>>AVRI DORIA: Hi.  I just want to start this.  It's 2:30 now.  Please find seats.  As you'll notice some of the GNSO Council is sitting up front, some isn't, because we don't have enough seats, and so -- and so the GNSO Council is basically given the choice and, yes, speak into the microphone.  Good training.
 So some of the GNSO Council is sitting here, some is sitting out there at their option, basically where they can find a seat.
 At some point, the only seats left may be up here.
 So thank you for coming.  I have got the agenda up there.  Basically there's two items on the agenda, other than public comments, and an update on the status of new gTLD policy, and it's basically an update on the GNSO draft for new gTLD final report, and I'll be giving that.  And then we'll -- Kurt Pritz will be giving a talk on the new gTLD implementation overview of proposed implementation guidelines.
 So in other words, while we have been still trying to finalize the recommended policy, the staff has already tried -- and this is working against a moving target -- to start thinking about how these things would be implemented, and it could actually be seen as running code to see if we've got something that might work.
 So I will start.
 So this is the update on the new gTLD PDP.
 So I will go through activity since Lisbon, a brief overview of the current status, over a couple of discussion points -- specifically protecting rights of others, what we have called in the reserved name group controversial names, issues dealing with rejection of an application, issues dealing with disputes and objections, some discussion on IDN very briefly, and then the next steps for the council in this process.
 So in Lisbon, one of the things that happened was GAC released a principles document.  And the GNSO committee met to discuss this principles paper in detail, and that meant reviewing each of the principles, looking at each of the proposed new gTLD recommendations in the light of our understanding of their principles.  Then we basically developed a set of questions for the GAC so that we could clarify our understandings to see if we understood what it was they were suggesting.
 And I mean, this is especially important because we know that -- that the GAC has its advisory role, and as opposed to waiting until the end of the process when their advisory role is in play with the board, basically the attempt was to try and take that into account as we moved along in the process.
 So we also held a conference call with GAC members to discuss the range of understandings of the principles.  We got -- in the conversations with them, we were able to, by and large, clarify the questions we had.
 Obviously, there's still the possibility of interpretations that will need further work, but, by and large, at the end of the conversation, I think we felt that we had some grasp on what their issues were.
 And basically we then took their recommendations -- or, rather, our understanding of their recommendations -- with the ICANN mission and core values which are, of course, must predominate in these discussions and the constituencies' priorities and basically started to try and make sure that the recommendations we were making for policy were consistent with all three, as much as possible.
 So since then, continued discussion among the GNSO constituencies in an attempt to reach consensus on all of the GNSO -- and this should be "recommendations," not "principles," though within our final document, we also have some principles as well as recommendations, and implementation guidelines.  And all of a sudden, I realized myself that I'm talking a trifle fast.
 The discussions are continuing here in San Juan, and this forum, opening meeting, is part of that continuing discussion.
 In terms of the current status, we believe that the PDP is nearing completion.  The related working groups, which had been initiated to try and delve deeper into some of the thornier questions have all completed their work and have reported to the new gTLD committee.  So that's the internationalized domain, the IDN working group, which actually completed its work and reported before Lisbon, the reserved name working group, and protecting the rights of others working group.
 And all of the reports from these working groups are available on the -- on the GNSO Web pages.
 And the new gTLD committee is now working to integrate all of these recommendations, as appropriate, and sort of fitting them into the right place.  Sometimes the recommendations are perhaps more appropriate for the implementation planning than they are for policy recommendations.  Sometimes their scope may have been broader than the scope of our current PDP process.
 But in so far as they were appropriate to the PDP that is being worked on, there was an attempt -- and there's a continuing attempt -- to integrate those into the final report.
 We're holding this open forum, and will continue this work through this week and after.
 So when we look at the draft final report that's out there, this is the updated draft of 16 June -- and as I talk about this and even when I quote from that draft, I'll be quoting from the 16 June draft even though we've had discussions since then, even though some of the wording is already being discussed and variations being talked about.  I'll basically base this one on 16 June.  If we get into some of the further discussions, we may find it useful to then perhaps try and allude to some of the work that's trying to be done on fixing the wording on some or continuing.
 So we have six principles in the final report.  These have been stable for several months.  Now, basically what we're doing is several times we've sort of walked through the whole document, going through the principles, asking, "Is this one still okay?  Are people comfortable with it?  Is there anything that we need to discuss?"  And we've gone through.
 So it's very much an iterative process where we go through all of these things, see if there is any commentary, see it if any new wrinkles have shown up, any new perspectives, any new information has shown up to make us reconsider, but -- so on the six principles, those have remained stable for quite a while now.
 We have 19 recommendations.  16 of them have been stable for a while, and we will probably walk through them at least one more time.  Three of them are still in discussion, and I'll go into more details on the ones that are still in discussion as I move through here.  And then there were 15 implementation guidelines which have all been stable for a while.
 And since the stable statements have already been talked about, in public forums before, I'm not going to go through those this time, though in Kurt's discussion, some of those will be discussed because what the staff has been doing is sort of taking the stable -- and even some of the unstable -- and trying to work around the implementation details around them.
 So there will be mention of those in his talk.  I'm not going to go through them because to go through them would take longer than the two hours we have.
 So as I mentioned, the staff has begun testing the principles against -- and it's actually testing the recommendations is the words I should change in this -- against the implementation.
 The staff and the new gTLD committee spent Saturday afternoon discussing issues, with the recommendations found in the attempt to understand the eventual implementation.  And a report on Sunday was made to the GAC, sort of talking about how the recommendations in the draft report associated or mapped to their principles.
 So coming to a discussion on one of the areas where there is still a good degree of discussion:  Recommendation 3.  Strings must not infringe the existing legal rights of others that are recognized or enforceable under generally accepted and internationally recognized principles of law.
 Now -- then basically the recommendation goes on to list, as examples, several different laws -- and these are only examples -- that may be relevant in terms of understanding the existing legal rights.
 At the top level, the current approach recommended the creation of a dispute process loosely based on the UDRP.  Looks like we've got a very crowded room with not quite enough seats.
 And so as part -- and this would be differing from the UDRP, in that this would be part of the application review process, and not a process that came out after the establishment of the TLD, and in terms of thinking about these dispute processes, Kurt will be going into that in more detail.
 At the second level -- and this was work that was reflected in the PRO working group document -- there is an ongoing discussion about creation of what we're calling "reference implementations" of rights protection mechanisms, for applications that may warrant the use of such a mechanism.  And there were various cases where there were -- in some cases, in terms of trying to avoid defensive registrations and such, there was a strong call for the need for some sample rights protection mechanisms that someone applying for a TLD might want to use in their application.  The applications do ask, "How do you deal with the issue of defending rights of others -- or protecting rights of others?"
 In some cases, though, and this is part of the discussion that's ongoing, the application may be such that the community for whom the TLD has met, and the rules around how an applicant is judged as being qualified for one of the -- for one of a second-level or third-level, if they're being controlled, may not require such a mechanism because the rules in terms of how someone is qualified for one makes the use of a specific rights protection mechanism not necessary.
 So discussion is still going on, and methods of reference implementation are still being discussed.  And as we get into the discussion, I'm sure that other members of the council may want to take what I've said and sort of characterize it, perhaps more deeply than I have.
 Oops.  So I went forward on slides without realizing it.
 Sorry.
 The next category is one that, in the reserved name groups has been called "controversial names" though perhaps as we discuss it, we've sort of moved on from that as a characterization, but I wanted to put that here so that people don't wonder, "Well, whatever happened to the controversial names discussion?"
 So recommendation 6, as it stands now, is:  Strings must not be contrary to generally accepted legal norms relating to morality and public order that are enforceable under generally accepted and internationally recognized principles of law.  As with recommendation no. 3, there were then listed -- and you'll find that in the draft -- several different possible legal standards -- treaties, declarations -- that may, indeed, be relevant to this particular discussion.
 In the work of the reserved name working group, the working group opted against creating a category of controversial names for reservation.  So there was a little bit of thought of, "Could one create such a list, how would one create such a list?"  And it was basically considered almost an impossible task to create a list of all the names that one would want to reserve against in controversy, if such a thing was even reasonable to do.
 So the current approach recommends creation of an objection process as part of the staff implementation effort.  There's still a lot of discussion around, indeed, what would be the criteria for objections, and I'll talk about that a little bit more.
 So -- and both -- and this feeds into the topic, the discussion topic of rejection of an application, and the basis for such a rejection.
 We currently have, in recommendation 20 -- and by the way, I said there were 19 recommendations.  If you look through the list, you'll find that one, 11, was eliminated, so therefore, 20 is the 19th recommendation.
 An application will be rejected if it is determined, based on public comments or otherwise, that there is substantial opposition to it from among other -- from among significant, established institutions of the economic sector or cultural or language community to which it is targeted, or which it is intended to support.
 Now, in the conversations that we'd been having since Saturday, we have discussed things like the phrase "based on public comments or otherwise," we have had significant discussions on "substantial."  We've had substantial discussions on the word "significant."  We've had discussions on what it means to be an established institution.  We tried things like "existing for N number of years."  Well, then the question of how many years is too long, how many years is too short, given the dynamic nature of the industry we're in.
