ICANN Brussels GAC Plenary Meeting with ALAC Tuesday, 22 June 2010 >>VICE CHAIR DRYDEN: Good morning, everyone. If we could take our seats, we'll get started. Thank you. Okay. So let's begin. My name is Heather Dryden. I'm the Canadian GAC representative and now the interim GAC chair. Janis has asked me to moderate this session, along with Sébastien Bachollet from the ALAC. The ALAC has proposed for this meeting today -- Could we please take our seats. Thank you. -- has proposed that we discuss today the DAG 4, in particular, morality and public order. And this fits well with the discussions that we just had in the GAC where we were discussing the same subject. So I'll allow Sébastien to introduce himself and make a proposal on how to continue. >>SÉBASTIEN BACHOLLET: Thank you very much. We appreciate very much to have this opportunity to exchange with the GAC members as at-large. I would like to make one quick introduction with one slide to show you the organization we are working with. And it shows you that as the GAC, we really cover the world. We're organized in five regions. Each region elects two people to the At- Large Advisory Committee, and we get one from each region coming from the NomCom. And our goal is to try and we hope we can have some help from your country to try to have at least one at-large structure in each and every structure. It will be a dream, but let's try to fulfill it. Now to go back to the substance, we really would like to discuss with you the DAG 4, and, more particularly, as the first item, MOPO, as Suzanne said the other day. I don't know if it's a registered trademark yet, but I want to give her the credit of this acronym. And I think we would like to introduce the subject, and I would like to give the floor to Evan Leibovitch. >>EVAN LEIBOVITCH: Thanks, Sébastien. At the At-Large Summit that took place in Mexico City ICANN meeting, there was a working group struck from at-large to debate issues regarding the gTLD creation. There were two substantive issues, one of which had to do with cost of access in some communities. That issue is being tackled by a separate group that in fact has a workshop going on tomorrow. But the other one was morality and public order, which was universally objectionable within at-large. Upon subsequent meetings, including some significant activity that happened in Nairobi, we originally met together with the GAC and in fact had a small subgroup that met together which we had our eyes opened and found that the current objection-based process is probably not much better-liked here than it is within ALAC. So we come to you today with the hope of trying to resolve this in such a way that recognizes that there are some universally objectionable strings, but tries to deal with their implementation in a way that isn't as wide-reaching and potentially confrontational and onerous as the current objection-based system. And so the idea of going back perhaps to what was originally conceived for this, which was creation of a list of restricted strings, that things, in order to get put on this list, would have to be universally objectionable on certain grounds, with a specific method to put strings on and to take strings off and do it that way before the fact of application rather than do it as an objection-based process, which is the way the DAG approaches it. So that's essentially the approach we'd like to talk to you about, is essentially trying to deal with this in a manner we think will work better and still accomplish the goals of identifying strings that we believe to be universally objectionable. >>VICE CHAIR DRYDEN: Thank you, Evan. Would anyone on the GAC side like to react to those comments? Please, Pakistan. >>PAKISTAN: Thank you very much. I think a very quick comment would be that, yes, the idea is fine to have a predefined list, but that should not close the door to have a mechanism where we can object later. Because, you see, it is, no matter how exhaustive you make a list, there's always a possibility, in most likelihood, that objectionable strings could make their way in. So I think that would be the only comment I have that -- whereas in principle, that may be the default to have a predefined list, but have the door open for appeal later on. Thank you. >>VICE CHAIR DRYDEN: Alan, please. >>ALAN GREENBERG: I may be the only one in the room who actually took part in the discussion that ended up with what we have right now. The concept of a list I'm not sure would work. The concept of subjects, I mean, may work, because any specific list, you can always come up with variants of it which aren't on the list but have the same intent. The rationale for the current morality and public order words there, as flawed as they are, the rationale was, if you have an objection process, you need to give the people who are going to adjudicate the objection some basis for making their decisions. And what you see in the applicant guidebook right now was an attempt to have -- to define what rules the adjudicator would use. So the real issue is, if that's not the right words to define those rules, what are? When we went through that process three or four years ago, there was -- we went through many iterations. Maybe we need to go back and look at the second to last iteration. Maybe it was better. But it was an attempt to find some rules which could be used in the decision-making process after the objection. >>VICE CHAIR DRYDEN: France, please. >>FRANCE: Thank you. As those who were here earlier, we already had a first round of discussion in the GAC on this. I think the more I think about it, the more there is a very, very strong distinction between two very different things. One is a given string not in the root at all because it has been deemed unacceptable, whatever the process is. Another thing is a string that is acceptable for some and not