NCUC gTLDS - Freedom of Expression Workshop ICANN Meeting San Juan, Puerto Rico 27 June 2007 >>ROBIN GROSS: Okay. I think we can get started. If folks want to wrap up their conversations and take their seats. Welcome to the civil society workshop organized by NCUC and ALAC. The name of this workshop is called "Keep the Core Neutral," and the focus is on the freedom of expression issues in the new gTLD policy that we're working on here at ICANN. In terms of format, we wanted each one of our speakers to give a presentation, a short presentation, and then we want to open it up to you folks to really hear what the concerns are from the ICANN community. So hold onto your questions and save them, and we'll -- you'll get a chance to ask them after the presentations. So our first speaker today will be Mike Palage, and Mike is an attorney in California and -- Florida. I'm sorry. I thought you were in southern California. The other sunny state. [Laughter] >>ROBIN GROSS: And ex-ICANN board member. So without further adieu, why don't I turn it over to Mike. Thanks. >>MICHAEL PALAGE: Thank you, Robin. And what I wanted to do here is actually just instead of go through a lot of slides, I wanted to keep this more, if you will, a -- a dialogue format. Prior to the Lisbon meeting, Avri Doria and I coauthored a paper called "Please Keep the Core Neutral," and the question behind this paper was the discussion within the community regarding the ICM application for the xxx domain. It's important to note that in this paper, we did not take a position on whether the TLD should be approved or denied. What we talked about were some of the things that, if you would, be kept out of that discussion. And what we were concerned about was the political influence of making decisions at the core which we thought were more of a technical nature. One of the things during my service on the board, one of the benefits I took out of it was learning from vicinity, Steve, and some of the other, if you will, founders of the Internet that always talked about how what made the Internet great was the innovation that took place at the edge. How having a neutral technical core allowed innovation. And what we proposed in the paper, Avri and I, was to allow that neutrality to continue at the core. We recognized that sovereign governments had the ability to regulate the actions of citizens, and what we proposed was hypothetically if a dot xxx or adult TLD was ever added, those governments that had expressed concerns regarding that would be able, under their laws, to block access, if you will, at a national ISP level. So this was something that we had talked about, and again, put that out there more for what we call thought leadership, to get people to think about it, and obviously the fact that we're having this forum and people are attending and people are speaking I think, if you will, the result was recognized and hopefully this dialogue will continue into the future because, again, ICANN has not yet finalized the mechanisms by which new TLDs will be added. So I think what I would like to do, as -- if you will, to continue the thought leadership is to look at two particular TLDs that might come forward, or two hypotheticals, to sort of walk through, of how free expression might be impacted. And this is one of the things that we've heard that ICANN staff -- they've actually been running through scenarios. The 26-step chart that they've discovered -- have proposed, their flowchart, they've been going through test case scenarios. How are these things going to work? So two scenarios that I would just like to briefly run through are a dot gay TLD and a dot cat TLD, although cat is already in the root, I'll sort of tweak the hypothetical in that scenario. So dot gay -- what I think is interesting in dot gay is there are certain countries in which homosexuality is punishable by death, so what would be interesting here in the proposed recommendation, in the reserved name working group, regarding controversial names, we did not require any supporting organization that would, if you will, chose to oppose or trigger that to require a consensus mandate. We actually left it up to each individual supporting or advisory organization to come up with their own mechanisms. Now, what was interesting to note in the GAC communique regarding xxx is, there never was a consensus position against the application. The communication communicated the strong opposition of many members that were strongly opposed and the other thing that was noted was that there was no GAC member in favor of the proposal. And I think that's something that is important, potentially, with a dot gay TLD because although there are -- as I said -- certain governments that do have strong -- I would -- let's say I think there would be certain governments that would have a strong opposition to a potential dot gay application in the next round, but I do believe there would be certain governments that actually would be in favor. So I think that is something that is important to recognize. My concern, however, is that we do not yet know the mechanisms by which supporting or advisory organizations may make challenges to TLDs that might raise moral concerns or the current standard. So I think that is important. You know, if I can, switching over to the other example of a dot cat, if dot cat was not proposed in the last round but was to be proposed in the future round, I do think that there is a scenario with the mechanisms that we are talking about by which the Caterpillar Corporation would have been able to file an objection to that string. That was a challenge that was not specifically provided for in the last round but is -- there is going to be a particular path by which a trademark owner might be able to challenge. I think that's -- as a trademark attorney, I find that very hard, and if you look at the precedent that has been established through the UDRP decisions over the years, I would probably say that using that precedent as a baseline, it would be likely that a trademark owner would probably prevail in that particular situation to block the dot cat TLD. I know Werner is disagreeing, but that's -- that's -- again, that's -- we don't know what the exact standard is, but I do think that there would be the potential, again, using the precedent of the UDRP -- and there is precedent where they talk about commonly used words or phrases or which people call generic terms, third-party usages. I think that there is the potential that that may have gotten blocked. But the -- it would just be the string. I think the Catalan community would have been able to come back with dot Catalan. They would not be denied access to the root. Again, these are just hypotheticals. I don't know. But when I look at free speech, these are the two areas that I sort of look at as potentially being problematic, going forward. Those that raise certain moral issues, such as a dot gay, and those such as a dot cat, a dot apple, we can come up with other -- if we use commonly word -- commonly used words and phrases that do have trademark significance for companies. So these -- these are the issues that I hope the community will be able to resolve collectively through the council, through board consultation, staff input, over the next coming months, because I do think it's important to get these processes right, if people are going to have -- if you will -- integrity belief in the process towards that predictability that we're looking for. So, again, that's my sort of initial comments, and I look forward to any of the discussions or questions that come afterwards. >>ROBIN GROSS: Thank you. Before I introduce the second speaker, I should introduce myself. I don't think I did that at the beginning. My name is Robin Gross, and I'm with the noncommercial users constituency, and run IP justice in San Francisco. So our second speaker will be a professor Wolfgang Kleinwaechter. He's a professor at the University of Aarhus, and he can give us a European perspective. >>WOLFGANG KLEINWAECHTER: Thank you, Robin. As a university professor, you know, let me allow to put this question of freedom of expression into the broader historical and theoretical context, because this issue on freedom of expression is not a new one. I want to start, but very brief, you know, with one of the first inventions in communication technology. This was when Gutenberg invented the printing press, and this was a revolution -- a technological revolution, but what are the consequences of that? First, the church was very happy about this and said, "Now we can print the Bible and distribute our message directly to the rest of the world." But at the same time, other groups came up and said, "Okay, we can produce some pamphlets now which are criticizing the church and can distribute it all over the rest of the world." And suddenly governments came and said, "Oh, we have to keep this into control," and they introduced the system of censorship, which is, okay, you cannot publish before you have license -- or, you know, confirmation by the -- by the priest or by somebody else, so -- and then came -- then came a man which was called John Milton, and he said, "Okay, look, what's going on here?" You know, if we look to! all the history on the Aeropag, this is the place where people come together, they change -- they trade goods and they change all their ideas and information, and he wrote in Areopagitica where he really developed the concept of the right to freedom of expression as an individual right, which is not given by the government to the individual, but which belongs to the individual. That means the owner of the right to freedom of expression is the individual. And governments have to make sure that this right is -- can be implemented on this individual level. And so that's why all the concept of free trade and free flow of information are very interlinked coming from, you know, this perspective. And if you go through all waves of technological development, you see the same reaction, you know. When the telegraph was invented, great new opportunities, but soon governments realized they have to keep this under control and the -- in 1865, it was Napoleon III who invited governments to Paris and they signed the International Telegraph Convention and Article 4 of the International Telegraph Convention from 1865 is more or less a censorship paragraph which allows the governments to intervene into the telegraphic traffic under -- if secrets or governmental interests or national security is concerned. So a paragraph which allows you to do everything. By the way, the ITU celebrates, you know, this date as its birthday, because this was the beginning of the ITU and sometimes I have discussions with ITU officials and say, "Be a little b! e careful if you celebrate too much the 1865 convention, then probably you know you support all the concept of censorship." It was clearly done in this 1865 convention. And only after World War II, when we, you know, had the new beginning of the United Nations, then the international community was able to create a broader legal framework which is reflected in Article 19 of the Universal Declaration of Human Rights, and here it's very clear that this was then the agreement that there should be the right to freedom of expression includes the right to receive, seek, and impart information and ideas of all kinds, regardless of franchise and via any media, and I think this is up today the international legal basis we have. However, there was also an Article 29 in the Universal Declaration which gives the nation states the sovereign right, you know, to create legal framework on the national basis. And this -- this combination of freedom on the one hand and some restriction -- justified restrictions on the other hand -- is reflected also in the convention from 1966 which has in its paragraph 19(3) f! our categories where it's justified to have restrictions. And these all are very vague -- vague reasons. This is national security, public order, this is moral and public health. So that means how you define internationally what national security is. China has probably a different approach than Denmark to national security. Or what is moral, you know. India has a different approach than Saudi Arabia or Germany or the United States. That means we have no standards, and -- which really made clear. And so far the conclusion is, you know, from the legal framework that you -- that there will never be an internationally agreed framework of clear principles, so that means you have to live with differences and what Milton said the day before yesterday in the discussion, then he said, okay, we have to look into the national legislation and one country with regulate within its country according to the -- to the Universal Declaration of Human Rights and the international covenant. But it cannot speak for the rest of the world. But it means we have to live with this -- with these differences, and if it comes to cases, then we have to look on a case-by-case basis, you know, whether values or legal goods are affected. This makes it, thirdly, much more difficult but there is no alternative. The final thing is that when the Internet emerged, the same discussion we had after the invention of the printing press and the telegraph and all this new media, now also is in the Internet world. So the same argument is, you know, coming from different circles. Governments say, okay, we have to keep this under control. Commercial groups say, you know, we have to channel the -- the money stream. And on the other hand, we have groups that say, you know, this is a great new opportunity for the distribution of our ideas. And if we go further down from the Internet to the core resources, and it was interesting what Michael just said, then you see the same arguments now around the introduction of new top-level domains, so that means you have these two kinds of censorship, some governments raise political concerns, some companies raise commercial concerns, and again, you know, there will be no consensus. To expect there is a consensus, in a general way, is unrealistic. What you have to do that's the only way, you have to build a community, you have to discuss, you have to create a consensus and -- at least a rough consensus, and to go case-by-case through this issue. With regard to the justified restrictions, let me add a final argument, and this is -- okay. So, yeah, I'm really speaking too fast, so... Normally, I need 40 minutes for this