 We've had different discussions on what the scope of such opposition should be.  Should it be narrowly limited?  Should it be wider?
 Those discussions are very much ongoing, and, in fact, the list is currently seeing a discussion of different wordings for this recommendation.
 Indeed, the current approach recommends, again, dispute and objection mechanisms, and you probably have noticed the trend that on these controversial things, we're basically advocating a policy, a set of processes and mechanisms, as opposed to any a priori judgments so sort of say, "This is in, this is out," but basically making it an application process event where these things are dealt with.
 [Speaker is off microphone]
 >>AVRI DORIA: It's fine -- so I --
 >>DENISE MICHEL: I apologize for the interruption.  We still have a lot of people in the hall that would like to be part of this forum.  I'm not sure why it was scheduled for this room.  The ballroom is empty.  I know it's an imposition, but if you'd bear with us, we'd like to go right across the hall and reconvene in the ballroom and finish this.
 >>AVRI DORIA: Okay.  So I will come back to this slide.
 [Meeting was moved to Geronimo Ballroom and continued as follows:]
 >>DENISE MICHEL: We're all set.  Again, I apologize for the inconvenience.  We're ready to go.
 >>AVRI DORIA: Okay.  Thank you.  So pretty much -- I pretty much had finished with this slide.  This has been, though, a critical topic of conversation inside the committee, in terms of realization that certainly there will be applications and there will be criteria for rejecting them, but trying to set those up and trying to set up the mechanisms that are appropriate for that has been, you know, an ongoing topic of discussion, and it's something that once we've seen the first attempt at an implementation of all of this, you'll see some of the issues that have come up in that, and then I'm sure that it's something that we'll be able to talk about more.
 And so as I said, the method that's being recommended for any of the objects that come under dispute, whether it's things like the confusingly similar, whether it's topics that under the rubric of protecting the rights of others, whether it's geographical considerations or controversial name category that includes the names that deal with sensitivities and morality and such, all of those will be dealt with a variety of different dispute mechanisms and objection mechanisms that would occur during the application process.
 Now, this is currently a subject, sort of an iterative discussion between the staff's efforts to try and understand the policy recommendations, to deal with perhaps some of the ambiguities that have been left within the policy recommendations, and the necessity to actually implement something that works, and also in the implementation we may see an aspect that we didn't quite think about or think all the way through, and so very much advantage is being gotten from the interactive work of sort of policy definition and policy implementation.
 So some of the open issues that are sort of very much objects of discussion is:  Who has standing in a dispute or objection?
 There's the scalability question.  If we decide that everyone has standing in the disputes or objections or perhaps in at least one of the types of disputes or objections, how do we deal with that?  How do we deal with everyone?  How -- who or how can these disputes, these objections, be preprocessed if we decide that everyone has standing to object?
 There's a discussion of fee mechanisms for objections for disputes.  Fee mechanisms for the applicants, fee mechanisms for the registering of an objection, and perhaps mechanisms in terms of how those work if someone, you know, prevails in an objection, if an objection is spurious or deemed to be spurious, how the whole issue of fees would work out.
 There's also -- if you look at the possibilities that there are for disputes or objections during the application process, how do you manage the priorities of this complex set of things?
 Do all objections of all types get dealt with at the same time?  Is there a precedence that at this moment, we're dealing with one type of objection and, later, we deal with another?  At this point, we're dealing with disputes of, for example, the confusingly similar and at another point we'll deal with objections related to sensitivity about various words, various names.
 So trying to sort of figure that out.  And especially because sort of trying to balance between the difficulty of doing them all at the same time versus the extension of the process and the increasing unpredictability you get in its duration if you do them serially.  So there's a lot of those considerations in it.
 We've talked about IDNs.  We continue to talk about IDNs.  We've registered a very strong community interest in including IDNs in the first round of the new gTLD applications.  We saw that reinforced in this morning's conversations.  It's still a question, I think, to what extent this will be possible and if anything is necessary in the implementation to facilitate that, and that's a question that's still on the table and -- but has not really been -- we talked about it in the Lisbon time frame.  We had some conversations with the chair of the ccNSO on Saturday on some of the issues.  The council has not yet had a chance, after that discussion, to talk among itself in terms of what, if any, the implications were for the processes we were putting together.
 So the next steps.
 Our goal is to finish the GNSO processes within the next two months.  In the world we would like to see, we would have passed on our recommendations to the board before L.A.  That is a goal.
 This -- the work to do that includes integrating the results of open forum discussions and other comments that have come in, in the appropriate places, as appropriate to the processes.  Are they -- are they recommendations?  Are they principles?  Are they implementation details?  Et cetera.
 Continued iteration with staff on implementation issues.  Complete review and integration of the working group results, as appropriate.
 In other words, the results from PRO, the results from the reserved name group.  And then to hold the council vote on the final report and its recommendations.  Basically to vote on the recommendations.
 So that's for my report.  And now I'll pass the connection over to Kurt, who -- Kurt.  Oh, I guess I don't have to.  He'll be plugged in there.
 Who will actually go on and sort of discuss where the staff has started to talk about the implementation.  Thank you.
 >>KURT PRITZ: Thank you, Madam Chair.
 As Avri mentioned, staff has been doing quite a bit of work in parallel and iteratively with the policy development GNSO team in order to test and feed -- test potential policy recommendations and feed back what it's learned into the policy development process.
 So in that work, we've taken advantages -- advantage of a lot of prior experiences.  Some were good experiences and some were difficult experiences, but they're certainly all learning experiences, and used those in order to provide our best advice to the policy development process.
 In particular, and lately, staff reviewed a set of discussion points with the GNSO committee.  So staff reviewed their most recent policy recommendations in their draft form, and distilled down a list of questions or issues to discuss with the -- with the GNSO committee on this issue.
 So it was mostly to describe any issues we -- identify or flag any issues we had with the policy, as far as its being able to be implemented in a clean way.
 Also, to make a check that we -- we were sure we were understanding what the council was saying.  We're reading -- we've attended meetings, we read written words, but we want to feed back what our understanding is, so that there's clear communication.
 And then as I said, there -- we requested guidance on certain issues.
 And in some detail in the meeting on Saturday that Avri described, we jointly worked through some of the details of a proposed objection-based evaluation process, and I'm going to talk a little bit more about that later.
 But it's a very good concept and it is a very complex concept and a significant amount of discussion about this issue and the implementation of new gTLDs.  In that discussion, one feels more and more constrained to a certain model that addresses potential problems that could arise, problems that we know of arising in future rounds or from past rounds or past experiences or trying to anticipate problem that we don't fully understand.  And the most complex part of this process are determining on what grounds can an objection to a proposed gTLD be filed.
 What are very objective grounds, and then once an objective is made, you know, what are the rules wrapped around deciding to resolve that dispute and the resolution of that objection.
 The second complex part of the objection-based process is determining who has standing to object to a proposed TLD and on a case-by-case basis that differs.  
 What happened?  My slides are not showing on the screen.  They're right here.  I see them.
 [ Laughter ]
 Those aren't my slides either.  That's my boat.
 >>AVRI DORIA:  It is a nice boat.
 >> That's your escape route.
 [ Laughter ]
 >>KURT PRITZ:  So this is where I discussed building on our past experiences and making comment to the draft policy.  This slide describes the general purpose of our consultation with the committee this past Saturday and then this is where we are, where we're discussing the development of an objective-based process and the efforts we went through to work through some of the details.
 As a result of those discussions, there are going to be some amendments made into the recommendations and then there were specific actions given to staff to undertake right away in order to provide more clarity, more information so that decisions could be made.
 So as I get into some of the detail about a potential implementation of a new gTLD process, I just want to warn everyone that this is a draft work and is intended for discussion.  No decisions are made.  And, in fact, this model is intended to test the feasibility of a policy draft and the work of the staff in working through different scenarios and doing different sorts of research will inform the policy development team as to whether certain models are feasible to implement or not.  So this is sort of an iterative stage.
 We also want to do some advanced planning and implementing of the policy when it is delivered.  So as Avri said, there are certain things that are well-settled and staff can do work now in advance of the final policy delivery along -- as far as implementation goes.  So we want to take care of that timing and then certainly we want to solicit input in forums such as these as a step to the research that we're doing.
 So there's a few words that have been introduced into the lexicon of the new gTLD process that should become part of our everyday usage going forward.  If you think about these three bullets as being a process, it is after an application review for completeness, it is essentially a two-step process to test new gTLD applications and process them and then finally delegate new gTLDs.
 The two phases are intended to expedite gTLD applications that are non-controversial or not reserved names or don't raise an objection.
 So in this matter, applications that are not controversial can be processed in a fairly quick way and gTLDs can be delegated in a fairly timely manner.
 We think most of our discussion focuses on controversial names and the difficulty in considering them and processing them, but there is also a thought that there is lots and lots of non-controversial, non-objectionable names that can be delegated fairly quickly.  But the discussion is about the hard part.