acceptable for others would maybe get in the root, but there still would be always the legal right and the legal capacity for one country, for instance, to say, "I don't give access at all to this whole TLD in my country." Let me say that actually the second case could also apply for a given ISP that says, "Well, I want to protect whatever concept my clients have, and I promote my access to the Internet as being not that." So these are two different concepts. Because not getting something in the root, we can have two extremes. One is as soon as one actor, let's say, for instance, one government says, "I have an objection," then it's out of the root. That's one extreme. And I say one government. It could be one other actor. The other extreme is, anything gets in the root as long as there's no technical danger for the system, there is no objections on the other grounds of trademark or -- or previous rights or confusing similarity and so on. And everything to deal with objectionable words is dealt with at the national filtering level, let's say. If you really try to think about the two extremes, they are not really satisfactory, because they are either too open or too closed or -- it really doesn't work. The second element -- But the distinction is clearly that. There will always be the sovereign right of one country to adopt blocking. And there will always be a desire to enter in the root as many things as possible and hope for universal accessibility for what is in the root. One thing that came to mind -- and I would be happy to have feedback on -- is, there is some similarity with the problems or the solution that has been found in the dialectic between article 19 and 29 of the Universal Declaration of Human Rights. When you think about it, there is a sort of underlying principle that seemed to emerge that we all want and have almost as a principle that anything that is in the root, in principle, should be accessible in all countries, and that any blocking or any limitation regarding content should be done at the most granular level to limit as little as possible the -- the access to information. And if you look at the way article 19 and article 29 function, what article 29 of the Universal Declaration of Human Rights says is, there can be limitations to access to information and whatever, provided it is done by law and that it is for the purpose of guaranteeing access, or -- surprise, surprise -- for purpose of morality and public order in a democratic society. And so what is here as a tension is that we don't know whether there are, indeed, universally objectionable strings. I'm not even sure that there are really things that would be so unacceptable by absolutely everybody. But let's suppose that there are. Maybe the question is, is there a way to formulate those? Is it, as Evan was saying, is it a finite list? Is it a type of subject? It's likely to be very complex. So having something that does not get into the root because it is so universally objectionable is, I think, rather unlikely also, because somebody who is launching a TLD has no interest of being blocked in every single country where he will be trying to reach people. What about a mechanism that says -- or what about exploring a mechanism that says there is actually very, very few or even no limitation on the question of morality and public order regarding entry in the root itself -- I'm explore hearing -- entry in the root itself. The only criteria are about stability, about who has the right to run the TLD, and that's all. And that on the other hand, any country -- if we take countries -- that wants to block a whole TLD has to do it clearly through a legal procedure according to very specific criteria that are explicitly explained. So an element of visibility and transparency in the process. And, three, when the application is being made, what about a mechanism that would allow for a period relevant governments to tell the applicant, "Listen, if you apply for that very string, it is so offensive for me that I tell you I will take a national law to block it on my territory. If you change the label in one way or the other to make it less offensive, then it will remain accessible and we will do the filtering for the content of the lower level." Because there will always be a challenge. I give you a very concrete example. We've been talking about dot Nazi forever. How should we treat a dot NSDAP? NSDAP is the name of the party. It's the same. And it could be a very valid TLD for historians who are dealing with the study of the second world war. And so in France, for instance, depending on whether NSDAP is used for hosting hate speech and promotion of Nazi memorabilia and things or if it is for the study of those things, it will be treated completely differently. So just to give a framework, we are really exploring at that stage. And I would be happy to get feedback from friends in ALAC. >>VICE CHAIR DRYDEN: Thank you, France. I have European Commission and then Alan. >>EUROPEAN COMMISSION: Thank you. I want to express some caution, actually, about an idea which emerged in our earlier session. And some of you may have been here. But I hope you'll forgive me for repeating it for those who weren't. That's the idea emerging about the fact that countries could perhaps block top-level domains at the national level. I wouldn't dispute that they have a legal right to do that. But what I would propose or suggest to you is it won't take them long before they realize they can also add top-level domains at the national level. If they have their own version of the root zone file which they're disseminating -- actually, I would be happy to be contradicted -- they'll realize they can add some. We can lead to a situation where countries realize they can have conflicting top-level domains. I can have one version of dot phish, and you can have