lecture, so -- [Laughter] >>WOLFGANG KLEINWAECHTER: Okay. So the final point I want to raise is the following one: Okay. Freedom is the principle and justified restrictions are justified, but if it comes to the restrictions, there are limitations for the restrictions. You know, I've grown up in a communist country where the article 19 was read from the different corner. First the restrictions, and then the freedoms. So -- but we have to talk about that justified restrictions have itself their restrictions. They cannot be used, even if, say if they are written in law to undermine the basic principle. And the basic principle of freedom is expression and this has to be clear. That means you can a little bit, you know -- to avoid the misuse of freedom, that's fine. Justification -- there are justifications for the restriction of freedom. But if you go too far with this justification of restrictions, you end up with a censorship system and this leads to a dictatorship, and I think we have to p! ut -- to ask a big question, if we go to the very detail like the introduction of new gTLDs, because this is what you can learn from history. The structure of the debate -- and by the way, also the arguments in the debate -- have never changed over the last 400 years, and I expect they will not change the next 100 years. So this will be a better field for the years ahead. Thank you. [Applause] >>ROBIN GROSS: Thank you, Wolfgang. Our next speaker this afternoon will be professor Milton Mueller, who is the chairman of the noncommercial users constituency, and founding partner of the Internet governance project. >>MILTON MUELLER: Thank you, Robin, and thank you very much for being the inspiration for this panel. As you may know, I have been goading and pushing and hoping that ICANN would be coming up with a new gTLD process since 1999. Be careful what you ask for. [Laughter] >>MILTON MUELLER: I expected the new gTLD process to be more restrictive than I would have liked it to have been. I think the idea of hundred thousand dollar entry fees, of financial and business requirements that a startup has to be operating an infrastructure like dot coms is excessive and unnecessary, but I knew that this is a much more conservative community and there are interests involved, and I expected some tradeoffs. I did not expect the new TLD process to openly and explicitly introduce a regime of global censorship. I did not expect standards of morality and public order to be inserted into the string selection process. And that, indeed, is exactly where we have gone. We are taking dramatic steps into new territory. I think we have to be very aware of this, that there is a precedent being set here. There has never been a global regime for the regulation of media content. The 1860 ITU treaty that Wolfgang mentioned did contain some restrictions but they -- that was a European game. It had absolutely no application to the United States or Asia at the time, and it wasn't that hard for the rather Mercantilistic and authoritarian states of mid-19th century Europe to reach agreement on the nature of what information should be regulated. What we're doing now is very different, very different. We are, in effect, saying that we are going to create a global institutional mechanism for deciding what kinds of words cannot be published. Purely on the basis of the strings, mind you. We're not talking about the content under the strings. We're not talking about the way the strings are used. We are saying that this community and the governments that participate in it will be able to censor particular words from every appearing in the domain name space at the top level. Now, if you're going to do this, at the very least be careful. Okay? And don't be one-sided. So for example, the original phrasing says something like -- is it recommendation 6? No, it's no. 3 says, "The process for introducing new gTLDs must make proper allowance for prior third-party rights and particular trademark rights." Now, why wouldn't you want to say also that it must make a man's -- or take into account prior freedom of expression rights? Simply as a matter of balance. Both of these are written into various kinds of national and international law. Why wouldn't you simply be balanced and say, as the NCUC amendment drafted by Robin proposed, "the process for introducing new gTLDs must make proper allowance for third-party rights, including trademark and freedom of expression rights"? What is so difficult or unreasonable about that? It's a simple message here. Freedom of expression rights must be recognized and must be taken account of in this debate, wherever you fall down on it. Those rights exist and are enshrined in law both at the national and the international level. How can ICANN not recognize them? Now, to move beyond that, I disagree and I think it's an important disagreement but certainly an intelligible one with Wolfgang when he says that the solution to these problems is to build a community around ICANN and reach consensus within this community as to what should be censored and what should not. I propose what I call the reformation principle, which is, you know, instead of building a religious community in Europe and deciding whether you're going to be Catholic or Protestant or dead that both sides stood aside, stood apart and said, "We separate state from church. We do not try to dictate who is the right religion and what are the right words and what is the proper public morality and order at the global level. When you're dealing with TLD strings, the principle is one of tolerance and minimal -- absolutely minimal -- forms of regulation. That is, if a Catholic kills a Protestant or vice versa, you know, they're guilty of murder, it's not about religion. So that might include some kinds of speech -- for example, conflicts with legitimate prior trademark rights, which are simply inimicable to public public interests to have going on, but it does not mean that you are simply trying to create a global consensus around certain kinds of norms or standards. We don't think that there will ever be a global consensus on norms and standards regarding proper forms of expression. Now, that's a hard problem. That's a very hard problem, because I think the only way out of it is for ICANN to simply be a highly neutral technical coordination body that basically looks, first and foremost, as to whether there is a string conflict or not, and permits people to register things primarily on that basis, and then allows national laws and applications at the edges, which people can choose to use or not, to determine what kinds of strings get through to people and which kinds do not. Clearly that requires more work. But it is the right thing to do. It's something we have to do to protect the freedom of the Internet. We cannot allow the core resources of the Internet to be used as leverage to impose a global regime of control and censorship. So there's two very clear approaches to this problem here. One says ICANN is the nexus of a global community that's going to impose its own norms and standards on the world, and the other one says ICANN is a neutral coordinator of lots of different people's views and should stay out of those kinds of disputes and let them be handled elsewhere. Now, there's something very dangerous here about the relationship between governments and ICANN. And let me publicly accuse governments, who -- most of whom are probably not here -- of laziness. [Laughter] >>MILTON MUELLER: Hi there. Yes. Bertrand is here. We know that Bertrand -- we know that Bertrand is listening, and is a good listener, and will have things to say later. We know this. But I accuse governments of laziness. You see here's the way it should happen. If governments want globally applicable norms regarding what strings or what words can be uttered publicly on the Internet, they have to get together in a room and negotiate a treaty, they have to negotiate specific words that are consistent with their own laws and they have to get those treaties ratified by their own democratically elected legislatures. They probably know as well as I do that may never happen; and if it does happen, it would happen 15 years from now. But it probably would never happen. Look at who you are talking about, the United States, China, Saudi Arabia, Iran, Denmark, the Netherlands. Can these people all agree on what are the proper norms of freedom of expression? So what's dangerous about this situation is that they can use ICANN as a way to avoid making real public policy, making real treaties and laws and they can simply express displeasure at something and then we all get intimidated and we back down and that regime of censorship is much more effective, the Chinese government knows this, that kind of encouragement of self-censorship is more effective and repressive than explicit legal censorship. But it's all, I think, about governments. How much time do I have? Couple minutes. Basically, governments, some of them are afraid that words that they don't like will be entered into the new TLD space. I have heard a particular government that you can probably guess worry, what if someone wants to register dot Jihad. I have heard we saw that some governments didn't like dot XXX. We have government that is are concerned about religious and cultural terms. And so, yes, these governments have legitimate concerns about terms that might be offensive to them, that might be sensitive in their culture, that might be even illegal in their culture, but the question is, how are they going to exercise this power globally? Or should it be global? I'm suggesting that they are going to exercise that power legitimately, that it should be exercised first at the national level. And if it is going to be global, it has to be done through a formal negotiated treaty in which they write down what they're doing. They go through democratic and accountable process and get it ratified by democratic national legislatures. And then and only then can they tell us what words can and cannot be in the top-level domain space. Personally, I think people put way too much emphasis on the existence of a string in the top level. There is a dot Jihad in the second level and third level. The word "Jihad" can appear on Web sites. It can be uttered in people's speech all over the world. It can be on bumper stickers. It can be on posters, television. The idea that the world will surrender to terrorism because there is a top-level domain is obviously a bit crazy. But it's this collective process of recognition that worries everybody. People think, well, if we recognize dot XXX, if we all go through this process of ratifying it, then suddenly we are sort of legitimizing what's there. That's not true. If you are a neutral technical coordinator of the Internet, your simply accepting applications for strings, making them meet certain criteria, and if they meet those criteria, they go in. And the regulators and governments worry about the other issues as we go forward. So that is my perspective. I appreciated Wolfgang's historical background. I think it was a very good contribution, and in this debate, I'm very happy to have the name sake of the British philosopher and poet who took one of the more extreme but, I think, one of the positions that really helped to move this world society forward in greater freedom and progress. Thank you. [ applause ] >>ROBIN GROSS: Thank you, Milton. Our next speaker this afternoon will be professor Christine Haight Farley. She is a law professor at American University in Washington, D.C. Christine? >>CHRISTINE HAIGHT FARLEY: Good afternoon. Let me begin by saying that I'm not a first amendment scholar and I'm not a regular participant in ICANN policy. I'm a humble law professor, and I've come all the way here just for this session and just with one simple message, and that is that trademarks and domain names are distinctly different animals, so policy that works for trademarks doesn't necessarily work for domain names. This is my chief concern with the draft report. It may be -- this caused me to think that it may be that trademark law is a little bit different from other areas of intellectual property law in that it's not so easy to tweak trademark law into the digital age the way we've seen this being done with copyright law and patent law. I think trademark law looks very flexible and accommodating but that is quite deceptive. Trademark law probably chief amongst all the areas of intellectual property law is very finally balanced and it can be set off kilter quite easily. And I think this policy would really do that. So what I have to do is in the next couple of minutes give you trademark law 101, but I promise you it is not boring. It is really rather interesting, so... I can identify three large distinctions that need to be made between trademarks and domain names. Trademark law is really focused on territories. It seems strange in this age to be focused on territories, but it is. And even though we have now quite developed international trademark law, that law still respects that fundamental principle that trademark law is territorially based. Even in international trademark law, these territorial rights and concepts are respected. We regulate trademarks because they have a significance within a region, that region may be the state. That may be a smaller region within the state. That may be in a particular niche market within a state, but that's the extent to which the rights extend. The rights do not extend beyond that. In order to have trademark rights from one state to the next, you need to do something to acquire those rights. The virtue of having a trademark does not grant rights internationally and that, I think, is something to remember here. When we start to elevate these trademark concepts and policies to domain name policy, we cross a very big threshold because then we grant these trademark rights internationally, something that trademark law -- international trademark law has never done. Trademark law is always focused on the consumer, right? We justify trademark protections because we're seeking to protect consumers from commercial deception. But all of the rights in trademark law, all of the doctrines, all of the tests in the law are focused on the consumer. The law always asks who is the relevant consumer for this mark, what do they know? What do they think? What would they see when they see this second use of the mark? Would they be confused? So it always considers the relevant