 So in the initial evaluation, a string will be subjected to essentially five tests to see if it can be delegated right away -- essentially right away.  One, that it meets certain business and technical criteria.  Think about the sTLD round and the business and technical panels.  Our endeavor is to write more objective criteria around that so that decisions can sort of be rapidly made and there is a black and white sort of decision process there.  We do recognize the need for innovation and that innovation may give rise to technical issues and technically competent panels may raise them.
 So we recognize that there might be some iteration there, but we want to minimize it.  There is another technical test that the string does not lead to technical instability or unwanted results in the DNS.  What come to mind or have been mentioned to me, because I am not an expert in this area are strings such as .http or .www or something that might impinge on the architecture of the DNS. 
 .PDF might be another one.
 Whether or not the string is a reserved name is certainly black-and-white test.  Less black and white test -- one we think can be objectively made -- is whether a string is confusingly similar to an existing or proposed string.
 If a string is found to be confusingly similar to an existing string, the application would be rejected.  If the string is found to be confusingly similar to an applied-for string, then the two strings would go into a process called string contention resolution where staff has designed a few processes around resolving that contention.
 And then, finally, if no formal objection is raised to the TLD application or the string, more precisely, then the applicants of that string can enter into negotiations and settle a contract and become delegated.
 But any issues that are raised in these five areas would cause a string to be subjected to an extended evaluation process.  Now, what's key here, I think, is the RFP itself has to identify that extended evaluation process for the applicant so that when the applicant enters into the initial evaluation, their eyes are wide open as to what the entire process may entail.
 The extended evaluation might require the payment of another fee so that non-controversial applications or strings don't have to bear the cost of those that might require this extended evaluation.  Sorry.
 So this extended evaluation mostly is all about dispute resolution.  So the idea about -- behind objections and an objection-based process and dispute resolution, it is intended to take these disputes outside of ICANN.  An individual or entity can file an objection to the string, and if they have standing to object and it's a valid objection, that resolution of the dispute can take place outside of ICANN and that resolution can be conducted by a dispute resolution provider that's retained for this process.
 We also want to draw a line between community comment and formal objections.  We want -- the process is intended to draw a bright line between those two so that it is easily identified when an application should enter into this extended evaluation.
 The extended evaluation, I've written here, is additional scrutiny but I should have said  "extended evaluation" to be perfectly clear.
 And the process provides four specific reasons for objection.  Avri mentioned them in her presentation.  You will get to hear them again in a second.
 And then who has standing to object is determined, I think, by the reason for objection.  That's the discussions we've had so far.  So if the objection is by a third-party rights holder who feels their rights are infringed by the string, that person would have a standing to object.  But only the rights holder would have standing to object.
 If somebody were to object that a string purports to represent a community and that applicant does not represent that community, an established institution from that community would have to object, not just anybody.
 On the other hand, anybody could probably have standing to object to a string that's confusingly similar to an existing string.  Whoops, sorry.
 So just to reiterate this process, and this is in very simple graphical terms, the initial evaluation has these five tests and they'll occur essentially concurrently.  The box that says "formal objection," that is the raising of a formal objection by a person that has standing.  It is not the resolution of that objection or the dispute resolution process.
 So ICANN would conduct the other four inquiries and, in the meantime, make available to people an objection process where they could file formal objections.
 Avri discussed these in her presentation, and they are summarized here.  Part of the exercise for ICANN staff is to take the 20 policy recommendations and shuffle them in the right order so that a sort of coherent, sequential process and procedure can be developed.  So these grounds for objections are gleaned from the 20 recommendations made by the council.
 So the objections are -- and I'm repeating -- but they're confusingly similar to an existing top-level domain or another applicant string.  Avri described -- And that objection -- that objection can be made by anyone.
 May infringe on the legal rights of others that are recognized and enforceable under generally accepted and internationally recognized principles of law.  So, clearly, ICANN wants to find existing precedent and rules in order to resolve conflicts that resolve -- revolve around this issue.
 We see this as potentially sort of UDRP-like process.  Of course, UDRP happens after the name is registered and this occurs before the name is registered.  And there is other differences in the rules that would have to be developed and applied.
 There is a reason for objection if the string is contrary to generally accepted legal norms relating to morality or public order.  This, of course, is fairly difficult.  Again, ICANN would rely as much as possible on existing law or existing precedent.  There has been, in our discussions, some treaties that describe this exist term, the term "legal norms relating to morality or public order."  And so we think there's some precedence or some sort of precedence wrapped around this definition where decisions had been based on those treaties.
 We might also look to the rules of an existing jurisdiction, a single jurisdiction that has well-developed rules or laws and apply those.
 And then the last reason for objection is that the string results in objections supported by substantial opposition -- Avri's read this before -- from an established institution or a community member.  We had quite a bit of discussion about that because many of those -- many of those words are vague and, perhaps, it would be better put to make it -- to take the opposite tact on it and say that an objection can be made by a community member or established institution because the applicant does not properly represent that community.
 So if you think about two library associations in the world, one is just -- one may be just as suitable as another for representing the library community.  So an objection to the library applicant would not be sustained because they're both reasonable representatives of that community.
 So we can -- in questions, we can go and talk more about those.  If you think about then the extended evaluation process that needs to be well-defined, it would take on these four areas and the biggest -- those squares all look the same size or rectangles, but the one on the right-hand side, the dispute resolution process is certainly the biggest square and requires the most work.
 The other two, technical and business operational qualifications, that is if the technical community, say, has questions for an applicant beyond the black-and-white questions or the objective questions posed in the application, that the innovative model raises some technical issues, those could get raised in either of the other two boxes, the technical and business operational qualifications or the technical evaluation.  So we wanted to provide that capability in the process to identify questions that came up for different technical or business models.
 Then after an outside dispute resolution provider makes final determination on whether a string should be delegated, we might be left with one or more strings that are still identical or confusingly similar.  So in this case, we would invoke a string contention resolution process.
 We had this box at many different spots in the flow, and we settled on this because we didn't want to disqualify an applicant through a string contention resolution process when one of the contending applicants might get eliminated under some of the previous considerations.  So that's why it's there.
 So these -- this flowchart -- these two flowcharts have been fairly well thought through.  Staff -- policy staff and ICANN operational staff have sat and run through many scenarios, many test strings to see how this machine would work.
 We have -- this actually, when mapped out, has a process of about 26 phases.  Each phase has several steps to it.  So a lot of the process and procedure has been thought through.
 What remains to be thought through is the subject mostly of our discussion here, the hard part, right?  And the next steps for staff.
 I just want to talk a little bit about string contention first and the process for resolving that.  Another reason for having string contention at the end is it affords the parties involved to resolve their dispute.  So ICANN will provide meeting space or other facilities to afford parties to resolve their disputes before they get to that.  If the dispute is not resolved and there remain contending strings, there's a three-step process we see to resolving that.
 First is that the parties may elect to enter into arbitration -- binding arbitration.  That would require the agreement of two or more parties to enter.  Just one party couldn't force the other party into binding arbitration.  In the cases where one string represents a community or an established institution, that party could ask that a comparative evaluation be made of the strings.  And this is one of the recommendations of the GNSO Council, was that some consideration can be given to the fact that some strings may represent communities or established institutions and that community space should be respected.
 So -- and in cases where that comparative evaluation does not resolve the issue or that the parties don't enter into that evaluation, which one or more of the parties would ask for -- whoops, did I do that?  Shoot.  Okay.  I don't know how I did that either.
 Absent that, we would conduct an auction.  ICANN would conduct an auction to award the gTLD.
 So there is a lot of work that's been done on mechanics.  Now a lot of work needs to be done on the implementation of the objection process as a process.  There is considerable legal research to be done to inform the policy development team regarding how the dispute resolution would occur, the set of rules and the set of procedure for the dispute resolution provider for each of the four areas of objection.
 We intend to write a draft dispute resolution process, and the intent of that draft is to test how well this can be done.
 And at the end of that, we'll deliver staff work to the council for consideration and also post that publicly.
 A third area of research is how to include public comment as part of the process.  We want a bright-line process that differentiates between public comment and objection but certainly public comment should inform the dispute resolution process and those that wish to comment on a string should be able to.
 So I would like to thank you for wading through that, and I'm willing to go back through that.  The last 19 minutes and 10 seconds represents months, months of staff work and probably person years of staff work sitting around a table and hashing this detail out and there is a wealth of material available online about this and there is other staff-developed documents that are going to be posted that flesh out some of the work here.
 >>AVRI DORIA:  Thank you.  I guess the line for comments is open.  Questions?
 >> Whatever you like.  You're the king.
 >>VINT CERF:  Why don't you go ahead.  That way I can trump.
 >> ELLIOT NOSS:  It is a very quick question.  I heard all of this, but there is a very narrow point that I wanted to -- I'm sorry, Elliot Noss for the scribe.  There is a very narrow point that I would like a crisp, clear answer to which is, could you give us the tight timelines that you expect to be moving towards here?
 >>AVRI DORIA:  I can only give -- as I said in the presentation that I gave, the goal of two months to deliver it to the board.  After that, I think the timeline is not ours.