another one. What I'm cautioning against here is that if the patient has a problem, we should make sure that the cure for that problem doesn't kill the patient. The patient here, of course, is the universal resolvability of the root, something that I think is very precious to those who understand why the Internet has been successful to date. And I just hope that in ten years' time, we don't look back at the last 20 years of being a golden age of the Internet where there was universal resolvability, and in order to solve this specific problem we have here now that we abandon that. So that's a personal view in some ways, actually. But I know it's shared by some other members. But I really do find the prospect of that part of the solution to this problem might be that countries choose to start a process of running their own version of the root zone file is something I find quite worrying. Thank you. >>VICE CHAIR DRYDEN: Thank you for that. Alan. >>ALAN GREENBERG: That intervention was a great lead-in to what I was going to say, or it could have been the other way around. The concept was discussed reasonably extensively at the last go-round when the GNSO was looking at the issue. That was a time when there was a lot of discussion of there being -- of people setting up alternate roots. And to be candid, the issue was discussed. I think I was one of the ones who raised it, as a matter of fact. And the concept of in an ICANN document we even talk about the concept of countries filtering out top-level domains and essentially setting up multiple roots was deemed to be so offensive that it wasn't something we were willing to do at all, even though it was a viable way of addressing the problem. So history keeps on going around. >>VICE CHAIR DRYDEN: Evan. >>EVAN LEIBOVITCH: I wanted to speak in support of what Bertrand was saying and add that there is already a sort of precedent within the framework of getting new TLDs. And that is the trademark clearinghouse, that when somebody is submitting a string, that it has -- that the proposal is that it go through a clearinghouse, and that the clearinghouse provide advice back to the applicant of how likely it is that their string is going to be rejected based on trademark grounds. So we already have something where there is a process of putting forward a string, getting advice back on that string of what is likely to happen should you proceed on it, and then it is up to the applicant to decide whether or not they want to go forward taking the risk or to rethink what they're doing. So I do think -- I very much like what you were suggesting and would comment that there is already mechanisms within ICANN on issues such as trademark where the same -- where a similar approach is being taken. >>VICE CHAIR DRYDEN: Any reaction to that proposal? Brazil, please. >>BRAZIL: Thank you. From our perspective, it brings a little bit of concern leaving the final decision of going forward with the controversial gTLD, leaving this decision with the applicant itself, because this can go to a kind of judgment regarding the capacity or the size of the market or Internet users in the country or the community that are -- that is offended. And it may cause a unequal relation of power on the Internet. So we believe that the consensus and criteria that could be developed on a consensus basis through all countries may be a better way to deal with this complicated issue. So we agree we need further time for discussion about this question. Thank you. >>VICE CHAIR DRYDEN: Any further comments? We haven't solved it yet, so.... Please. Adam. >>ADAM PEAKE: I was wondering, many governments around the table. You are sponsors of ccTLDs. So how do you deal with this issue at the ccTLD space that you have sort of control over in many ways? I mean, what policies do you have on morality and public order, if any? And if you don't, well, why the concern in the gTLD space? But, more importantly, what lessons can we learn from how you do implement this in your own ccTLD areas? >>VICE CHAIR DRYDEN: I see someone in the audience. Please. Oh, yes, Greece. >>GREECE: In Greece, the national regulatory authority that's responsible for telecommunications and domain names approves every domain name before it is actually registered. So the ccTLD is one, and there is no question of approving the string for the ccTLD. This has been here. This is dot GR. But there is a procedure as it is out of the Greek national law in telecommunications by which, in a very quick way, this takes less than a week, all domain name applications are looked at and are approved by the NRA. The experience here is that we will probably not approve a very small number. Why not out of 10,000 -- that is just a figure. And sometimes we receive claims by competing interests and we have to resolve who has the right on a domain name. But there is a procedure in Greece whereby we deal with domain name management. It's not at the level of approving the string of the ccTLD. >>VICE CHAIR DRYDEN: France, please. >>FRANCE: Thanks very much to Adam for raising the question in those terms. Actually, it's a very interesting angle to take, because when you think about it, the Domain Name System is what, in complex system theory, you call a fractal system. It's a self-replicating mechanism at different levels. I.e., if you own a second-level domain in dot com, you are the master of your domain for the third level. Like I am IBM.com. I am responsible for what comes below. And if I register something in my own third level or after the slash and this content is unacceptable in one country, you know it will be prevented, it would be not accessible in that country. If you get a TLD, that's the same, regarding what happens at the second level domain. And what we're talking about here actually is ICANN behaving almost as the registry for dot root and actually