______________________________________________________________________ DRAFT TRANSCRIPT Opening plenary Wednesday 1 September 2004 9.30am ______________________________________________________________________ PAUL WILSON: That was really fantastic, wasn't it? We haven't had a welcome like that at an APNIC meeting before. That was laid on by the Sheraton for us. Thanks to them for giving us such a great welcome to the meeting. Welcome to you all as well to the opening plenary of the APNIC AP local policy meeting. Thanks for coming. It's a good turnout for 9am Fijian time, particularly after last night's social. It's great to see. Thanks for coming on time. Thanks also to TransTel for last night's social. The meeting hosts are Telecom Fiji and Connect Fiji. We've been working hard to make it possible. Thanks to the hosts. Today's meeting is also sponsored by Softbank BB and Internet NZ. The support we get for the meetings - financial and any kind - is critical to making the meetings as accessible as possible for anyone who comes to reduce the cost to allow us to do things like the online participation - we've got web cast happening right now. So welcome to those who are joining us on the Internet. This morning's opening plenary, the agenda is on the screen. I would like to introduce Winston Thompson, Jordi Palet from Consulintel will be giving us a presentation on IPv6 as an innovation opportunity. We had Phillip Harris from Cisco scheduled, as you may have seen on the agenda. He's been unable to attend. There have been a couple of last minute changes. We're going to be hearing about ICANN and WSIS from Mohammed Sharil, the chair of the Government advisory committee. We'll hear about the AfriNIC status report. Finally we've got a public presentation of BGP, the movie, which Geoff Huston will be bringing us. That's a fairly auspicious event which has been moved from the routing SIG into the plenary. Without further ado I'd like to ask Winston to come and give us a few words of welcome from Telecom Fiji. Thanks very much, Winston. WINSTON THOMPSON: Thank you, Paul. A very warm bula to you all. Bula, for those who haven't been here before, is the Fijian word for welcome. It encompasses the many meanings, not only good health, but how are you, cheers, virtually anything to do with lightening the spirits and conveying a sense of joy and conviviality. That's what bula represents. So wherever you go in Fiji you will probably be offered the word. That conveys a whole host of meaning. Akinori Maemura, the chairman of APNIC, Paul Wilson, the Director-General, regional Internet registry heads, ICANN representatives and delegates, a warm welcome from Telecom Fiji to the 18th open policy meeting. Telecom Fiji and Connect Fiji who are your hosts for this meeting here are pleased and honoured to be given this opportunity to host not only what is a very important meeting, but also we understand the first of its kind in Fiji. The amalgamated Telecom holding - Telecom Fiji, Connect, TransTel and FINTEL - have been working over these past weeks to make sure that this opportunity is taken advantage of to demonstrate that here in Fiji we have the telecommunications infrastructure to service and coordinate a meeting of this sort. It is a development that is changing the face in Fiji - in fact, right now there is quite an intense public debate and argument going on in Fiji about the position, functions and costs of telecommunications in the wider community. It is one that we are very closely involved with. This sort of meeting helps to highlight how important is this aspect. We have been very fortunate that APNIC and the Pacific Islands Telecom Association, along with other stakeholders sponsored a networking workshop last week here in Nadi which for some delegates was a lead-up to events this week. This made it easier for all concerned in the organisation of these meetings to coordinate on a back to back basis in one location. I understand that APNIC's policy meeting - open policy meetings - look at policies that will address the changing environment of this industry and it will also provide an awareness and understanding of the use of the Internet protocol version 6 in the Pacific region. As you well know the Pacific islands are very scattered and in most cases very small, in many cases tiny. So the population and the peoples and the governments are looking to the Internet - this is becoming evident every day in our meeting - to provide the means not only to connect them with the rest of the world, but also to provide the wherewithal so that they can find an economic place in the world of the future, which, because of their isolation, their smallness and the paucity of resources, they have very few opportunities. So there's a great deal of expectation in what the Internet can provide for them. So we're looking forward very much to a fruitful and successful meeting and hope that you will enjoy the activities that have been organised and that everything goes well. If arrangements are not to your satisfaction, we would seek your indulgence and hope that you will look at it kindly. While we have done some of these in the past, we're not very proficient and professional at it and I'm sure there will be hiccups along the way. Please bear with us. Finally I urge you, for those who have not been here before - even those who have - to take the time to look at the wider picture in Fiji. There is much to offer. There is much that you can offer in the contacts that you make with the people in Fiji. We hope you take this opportunity to experience what there is and the diversity that is here and that you will leave this meeting with a greater sense of understanding and awareness of what a Pacific island is like and there are many islands that are less well endowed than Fiji. Even here you see the stark contrast with many parts of the world. Thank you again. We wish you a very enjoyable and successful meeting. Thank you. PAUL WILSON: Thank you for that welcome. I'd like to say that any lack of professionalism that may have been seen - and I can't think of