 >> ELLIOT NOSS:  Excellent.  How fortuitous.
 [ Laughter ]
 Could you give us the timelines?
 >>KURT PRITZ:  Elliot, there is a project plan that has a timeline that takes the policy development process being done in two months and then looks towards completion of the implementation plan about this time next year.
 >> ELLIOT NOSS:  If you were me would you make this a 2008 budget item?  2009 budget item?  2010?  '11?
 >>KURT PRITZ:  I think the latter half of 2008 depending where your fiscal year ends.
 >>VINT CERF:  Elliot, are you a gambling man?
 >>ELLIOT NOSS:  Having taken my money in poker, you know the answer to that, Vint.
 >>VINT CERF:  That was interesting.  I hadn't thought you would bring that up.  Let's see.
 [ Laughter ]
 As the entire discussion dissolves into something or another.  First of all, one thing that caught my eye very, very quickly is that there are some interesting possibilities --
 >>AVRI DORIA:  Hard to hear you.
 >>VINT CERF:  Sorry.  There are some interesting possibilities that seem to be allowed for in this process.  For example, city names look like they could easily fit into this process and and I know there has been publicly expressed interest in that.  On the other hand, there are sure a whole lot of cities.
 Is there somewhere in there consideration for geographic issues?  
 In the past, the GAC has brought up some concerns about the use of geographic references in top-level and second-level like dot info.  I don't know whether that already factored into the spreadsheet -- or into the evaluation sequence.  Would that be a reserve word problem if that were the issue?
 >>KURT PRITZ:  Avri is looking for the specific recommendation, but it is intended to be captured under the community set of objections.
 >>VINT CERF:  Or it might be captured under an even earlier one if it turned out to be a reserved element.
 >>KURT PRITZ:  That's correct.
 >>VINT CERF:  That's fine.  Just curious.
 >> MANUEL VILAS LOPEZ:  Hello, My name is Manuel.  I'm a member of dot gal. It's an organization who promotes a domain for the Galacian language in a similar way that Catalonia got dot cat.  So my first question is similar to Elliot.  When we are going to know when ICANN is going to publish the call for proposals because we have been working like two years already, and there is a huge expectation in Galicia related to this question.  
 And the second question I have is whether all the applicants are going to pay the same fee, whether they are cultural domains or domains related to industrial or corporate.
 >>KURT PRITZ:  So the project plan intends to complete this work in about a year from now.  There's several contingencies in that schedule.  One is the development of these dispute resolution processes in a way that can be implemented so that might require some iteration or rethinking.
 The fees are not -- the fees in accordance with the policy recommendations are intended to recoup the cost of evaluating an application evaluation.  So that we think there will be two levels of fees, one that will apply and, perhaps, another fee if the extended evaluation is required so that the straight-forward applicants don't have to bear the cost of the others.  There has been discussion, and Avri can probably talk to this better, at the council level of having another differentiated fee schedule where those that might not be able to bear the cost of fees or where fees are a barrier to entering the market are considered.
 Since we don't know how much it is going to cost at the evaluation at the end of the day, we will kind of hold those decisions off until we have a better idea of what the costs are.
 >>AVRI DORIA: But in general -- and anytime someone else in the council may want to add, the fees had been divided among those that went through a quick process and those that needed the longer evaluation process.
 The recommendations from the GNSO had basically recommended that there be perhaps another process to basically assist in the funding, but that the funding -- but that the pricing not be differentiated within the applications.
 >>MANUEL VILAS LOPEZ: Okay.  Thank you.
 >>BERTRAND DE LA CHAPELLE: My name is Bertrand de La Chapelle.  I am the French GAC representative.  I already made a few comments on the -- on these issues related to new gTLD creation.
 First of all, I want to nonetheless commend the work that has been done in the -- in the GNSO.  I've done that in the joint GAC/GNSO discussion yesterday.
 The present report is a very clear and very effective presentation of the approach that is being taken.
 I also appreciate the work that has been put in by the staff in designing appropriate implementation modalities for this approach and the whole thing makes a lot of sense given the approach that is taken.
 My concern is that in certain respects, the approach itself seems to present a certain number of elements that I'm not absolutely sure I agree with at a higher level, in terms of the balance between the development of a market, the management of this market, and the public interests that are related to the management of the global domain name system in the TLD space.
 It would be too long to discuss it here.  I just wanted for -- for the record to point to a very specific question.
 Provided that somebody comes with an application for a string, and that there is no reason in particular to believe that this string is wrong or that there are major objections to it, as you know in the GAC principles, we made a clear distinction between introduction of a string, delegation of the string, and operation rules for the string.
 Provided somebody comes with a TLD string application and it doesn't cause any problem, has the possibility of opening up a call for other actors to submit their own application for that string so that the best public interest is covered, has this been taken into account, and if yes, what is the counterargument for not doing this?
 And the last point is:  What is the mechanism that is going to be put in place by the staff and the further discussion for interaction with the different constituencies, including the GAC, on the implementation modalities?  Thank you.
 >>AVRI DORIA: Chuck, you wanted to --
 >>CHUCK GOMES: Sure.  Bertrand, with regard to the question regarding the scenario where a string is selected first and then multiple applicants could apply for being the operator, that was an issue that was discussed quite thoroughly, I think, very early in the new gTLD process, and the new gTLD committee moved away from that to the model you see.
 So the answer to your question is:  Yes, that was considered, and the committee elected to go a different direction than that.
 >>AVRI DORIA: Right.  And if I can add, as I remember, some of the considerations had to do with the difficulty of what essentially ended up a beauty contest between who supports the public interest best, and therefore, it went back to the notion of within around a first come, first served.  Now, obviously everyone within that round is -- is first, but, you know -- and are equal in terms of timing.  But essentially maintaining the first come, first served notion.
 And I don't know if you wanted to respond to the question?
 >>KURT PRITZ: Well, I think given that -- and it's a very good question -- the key to making that work is an effective communications plan.  And that's -- when you talked about review of the implementation by the GAC and the supporting organization, part of the implementation has to be a communication plan where the launching of a TLD round is clearly communicated everywhere in the world, and that the applicants that have applied for a TLD are clearly communicated, so those that might wish to object to the delegation of a certain string are knowledgeable of the process and can object.
 >>BERTRAND DE LA CHAPELLE: I just wanted for the record to mention that the balance between the market mechanisms and the public interests in the creation of new gTLDs in this mechanism seems to be biased in favor of the pure market mechanisms rather than the public interest.  But...
 >>CHUCK GOMES: Now, one other thing you need to recognize.  There's nothing to prevent multiple applicants applying for the same string, but they wouldn't be given inside information in terms of who else is applying for that.
 >>BERTRAND DE LA CHAPELLE: I know.
 >>AVRI DORIA: Yeah.
 >>MAWAKI CHANGO: If I may add, would you consider, for example, that another approach to address your concern would be to open up a little bit wider the gTLD application process itself, so that completely different strings competing to serve the same purpose could be considered?
 >>BERTRAND DE LA CHAPELLE: As a matter of fact, without getting into too much detail, two elements.
 The first element is, I do not believe that all strings will be called.  Some will have different functions.  Some will be limited to certain communities, some will have a pure commercial function, and this is fine.  I have absolutely no problem with that.  I'm just afraid that addressing all the different TLDs with the same procedure, irrespective of their functions, is going to create imbalances.  That's the first element.
 The second element is a notion that I've mentioned already in Lisbon, I think.  It is the notion of categories of TLDs.  I think -- and I hope -- that there will be a lot of interest for the new TLD system.  And I think some categories are naturally emerging.
 Dot Berlin and others are, as I mentioned, obviously outlining, you know, a category that can go be called cTLDs.  I heard of another application that might come up for intergovernmental grouping of some governments in Latin America who want to have a TLD for that, and they have agreed among themselves it will be governmental proposition.
 I don't think it will be addressed in the same way as a dot cars or a dot whatever else.  So the notion of categories of TLDs is something that I believe could be used not only to streamline the process of analyzing the applicant -- applications, but also in order to delegate and organize in a very efficient manner the treatment of several applications of a relatively same sort.
 But I'm happy to continue it.  I don't want to monopolize this.  I think the notion that you try to outline categories of TLDs and not only the position to a string, I think the whole process here is very useful.  It is excellent in terms of position.  In terms of evaluating the contribution and the various criteria, and I think there's an additional dimension that could be added.  Let me put it this way.
 >>AMADEU ABRIL i ABRIL: Hi.  I'm Amadeu Abril.  I would like talking to each and every single point, but I will not do that.
 [Laughter]
 >>AMADEU ABRIL i ABRIL: At least not here.  
 Just one suggestion that may sound very dumb, but I think it's not, Kurt.
 The roadmap.  The most important thing for having a long roadmap in the implementation is complete resolution.  We all agree on that.  We even don't know all the sort of conflicts that may arise, and we don't have the rules for solving them.  Good.  
 Then -- and we know that it would take perhaps one year.  Good.  Or perhaps five.  Who knows?  
 But is this good enough to stop everything?  Because perhaps we will have conflicts in 20% or 50% of the applications and no conflicts, no challenges, for the others.  