defining the second level registration policy for that root. In that respect the interesting point that Adam raised is that at the national level, in France, I learned recently, actually, that dot FR, the ccTLD run by AFNIC, indeed has a list of terms that are restricted precisely because of the national legislation on hate speech. I don't have the list in mind, but I discovered that there is, indeed, a list. The key problem we're facing here -- And it's a pretty good list. The key problem we are facing here is this works at the national level because there is a complete feedback loop between a national entity dealing with registrations according to a national law, enforced by the national jurisdictions. If we were to apply this system at the international level, we will be confronted, even if the clearinghouse notion is interesting in terms of a warning signal, if we were to apply this as rigidly as the national ccTLD does at the international level and combine all the different names and words that are inacceptable in one country, at least, we would end up with an incredibly controlling mechanism that would basically guarantee that we have the universal accessibility, but at the same time we have hampered free speech in a somewhat strong manner because we would align to every single objection in every single country. And so I think the analogy is interesting. If we combine it with the notion of an indicative clearinghouse, so that if you are an applicant, maybe you could check in the thing that, before applying, there is a word that is likely to be really objectionable. It can be almost almost declarative by governments or other entities but by governments in particular that say indicatively, there is a list of words that I would consider absolutely unacceptable at the national level. But I just want to finish by answering both Bill and Alan's comment regarding the risk of having a blocking at the TLD level. I didn't think about what Bill is mentioning as the risk of replicating basically different files. I think it would more be done by giving instructions to ISPs on the given country to not give access to that content. So I'm not sure it would necessarily lead to alternative roots or alternative TLDs. But here again, we can think in sort of fractal manner because the problem we are facing here regarding blocking is not that different from the problem we are facing on globally hosted content platforms. When, for instance, you have a problem on YouTube in one country because of one video, and, for instance, if I take an example I know that in the Netherlands, I was discussing with a company in France called Daily Motion, which is the equivalent of YouTube in France, and there was a video that was made at one moment in time that is illegal in the Netherlands. And so they took the measures to make sure this video is not accessible for somebody who comes from the Netherlands. It's not perfect based on the IP addresses, but there is some element in that. And so this notion of granularity, and the two principles I think that are emerging are the general principle of universal accessibility of all the TLDs, and the second principle is whatever measure of restriction should be done at the lowest granular level possible. And maybe we can go as far as saying under almost no condition, or an exceptional condition, can a very specific, explicit, legal measure to block a whole TLD be taken. I don't know, it's -- Again, we are trying to find boundaries there. >>VICE CHAIR DRYDEN: Thank you, France. Sebastien, please, and then New Zealand. >>SÉBASTIEN BACHOLLET: Yeah, as an end user participating to the advisory council within AFNIC to set up some part of the policy, I am sure that Bertrand can't know by heart the list because it's very, very long. But at the same time, as AFNIC was just revealed by the government to be the official registry, we will have to go through this list because something who was through ten years ago, it's maybe not through anymore, and at the same time maybe there are some terms we have to add. And it's an ongoing process we have to go through. As we will not solve this problem here today, may I suggest that we think if we can, GAC and ALAC, set up a working group to further discuss this issue and maybe to come to some idea before the next meeting. At least if we can do prior it, would be good to have that, insert it to the DAG 4.5 or 5 or final one, whatever number will be. And then to help the process to not to derail from where we are today. Thank you. >>VICE CHAIR DRYDEN: New Zealand. >>NEW ZEALAND: Thank you, Heather. I just want to put something on record on behalf of New Zealand that's come came out of the (inaudible). Just as Bill on behalf of the European Commission has highlighted an issue of breaking the Internet through different types of roots, and thus breaking down the universal Internet, for many countries, and mine is one, principle of free speech is extremely important. And the idea that the government would be filtering -- compulsory filtering part of the Internet on the basis of a domain name is absolutely repugnant. I would be very, very disturbed if the GAC or ICANN or DAG 4.5 or 6 or 10 proposed that as a mechanism for getting around this problem. Thank you. >>VICE CHAIR DRYDEN: Additional comments? Pakistan, please. >>PAKISTAN: Thank you, Madam Chair. A very quick one. I think we should also remember that blocking is not the answer in some cases. For instance, the honorable colleague from France gave an example of a site called Nazi Forever, I believe, or there could be, say, a region that warrants independence from a country and declares a certain country code top-level domain equivalent. Now, maybe half the countries in the world or some countries are supporting that independence, whereas some other countries are totally against it. So blocking in