consumer. Again, that's geographically based. It may be market based. It would be impossible to consider an international generic universal consumer. It would be impossible to ascribe knowledge, sophistication, attributes to a universal consumer. What would they take? Because as you know, when you travel, you take different information with you. You perceive things differently from place to place. Finally, all trademark rights are considered in context. In fact, trademark owners have fewer rights than you'd imagine. They don't own the trademarks in any reason sense. They have limited exclusive rights against others, only in situations in which another would cause confusion of the relevant consumer. That's the only case in which a trademark owner can assert their rights against someone else. So we have to look at the particular context of use by third parties to see whether the relevant consumer would be confused. We don't just look at the words. We look at the market situation. How are they displayed? What other information is on the label? What other information is in the buying experience? What kind of a product is it being used on? We looked at all of these things to determine whether or not a consumer would be confused? It is highly contextual. It is not in abstract. It is not in theory. These are a very limited set of rights. So to take these ideas and apply them to some of the recommendations, I want to begin by looking at Recommendation 2 which says strings must not be confusingly similar to an existing top-level domain. Now, that "confusingly similar" is a trademark concept that's being used. But it's used here with an understanding -- with an incorrect understanding that that is the trademark test of infringement, confusing similarity. It is not. It's the first step toward an analysis of infringement. It is the first step that we take when we consider whether the trademark owner has a prior right. We look at that as we look at the two words -- of course, trademarks are not only words, they're all kinds of things now. We look at the two words and see whether the identity is so close that we would say they were confusingly similar. But as I've just said, the analysis goes much beyond that. We look at the kind of consumer that would buy this kind of a good, how sophisticated are they? How much care do they take in this buying decision? How is this being marketed? Let's look at the media uses? Look on what's on the shelf next to it. We look at all sorts of context actual things so words we may initially determine to be confusingly similar may not be an infringement. That is, they're not likely to cause confusion. And that's why we allow many parties to use the same words as trademarks, as legally protected trademarks, right? I give the example of delta. There is Delta Airlines. There is Delta Faucets. There is Delta Dental and probably numerous others. If you do any search of a trademark in any database of any country, you will find numerous parties registering the same terms because they are used in different context. They may be confusingly similar but they're allowed to be used. I guess I just want to add one more point here. The idea of making an international right in a trademark has never come to fruition in the history of trademark law and the history of international trademark law. The closest we've come to attempt that kind of thing has been in the protection of what's called geographic indications, and this is a source of international debate presently. That is should the word "cognac" be reserved against all other uses internationally. You can imagine some countries would favor that policy. But international law does not protect that idea because it recognizes that cognac may mean something in France, but it may mean something entirely different to consumers in the United States who may regard that term as indicating a fine brandy but not necessarily a product of a particular region. And so we allow those kinds of terms to be used and we allow the governments of different countries to regulate their use. There was an attempt to have a kind of universal protection for these terms, but that is not the case. Okay. Recommendation 3 says strings must not infringe the existing legal rights of others that are recognized or enforceable under generally accepted and internationally recognized principles of law. Well, let me tell you that everything that you can think of is protected as a trademark. I would challenge you to come up to me afterward and give me a name, a word that is not protected as a trademark. Even some of the words that have been discussed as possible good candidates for new top-level domains are protected as trademarks. We are sitting in the Hilton Hotel. I guess, earlier there was a conversation about city names like Paris being a good candidate for a new top-level domain. I can tell you that there is somebody who has registered the word "Paris" connected with the Hilton Hotel chain and family. So really -- and I will also say that once a policy like this is really not draft but final, there will be a situation of trademark squatting. So if by chance there is not the word "books" registered as a trademark, there will be because in most countries of the world it is very easy to get a trademark registration. You don't have to necessarily be using the trademark in commerce. Very few countries, like the United States, have any kind of restrictions like this. It is very easy, and in some cases cheap and quick, to get a trademark registration. My guess is there are numerous trademarks for the word "book." If there aren't, there will be. This is really an unworkable standard. When you think about it, the fact that someone has registered the word "Cherokee" already and, in fact, numerous commercial entities have registered the word "Cherokee," why is it that we would consider these parties to have important prior rights that we would want to respect? If, for instance, the Cherokee nation thought it would be a good idea to have "Cherokee" as a top-level domain. All right. Now to the thing that -- oh, and one other point. Since I'm so focused on the idea that trademark law really doesn't track good domain name policy, I haven't been focused on the fact that there are all kinds of defenses and limitations in trademark law that would need to be taken into account. One such limitation is the Fair Use Doctrine which says if a trademark owner has captured a descriptive word as a trademark, they don't have the right to exclude other competitors who would want to use that word in commerce with a use that is not a trademark use. So that doctrine alone would allow top-level domains, which are inherently not trademark uses, to exist even where there would be confusingly similar to words people own as trademarks. All right. Recommendation 6, about the public morality and public order, I can understand why the committee would have been thrilled to see that international trademark law does address this idea that there could be some regulation of immoral words. But I think there's a misunderstanding of what international trademark law does here. We have a very recent and huge international trademark treaty, and that is the WTO's TRIPS agreement. This is the newest multilateral agreement on trademark law. It doesn't mention this. It certainly was a possibility to include but the TRIPS agreement does not include this provision. The Paris Convention does include a provision relating to this kind of restriction, but it is important to note that this is not a minimum standard. It is not a minimum standard that countries must regulate immoral words as trademarks. The Paris Convention merely permits countries, if they so wish, to make a kind of regulation restricting words as trademark registrations. The next point -- and United States is one of those countries that has some kind of legislation restricting immoral words as trademarks. But it is very important to note that this is not a restriction on use of a trademark, and this is not even a restriction on legal protection of a trademark which could be deemed to be immoral or scandalous as the trademark act mentions. This is only about federal registration as a trademark. So if you wish to adopt an immoral word or a scandalous word in the United States as a trademark, you may be prohibited from federal registration. You may not be prohibited from using that word or from deriving legal protection through federal common law, which is built into the trademark act over the word. It is not as grand a restriction as you might think from the committee report, and it is really little used. Again, this restriction is used in context. So words that would appear to be scandalous or immoral, in many cases in decisions, the court has said, Well, when we look at the context of the use of this word as a trademark, when we look at the marketing materials, when we look at the label of the goods, we see it is a double entendre and it is really not scandalous. We don't take the word alone even under the doctrine. It is in context. And it is important to note that even though this provision exists in U.S. law, by and large, it is only utilized by third parties who are raising objections. It is not something in which the trademark office is asserting the standard against would-be trademark registrants. That's all I have to say. Thank you. [ applause ] >>ROBIN GROSS: Thank you, Professor Farley. That was wonderful. Our next speaker will be professor Wendy Seltzer from Northeastern University Law School, and she's also a fellow at Harvard Law's Berkman Center for Internet and Society. Wendy? >>WENDY SELTZER: Thanks very much and thanks, everyone, for coming out to engage with us on the subject of free expression. I will try to keep my remarks short so we can also engage in some discussion with the audience here. When I went to look at the proposed new gTLD process, I was sent to a page with a long PDF, with a 26-step chart through which applicants and applications would go. I tried to print this out but, unfortunately, I ended up with one page with a black smudge down the center as my computer tried to shrink this to fit one page. Printed on multiple pages so as to be readable without a microscope I think I would have had to bring a stack of paper an might have exceeded the weight limit on my flight over here. The problem is that this is a thick process that's being developed where applicants for a new gTLD string have to go through multiple steps of vetting and opposition, including the possible opposition in this policy Recommendation 6. Strings must not be contrary to generally accepted legal norms relating to morality and public order, enforceable under generally accepted and internationally recognized principles of law. The problems are, A, that there is no such internationally recognized principle uniformly to be found and, B, that even were there such a principle, the costs of finding it out and of determining whether your proposed string met or did not meet those criteria would be prohibitive. So what we have is a thick process, and not just thick in number of pages it would take to print it out, but thick in the costs of engaging with it. It is a process that's open to the heckler's veto, a competitor who sees competitive advantage to keeping your string out, will always be able to dream up some objection under claims of infringement or claims of morality. And so that thick process creates disincentives to innovation and creative uses of domain name space. So instead of this thick process, I would much prefer to see a thin process or borrow in terms that David Isenberg has used to talk about the network itself, a stupid process, a thin process, a very light-weight process, a process that could be as stupid as compare string to string already in the database. If string is not found, add new string to the database. Lightweight, easy to implement, inexpensive, doesn't raise these barriers to innovative use and this has worked very well for the network itself. After all, the Internet -- the reason we're all here -- is just such a stupid network as David Isenberg and others have described, the network doesn't care what kind of content you're using there and it doesn't care what kind -- whether the network itself doesn't care whether your content infringes, whether your content is immoral by somebody's definition, whether your content is valuable or valueless, whether your content or your application is useful or useless, whether it is good or bad or cheap or expensive. Instead, it let's the end users, the people at the edges of the network decide those things for themselves. And it let's governments at the edges of the network regulate what will be permitted on their particular piece of network. So governments engage in filtering regimes and some of us may like that and some of us may dislike that, but if a government has decided that certain content is permissible on their restriction on their territory, they can attempt to block that locally. And if another government has a different, more expansive view of what's permissible, they can permit that. The stupidity of the network has permitted flourishing of independent creativity of unanticipated content and applications, the designers the network didn't know that, people would start using it for voice over Internet telephony or for video or podcasts or audio or Web logs or RSS or any of the new kinds of content and services and applications available. Keeping the network thin, allows people to add any of those services on top of it. Moving back to gTLDs, some people think that new TLDs will help with the development of new kinds of content or new applications or new kinds of services. Some people have ideas for businesses that they could business around new kinds of content or applications or services. We've heard from some of them using domain name service for things other than just resolving to Internet locations, the dot tel proposal to use DNS as a storage location for contact information or the dot post or dot mobi domains' proposals. We hear new TLD proposals for a particular community, a particular language, a particular application content and some of those may be great ideas and some of them may be terrible ideas, but there's no reason we should be determining in advance which of those are going to be good or bad, successful or unsuccessful, useful or useless. If we have a thin process in place, we can allow people to try them, see what's valuable and add