 So why don't we start in January, and everybody agrees that by submitting the application, if there's any sort of challenge, it's blocked until we solve the problem.  Those that arise some challenges will have to work, but will be exactly in the same position as applying in June.  Those that are not, you know, causing any challenge will go forward without waiting for Godot of this universal solution for peace and love that we are in search of.
 The other question regarding -- and it is a serious proposal.  It's not just a joke, yeah?
 I mean, let's not make the majority wait for the minority all the time.
 Second, because nobody's worse off, and some people is worst off.
 Okay.  Second point, what do we need for conflict resolution?  Sometimes we don't need conflict resolution.  We need something else.  Mandatory expert opinions.  That is, for instance, a ccTLD or language and cultural TLD or intergovernmental proposal, and then ICANN and the evaluators find themselves in the position of deciding, you know, what are the relevant institutions there is a challenge for from, you know, Amadeu Abril or the government of that city or Vint Cerf or whatever.  But there are people that know authority institutions in that city, there are people that know what's the language, who are the relevant institutions regarding the language and culture.  There are people that know, you know, what's the structure of the international organizations and the bodies representing each one of them.  So what ICANN is to do today is decide whether -- I mean, what's the challenge here?  That is, in case of Amadeu Abril, an individual opposing the TLD is enough, or not?  
 If it is, we have them waiting for the solution.  If not, well, if that individual is opposing, it doesn't douse anything.  
 But what happens if, for instance, the city hall of Berlin opposes dot Berlin or the Academy of Language for the Aleutians opposes dot gal.  That problem is significant and they should go to this resolution process.  And sometimes even not going to the resolution process, we should agree beforehand that for certain TLDs that represent certain communities, a relevant institution's challenge is enough to buy them some coco locos and mojitos and get into a room and say, "Find an agreement or you're not in," and that's all.  And we can do that today.  Well, today.  This week.
 >>AVRI DORIA: Yeah, okay.  I'd like to make two points and then perhaps other people will want to comment.
 The idea that the easy ones should just go now, the ones without conflict, does have a certain amount of -- sounds very appealing.  One of the principles, one of the recommendations, though, that we've been talking about is sort of the predictability of the process.  It's very difficult a priori to say that this one will -- these will not have any complexities, and, "Oh, yours just ran into a complexity.  Well, we haven't quite figured out the process yet for dealing that."
 So I think under the rubric of the -- all applications evaluated against predictable criteria, it makes it difficult, if we don't have the predictable criteria before we start.  So I think that was one of the points for why not just start with the easy ones first.  Does Chuck -- you wanted to add -- or -- 
 >>AMADEU ABRIL i ABRIL: Yes, you are correct, except the real problem is not this.  It's putting into place -- I mean that's what will delay for at least six months, putting into place the resolution system, and this is the part that doesn't make sense for that to wait all the time.  And look what happened before.  I mean 2004, 16th March, 10 applications arrived.  Do you know when they got the contracts and were approved?  One year and a half, two years later.  So I mean people is used to wait afterwards.
 >>CHUCK GOMES: Amadeu, your suggestion is appealing and be -- and I can assure you that the new TLD committee really wants to get this thing moving, contrary to what people might think by how long we've taken.  But it's been a lot of hard work, and we're getting close.
 But there's one principle that I think we unanimously agreed to, fairly early in our process, and that is there should be full -- fully disclosed information about the process, including the evaluation procedures, the challenge procedures, before it ever starts, so that every applicant knows exactly what they're getting into.
 And so to do what you're saying would mean we would have to start the process without some pieces being disclosed, and that would be contrary to what we agreed to.
 Now -- and we think, because we want it to be as objective and open and transparent as possible, that it's really important to be fully disclosed.
 For example, you know yourself that the -- the base contract happened after the process started in some -- we think that's not a good way to go.
 Any applicant that's considering to apply should know as much -- they should know all the detail about the process up front, to make the most intelligent decision and develop the best plan they can.  And so that would be -- doing what you're suggesting, as appealing as it is -- and I have sympathy myself for that -- that would go against that agreement that we had I think unanimously as a committee.
 >>AVRI DORIA: Yeah.  It has been one that's -- when I spoke earlier of recommendations that had been stable for a long time, that was one.  The other thing I wanted to mention, your second question about the conflict resolution mechanisms, I think that what you essentially suggested is, indeed, very similar to -- to what we've been talking about that when the conflict has to do with a specific language community or a specific type of community, that the specialists that -- that understand and know are, indeed, the ones that are asked as experts for an evaluation.
 So I think that's very much in consideration of what it means to have the dispute resolutions and such.
 >>PHILIP SHEPPARD: Avri, just before we have the next question, if I could just one observation on the same topic.
 I think the thing I find very unappealing about Amadeu's suggestion is what VeriSign very eloquently has demonstrated to this market, which is the importance of first mover advantage.  And I think if we introduce the process that was biased towards a certain type of application, and maybe led to uncertainty and an uncertain delay for other types of applications, that would be introducing the bias, which I think would run even further away from the public interest perspective that was so elegantly outlined by Bertrand earlier.
 >>AVRI DORIA: Okay.
 >>MILTON MUELLER: All right.  Well, I'm not a big fan of the philosophical basis of this process, and those kinds of concerns will be aired a lot more comprehensively at a workshop that we'll be doing on Wednesday afternoon.  ALAC and NCUC will be jointly sponsoring a workshop.
 But let me try to minimize what I see as the damage to freedom of expression that is -- this proposal is going to do.
 And what makes it so difficult is that I really think that the people who have done what they've done are really trying to have a kind of a consensus model of policy-making.  And are really trying to do the right thing and simply have no clue about the free speech implications of it.
 So just as a philosophical principle, let me begin by pointing out that diversity and plurality in democratic societies presumes that you don't have to have consensus to express yourself, that people can disagree with you, even violently disagree with you, and you can still speak.  You can still say something as and that is why the second level of the domain name is so robust because you don't have to ask anybody to register a second-level domain.
 But again, let me focus here on this specific recommendations that can improve this.  I know we're not going to reach full agreement on, you know, that particular issue, but there are ways you can minimize the damage.
 Number one, in recommendation 3, you simply have kind of a grammatical problem.  You say, "Strings must not infringe."  You should say "strings and the string selection process must not infringe," because when you mention freedom of expression rights, which do exist in many countries, believe it or not, there's no way for a string to violate a freedom of expression right.  The string selection process can violate those rights by, for example, eliminating all controversial words or censuring certain kinds of words, so you really need to modify the language there.
 Secondly, I would point out that there seems to be a contradiction between recommendations 6 and 9.  I can't -- I think it's 9 that says there must be a clear and pre-published application process using objective and measurable criteria.
 Now, how can any applicant measure and know in advance the degree to which their string is going to be offensive to any of thousands of governments, religious groups, ethnic groups, cultural identities, and so on?
 So you have to use an objective criteria there.  That means -- fundamentally, it means law.  And this is where we get into the deep philosophical implications and political implications of this.
 When I talk to people about this, they say, "Well, you know, there's a lot of governments around the world and they have diverse standards and some of them just don't like certain words.  They don't want them to be in there."  And my answer to that is:  I think this is not a complicated problem at all.  This is not hard to handle.  If governments have an objection to a string, the issue is, is there a law against it in their country or not.
 If there is, or even are they willing to pass a law against it.  In other words, are they going to go through democratic due process to pass a law against it?
 If that happens, and the applicant is under the jurisdiction of the objecting government, case closed.  They're dead.  We'll accept that.  I'll accept that.
 If the application is not in their jurisdiction, there's no reason to kill the applicant or the application, I should say --
 [Laughter]
 >>MILTON MUELLER: Some people might want to kill the applicant.
 [Laughter]
 >>MILTON MUELLER: But still, it's not very difficult for those governments to block a TLD in their jurisdiction or to pass regulations blocking a TLD.
 So, again, they have to follow due process.
 What I really object to is the idea that governments make something called public policy by expressing opinions in a GAC meeting.  I'm sorry, that is not law.  That is not anything that has any legitimacy.
 If governments don't like a word, either they have a law against it or they don't.  If they have a law against it, we will fully support their ability to enforce it.
 If they don't have a law against it, there should be no roadblock or obstacle in the ICANN process.
 So I would like to see some kind of strong statement to that effect in the report.  Even if it's a minority position, we would like that viewpoint to be strongly expressed in the report.
 One other point relating to this same theme.  In -- I think it's recommendation 6, you have the limitations on article 19, which are in Article 29, and I -- again I would point out that it says in the exercise of these freedoms and rights, everyone shall be subject only to such -- only to such limitations as are determined by law, so, again, you're thrown back on law.  You can't just say that you can limit free speech rights of article 19 because of opinions or wishes or objections.  You can only do it on the basis of law.
 So that's the basis of my concern, and I hope -- I think I've outlined ways -- again, realizing that we're not going to fundamentally everybody agree on this, I've outlined ways that the report could be modified to take these concerns into consideration.  Thank you.
 [Applause]
 >>AVRI DORIA: Thank you.  Would anybody on the panel like to comment further on that?