itself does not solve the problem, and I think we have to look for a more concrete solution that the domain, notwithstanding the earlier comments from our colleagues from EU and New Zealand said that blocking will also fragment the Internet. So we have to look for a more robust solution where such controversial names do not appear in the TLDs. Thank you. >>VICE CHAIR DRYDEN: Indonesia, please. >>INDONESIA: Thank you, chair. First of all, the ccTLD which we are having today, it doesn't give a lot of problem because basically it is a country name shortened by two words, and so on. Indonesia for ID, Malaysia with MY and so on. So it doesn't give a meaning at all. But gTLD might be a name. And that can be all right for some countries but may be very bad or extremely bad, extremely unacceptable for other countries. Of course each country will always be able to block a TLD using their own regulations; however, if there are many countries blocking a particular TLD, from my point of view it will be seen as a failure of ICANN, especially for the GAC, to look after or accommodate the public or global situation for this. [ Alarm sounding ] >>INDONESIA: Oh! [ Laughter ] >>INDONESIA: Okay. Well, my friend Bertrand mentions there is a sensor here, and if you say "a failure of ICANN," you know.... [ Laughter ] >>INDONESIA: Just a joke (laughing). Don't take it too serious. Just a joke. Just a joke, of course. Well.... [ Laughter ] >>INDONESIA: Okay. But again, what I mentioned is that ICANN, as a global organization, worldwide, where the membership -- [ Alarm sounding ] [ Laughter ] >>INDONESIA: Well, let me try to rearrange my words. [ Laughter ] >>INDONESIA: It will be a sin to say something bad about this. Another one will trigger another sound like that. Okay. Well, what might be seen as a good effort of ICANN is that it can also accommodate the public -- the global public situations. So it is, in this case, the reason why we need to set up some sort of standard operating procedures for the so-called gTLD. In the previous meeting I mentioned about the possible trial period, complaint procedures and so on. Now, bearing in mind that if a GAC member, if a GAC member said yes or no to a gTLD name in this meeting, he or she must also stand to defend his or her statement back in the country. In a country like Indonesia with 240 million people it is not so easy to defend what I am saying here. And so actually what I am saying is that before we really give -- affix gTLD names worldwide, then we have to have an SOP to make sure that gTLD is basically worldwide globally more or less acceptable, and will not cause any problem or expectable problems in the near future. Well, this is for gTLD, but actually this is just additional, not related to gTLD. Perhaps we can discuss about possible rating of the Internet content, just like in the movies, for example. In the movies you always put XXX, whatever, or "with parents guidance," or things like that. So if GAC can also consider discussion like this in another location. Thank you. >>VICE CHAIR DRYDEN: Thank you for that, Indonesia. Are there any reactions or additional thoughts on this topic? No one wants to try to set off the alarms? Oh, Alan. There we are. >>ALAN GREENBERG: I have a question. I was just wondering are there any existing gTLDs or ccTLDs which in any country or language are significantly offensive? And if so, how is it handled? How is it addressed? Is it ignored or is it, indeed, addressed? Among all the two-letter combinations and the few gTLDs we have, there must be something that's offensive somewhere. >>VICE CHAIR DRYDEN: Evan, please. >>EVAN LEIBOVITCH: Actually, within the circles that some much us have been talking, just as dot Nazi has been bandied about as an example of something that would be universally considered objectionable, we have also heard use of, for instance, dot gay which is a genuine attempt at an application under way right now as something that might be partially offensive and is the kind of thing that we need to deal with, of a TLD that might be highly objectionable in some societies but totally fine in others. And this is the kind of thing that I think we need to deal with. Under the current objection process, there was a significant concern that an objection by some societies might force something to be globally unallowed. And this was a significant concern. So there are definitely some concrete examples, I think, of TLD attempts that we need to consider to make sure that, you know, that national interests are preserved while, at the same time, allowing those societies that want something and are okay with it to let it go. >> There's an issue here of the string itself being offensive, and I don't think you can say the letters G-A-Y are in any way offensive. They are not rude words. It's not the sort of thing you say when you stub your toe, and that's an offensive word. I am talking about swearing and what have you. That's probably offensive. We are talking about the content that we're anticipating under the TLD that some people would find offensive, and it gets a little bit more -- I don't know. What are we talking about here, really, I suppose? Are we talking about offensive words, the actual string itself, the combination of letters that are offensive, or is it the anticipated content that's underneath? >>VICE CHAIR DRYDEN: I have France and then Fouad, please. >>FRANCE: Thank you. First of all, I would like to share how good this discussion is, because, actually, we're addressing a certain number of topics that are very delicate, that touch very important national and sovereignty issues, very important philosophical, ethical and rights problems. So I think it's very fruitful. And the examples that we are taking help understand. Actually, Adam is right to remind us that this is about labels, and that there is a general, as