that value to the network for those who find it valuable. The consequences of doing otherwise, of putting these restrictions into the new TLD process are not only a slow, expensive, easily vetoed process, but also a race to the bottom in terms of what expression is permissible. Every time we try to harmonize the permitted strings by various local moralities and various local trademark juris rules, we end up with less and less that's permitted and just as if we allow every individual who has a complaint to veto the process, if we allow every government that has a complaint about a particular string, to veto them, we end up with a very small pool of interesting ideas from which to draw. I see some of this -- the variety of international law on expression through the chilling effects project, a project that I run at chillingeffect.org where we collect complaints about online content and requests that have been made to various providers of content for the removal of that material. One of the sources of complaints that we receive in our database is Google, the internationally available search engine that as I'm sure I don't need to explain to anyone in this room will provide you with responses to any search query you type into it in nearly any jurisdiction in the world. Now, Google gets complaints from people about material that's made available through the search engine. And depending on the jurisdiction of the complaint and the jurisdiction of the complainer and the jurisdiction of the Google servers responding to those complaints, Google may or may not remove the material that's being complained about. And so if you're in the United States and you make a request to Google to remove material that you claim infringes copyright, Google will remove that link from its search index. But if you make a complaint to Google that a link returned in a search query defames you and you're in the United States, Google will not remove that link, because United States law says that an intermediary won't be liable for the defamation that some third-party user has made accessible through it. And so in the U.S. you have one view of the Internet that gives you access to quite a lot of material. If you're defamed, your response has to be either to go after the person who has defamed you or to post your response to be found equally by the search engine. By contrast, if you're in France and you make a complaint that you've been defamed there, Google will respond to that by removing the complained-of page. And so we see complaints enforceable in various jurisdictions, not just of defamation or copyright but of publicity rights, of hate speech, complaints of sexual content, of obscene content, where the definition of "obscenity" varies widely from country to country. Some countries have laws against incitement of varying kinds. Some countries ban particular political parties or particular opinions about political parties or about historical events. Some countries ban holocaust denial. Some countries ban religious content. Some countries ban various kinds of propaganda. Some countries ban material based on competition policy or trademark policy. And so each country, with this stupid network layer in the middle, gets to set its own policy about what will be permissible there. And if you do a Google search from France, you will see different results returned from what you see doing a Google search from the United States. And the Internet hasn't broken, as a result of that. Each country gets to -- to set its own laws and policies, and eventually perhaps they may harmonize, but the disharmony hasn't caused terrible problems. By contrast, if we try to respond to all of those concerns at the level of pre-approval of new domain strings, we would have to engage with every one of those different complaints. Even the complaints of countries that don't permit human rights advocacy on line. So the stupid network runs on what have -- at the network level, runs on abundant infrastructure, an underspecification of what might go on over the network, and interoperability, Eisenberg tells us. And that has sort of allowed us here and elsewhere to figure out what sorts of things we want to plug into it. I think a stupid policy for addition of new strings to the root, making it cheap, telling in advance very little about what's required and making it easy to inter-operate and to build upon that layer will similarly allow us to experiment, find new and valuable uses for the domain name system, and new and valuable uses for the Internet in ways that sort of enhance all of our freedom of expression. [Applause] >>ROBIN GROSS: Thank you. Thank you, professor Seltzer. We may have a couple other speakers on our list who may have come into the room since we first got started, on let me just check to see if any of our other speakers are here. Joi or Ram or tin? Any chance you've come in the room? No? Okay. We also wanted to bring Dan Krimm up here to announce the "Keep Core Neutral" campaign. Dan is a member of the noncommercial users constituency and he's going to explain this initiative. Thank you. >>DAN KRIMM: Hi. I just wanted to let everyone know that we are building a formal campaign around the "Keep the Core Neutral" concept. We have a coalition in force. We are launching that campaign today with this workshop. We'll be presenting a press release later this afternoon. We have a petition which we encourage everyone who supports the idea of core neutrality to sign. We have 58 signatories already at this point. A lot of major organizations and prominent individuals. You can find the Web site at www.keep-the-core-neutral.org, and I would welcome everyone to visit the Web site. We have a lot of background information and frequently asked questions and that sort of thing, and we would welcome everyone's support. Thanks. [Applause] >>ROBIN GROSS: Okay. We do have a few other speakers here. Let me turn the mic over to Ram Mohan, who is the vice president of business operations and CTO at Afilias, limited. Go ahead, Ram. >>RAM MOHAN: Thank you, Robin. I probably don't have as many things to say as the others before me. Part of it is because some of what I wanted to say has already been covered, and part of it is also because it's getting to be late in the day. So I have just a few observations to make, and I'm going to make most of my observations today with a few specific points of view. One is a -- since this is something that we're talking about, gTLDs, I'm going to focus on gTLDs. And since there has been some discussion about the internationalizing, the -- you know, creating multilingual domain space, I'll have a little bit to say about that. And probably more of my comments are going to be in the technical area than in an only-policy area. So I'd like to say this to Wendy. You know, I heard Wendy talk about thin networks and I think that's a very good idea, so long as a thin network and a process that continues such a thin network is congruent with a stable core. And when I hear suggestions and when I hear comments that say, "Let us underspecify how new TLDs should go in because we can experiment, we can put them in and then we can figure out what goes wrong and then take them down," the caution flag that goes up for me is: You do real harm when you put a TLD in and then you decide somewhere down the road, "Let's take it down." You know, there are real users who get affected, people who truly have content and expressions that get affected. So what I would like to suggest is that when we talk about a thin process, a thin network, that we also keep in mind that it has to be done in a way that is congruent with a stable core. Now, the core ought to be pretty thin, and some of the earlier principles that have been espoused talk about, you know, trying to keep as much of the innovation on the edges, and keeping the core pretty stable and, frankly, pretty boring. And that's not such a bad goal to continue to keep adhering to. Having said that, I think IDNs have assisted in removing some of the boredom from the core. [Laughter] >>RAM MOHAN: We've -- you know, I've heard various proposals to continue the old legacy of, you know, ccTLDs are always two characters in length, and therefore, somehow IDN ccTLDs, such as it is, should also be two characters in length, even though there are probably cases where you can represent an entire territory in one character. You don't know what to do with the other character that's now free for you. Or in other cases, you have territories where, you know, the artificial truncation to two characters causes some amusing results. I'll use my own country of origin, for example, Bharat, is the Hindi representation of India. If you take Bhar, that has meaning. That means burden. If you take Bhat, that, too, has meaning. That's rice. I don't think India would really like to be abbreviated to either a burden or rice when it comes to, you know, an IDN TLD representation. And there are those who say it's merely semantics, it's just a mapping. Just take whatever is ! represented. You know, nobody ever said initially dot IN was supposed to be India. People will figure out that "burden" or dot Bhar is actually India and people will figure out that India is burden. But, you know, I don't think that's necessarily an appropriate place to go, because the legacy of a two-character restriction doesn't actually translate well into a multilingual world, and the technical reasons for a two-character limitation, I fail to find many technical reasons. I find many convenience reasons, reasons of legacy and history, but this is something -- this is, for example -- for example, an area where we ought to actually build a new mind-set about what is an IDN ccTLD. I'll state a concern that many gTLD -- existing gTLD registries have. It's a concern that has been stated at least two different ways. I will state both the ways, for completeness, and then I will state the way that I actually think is the right way, or a good way forward. And this is not just a gTLD issue. This is something that works for all TLDs that are currently represented in the ASCII repertoire. There is a -- there is a gTLD registry that is on record that has said that their representation of a name in -- that is currently in ASCII, dot travel, they have stated that they believe that as a result of having been -- of being the registry operator for dot travel, they ought to be the rightful home for dot voyage, which, you know, you could also pronounce it "voyage." Now, there is not -- so that is a semantic representation or a semantic equivalency argument. There is another point of view -- point of view that says we have a TLD. Let's say it is dot info. Let's say it's dot DE or dot SA. And the concern is that a representation of -- in some other script that looks similar to INFO or DE or SA, not sounds similar but looks similar, can cause real harm. You can imagine any number of phishing and other spurious Web sites that can come up that to the unaided eye look exactly the same as a TLD that is popularly accepted to be a trustworthy TLD. And this is not a gTLD issue. This is just a pure ASCII TLD issue, and it cuts equally across gTLDs and ccTLDs. And I think caution has to be exercised here. It is -- I very strongly endorse freedom of expression. That's a very good thing. But I think we have to balance some of those with ensuring that there is not intentional confusion that is introduced into the domain name system by the introduction of domain -- of top-level labels that look virtually identical, look similar, to an existing TLD, be it a gTLD or a ccTLD. Real harm can be done and will be done. I think if you take some of these core principles which I -- in my opinion, do not make for a fat network, continue to enhance stability at the core, if you take some of these principles -- and I certainly don't claim that these are the only principles that should be there. I think there is a large community of active minds that can contribute to more strong principles that have real harm, real technical and resolution and universal resolution issues that come through, you can actually define them. You can show what the harm is. And then you -- I think once we define those areas, those ought to be marked "Do Not Go There." And if you -- once you leave those out, then you have, I think, a wide area that you can -- that as a community we can feel free to experiment, innovate, and introduce TLDs in, without worrying about the ill effects of a poorly considered and hastily introduced mechanism for new TLDs. Thank you. [Applause] >>ROBIN GROSS: Thank you. And our final speaker this afternoon will be -- I'm sorry. Our final speaker this afternoon will be Tan Tin Wee, the professor at national University of Singapore. >>TAN TIN WEE: Thank you very much, Madam Chairman. Apologize for being late, and I was just outside with -- talking to quite a number of people about something along the lines of thin networks and thick networks and so on, among others, John Klensin pointed a finger at me and said -- is John here, by the way? Well, he pointed a finger at me and said he invented the IDNs and we've -- and, look, I have got to clean up the mess he made, right? [Laughter] >>TAN TIN WEE: So anyway, that was as long as 1998, and I'm so sorry to raise this whole issue up again. It's because after all these years, nearly 10 years of struggling through -- push IDN through to its rightful fruition, it's been -- it has taken far longer than what I had expected back in March of 1998 when I put together an implementation of IDNs and demonstrated it subsequently to the IFWP, later in that year, the precursor of ICANN, and I guess nobody took any notice about that. And I'm really glad that ICANN today is taking serious attention about this whole thing about IDNs. And I just want to remind everybody of why is it that I'm so passionately involved in IDNs. Because let me illustrate by an example. I am a professor in a university. I look after student hall of residents in the evenings, where I have 500 students, and one of the projects they do as part of the student projects is an expedition to Cambodian. These are Singaporean students from National University that go out every year to Cambodia to teach students in a little village in Prey Veng outside of Phnom Penh. These kids don't learn English when they start out. They speak in Khmer. My good friend, Norbert Klein, just tells me that finally the government has adopted a standardized keyboard system, computerized mechanisms