 >>ROBIN GROSS: Yeah.  I just wanted to pick up on that point, because a number of us are very concerned about this recommendation 20 that's -- that can open up this free for all process for people to object, because they don't like an application.  So if we're talking about in our principles we need to have objective criteria that's known in advance.  We're completely losing that with this recommendation 20 that's going to allow for people to object for nonlegal reasons, for nontechnical reasons.  Maybe they're a competitor and they want that string.  Maybe they're a critic, and so they -- they don't want their -- the person they're criticizing to be able to have that string.
 So we're opening up this big process for lobbying amongst various factions over words and over strings, and so a number of us very strongly believe that we should not have this public opposition process that just opens it up for a free-for-all for people to be able to say, "Well, I don't like that, I don't think they should be able to have it."  Driving up the costs to the point where it's prohibitively expensive to even apply for a string that's controversial.  Thanks.
 >>AVRI DORIA: Okay.  Thank you.  Next, please.
 >>WERNER STAUB: Hi.  Any name is Werner Staub from CORE.  With respect to the perfection that we're achieving with this gTLD process, it is interesting to note that when ICANN, in 2000 -- this was in Egypt -- decided that we should possibly look at new TLDs on the basic assumption that we don't want them, it took about a couple of months to have a process.  A little bit later, ICANN decided possibly it would want some gTLDs -- namely sTLDs -- and it took about two years to have a process.  And now when we've decided five years ago that we definitely want new gTLDs look where they are.  The problem is that we may be believing that we can really create -- how should we say -- predetermined process, but maybe if we really believe that, what we should do would be to play in the stock market.  Is that really -- can we really get to that level?  Doesn't it at some point actually just achieve the opposite.  The more we create an objective process, the more we create something is g!
oing to be.  At some point.  Just because it is predictable.  We can exactly get the thing that causes the problem.  So I think we have no other choice but, to a degree, do what Amadeu suggests.  It doesn't have to be formally like it's all open, but we have to encapsulate some of the open issues into a process more.  This was a challenge.  You know, we don't know everything that is going to happen in the challenge.  It is impossible to know everything that happens in a challenge.  But we can also work on the other aspect, how likely it is that somebody would -- would start now, and this is basically where I certainly repeat myself.  Let us make sure that we just don't look at the single round that starts somewhere next year, it seems, but that we announce a rotation -- that is, a round that's going to be in a preannounced fashion by second, third, fourth and so on round, that people know they don't have to jump on the first one and expect desert another five or 10 years bef!
ore something else would happen.
 >>AVRI DORIA: Thank you.  I think -- and I'll check.  One comment I wanted to make is I think that part of the concern with a predictable process with being as objective, fair, and transparent as possible came from the reviews and the commentary after the past rounds, and it's basically concerned with lessons learned from the past experiences and such.
 So I think that that's a big part of why there's an attempt to not make it perfect.  I think everyone has acknowledged that perfection isn't going to come out in any set of processes, but to understand and learn as much as possible from the past and get it as right as possible.
 I don't know if anyone else wanted to add something.
 >>CHUCK GOMES: Just real brief, and it's been a long time since we talked about this, but as I recall, part of our recommendations -- and it's not one of the main recommendations but I think it's more implementation -- is the plan is to announce not only the first round, but the next round, so I think, Werner, that that is the intent, and we understand why that's important.
 >>AVRI DORIA: Okay.  Thanks -- I wanted, before Vittorio starts, to mention that we do have a hard cutoff in about 26 minutes so we'll continue going through the line and discussions as we can but we do have a very strict cutoff.
 >>VITTORIO BERTOLA: Okay.  Sorry.  I'll stay in 25.
 >>AVRI DORIA: No.
 >>VITTORIO BERTOLA: I have an observation on the question, so... 
 The first observation is that of course I really think we should go forward with this as quickly as possible.  On the other hand, I am concerned by the idea that I sense that many good applications are going to come in this first round, maybe 20, maybe 50, maybe a hundred, I think we are talking about these kind of numbers.  And so most of the better strings could go away in this first round.
 And so was there any thought about how do you publicize?  Do you make these opportunities known, since this is a new process?  And if it's actually too quick, people might not realize that it's finally happening, so it might end up being mostly a group of insiders, maybe some governments or local authorities from developed companies and companies that have been participating in ICANN for a lot of time and maybe you would get complaints later, maybe in two years from now when people realize that this already happened and it's too late so maybe there should be some real effort in making everything really well-known, some outreach effort, I don't know.  Maybe it's more of a question for the board than for the GNSO, but may -- it's a problem that should be considered.
 And the same -- on the same line, I think we have to think of the problem of fees.  Well, the ALAC actually had some suggestions, you know, already four years ago, and I won't get into -- again into that, but I'm a bit troubled by hearing that, okay, we go with a flat fee for everyone and then sooner or later there will be a fund to sponsor applications from developing countries or from nonprofits or -- because in the end, when this fund will ever come, it might be too late.  So you're actually creating a sort of second class.  So I think that there is the need to discuss that before starting the first round.
 The question is about objections, so for example let's pretend that dot cat does not exist yet, and the company that has a trademark on dot cat comes and applies for dot cat.  Yeah, Caterpillar, for example.  And as I see the process, you're saying that only institutions that have -- are established in that field can object, so would the government of Catalonia be able to object to dot cat application for caterpillar because they feel that even if it's a totally different field, that's the string they are known under and so they would like -- maybe if they didn't present an application yet, but they are considering it, so they would like to stop the application and discuss, would that and acceptable objection or not?  And the other way around -- so I understand it, the other way around would be, okay, so if Catalonia applies for dot cat, caterpillar can object since they have a trademark, so these are the rights of others, so would it work the other way or would trademarks !
be treated better than other governments or did you have any thought of that.
 >>AVRI DORIA: Thank you.
 >>KURT PRITZ:  Thank you, Vittorio.  There is two questions I heard.  One was about a communication that I hoped I relayed during my talk that we think -- we think that the communications plan is a very important aspect of rolling out gTLDs.  
 Can you hear me?  ICANN has put together a line item budget for this fiscal year for addressing the upfront costs in launching the process.  The communication line item is significant there.  We understand communications need to go far beyond what's capable by ICANN staff, ICANN contacts and network in the regional liaisons and extend to newspapers, trade notifications, notifications to governments, notifications to IGOs, so a very important part of the implementation would be a complete a communication as possible because we want this to be available as you say everywhere.
 The question of the objection process and the example of dot cat and dot cat is very interesting.  And, also, when Avri and I discussed that recommendation, we mentioned that there is considerable discussion about potentially amending that.
 So I would think and this is my personal sitting around the table talking about this, if dot cat as in caterpillar were to apply for a string, it would not say, I am part of a caterpillar community but, rather, this is a commercial string that I would like to purchase.
 I don't purport to represent any community or anything of that nature.  So that would make it difficult for Catalan to object to that string as being not representative of the community because it doesn't purport to.
 If the present dot cat, puntCAT, put in an application for a TLD and Caterpillar Corporation objected to the string because it did not represent that community well, then that objection would potentially lose because dot cat does represent the community and, also, Caterpillar would not have standing to object, not being an established institution representing that community.
 Finally, we would have the case where they both applied for a string and they would enter into a string contention round and a comparative analysis.
 Now, as I understand the recommendations of the council, there is considerable weight in that comparative analysis given to who represents an established institute or what established institution represents a community.  So in that weighing, Caterpillar would not necessarily win because they don't represent a community so it might not bring as much value to the name space.  You can also just imagine sitting around our conference room table and discussing 100 of these scenarios.  It is pretty interesting.
 >>AVRI DORIA:  Any other comments?
 >>SLOAN GAON:  Good afternoon.  My name is Sloan Gaon.  My question is surrounding one of the recommendations around the use of registrars for all new gTLDs.  I know this has been quite a contentious issue.  The question I have is:  Has the council considered smaller registries which wouldn't necessarily warrant the use of registrars?  
 Thinking of relatively small registries, for example, if Vint wanted to register dot Cerf and he just wanted it to be for his family, would there be a requirement that Net Sol and Go Daddy offer dot Cerf if they wanted to?
 Once again, it is a question of a requirement for new registries to use registrars in the ongoing process.
 >>AVRI DORIA:  There has been some discussion.  Did you want to --
 >>CHUCK GOMES:  Sloan, the registry constituency representatives on the committee absolutely raised that issue and the committee elected to go with the requirement we have now, the requirement to use accredited registrars.
 What's going on behind the scenes right now is -- I'm not saying anything will come of this, so don't read that into that, but registries and registrars are talking about this in the case of small registrars, so we're still looking at it.  Whether anything will come of it, I don't know or whether it will come in time for this new process.  But we're still talking.  Not as a committee but people on the side.  If registries and registrars came up with some ideas -- and some have been floated -- we would put that to the council and the committee, and it will be looked at.
 >>SLOAN GAON:  Thank you.  I believe it is important that there is an exception policy.  It doesn't make sense in some registries to have that model in place.  It drives up costs for consumers, which is essentially one of the things we want to avoid.  Thank you.