I think I was saying earlier, there is a general principle that the registry, even less than the hosting platforms and so-called intermediary liability situation that, the registry is not responsible for the content that is on the pages that are on the second-level domain. So it's a label. And so the question of why is it objectionable, really, that a word is being used is a real question. This is why I was wondering whether there is even something that is universally objectionable in terms of the wording -- the word itself. Another element is that if you take dot gay, for instance, you can have exactly the same TLD and exactly the same content behind with a dot LGBT for lesbian, gay, bi and transsexuals. And who is going to say that LGBT is a combination of letters that should be unacceptable? So it is the same purpose. And I think there is an educational role also for all of us, including in ICANN, to understand the role of a label and the fact that there is a strong distinction between the registry function and the content. And I do agree that it will be difficult to understand also in the legal frameworks of a lot of countries. But maybe it's an objective. Another element I want to put in the discussion in terms of example is I had very, very lengthy discussions a long time ago with Amadeu Abril i Abril regarding introduction of dot cat and the extreme attention that he paid and that they paid to explaining and to building it as dot Catalan and not dot Catalonia, which is answering part of the comment that our colleague from Pakistan was making. There are very sensitive issues regarding regional, subregional parts of territories. On the other hand, the beauty of the Internet is the capacity to link communities of language. And in many cases, some orientation, some presentation, some choice of words for the TLD can allow the TLD to be accepted because it is presented in a way that takes into account the sensitivities. On the other hand, let's be honest, if somebody wants to use the TLD name as a provocation, then there will be a problem in general. So the question that people have to ask and the regime should encourage is to make sure that freedom of expression is fully supported, and that it is made in a way that makes this expression acceptable as well. And so if you want to do a TLD -- I mean, if instead of doing dot cat, people speaking Catalan said dot free Catalonia, it probably wouldn't have faired very well with the local authorities and so on. And the purpose would be the same. So so it's just a matter of managing the situation we're in which is as the Internet expands the diversity of actors and the diversity of people and the diversity of value systems expands as well. And strangely enough, the more diverse, the more we need rules, and the more diverse, the harder it is to find universal rules. So this is the challenge we are facing here and it's not new. But I think this discussion, again, is trying to highlight the different distinctions that will help us move forward, and the more we dig into it the less the mechanism that is proposed in the DAG seems appropriate. >>VICE CHAIR DRYDEN: Thank you, France. Before I give the microphone to Fouad, anyone else wanting to speak to this topic, if you could signal in the next few minutes. We're going to move to some other issues related to gTLDs. The ALAC has indicated they would like to raise some additional issues. So Fouad, please. >>FOUAD: The issue that sometimes is discussed around the, for example, of dot X, for more than like around 50 countries, the issue stands as a matter of social, culture and religious norms and beliefs. And with such a TLD objectionable by, like, a certain portion of the world which represents something like 1.5 billion people, I'm still wondering how would governments be looking at that issue? Because it's the sort of issue, we even have the same issue with the human rights declaration that it's not fully accepted within the context of, like, being a universal principle. And once again, if you look alt the point of filtering and blocking, some people see it as an opportunity that since everything -- content like that will be categorized that way. Would you call it an organized categorization or a natural categorization? And it would be much easier to block such content and access to it. So it would require some, like, you know -- I don't know, some comments on how does the world see it this way. Like how do the governments see it this way. Thank you. >>VICE CHAIR DRYDEN: Thank you for that. I have a request from the side there. If you could move to the microphone and introduce yourself. Thank you. >>BOB BREUN: I am Bob Breun from KnujOn. I just want to point out how difficult it is to decide words that should be banned. A gentleman in Virginia in the United States had his license plate taken away because it was 14CV88 and it was code word for white supremecist. 14 word statement, CV for Confederate Veterans, 88 was the HH in the alphabet for "Heil, Hitler." He had it for a number of years until someone figured it out and then they took it away. If you try to do the coding of all those kinds of things, it is going to be almost impossible. So you are opening up a very, very, very large can of worms. >>VICE CHAIR DRYDEN: Very interesting point, and I imagine that in IDNs that only gets more difficult. [ Laughter ] >>VICE CHAIR DRYDEN: So I don't see any requests from the floor on this issue, so let's leave it there, and -- Indonesia? You do? Please, go ahead. >>INDONESIA: Just to comment a bit about our colleagues, because you mention about a social and economical problems in the Internet. I think one thing we have to realize is that in the world today, you have two space. One is the real space and one is the cyberspace. The real space is basically looked after by discussion in the United Nations