for people in Cambodia to type in Khmer. And so my students are going to these schools and trying to help them bridge that digital divide. Till today, they cannot key in Khmer -- they cannot read Khmer language Web sites unless they actually master the Latin character set or the ASCII set or the LDH that is so familiar with us. It's totally transparent to us but to these people it is a ! real thing, it is a real impediment that causes and perpetuates this digital divide. So the reason why I'm -- I like to bring to this forum is that if this issue of keeping the core neutral can help accelerate the process of introducing new IDNs, then I'm all for it. Will thin networks work better or thick networks? I don't know, really. And probably all of us wouldn't know unless you have some special foresight into the future. But I'm willing to try, if it will work out, because I've tried almost everything else in the last 10 years to get IDNs through. Ram mentioned real harm. Real harm that is done to the system if we take necessarily haste steps going forward. I would posit that that statement should be rephrased as: Real harm done to the have's. Never mind the have not's who are struggling out there, the millions of the have not's out there that are being left out of this digital revolution that's transforming the world in this 21st century. And here we have a bunch of students of mine going out to Cambodia to help these young kids out there bridge that digital divide because these are the citizens of the future. They need to have the tools. They need to reduce the number -- the barriers -- the access barriers to help them become citizens of the digital era. The longer we keep them away, it's going to bite back at us. The world has become so small and so flat, as Thomas Friedman has mentioned, that we do not -- we can pretend that we are discussing real harm in isolation of the have's, and totally ignore the have not's, but you will come back and bite back at us, and, therefore, it is totally in our interests that we seriously address this issue of the have not's out there, where serious harm and real harm is constantly being inflicted because the longer they stay out of this -- stay on the other side of the digital have not's, that harm will eventually get to us. So if keeping the core neutral through this process will help us achieve this goal of bridging that digital divide, then I'm certainly all for it. I'd now like to come to the next point that I'd like to make, and the points that were stated about how to keep the system stable because the Internet infrastructure has become such a public good that we have a serious responsibility to keep it stable and secure going forward. And I fully subscribe to all the points that Ram mentioned earlier, that we have to take the responsible way forward and not to hastily introduce anything that does real harm. While we all know that the more complex -- we know intuitively that the more complex we make a situation, the more parameters that we have to control for, the more complex the situation becomes in terms of controlling the thing, and we can try to pretend that we can be God and cover all corners -- cover all angles. So it comes to the old -- age-old formula KISS, keep it simple silly. Can we keep the system secure and stable and yet at the same time keep it as simple as possible without having to magnify the complexity of the whole process. So far as you can see, the ICANN process to any newcomer is getting pretty complicated, right? We are tying ourselves into a lot of knots because we want to play God and try to make sure that everything is safe, secure, and stable going forward, and that it does no real harm. Well, not exactly. No real harm to the have's. Never mind the have not's. Maybe keeping it thinner, keeping it simpler, may well help us achieve our goals of making the Internet the Internet for everyone everywhere. This is the investment we are making for our children and our children's children, and it's worth this effort and I thank all of here in the audience that are spending their lunch hour to actually come in here and seriously talk about how we can keep the core neutral, if that could be a way -- whether it be for free speech or for helping create a more democratic process whereby the have not's have a real democratic chance of surviving in the future of the digital revolution. Thank you. [Applause] >>ROBIN GROSS: Thank you. And keeping on the line of recognition that there is a diversity of viewpoints in the world and certainly a robust diversity of viewpoints within this room, we'd like to invite people up to hear what their comments are and hear what they have to think. So if you want to form a queue. >>WENDY SELTZER: I just wanted to start by conveying the regrets from Joichi Ito, who apologizes that a miscommunication prevented him from attending the panel as he had hoped to. >>BRUCE TONKIN: Ready to start or -- okay. My name's Bruce Tonkin, and I think like most in the community, I applaud the freedom of speech and I notice that everyone up there is good at making speeches, and that you're all free, so that's a good start. [Laughter] >>BRUCE TONKIN: The -- one of the things that concerned me in the early parts of the discussion about your solution to I guess disputes on the Internet sounded to me like it could do great damage to freedom of speech and I really wanted to caution you in actually using that solution. And that's the idea of doing some kind of countrywide blocking of TLD strings. That isn't something that you can do precisely, and you're actually more likely to block a lot more than just the string, if you're not careful. Now, I know in Australia, we've had this debate around some content regulation, so for example it's illegal to host child pornography on a computer in Australia and that sounds fairly sensible. But there are computers that are outside of Australia that do host child pornography content. Now, one of the temptations that the government has had is to say to the industry, well, you know, you need to now block that. Now, the industry has pushed back because it's actually not as simple to do as some people think. And the industry has pushed back on that very strongly, so there is no laws in Australia requiring ISPs to block anything. There is software provided and the government has encouraged that by a number of software companies that will provide software that the individual can use, and that individual can use software that blocks various types of content on the Internet, but to actually suggest that you encourage a country to start blocking, which is generally a wide-scale network issue, one is it's quite expensive for network operators to do that's correct and two, the implications could be that you start creating an environment where the Internet's not connected to that country at all. And we have seen instances of that, so I really strongly encourage that that not be proposed as a solution, because that would cause a lot more damage than I think you're trying to solve. Secondly, the examples that were given, I think there's a misconception -- I'll now just talk, I guess, as chair of the committee that's been working on this, and let me say that the committee -- at the beginning of the process, probably half the committee did not want new gTLDs at all, and half the committee did, and we've moved a great distance there to getting an environment where I think all the committee now supports adding new gTLDs, but with some checks and balances. So let's be careful that we don't derail the process or slow the process down, so we don't get any at all. Rather, we should be looking forward and saying, "How do we get new gTLDs and what sort of protections do there need to be in place to ensure we get the best solution?" The examples that were given, dot gay, dot cat, I see neither of those with the current rules would be prevented. I can't see how Caterpillar, the corporation, could block the Catalan community for having dot cat. They'd have no trademarks that have anything to do with the Catalan community. I can't see anyone blocking dot gay. To give some examples that -- names that may be blocked would be dot kill all gays. So there are a number of laws that are now in Australia -- there's racial vilification laws, there's laws about sexual discrimination. I don't think they would support dot kill all gays, but dot gay is just a single word. It can be interpreted in a range of context and probably matters more what's in front of it. If it's kill dot gay, and the Web site is www.kill.gay, then you might have a complaint about that second-level registration called "kill." But I don't see how you could block dot gay under the current set of rules so that just sort of gets some context there. I greatly appreciated the talk from Professor Farley, and I think that all the things she talked about was actually exactly what recommendation 2 was intended to do. Bear in mind the recommendation is just shorthand and you need to understand the context within which especially I noticed because you were saying understand context, you've got to understand the context within which that recommendation is being written. You're trying to sort of put in a single dot point a concept that there is a need to protect the rights of people, but it's not a case of me, to use the dot cat example, on the Caterpillar corporation, I put in the complaint so it stops the TLD. There's certainly no intention of that. In fact, that would go through exactly the process that you talked about which is used in trademark law, which is understanding the relevant community, making sure the rights are considered in context, looking at it from the point of view of protecting the user from confusion,! all of those things. So the actual dispute process would consider all the context that you had talked about, and I think probably what would help because what we've asked the staff to do is actually to flesh out the procedures and the rules that would apply in resolving that dispute and I think a lot of the material you've provided in your speech today will be very helpful for the staff in actually documenting that, because the way you describe the methods of protection is exactly what the committee wants to see happen as well. So I think it is entirely consistent. There seems to be a them and us approach here rather than you're part of the team. You have members standing up to say -- or sitting up on the stage today that are part of that committee and have absolutely the ability to provide that committee with assistance to achieve the results that you're looking for. So I will leave it at that, but thank you very much for your speeches. [ applause ] >>MICHAEL PALAGE: Bruce, if I can respond to a couple of the comments. One, in the paper that Avri and I wrote, we do not encourage governments to block strings. However, we recognize they do have the right to do under their national law. I am not encouraging governments to do that, but I think if we tell governments they have -- ICANN is not in a position right now to tell governments that they need to make every Web page on the Internet accessible to their citizens. It's knowledgeable right now that there is certain government that is do block certain access so that was kind of the things I fully agree with you. I speak for myself, I don't know if I speak for Avri. We are not encouraging governments to block TLD strings. We just recognize that they have that right under national law to control what their citizens access. That's point number one. Point number two, with regard to dot gay, I fully support that string should not be blocked. The reason I use that as an example is I wanted to discuss some potential concerns because what we have discussed in the reserved name working group with regard to controversial strings is that each Supporting Organization or advisory group will have the ability to propose their own mechanisms by which they may submit a, if you will, concern regarding a string if it falls within that controversial definition. And what concerns me is that the original draft of the proposal required consensus that a Supporting Organization or an Advisory Committee had to have consensus before they would propose that. That language was taken out and it no longer appears. We now leave it up to the individual supporting or advisory group to do that. So the reason that I'm concerned is if someone would want to block a dot gay for whatever reason that may be, I think it's -- it would just be interesting to note from an open and transparency standpoint what members of that organization or what level was blocking it. So, again, I'm in agreement. I don't think the current rules should block. I think if an application for dot gay came forward and was represented by the appropriate community, that should go forward. With regard to dot cat, I guess, I have a potential disagreement on where the rules are going right now as to the ability of a trademark owner to potentially block that string. But we could have that discussion over drinks possibly tonight. >>BRUCE TONKIN: The point is to make sure the rules don't do that, Mike. What we are trying to do is exactly what you're trying to do. Why don't you help us do that rather than assuming that's going to go the wrong way. >>MICHAEL PALAGE: Well, we can actually read Bruce's comments there. The scribes have very good hearing. [ Laughter ] >>MILTON MUELLER: I wanted to respond. >>WENDY SELTZER: I just wanted to say, I also did not intend for anything that I said to be construed as advocating government blocking but merely recognizing that it is a fact of life. >> That's what I hear. [ Laughter ] >>WENDY SELTZER: And as between that and blocking at the core, blocking at the edges does less harm to global freedom of expression. >>BRUCE TONKIN: There are some big edges. There are some big countries out there that you are blocking. >>MILTON MUELLER: Bruce, I don't think you can blame -- if governments want to block, they're going to do it. And you can't blame an open policy at the core for causing more blocking. I guess the implication is we should do the censorship for them so they don't feel motivated to block? [ Laughter ] >>BRUCE TONKIN: All I am saying is not encouraging blocking, Milton. >>MILTON MUELLER: Nobody is encouraging blocking, except that end users do have the right to block. >>BRUCE TONKIN: Absolutely. >>MILTON MUELLER: At the domain name level, it is easier to say my resolver will simply not allow