan would be for that council of 15, as originally proposed in the LSE, would be that the registrar constituency gets to elect two, the r!
egistries get to elect two, and then the registrars and registries together need to come to an agreement to elect the fifth.
 The five business constituency representatives, instead of, if you will, forcefully merging the ISPs, the BC, and the ISPs together, you would allow them to maintain their autonomy and let those three constituencies come up with a process to elect the five business stakeholder representatives on the council.
 The five noncommercial could be elected through the NCUC and the ALAC together coming forward.
 So what we would have here is instead of coming up with a new layer of management, it would just be representatives that would be elected to the council through possibly a different structure than they are today, while still recognizing the autonomy of the existing constituencies.
 What I would do is I would fix those constituencies in place for three years.  They would be the ones that would get to elect those 15 on the council, but what would be encouraged is for other self-forming constituencies to come forward over the next three years.  So if there wanted to be an individual users' constituency, they would have the ability to self-form and participate in the noncommercial, if you will, group.
 ICANN would be able to -- would be able to, if you will, provide them support to see if they could self-form.  At the end of three years, there would be a determination were they carrying their weight, were they doing what was necessary.  Then possibly, if you will, recognize them formally.
 As I said, I -- as Mike Rodenbaugh has said, when you begin to merge and do things, it really is a sign -- it's really a sign of last resort, so I think we need to let these constituencies try to work.
 And this goes to just my final point.  ICANN, with its ever-growing budget, has allocated a lot of money to global outreach, and the question that I am going to -- that I will be asking during the public forum later this week is, 91% of the budget, the 40 -- the 46 or $50 million budget, is attributed from revenues from gTLD registrants, gTLD registrars, and gTLD registries.  And my question is:  What is ICANN doing in its global outreach efforts to bring people in?  There's a lot of criticism that says, "Well, the ISPs are underrepresented."  They're not -- you know, they are not representative of the ISPs.  My question would be:  What is ICANN doing when it goes out and hosts all of these things to find people to be participating in the GNSO?  Whether they're volunteers, whether they have a financial interest.  You know, this is not just a one-way street.
 >>ROBERTO GAETANO: Time.  Thank you.  Thank you, Mike.
 Can you go to the last page?
 Okay.  We have also other questions and other things, but there's no time to relate to those, and so the -- the slide will show the e-mail address for providing comments.
 I will just go over quickly what is happening.  First of all, we have -- the other part that we needed to go over was the relationship of the GNSO to other structures, and that's -- the question is that are there steps that the GNSO can take to enhance its relationship with staff and the rest of the ICANN community, and are there steps that staff can take to enhance its relationship with the GNSO and are the chairs of the other two supporting organizations, the GAC and the ALAC, interested in more coordination.  That was trying to figure out whether there is some relationship and communication problem between the different parts of the -- of the GNSO.  This is not something that is internal to the GNSO, but is the relationship with the other parts of the organization.
 So also on this, your comments will be welcome.
 I think that I'll close it here.  I'm sorry for having gone very much over the time allocated, but we also started late, so you are now free to attend the other activities of the evening.  Bye.  Thank you.
 [Applause]
 >>ROBERTO GAETANO: Yes.  Go and multiply.
 [Laughter]