organizations. The cyberspace, this is the place. So can we imagine how important it is to ICANN looking after the cyberspace, a parallel world of the real space. So that's why we have to talk about everything, from the social, technical, economical, moral, and so on, you see. Because it is the space where cyberspace is discussed. So I hope there will be study more about this, how cyberspace can change -- can be in parallel with the real space, and that's why we have the problems that you mentioned. Thank you. >>VICE CHAIR DRYDEN: Thank you for that. I have Portugal and then Evan. Please. >>PORTUGAL: Well, just for a short remark addressing this question of the kind of procedures and restrictions that could be considered to solve this problem on one hand and the freedom of speech problem that was raised. Well, from our point of view, what concerns us is that certain gTLD strings could be objectionable based on law of some jurisdiction and, therefore, could lead to fragmentation of the TLD space. This is the basic concern. That has nothing to do, from our point of view, with freedom of speech. Because freedom of speech is related to content. And to be very clear about that, even without the new gTLDs, there is no restriction whatsoever that comes from using the ones that just exist related to freedom of speech. So I think these are very different questions. >>VICE CHAIR DRYDEN: Thank you for that. Evan, please. >>EVAN LEIBOVITCH: I just wanted to make sure we didn't leave this issue without coming up with some kind of follow-up action item for this. It's an important issue, and I know you want to move on to other things. I just want to make sure we didn't without figuring out some kind of follow-up on it. >>VICE CHAIR DRYDEN: Would you like to consider the working group proposal or did you have something else in mind? Okay. So I invite to you reflect on that particular proposal. Okay. So we can move to Adam, I believe. >>ADAM PEAKE: Well, I wanted -- the subject I wanted to raise -- and I'm going to put Bertrand -- sorry, you shouldn't smile at me. You raised something yesterday, Bertrand. But perhaps if we're going to have a DAG that is ever enforceable, there needs to be a mechanism where people interested in this work get together and turn it from a large -- what do you call it? -- to operationalize the DAG so it can be voted upon, implemented by the board. And I wonder if you would wish to address that, probably in a personal capacity rather than as France. I don't know, but.... >>VICE CHAIR DRYDEN: France, please. >>FRANCE: Thank you. Actually, it's -- the idea -- the initial idea was an intervention by Jon Nevett, who basically suggested the idea of a gTLD summit in the discussion this -- yesterday. And his idea was to say, fundamentally, the situation we're in is heading towards a case where the board will have, or will feel in a situation where it wants to close, like, set the cursor on a few remaining issues so that the program can move forward. And his idea was, it's up for the community to basically close the program or to finalize the gTLD program. And he was suggesting the idea of a summit or some getting together of a portion of the community to iron out the last elements that need to be ironed out. I happened to be in the queue at that time. And it actually connected with another thought that I had, which is the following analogy -- and apologies for those who have already heard it. In any national process, when you draft a law, you do preparatory work. And the preparatory work usually produces a huge amount of paper, up to the size of the Eiffel Tower sometimes. I mean, it's tremendous. But when you draft a law itself, hopefully, the law is a relatively shorter document so that it can be used, manageable, and it doesn't have ambiguities or things that are not clear. And I've been wondering whether, actually, the DAG or the ancillary documents and, basically, the more than 700 pages altogether that constitute the current documentation should not be considered as preparatory work. And what we actually need now is to finalize the law or the regime. And a 20-page, 25, whatever, document that would have basically three parts. The first part would deal with everything related to the string or the applicants, separately, like, is the string a reserved name? Is the string confusingly similar? Is the string a trademark? Is the string blah, blah. Or for the applicant, is the applicant technically viable, financially viable, and so on? That's the first part. If the string is okay, whatever the rules are, if the applicants are allowed to compete, you get into the second stage, which is the designation of the appropriate operator. If there is one string okay with one applicant, okay, move to the next one. If you get one string with several applicants, then you kick in the community evaluation, the whatever beauty contest we can put in. You choose the delegated operator. And then the third part is the most important, actually, is the ongoing operational rules, currently, the registry agreements, plus the different elements related to monitoring, enforcement of the regime in the long term. And so the idea that is emerging is the following. And I would be happy to have feedback from colleagues on the ALAC and also the GAC, because we didn't have a chance to discuss it. If such an approach were to be envisaged, is there a possibility to see that, for instance, in the autumn, before Cartagena, there is a real period of time that is dedicated for a smaller group to finalize and produce this kind of document so that in Cartagena, there is a final document. Because otherwise, the situation we're in is there will be a staff finalization, then a final board finalization, and there will not necessarily be a complete buy-in. I understand that Peter Dengate Thrush, following those discussions, has already