domains that have the dot kill all gays extension to come in. >>BRUCE TONKIN: As long as that's at the end user and not at the country level. >>MILTON MUELLER: And I think under the rules as I read them, when you are talking about generally accepted principles of morality, that a dot gay could get into serious trouble not from western liberal countries but from other countries in the world. I think you're opening a door to that. >>ROBIN GROSS: I think also the point is not to encourage governmental censorship but to recognize that we live in a world where that happens and we don't want to extend the censorship in one country to the rest of the world. Let's try to limit it as much as we can. Anyone else? And since we have such a long line and -- why don't we try to keep the comments to three minutes. Thanks. >>CHERYL PRESTON: Hi, I'm Cheryl Preston, Brigham Young University Law School. I take issue with the statement that the freedom of expression arguments have never changed over 400 years because I don't think Milton -- not current Milton but historical Milton -- would ever have imagined -- [ Laughter ] -- would ever have imagined an Internet where children are exposed to graphic, explicit sexual conduct in images. And so I think we got to just divide two concepts here. One is the argument of a thin Internet and the argument that ICANN ought to stay completely out of any kind of policy value system and simply make approvals on a technical basis. And the other argument is are we really encouraging or choosing a, quote, free speech position which is going to create barriers down the road for everyone who's concerned about things as for which there is surely a consensus, like child abuse, child sexual exploitation, trafficking children for sexual purposes. And so if you look at the petition that keep the core neutral, the title is neutrality. But if you read it carefully, it appears to continue to advocate an extreme free speech position, and that doesn't represent the interests of the consensus. Certainly not the interests that have been recognized in any government. Finally, current Milton said that he thinks there is a consensus in every country -- well, every democratic country about freedom of speech. But freedom of speech is such a loose term, I don't think there is any democratic country that sees freedom of speech as includes exposing children to graphic sexual conduct. And so I think when we use that term, there are some basis, there are some domain names that really would be offensive in every country. And the higher standard, I would think, would be for ICANN to become a world leader in drawing some bare minimums so it doesn't have to be blocked in every country. Certainly, the concept of free speech is protecting unpopular political speech, human rights speech. That's something that we can agree on, but there is a point at which speech goes to issues that the historical Milton would never have imagined would come under the rubric of freedom of speech. >>ADRIAN KINDERIS: Thank you. [ applause ] My name is Adrian Kinderis. I will keep this brief. Call me naive -- and I have been called worse things. But during all that rhetoric -- and I appreciate the sentiments, when you talk about neutrality and you talk about keeping it simple and you talk about a stupid network and so on and so forth, that implies that the network itself or the core is neutral. And speaking as an Australian and also someone that's worked in the Middle East for the past year, it's quite apparent, at least, that the ownership of the Internet isn't neutral and potentially given the basis of ICANN and where it's come from and the Department of Commerce's long reach that still is around ICANN, it could be argued that it's not neutral. My point being at the end of the day if we open the door for any TLD and didn't care about anything -- and I'm not necessarily saying I disagree with that sentiment, but are you going to -- there is not going to be one senator in the U.S. that's going to not try and assert some influence when anti-USA -- or dot antiUSA gets put forward. We said it with dot XXX, that that's not going to impact someone within ICANN to make it want to stop. >>MICHAEL PALAGE: Can I answer that? If you will, in a separate I wrote as part of the WSIS who is called "DNS Detante" -- one of the things -- I agree with you, I believe no one government should have any blocking authority. To me, the root is a community resource. And one of the points of divergence in writing this paper that Avri and I had was it was constructive disagreement. I was actually of the opinion that in order to be in the root, the registry operator, we talk about how the registry operators must be in sort of legal compliance with their laws. I was of the opinion that a registry should go to their government and say "here's our business model, we want to operate under the laws of your country. Will you please, you know, if you will put me in the root?" I believe every government has the ability to put something in the root. That's my personal viewpoint. So I agree with you. I don't believe that the United States government should have a -- the ability to block, if you will, a dot anti-U.S. It is that type of, if you will, monopolistic control that I think has led to the criticisms or the current situation that we find ourselves in. So if it is truly a public resource, every government should have that ability. >>ADRIAN KINDERIS: Right, exactly. I think that sentiment certainly still exists and we can argue all we like about stupid networks and everything else, but it ain't stupid. It is owned by somebody and it is it the U.S. government. >>MICHAEL PALAGE: To follow up, that's the whole idea of keeping the core neutral. There is no single point of control. >>MILTON MUELLER: Your argument is completely consistent with the theme of this panel is that it should be neutral, right? And just a quick response to Cheryl about is there agreement on content issues, for example, child pornography, that again is an example of sort of what I was saying. The current Milton, as you put it, you are talking, first of all, about content, not about domain names and there is, in fact, an international convention on the rights of the child which explicitly bans child pornography and other international laws and all kinds of national laws which involve international cooperation around child pornography. That's my point about governments taking the responsibility, if they want global standards, to do it right and not do it indirectly by the raised eyebrow. It just sort of saying we don't like these words. Pass a law about it. Tell us what those words are and what makes them illegal. [ applause ] >>ANNETTE MUEHLBERG: Annette Muehlberg from the At-Large Advisory Committee. First of all, I want to thank you all very much for these enlightening speeches, and I have two remarks. One is it has been mentioned all the time before, it was a little bit too relaxed for me as the argument, the nations can block. I think to keep the core neutral, we should be a little bit more careful with that argument and, you know, to have freedom with the argument you can take it away over there anyway, that is always a bad argument. So, but I can see your intention -- I share your intention. It is just the way you argue it, I think, there should be some more work on that. The other thing is we're talking about control and power and something which was not mentioned at all, the fact that the introduction of new gTLDs normally is -- these are restricted gTLDs. So now we are -- we're going to have more and more TLDs with general meanings, you know, like real words. And if these real words like we already have dot jobs, dot travel and all that, these top-level domains are run by people who have the right to choose who can register. And I think this causes a lot of problems, also political problems, and you would take a lot of pressure out of the whole situation if top-level domains would not be restricted. Then if everyone can express his or her opinion and register whatever domain they want to register, you take a lot of pressure out. For example, if you had -- I gave that example before, if you had a dot God and the one who runs it is a member of the Ku Klux Klan or whatever or just one special religious group, they say only those people can register, then you're in trouble. But if you have, you know -- if anyone can register a domain, it is totally different. So I think the matter of having restricted domains are, you know, free for registration, that plays an important role, too. >>VITTORIO BERTOLA: Vittorio Bertola speaking absolutely personally. Well, first of all -- >>ANNETTE MUEHLBERG: May I add I am speaking personally but I am a member of the At-Large Advisory Committee. >>VITTORIO BERTOLA: And I am not. I am not really, it's true. I was going to say, first of all, that I don't have a final opinion on the entire matter so I came here to listen because this is really complex. But the first thing I think needs to be said is that for me there is no such thing as a neutral core. The decision to have -- to allow everything into the root mechanically or stupidly so that you can have dot Nazi or dot Jihad or whatever blasphemy you want to say around whatever God, it is a decision. It is a choice. You want to say it implies certain values and hurts other people's values. It is not like you are choosing. You are actually making a very overloaded choice. And this is actually the problem. The problem is that the Internet is a first real instance of a global society where all the different cultures of the world are in the same place. And this creates the growing tension between enforcing the human rights, not just the universal human rights because the universal ones should be agreed by everyone, even if the interpretation is not agreed. But your rights, the one your society gives to you but other societies do not give to their citizens, this creates tension respecting the differences of these other societies which are at different stages of their evolution or evolved in different ways. I think they are totally entitled to have different values than ours. And so actually, what you seem to be trying to do is to trying to push the western liberated values on everyone else aggressively. You are trying to say since we want to have the right to discuss the things in whatever way possible, including the most visible way at the root level, not just in a Web site somewhere but at the root level of the Internet, then everyone else has to come up with that and accept that and does not have the right to object. And whether this is fair, to me it's not fair, but this is a personal call because again it is a matter of values. The other point I want to make is, where does it lead? I think there are two possible consequences of this approach. The first one is that if I can say we do not want to make political judgments, then someone else will make them. The rest of the world does not believe that these judgments should be purely technical, just go through the ITU or whatever, create something else to make these decisions. There is no way that they are going to say, Okay, we accept it is, it is just technical and we go away. No, that's not going to happen. The other possible option which is what you're actually saying in your document is that this is going to become -- dealt with at a level. We are not encouraging them to do this, but it is in their power to do it and it is true. But you are actually putting them in a position where they do not have any other alternative to enforce their laws and their values. Even in those countries where the laws are democratic, because dot Nazareth would be forbidden in many countries in Europe. We are not just talking about Arab countries or China or the usual suspects, if they are suspects. So actually what you are proposing would lead to more censorship of the Internet. Would lead to less dialogue. The real value of the Internet is actually people start to talk to each other and that's the only way that these cultures can over time change their values and accept these rights that you are proposing. If you stop dialogue, you just create fireworks and they are never going to change. For a world that has this huge problem of engaging certain parts of the world with different values that are different from tanks and bombs, it is really a pity. I think it is a sad proposal. Thank you. [ applause ] >>MICHAEL PALAGE: Vittorio, part of the western philosophy -- I mean, one of the things that I had just stated before is I'm of the opinion that -- my personal opinion, if a registry operator has a business model, they need to work with their government upon which the laws that they're regulated, because the last thing I want from a security and stability standpoint is someone operating a business where that country could change the laws and imprison that person. Okay? So to me my personal viewpoint here is that if someone wants to operate a business, they need to work in a collaborative fashion with their government to make sure that they have government support. And I would even go -- I don't want to say that they would need to sponsor an application before being submitted to ICANN, but going back to the comments about keeping the core neutral where every government should have access to this global resource, I don't think that's necessarily a western philosophy. This is a viewpoint that, I think, every government should have the right to work with businesses in their community to put something on the global Internet. So I don't think -- you know, again this might be -- I am not speaking for others on this panel that may have a more, if you will, different viewpoint on free speech. But I just want to try to sit there and articulate my concerns of the impetus of when I was involved in writing that paper and the previous paper that I wrote entitled "DNS Detante." I just wanted to comment on that. >>MILTON MUELLER: I have to thank Vittorio for putting the issue so nicely. This is really the most fundamental issue and it is what I called before the communitarian approach versus the reformation approach. You are saying there is no such thing as a neutral core when you put all of these diverse TLDs in on a neutral basis, you are basically creating a kind of western liberal ethic which is being diffused to the world. Now, that might be a coherent argument but then you're contradicts yourself and say the countries will censor at the national