 >>AVRI DORIA:  Thank you.
 >>THOMAS LOWENHAUPT:  Good afternoon.  My name is Tom Lowenhaupt.  I'm here on behalf of Connecting.NYC, a non-profit organization formed to secure and develop the dot NYC TLD.  I wanted to congratulate you.  You went through so many questions.  It is amazing how you are coming up with these answers.  It seems like a reasonable process.
 I just wanted to mention when our application gets before you will be from an organization that has a good representation.  That's a thing we spent a good deal of time of these days, is trying to figure out how we represent the many interests in New York City in a design and development of a TLD.
 And I don't know if it goes into your evaluation process, but it would seem to me that it would be a key criterion in deciding whether we were an appropriate organization to operate a TLD and so I just bring that point to you for your attention.
 And relating to that, with Dr. Grosstein, I published a paper called "Towards City TLDs in the Public Interest" which you might be interested in.  It touches on the issue.
 Also, we are going to be having to follow up on this governance issue.  We're intending to have a conference in November in New York City focusing on what is an appropriate governance process for a city TLD.
 And we will certainly be looking at the experience that ICANN has had over ten years in designing a governance process for a much more complex world but one that seems to work in many ways, so I congratulate you and thank you.
 >>AVRI DORIA:  Thank you.
 >>PETER DENGATE THRUSH:  Peter Dengate Thrush, member of the board and also a representative from time to time of InternetNew Zealand but speaking here entirely on my own behalf.  Let me begin by confessing, I am really a trademark lawyer and so I want to begin by thanking you for creating probably generation of work for generations of lawyers.
 [ applause ]
 The little box called "dispute resolution" will keep us going for an awfully long time, so thank you for that.  But more seriously -- and I have to say, I think Milton has said some of the things I have been thinking much more clearly.  I just come to this obviously very late.  I haven't participated in your debates and probably should have.
 But you do say a couple of times that we try to learn from the past experiences, and I can't see any evidence of that.  That seems to me that we haven't particularly learned from the highly contentious UDRP process that brought a lot of us actually into the ICANN process in the first place.
 Remember that in the end it was successful and has been successful only because it had extremely limited scope, far less than was originally contended for by the largely U.S. legal protagonists who were at that time facing enormous problems in cybersquatting.  We ended up with a workable solution but, remember also, that ICANN was enormously criticized at the time for what was seemed to be an attempt to try to make international trademark law.  To be fair I think ICANN has stayed away from that kind of an exercise since then.
 So I would like to come back to, I think, Milton's question which really wasn't answered and say what was the theoretical basis for allowing anyone to object to another's application of a gTLD?  What is it that gives someone a right to oppose in the first place?
 And what you seem to be doing is extending the UDRP to a whole lot of other issues, most of which to seem to me to favor incumbents, existing banks, businesses, religions, for example.  They have to be given standing.  And it seems to me power to buttress their incumbency.  I think this is the point that Milton is making.
 What is the law?  What is the certainty that allows someone to stop someone else from saying or doing something?
 Again, what is wrong with allowing the national systems of law to take action against anything in their country which breaches a national law?  Is it wise to try to provide legal remedies globally where national law and the treaty systems that are set up to deal with these issues are struggling and cannot provide them at the moment?
 But you also said that you may find when you're developing that you will look to a country who has laws which are developed in this area.  That was another reason why I think many of us got involved because the laws that were being looked at at the time for the UDRP weren't acceptable in many other countries.
 There is no international legal solution at the moment to the standing and jurisdiction problems in some cases.  Perfectly lawful acts in some countries, completely illegal in others.  And I'm just completely alarmed -- and it would be funny if it wasn't so alarming -- to hear -- This is with great respect to you, Kurt, because I have respect for you and I know the work you do and you are repeating some of the things that the council has said.
 But to hear panelists today talking about attempts to develop systems that give rights to communities.  Just take that, there is going to be a thing called a community, right, is there to object to someone doing something?
 You talk about the value added by a string.  There is no such thing as value in a string.  There is only value when a market develops around the uses which hopefully entrepreneurs will make to the thing.  To have high-priority decisions about value is, in my suggestion, completely hopeless.  
 For me in this exercise, I would need to have a theoretical basis for the objection at all to be very, very clearly articulated, otherwise, it seems I have to come back to Milton's suggestion.  It seems an attempt at consensus decision-making where in an area where clarity and precision is required.
 >>AVRI DORIA:  Thank you.
 [ applause ]
 >>WOLFGANG KLEINWAECHTER:  My name is Wolfgang Kleinwaechter.  I'm from the University of Aarhus.  I am a academician. As academician, I enjoy always hearing discussions.  When Peter says that you produce work for generations of lawyers, I can add you also produce good stuff for hundreds of dissertations which can be written about this.
 But I think at the same time where we have here this debate, in the reality, practice is moving forward.  And so far I think, you know, while we -- it is absolutely justified to have, you know, this framework discussion and try to force all possible cases, objections and other things.  Probably it is also helpful to look already -- or to start now already to look into concrete cases because we can learn something also from the cases which are already under discussion.
 In our last ICANN class in Prague we had a panel on geographic TLDs.  It is really amazing if you go in this direction how many projects are already discussed with dot PTN for the Patonia (phonetic) dot SIM for Wales.  There are plans for dot SCO for Scotland.  People who have an idea say they can do something but then they are confused as soon as they are approaching the ICANN process.  They have no clue, should we start now?  Should we wait?
 How this will go.  And probably also a political decision on certain categories would be helpful.  And so far -- I invite you tomorrow, we have organized a workshop on this issue and we have very practical cases, so we have dot cat as one example where Amadeu will present some of his experiences and then we have Swede projects which already a little bit advanced, dot Paris, dot NYC from New York City and dotBERLIN.  
 I see already a long queue.  Though, probably it is helpful to look into the concrete practices and then to have already now an interplay between the creation of the frameworks and a new emerging project so it means if we have a better communication among this, this will help to move forward a little bit faster because if I see the timeline here, though I remember what Werner Staub said.  In 2000, we needed three years to get the first round of new TLDs.  But now we are discussing this already for five or six or seven years.
 I think this is really too long and we have to find a way where we are more practical oriented, even as I said earlier, I enjoy very much this very good debate.  Thank you.
 >>AVRI DORIA:  Thank you.  I want to cut the line where it is now.  So you are at the end of the line.  Did anyone want to comment?
 >>WENDY SELTZER:  Thanks. Wendy Seltzer.  I will keep this very short because I can mostly just subscribe to the excellent comments that Peter made.  Like him, I don't understand what it means for a string to infringe someone's legal rights.  Infringement in any area of rights depends on the context and the opportunity that ICANN still has in defining a new gTLD process is to describe the context and the semantic meaning that will or won't be given to names, a name in -- a string floating in space is not the equivalent of anyone's trademark or the equivalent of anyone's banned word or harmful religion.
 And so I think these criteria are simply a veto to competitors and hecklers that we don't want to be creating.
 >>AVRI DORIA:  Thank you.
 [ applause ]
 >>KHALED FATTAL:  Thank you.  My name is Khaled Fattal.
 >>AVRI DORIA:  Into the microphone, please.
 >>KHALED FATTAL:  Closer?  All right.  This is this better?  One more time, Khaled Fattal speaking here.  I am speaking to you on the many hats that I wear, speaking as chairman of MINC, speaking as an active member of the Arabic community.  First to clarify, I think what you guys are doing is superb.  I don't want to make some of my comments appear as if they are not thankful for the effort.
 But I think the challenges far greater and far bigger than just dealing with whether the string has value or no value.  I think the challenge is when we consider the relevance of new gTLDs and the authorization of new IDN gTLDs.  Now we're talking about totally new things.
 Without sounding like I am lecturing, IDNs does not -- and I remember the first IDN session that ICANN held which was in K.L. -- Vint would recall this very vividly.  One of my presentations was differentiation between IDNs as a product versus a step towards multilingualizing the Internet and many at that time laughed.
 But the principle here is IDN is not just a product, it is part of multilingualizing the Internet and involving local community, languages, cultures, et cetera, et cetera.
 The challenge for you, ladies and gentlemen, is far greater and far bigger than what you can come up with because what you have been asked to come up with is a mechanism that authorizes.  It is far easier to become an enabler rather than an authority.
 I pass the same challenge to ICANN and its board.  When you're dealing with IDNs and future gTLD deployments -- I will call them deployments -- imagine asking the Arabs to come in and submit an application and standing at the doorstep of Marina del Rey asking to be authorized so they can have an Arabic DNS.  That's not the way this process is meant to go.  This, perhaps, does not pertain to your role within the GNSO but your part of the structure of ICANN that is deciding on how to authorize.
 So what it may work in ASCII or in English or it has worked and some people might challenge if it has, in IDNs, I can assure you it will not work because a group of people that have the best intentions trying to come up with a solution for everything is not going to be able to come up with a process that will show respect to local cultures.