sent a mail within the board to say, "What do you think of this idea of a summit?" So on. It's an open question. But I think it deserves attention. And I think Jon was right to raise this topic. How do we finalize? How do we get to closure? Instead of kicking the ball down the line all the time. That's the general idea. And it's open for proposals. >>VICE CHAIR DRYDEN: I don't see any requests for the floor. [ Laughter ] >>VICE CHAIR DRYDEN: I think we have two proposals for various kinds of follow-up, one related to the morality and public order discussion we had for a working group, and Bertrand's proposal, more broadly, regarding the DAG 4 and its next iteration. I suspect that people need some time to digest. So I would suggest that the GAC think on this. And if we arrive at some sort of conclusion or suggestion for the ALAC, that we communicate to you a bit later. Failing that, I think we have about ten minutes left. And if there are no requests for the floor, we could conclude a bit early. Sébastien, Cheryl. >>SÉBASTIEN BACHOLLET: No other point you want to raise, at-large? Elaine. >> Is it on? Yes. Okay. I just wanted to point out that right now, we have blocking of top-level domains all the way down to village level, where an ISP provider can simply say, "I don't want to deal with dot NG, because there's so much crime associated with that string." So the idea that the root is fragmented because there's local censorship over a top-level domain doesn't make sense to me. And if someone could elaborate on why that would be true that the root's fragmented if there's just not access to it, I'd appreciate that. Or if you would just consider that in your deliberations. >>VICE CHAIR DRYDEN: France, please. >>FRANCE: I'm sorry, I'm abusing. I'm taking just the last minute to share something with the ALAC regarding the discussion on categories of TLDs. One of the things that could be explored -- and I mentioned that briefly in the new TLD session yesterday -- is that irrespective of whether we can introduce the use of categories or categorization in the DAG itself, a very interesting avenue seems to emerge, and I felt that Kurt Pritz is open to thinking about it, is the use of categories in the batching process for the evaluation of the different applications. The idea is basically, once ICANN receives and closes the call, receives the list of, as they say, 500, plus or minus 400, applications, you take the list, you make four piles, and you just throw them by categories with the six or fifth pile for others. And you first determine whether there is basically 90% or 20% of the applications that fall in those categories, and the staff would organize its own teams so that at least it's the same people who deal with all geographic type of application, the same that deal with the brand type of application, and so on. And so the batching can ensure that you're not prioritizing one category versus the other, but try to make an appreciation on how ripe the different applications are. So just to share this. It's worth further discussion. But we will explore it as well in the GAC. Thank you. >>VICE CHAIR DRYDEN: New Zealand, please. >>NEW ZEALAND: Just picking up on the question that was asked, nobody's suggesting that filtering out a particular domain name is breaking the root. My point about filtering was that any form of filtering on domain space is compulsory filtering. And compulsory filtering is a problem for New Zealand, and I believe should be a problem for most people. That was the issue with that. The breaking of the root was another solution which, I think, Bill dealt with in his comment where people could have their own version of the root which was not a universal version of the root, and that would break the Internet. Thank you. >>VICE CHAIR DRYDEN: Thank you. I believe you are our last speaker today. So with that, I -- Sébastien would like to make some remarks before I conclude. >>SÉBASTIEN BACHOLLET: Yeah. I will not discuss the topic we have raised. We wanted to thank you very much to have us with you today. I think it's a very important part of at-large and ALAC work to be able to exchange with the GAC formally in those sessions, but also informally, and even formally intersessionally. And we are very happy to have this opportunity. I would like to raise two points. The first one is that we have a dongle with all our documents in three languages, in French, Spanish, and English. And if you want to have this type of dongle to see how we work, what was the document we are using for our work, we will be very happy to give you. And, I guess in the back of the room, we will be able to give you if not immediately, before the end of this meeting. And I give one to my co-chair today. And the other we'll have to ask. My second and last point, it's to say specifically more to the European representative, but it's open to everybody. We, the EURALO, the European at-large regional organization, a showcase tomorrow where we'll be showing all the organizations at the European level from 10:30 to 12:00 at this arc foyer at the third level. And we will have the chance to have Antti Peltomäki, the Deputy Director General from the Information Society and Media Director General from the European Commission as the keynote speaker. And we will be very happy to have you if you are able to come and to exchange with us at that moment. And, once again, thank you very much to have us here with you today. Thank you. >>VICE CHAIR DRYDEN: Bravo. [ Applause ] >>VICE CHAIR DRYDEN: Thank you, Sébastien. And thank you to the ALAC for meeting with us. Joe, I believe, has a quick announcement to make. >>JOE TABONE: Yeah, if I can remind the Commonwealth representatives to stay behind for about five minutes, and we'll head over to the far corner of the room, if you could. Thank you very much. (Morning session concluded)