level and that's bad because that's not liberal enough. What we need to do is engage in a -- some kind of dialogue at the global level that encourages all these countries to be liberal. And that to me is an inconsistent argument. I also don't except the idea that the core can't be neutral. There is a very clear -- maybe we should be more precise about what we mean and say that the TLD process should be content neutral insofar as possible. Other than the visually confusingly similar part of it, you can be neutral in a sense that you're saying we're not going to eliminate this TLD from the root or eliminate its chances of being put in the root because of its meaning to certain people that don't like it. You can do that. That is not impossible. Again, I just -- to say that that puts countries in a position where they have to impose more censorship is, to me, a contradictory argument. Even the censorship. occurs at the global level and it is uniform and more repressive or it occurs in spotty fashions at the national level. And at the same time, in order to have this principle of neutrality fomented, we do have to have a global dialogue about why that's a good thing and about how it creates more opportunity and more freedom. In that sense, yes, we are encouraging greater freedom for the Internet. I don't think we are imposing any particular cultural norms on countries that don't want them. >>WOLFGANG KLEINWAECHTER: Yeah, thing. Philosophically, freedom is always linked to responsibility. So I think this is really important. It is merely impossible to definite fine exactly what the relationship of freedom responsibility is in different cultural contexts. I think this is an agreement -- and we should remember this agreement -- while on the legal level rights always links to duties. You have no absolute rights. Each right is linked also to a duty. And in doing so, I would not agree what Michael has said, countries have the right to control content. Yes, to a certain degree but, you know, in the Article 19 of the international covenant, it said absolutely clear that restrictions have to be justified, and the justification comes because restrictions are aimed to protect other values and the values are described. It means you have no carte blanche to control content. It means if you control content, only under certain circumstances to protect certain values which are values in the society and then you have to look for a balance. So if there is a conflict that different groups have a different interpretation about the balance, then you have to go to a neutral third-party, to a court, to say, okay, is this value higher than the other value? That means a country without a neutral court system, this is -- leads then to total control. Let me make a final point. Some of the debate reminds me here on discussions which took place in the 1970s and 1980s in the outer space committee of the united nation when is they debated satellite television, broadcasting of satellite programs via television. It was pushed by the Soviet Union. They wanted to have a control systems. They asked for prior consent that can send television signals only with the consent of the receiving country. And then they there were all sorts of cultural arguments. Bull fighting in Spain is very popular but in India, the cows are holy animals, so it would violate cultural values of India if you would have bull fighting on television from Spain to India. To find a compromise and resolution then -- non-binding resolution was we need prior consultations. That means we have to talk to each other. But we shouldn't give the right to anybody to make the decision. Talking to each other to try to understand each other and in the light of the arguments of the others, then to make a responsible decision that is different. So there is no golden way. You have to figure out the balance. And I think this is important. We have to strengthen a responsibility behavior, how we use freedom in this bottomless cyberspace. >>MICHAEL PALAGE: If I could respond to Wolfgang, I appreciate those viewpoints and one of the bases for my opinions of why I sort of advocate the closer relationship between a registry operator and its national government goes back to a conversation I had with Phil Sbarbaro, former deputy counsel -- or legal counsel for NSI who in my opinion is probably one of the most brilliant minds in this space and I remember having a conversation with Phil one time and I said, "Phil, when you have courts in different countries telling a registry operator to do different things, how do you go about resolving how to resolve that?" And he says, "It's real easy, Michael." He goes, "The one court that could put your officers in jail is the one you have to defer to." [Laughter] >>MICHAEL PALAGE: So -- and that is why, as I said, my approach here, although not as -- maybe as based on philosophy, it's more pragmatic, so what I look at is, I want -- I think it's important for a registry operator to have -- operate in a country where there's confidence that he will not be put in jail, that the operations will not be disruptive, and that's why my -- again, my personal viewpoint is, I think it is important to look at, if you will, where a registry operator is proposing to be incorporated and to make sure that that business model is not inconsistent with the laws of that country, because otherwise, I think you may have unintended consequences. So, again, that's just a little historical perspective on my basis. >>ROBIN GROSS: Bertrand? >>BERTRAND DE LA CHAPELLE: Okay. My name is Bertrand de La Chapelle and I'm the French GAC representative. I first want to say how happy I am that you organized this workshop, although it looks a little bit more like the presentation of the campaign rather than a discussion on the substance, but it doesn't matter. I'm very happy because it shows how obviously ICANN is only a technical management body. I mean, the issues we're discussing here have absolutely nothing to do with higher level, so it's great. It illustrates the dimension of policy that ICANN is handling, and is good. But I want to first stress that I have made no mystery in the past, and in previous Fora, of the concerns that the French government that I officially represent in the GAC has with the present new TLD introduction process. This is a different issue, and I'm not speaking here on behalf -- in an official position. I've spoken and I will speak on other occasions. I'm not speaking either as a Frenchman, although there are many references to cone knack, to dot Paris, to the Paris convention, the Paris treaty would make me -- [Laughter] >>BERTRAND DE LA CHAPELLE: Basically Paris Hilton as well, yeah, but she's not connected. [Laughter] >>BERTRAND DE LA CHAPELLE: Anyway, more seriously, I want to say that I'm making this intervention here on a personal basis, as an individual, that has invested and devoted the last seven years of his life on a professional basis, and I think on a personal basis as well, to the multistakeholder approach, and to the promotion that this is the only way forward to address the public policy issues related to the Internet, and the way to make global -- or Internet governance, first, and then global governance work. And in this respect, at a first level, the approach that Milton is pushing forward, with all due respect for all of you, that basically there is the "we" ICANN and the "they" government is completely misguided. It shouldn't be the case. I'm working as hard as I can so that the GAC members participate in the full ICANN activities and in policy development process, and this is a major issue. If there is one issue where the interaction between governments and the rest of the ICANN community is essential, it is this one. I'll come back to that later, but it's an important point I wanted to make. Now, more to the point, the first thing is before the latest additions, the cultural diversity of the panel, I would say, points towards a discussion that looks like not a freedom of expression discussion but a First Amendment discussion. [Applause] >>BERTRAND DE LA CHAPELLE: Excuse me. That's the truth. That's a fact. This is not a First Amendment discussion. And by the way, in the new TLD processes, and all the trademark discussion, I fully understand -- and I speak on a personal basis -- that the liability for ICANN as a whole, because of the location of its headquarters in the United States, that the liability would be towards the U.S. courts and to the U.S. legislation is bringing a large part of the rules that are going to be applied in the selection of TLDs. It's not a criticism; it's a fact. And it's a problem that we are facing. This issue is broader than a First Amendment or trademark law applicable to ICANN. But now I want to come to the other element. You're talking about the core, and keeping the core neutral. As far as I understand, from the discussion, what you mean by the core is the TLD space. If that is the case, then I would argue that your campaign is not about keeping the core neutral, but making the core neutral. Because at the moment, the core is not neutral. It is a limited number of very few words, a tightly-controlled thin core. It's not a neutral core. It is a thin and efficient core. If you make a campaign keeping the core thin and efficient, I support it. Making the core neutral is a perfectly valid campaign, but keeping the core neutral is probably not the right word. I think it's a big difference. As a side element, I would like the presentations a bit in line with something that has been said earlier, the presentations to be done with dot nigger, for instance, or dot fag instead of dot gay, and for the sake of competition, you would have dot nigger and dot Negro and try to run your arguments with that kind of string and see how you feel in terms of what kind of benefit it provides for the global community. But more seriously, I want to come to two points. And I know I'm long, but it is sufficiently important for me to just explain clearly. The first point is, the whole argument about the regime for actually content control and the articulation between the national legislation and the neutrality of the system actually I don't think applies at that level. Because the reality is that the questions that you're addressing -- and maybe the solution that you are proposing -- is a perfectly valid thing at the level of the content, and it is only applicable to TLDs if you consider that, in fact, the TLDs are the equivalent of a content labeling string. A bit like the problem that was used in the dot xxx. As the dot xxx was used as a labeling or potentially labeling thing, then it was triggering a potential control. The TLD string, as far as we are all concerned, has not been discussed as a labeling scheme, and so the content should be addressed in a different manner. It's not a TLD string. And to come back to -- quickly to the core, the innovation that is emerging at the margins and at the edge, if the core is the TLD, then the second-level domain and the third-level domain and further upwards is the edge, and the whole exhibition here, each booth, is showing how innovation is emerging at the edges. Not in the TLD, but in the second domain and further on. So it's not a freedom of expression issue or labeling issue, I believe, and the core is a thin core and not a neutral core. But the last point I want to make, which is the most important, this is about what we believe the TLD space is, and the TLD space is, as Bruce Tonkin has said in the Lisbon meeting -- and that was the clearest expression -- the TLD space is a vocabulary that is used by communities around the world to identify themselves on the cyberspace, and the real challenge that we have today is the governance of those communities on line. In a certain way, the ICM proposal for dot xxx had a whole dimension of internal governance among the actors who will be registering or using this space. The key challenge is not so much the string only, but how the governance will be organized. We have a billion people on line. They represent the global community and they are organizing in social networks that create global polities. The challenge of the TLD space and the ! new TLD introduction is how do we help the global community organize not in physically-bounded communities, but in fractilely connecting communities. We have overlapping territories, and this is about organizing virtual communities and virtual territories worldwide, and not about discussing freedom of expression at the level of the string. I'm sorry for the length but -- >>ROBIN GROSS: Thank you. Do you want to -- [Applause] >>MICHAEL PALAGE: Wendy could go. >>WENDY SELTZER: Thanks, Bertrand. A lot in there and I just wanted to respond on the very last point, where you say TLD space is a vocabulary for communities. I would think of it actually not as a semantic vocabulary, but as a set of hooks upon which communities can build or application developers can build or service developers can build or content developers can build. >>BERTRAND DE LA CHAPELLE: Absolutely. >>WENDY SELTZER: And so my notion is to keep the barriers to developing those hooks minimal by keeping the criteria minimal, so that people including communities can develop there. >>BERTRAND DE LA CHAPELLE: We are, therefore, in line. >>MICHAEL PALAGE: And just to follow on a personal level, the dot Nazi and the dot nigger, those were two of the strings that you used -- dot Nazi or -- those are two strings that I personally find offensive, and personally I don't -- on a personal level, I don't believe they should be in the root at the top level. That's my personal viewpoint. I think what -- part of -- the reason I used dot gay, and I think this is a good example and again may go back to some of my other comments about where a government would support, right? When we looked at the xxx debate, one of the things that was key in the GAC communication was, there was no government that supported that application. I would hope that if there was a dot gay, there would be a government that does recognize same-sex rights or other things, a more progressive government that would be willing to, if you will, recognize the organization of that registry under its national laws and would be supportive of it. I think in the -- in the two examples that I used previously, I don't think you would have a government that would be supportive of those two strings that I had used before, and hopefully the -- hopefully the mechanisms