 And I am not talking about strings.  I am not talking about relevance.  So dealing with the issues at hand, I strongly urge -- and I know Vint is working on his computer.  He has something that is probably urgent.  Perhaps this is something that ICANN can actually take lead on, especially before you leave the board, Vint, perhaps ICANN can become -- and announce itself as an enabler to these communities and to the Internet rather than an authority.  As an authority, you are going to have a challenge from getting participation from local communities and believing in the process rather than saying, well, we are an authorizer and these other people -- we get people to come in and object or not object.  That's not going to work, not in IDN.  It may still carry on in working in ASCII or in English, but I can tell you it is not going to work in that process.
 So there has got to be a deep-thinking requirement to look at the overall process of how we move forward and enable these communities.  So on that note, I will leave you to think about this.
 >>AVRI DORIA:  Thank you.
 >> DAN WARNER:  Good day.  My name is Dan Warner from fabulous dot com.  Yes, I am from Australia.  What I would actually like to talk about briefly is a bit about commercial reality, about finance and technical requirements of actually running a registry.
 I found it interesting that the technical requirements actually getting past that step is quite late in the process.  Even though we would all like to talk about esoterically that every possible extension should be entertained, you know, personally I would like to have dot Dan be wonderful.  All the Dans of the world could unite.
 But the reality is there is certain financial and technical requirements that you really have to have in place in order to ensure the security and stability of the Internet and of a community that is created, even with registrars recently in which they -- people have folded, that you create all sorts of havoc in the Internet.  What happens when a registry actually folds?
 I suppose I would like to see that the financial and technical requirements are both explicitly defined on what is the minimum requirement of a new registry, that the existing registries actually have to meet the requirements so you are not just going to create something that is an arbitrary new standard to be followed and then it's along the lines of a service-level agreement so you understand exactly what is required.
 Now, that is for a number of different reasons.  I'm ashamedly commercial myself.  If I would entertain the idea of me going after one of these things myself, the first thing I would want to know is what is it going to cost?  What is the hard cost to have a sustainable infrastructure, sustainable customer service?  How am I going to be able to serve the community?  What does it really cost to have a dot Paris?
 We can't all just have every extension possible and not be able to service it in a technical and financial level, and those levels need to be ascertained quite early in the process, probably the first and second steps.
 So can you answer the question as to what is the explicit requirement as to both financially and technically that's going to be required to have a level of service that is a base-level requirement to run one of these new extensions?
 >>AVRI DORIA:  I am going to say something quickly and pass it.  I think that the -- in the presentation given, those analyses are, indeed, done right at the beginning by ICANN in terms of the technical, financial, operational and there will be specifics on it.  It was only later in the process that if there was a problem that it went further and it went into the extended.
 Kurt, would you like to add?
 >>KURT PRITZ:  Avri has it right, but the other points you make are excellent and intended for incorporation.
 >>CHUCK GOMES:  Avri, just a real brief comment.  Unfortunately because of the limit of time today, we haven't really been able to talk about a lot of the recommendations.  We focused on three that generated a lot of controversy.  I certainly encourage people -- I know it is long, but there are certain elements of the report you can look at.  Some of the things I am hearing people say, I can tell they haven't looked at other aspects of the report and I encourage you to do that.
 The comments are still welcomed, but look at the bigger picture.  Most everything that's been brought up has been considered over and over again by the committee.
 >>MICHAEL PALAGE:  Thank you, Avri.  Michael Palage.  I speak today as someone who has a rather unique experience with new TLDs.  Over the years I have worked with dot info, dot coop, dot mobi, dot Asia and dot post.  I have a rather unique experience and actually continue to work with applicants that are looking forward to participating in the next round.  I have to admit that when I heard Kurt's statement about implementation being one year from this point in time, I was a little disappointed because I think that represents sort of a slippage.
 However, participating in ICANN since its beginning, I realize trying to push that date forward, resistance is futile.  So we are not going to push that date forward.  What I would like to do in making this statement is to try to sort of, if you will, put that in concrete as opposed to sand which tends to, if you will -- those lines constantly tend to get redrawn.
 And I think what's important here is if you look at the green paper, the white paper, various iterations of the MOU and even, in fact, the new joint project agreement which ICANN has entered into, there has been continually references to a new TLD process, predictable.
 And I think what's important is I would like to sort of make the two challenges.  Avri, as someone who has worked on the PRO and the reserved name working group, I will do everything I think we can to give the council the ability to make its decision in two months.  I think the challenge really falls upon the board to take action, hopefully at its annual meeting.
 I think, if you will, action in November is important because as was referenced in the opening ceremony today, Vint Cerf's term as chairman will end at that particular point in time.  I would hope after seven years as chairman of ICANN we would be able to sort of have as a nice little footnote, and, by the way, we did everything that the green paper, the white paper, the MOU and a joint, if you will, project agreement said with regard to introducing new TLDs.
 Again, what I would like to do, the challenge is let's get the council, let's get it done in two months.  Let's get it to the board and after the board is done, as we saw from the presentation today, ICANN staff has $1.6 million.  So I feel confident that $1.6 million should be enough to execute on implementing a predictable strategy and process for introducing new TLDs.  Thank you.
 >>AVRI DORIA:  Thank you.  I think it is a great challenge.
 >>BERTRAND DE LA CHAPELLE:  Hi, this is Bertrand again.  Very briefly, I couldn't resist after some of the comments that were made to come back on a very, very important word which is "value."  With all due respect to Peter and his presentation, I think it is not possible in this subject to consider that there is no other value than economic value.  There is economic value.  If you listen carefully to what Milton has said, irrespective of whether you share or not the position that he is defending, what he is saying is that this is also about the social value of the creation of strings.
 All I have been saying in this discussion is everything that has been done is a wonderful exercise and an excellent exercise that allows us to move forward in dimension of managing all the strings that are mostly for commercial purposes.  But I am not sure, though, that this will allow you or us collectively to handle in the same way those that have a more social purpose, community serving or whatever.
 I just wanted to highlight this because this notion of value is value that is added to TLD space and value added to the community on both the social and economic dimension.
 [ applause ]
 >>AVRI DORIA:  Thank you.
 >>SUBBIAH S.:  Hi, Subbiah, IDNS dot net.  I will try to keep this short.  I know Avri was worried I was going to ask this question maybe.  My reading of the document says many of the issues had recommendations in the back and there is a lot of discussion, usually a paragraph or so about each of the points.  But there was one issue, there was a lot of discussion, I mean, several pages, I think, a lot of footnotes and stuff.  That had to do with the issue of confusingly similar and in particular with respect to cross-languages IDN.
 And I was part of the contributed to the IDN working group that sent in some recommendations.  And the recommendations didn't treat the various types of confusingly similar, meaning something that is the same meaning across languages, something that is just visually similar, something that might be phonetically similar.
 I mean, I guess we could go to smell or something.  But the point is that these are the three that were discussed and I think the IDN working group came with some recommendations that treat these three things differently in the IDN context.  There was some discussion on that topic.
 As it seems to have been discussed in this document where, you know, there was an attempt to elaborate what IDN means in the context and that's what ran into five or six pages, there is discussion about patent law across countries so on and so forth.  Where do we stand on that?  It seemed unclear.  Is there an agreement that has been arrived on that?  What is confusingly similar?  And what is in these three things, or is this -- this is just input in terms of background and then it will be decided in some other way later?  That's my only question.
 >>AVRI DORIA:  Thank you.  I believe -- I mean, some of that is being dealt with in the implementation in terms of ICANN being -- and the staff there that are accepting the applications, being the first group that has to basically evaluate the -- are strings similar, are they confusingly similar?  I think a lot of it is also discussion and it is going to be in a dispute process where it's going to basically take other people to sort of point out often the confusing similarity.
 I don't know if anyone else would like to sort of comment beyond what I've said on it?  Though...
 >>PHILIP SHEPPARD:  I think the short comment is, I accept everything you say in particular when we get to things like IDNs, it is going to be a judgment, very difficult.  But our hope in a lot of these things that have raised concern and discussion to date is that we felt a need that needed to be there but our profound hope is that intelligent applicants will avoid going in that direction because of the richness of all our languages and all our imaginations in the short, simple, fast way you can apply for a TLD if you do it right.
 >>SUBBIAH S.:  So that means -- a take-home message on all three counts, it is possible to be confusingly similar across languages, meaning, you know, phonetically sounding the same, looking the same as in typing or something and also meaning the same when it is translated in some way to another language.  Is that my take?
 >>AVRI DORIA:  Yes, it is.  Thank you.
 >>CHUCK GOMES:  One of our objectives there was to make sure users are not confused.
 >>SUBBIAH S.:  I understand that.  Okay, thank you.
 >>AVRI DORIA:  Okay.  I guess, you had -- okay.
 >>ROBIN GROSS:  I just wanted to make a brief announcement.  For those who are concerned about going into greater depth on the freedom of expression issues in new gTLD policy, we've got a workshop on Wednesday at 1:00 in this room that will go into much greater depth on the free expression issues and possible solutions.  So you're all invited.  Please come.
 >>AVRI DORIA:  Thank you.  Any other closing comments?
 Okay.  I want to thank everyone.  I'm going to not say much because we have gone over our time.  Thank you for your comments and thank you.
 [ Applause ]