that we are talking about or that we're discussing right now in the ICANN community would work and prevent those strings from ever resolving. Again, that's my personal viewpoint, and again, may not be representative of some of the more extreme or absolute viewpoints. >>ROBIN GROSS: Okay. We're so late on time, let's move on to our next speaker and if -- if I can please plead with our speakers to keep it to just a couple of minutes, because we've got a bit of a line here, and not much time left and we really want to hear something from everyone. >>WERNER STAUB: Okay. My name is Werner Staub. I speak for myself and those, I hope, who are also working on TLDs for communities whose needs are not likely to address any possible free speech issue at all. And in this context, I would like especially those who campaign for free expression, to consider what it means if, as a result of this discussion, which has now gone on for some time and is now coming back with some vengeance, we have further delays. Further delays in something that shouldn't have been delayed as long, and what kind of free speech is the absence of any speech. I mean, not just the prohibition of unfree speech, so to speak, but the fact that we suppress speech during that time altogether as far as new TLDs are concerned, because we haven't quite agreed on what kind of safeguards there should be. Now, I tried to think what the motivations could be to insist so much on delaying now the process for this, and I came to the conclusion that possibly the word "precedent" is what worries you most. Indeed, by the fact that there is a rule now that says that ICANN would be self-censoring, so to speak, with the introduction of new TLDs, and indeed have rules that would actually allow to rule out dot Nazi or -- and other offensive strings, you see the fact that this is present, you know, would cause harm. Well, compare this to the other possibility. Compare this to the possibility that we now spend another two years discussing it, come to some conclusion, and indeed do manage to remove those restrictions that we right now have in the proposals. We will then have a predictable process, a predictable process how to introduce new TLDs, and that's going to be the wrong predictability. It's going to be the predictability by which all those who really want to create damage will! come forward and propose strings such as dot Nazi, that will indeed go through the ICANN process because there will be no rule against this, because we are predictable, and at that point, it's going to be stopped elsewhere. It's going to -- in the current setup, it will possibly be -- most possibly be stopped by the U.S. government because the U.S. government would then, so to speak, the intervention of last resort, stop it before it goes into the root. I would at least hope so. But then again, that is a very bad development and is not conducive to free expression. >>ROBIN GROSS: Thank you. Mawaki. >>MAWAKI CHANGO: Yes. Thank you. Mawaki Chango. Shall I say that I'm not speaking on behalf of NCUC? [Laughter] >>MAWAKI CHANGO: Okay. I totally agree that ICANN -- there's a need for ICANN to keep itself to technical coordination of Internet core resources, but the thing is, it seems that we can't avoid ICANN to take care of policy because there is a void. And that's a more longer-term -- a longer-term problem to address how to transform ICANN or to give ICANN the legal basis and grounds to be able to make policy. My main point here about the whole issue you are discussing, which I would prefer to approach in terms of technical coordination because the free speech or free expression, as you can see, may raise some -- some obstacles to some people who have supported the technical coordination arguments, but may have some issue with an absolute -- absolutist approach to free expression. The main problem, as Bruce and some others have pointed out, is that I don't think -- it seems to me that law is not necessarily the only way that social issues can be resolved. We talk in this context of international regimes. We talk a lot about soft law, hard law, and norms, and I believe that there are issues that legislations don't address or don't take into account, necessarily, or it's not felt that there's a need to go all the way to legislate in order to address those issues. So in those cases, do you think there's a kind of soft law fashion to address the issue you are raising instead of dumping everything to national legislation and relying on national law to address the issues? And a second point. The ICANN decision-making processes or procedures. If ICANN -- if the board members of ICANN eventually have to vote in order to decide whether to issue -- authorize a top-level domain name, I'm afraid that no matter what, even if you succeed in the campaign of keeping the core neutral, the individuals who are going to vote on that decision, they will vote with their perception, with their values, with their culture, their norms, and there's no way to control or to make sure that the core is really neutral. So are you proposing, for example, in order to authorize TLD that ICANN or the board members of ICANN have a checklist in order to check whether the application meets some criteria like operational criteria and technical criteria, instead of voting on the TLD? And last thing, about IDN and confusingly similar. I really much appreciate the presentation about trademark, and for me it clarifies a lot. We have been struggling with this term "confusingly similar," and one of the main arguments was that we are taking it from the Paris convention or trademark law, so we can't change the term. Otherwise, we may run the risk of setting up a new legal norm or people may understand that that way. So I would like to ask if it's possible, if you think this is feasible for us, as a GNSO Council, to qualify the confusingly similar and specify that we mean visually or graphically confusingly similar. Thank you. >>RAM MOHAN: Mawaki, thank you. To your second point, I think absolutely the evaluators should have some checklists that they should go through. On the technical side, there are some pretty clear bright-line things that they should look at. They should look at whether the application will continue a process of universal resolution of the domain names that will be registered under that label. Will there be continued interoperability. Those are pretty clear bright-line principles that ought to be there that perhaps aren't there in an explicit manner right now, and that is important. On the -- on the second point about confusingly similar, perhaps the right way to talk about it is visual -- visual confusability, and we don't have to wordsmith it here, but the real idea and intent isn't whether info is info in another script but it's really about when it looks a certain way, does it look exactly or very similar to the same label in another script. And visual -- visual confusability or visually similar perhaps are good terms to use. Thanks. >>ROBIN GROSS: Christine? >>CHRISTINE HAIGHT FARLEY: If I could just take a minute to say: It may be correct that the phrase "confusingly similar" may come from international law, but that's a piece of a larger phrase that sets out the legal standard, and the legal standard is confusion. So it's confusingly similar so as to be likely to cause confusion. So I think if you're looking for a shorter phrase to capture the legal standard, it would have to be, "Likely to cause confusion,." So the legal standard is confusion, it's not confusingly similar, and we can talk about it more afterward. >>ROBIN GROSS: Go ahead. >>TAN TIN WEE: Could I just clarify this point. Does that include phonetic similarity that would likely lead to confusion and would that include the possibility of using semantic similarity and so forth that will lead to confusion and is that what GNSO's recommendations in using the term "confusingly similar" is intended to mean? >>CHRISTINE HAIGHT FARLEY: Okay. So "confusingly similar" does relate to all kinds of confusion, phonetic, visual, semantic, but again, it's only the first step of a larger analysis about the likelihood that will cause confusion. So it's just a piece of it, and it's not the legal standard under any law. >>RAM MOHAN: And in the GNSO IDN working group, the -- when this topic was discussed, it was actually parsed into a very specific piece where there was agreement about visual confusion rather than phonetic or something that sounds the same, so there was actually further granularity in what that was intended to mean, and clearly some more wordsmithing is required. But it is not the intent and the broad agreement was not in, you know, generically saying, "Semantic confusion" and, you know, "phonetic confusion is included." There was some sport for it but there was not broad agreement. >>CHRISTINE HAIGHT FARLEY: Then that would be creating a new legal standard, which may be appropriate, but it's not what the law currently states. >>MAWAKI CHANGO: Just a note. What you're saying, Ram, is in the IDN gTLD report, which is subject to the new gTLD report itself, and in the new gTLD report, there's no mention, there's no qualification. One thing I -- one word I just forgot to pronounce is that we are working here in trans- -- let's say, a multi-script context, so I just wanted to point that out for you. Maybe we'll discuss it later on. The Paris convention, the previous legal frameworks, maybe were intended to one script context, and here we are dealing with multi-script contexts and we need to really make the distinction and qualify precisely what we mean by "confusingly similar." >>CHRISTINE HAIGHT FARLEY: Uh-huh, uh-huh. >>MAWAKI CHANGO: Thanks. >>ROBIN GROSS: Chuck. >>CHUCK GOMES: Thanks, Robin. Chuck Gomes. Let me comment on the "confusingly similar" issue first. That's not why I got in line, but thanks for the clarification on that, because the new gTLD committee is not trying to create a new legal standard. We are trying to accomplish exactly what you said. In fact, your language may be better language than we have and we thank you for that. So I really wanted to get up here and make a content-neutral -- I think content-neutral statement. ICANN is supposed to be bottom-up. And I think we're all a part right now -- even this very moment -- of a bottom-up process that is trying to incorporate views from all over the world, from all different cultures, from all different legal structures, et cetera. And the input that's coming today from this group is a vital part of that process, as is the people that got in line and disagreed, and those that aren't even here today. The endgame, though, is for us, first of all as a new gTLD committee, secondly as a GNSO Council, and ultimately the ICANN board, to try and -- and hopefully in not too untimely manner come up with some positions that most of us can live with that try to accommodate all of these concerns. And so I thank you for the input today, as well as everybody that came to the mic, and that all of the input that's coming into this process, but let's all work together to come up with something -- we're not going to all be perfectly pleased -- that we can agree on and we'll probably have to improve it down the road, but let's try and do that in a timely manner and work together to make that happen. [Applause] >>ROBIN GROSS: Thank you. >>MILTON MUELLER: Just a brief thank you to Chuck for making that point and I just want to make it clear that, you know, I've been in these new gTLD -- pushing for that for now like eight years, and I'm not trying to hold anything up. And I think it's very easy to accommodate some of the concerns we have with some changes in the language of the existing report. It could be done tomorrow. >>ROBIN GROSS: Becky. You have the last word. >>BECKY BURR: I hate -- some of you may be surprised with what I'm going to say but I'm actually very disturbed that we're having this conversation in the context of Nazi and nigger. I mean, there is an international convention that addresses that as well. So when I think about this -- I was very heartened to hear Bruce say that dot gay would go through, and the question I have is: Why is he confident that that would go through? What is the standard by which he is -- that he's going to give to the panel that resolves the dispute? So long as I know what the standard is, and so long as it's a little bit more helpful than morality and public order, we can evaluate it. So I think one of the things we may be getting hung up here is that there's this sort of big vague thing out there and we don't have the -- a clear sense of what the standards against which it's measured are. So maybe the problem is that the work of the committee is not done. I don't know. >>ROBIN GROSS: Thank you. Any final remarks from the panel? All right. Well, thank you all for -- I'm sorry. Wolfgang, did you have your hand up. >>WOLFGANG KLEINWAECHTER: No. I'm looking forward to the next 100 years. [Laughter] >>ROBIN GROSS: Okay. Go ahead. >>RAM MOHAN: Thanks. I did have just one thing to say, which had to do with some discussion about the term "real harm." When I say "real harm," I mean real harm in terms of universal resolution of domain names, and real harm from the fragmentation or potential fragmentation into a non-interoperable network. I mean we're united in our goal to bring disenfranchised communities or communities not yet on line on line. We're actually quite united in that. And I think categorizing concerns about issues like bank fraud or identity theft or cyber crime as too complex or a -- you know, something that affects the have not's, I don't think it does anybody any service. We're together in moving forward, in bringing all of these communities on line. But let's not forget that while we're doing that, the core continues to be stable, that the communities -- both the have's and the have not's, as Tan Tin Wee was saying, can actually rely upon what they're seeing, that when they click on something, what they're going to is what they thought they were going to, that it's actually reliable and stable and se! cure. I think we -- we have to continue to ensure that that kind of fundamental perspective isn't lost in the general "Sound and Fury" about have's and have not's. Thanks. >>ROBIN GROSS: Thank you. And thank you all for coming and participating in this discussion. We hope this is not the end, but the beginning, or it certainly will continue for a while longer to continue. Thank you so much. [Applause]