any - is far outweighed by the friendliness of the welcome we've seen here. I think that friendliness is preferable in fact to the professionalism that you might feel is lacking. I can't think of anything. If there is anything about the meeting arrangements, I might add, that require our attention, then please do certainly talk to the hotel or to APNIC staff. We're here in number. As I said last night, the APNIC staff are here to serve you, the members, to who make APNIC possible. So please don't hesitate. I'd like to introduce the next presenter, Jordi Palet from Consulintel, who will be speaking to us about the IPv6 issue. He will be one of the identities very much associated with it around the world. We're very glad and grateful that he was able to step in and provide this presentation to us on short notice. He's also helping out with some presentations in the IPv6 SIG later on. We'll hear a little more. But this is a general overview of IPv6 for innovation. Thank you. JORDI PALET: IPv6 is an innovation opportunity. While in this presentation what I am trying to say is that we started with IPv6 for a number of reasons. We started with IPv6 because we perceived the need for more addresses basically because we are already looking into the growth of Internet - a lot of new devices, new users, and especially because we have always new technologies making use of the Internet all the time. But IPv6 today is no longer just about the number of addresses. If we look to the deployment of broadband - I have some numbers here from the DSL forum. They are not very updated because the last figures I got was from September 2003, but I think they are already quite impressive. They clearly say that in the last 12 months starting to come from September 2003, 25 million new global DSL subscribers and that at the end of that month it is expected to be around 55 million. I'm not sure right now how many millions of DSL subscribers we have, but I'm sure we already have 25 per cent or 30 per cent more than that. This is only figures looking into DSL technology. But we have today a lot of additional technologies that allows us to connect to Internet, not just broadband Internet, because unfortunately not everyone in the world has access to broadband technology. The meaning of broadband is not 2 MB for everyone in some countries. So it's really difficult to decide what is broadband in general but the important thing is getting connected. We have countries like China and Latin America in general, which have very interesting growth. We are sure at the end of 2005 we will have with just DSL 200 million DSL subscribers. If we want in every home can we provide 25 addresses for each of these subscribers. That's an interesting question. We look also to new access technologies, because we have now technologies like power LAN. Power LAN enables you to use the electricity network not just for access but also to connect a lot of devices inside every network. Having electricity everywhere means we have networks everywhere. Same with wireless LAN. If we look at the features of IPv6 I'm not going to go through them because I assume you all know the advantages of IPv6. But looking at the concept of out auto configuration - plug and play - this is a must. We cannot ask people at home to configure those devices manually or to have means to do all the devices like we do. We really need IPv6 for that. The advantages of power line and wireless LAN are clear. We need to look for traffic lights, information panels on the streets, metering systems, whatever security or surveillance devices we have. We have today a lot of Internet cameras and more are coming. Vending machines - we can guess there will be a big explosion of vending machines using power LAN. We have 3G - the basic stations need electricity. It is ridiculous to have two links - electricity and a data link when we can use just one for both. What about the wireless LAN points - we connect them to the electricity network and the data line. Again if we can combine that, that's impressive. It needs to be done with IPv6 because the number of devices we are going to connect not just in private networks, also in public networks, means there's no way we want to have private addressing a space for connecting all these networks. It will not be difficult to manage. It will be really difficult to manage. Same with wireless LAN. I read yesterday Amsterdam has a plan to deploy WLAN all around the city. I guess it will happen everywhere - not just in Europe. We want to have some time for fun, but also to have some time for business and probably combine more and more of that. That means we need to keep our access to the Internet all the time. So IPv6 is about addresses. While today we make use of NAT, PPP, a lot of technologies to share these addresses. But NAT doesn't work for a large number of peers - for example, when devices want to call other devices. A good example is IP phones or video conferencing. That needs the deployment of new applications and services. NAT compromises the performance, robustness, security and manageability of the Internet. New types of applications and new types of access need unique addresses. We cannot do it all the way around, which IPv4, even if we use private addressing a space, I mean NAT. We have addresses available - that's clear. Not maybe for everything that we would like to do in an efficient way. I mean, yes, we can still keep going for so many years that probably cannot count IPv4 but is that really what we want to do. It is really going to be so easy like now to obtain as many IP addresses as we want. Probably this will change. I guess we need more and more addresses - we will need to adapt again our policy. Maybe it will be much easier. Maybe because we are starting to deploy IPv6. So today IPv6 is about restoration and innovation. It does not matter if we have or not enough IPv4 addresses. I am convinced about that. We will never run out of IPv4 address space but probably that will happen because we are already starting to deploy IPv6. What will happen in addition with IPv6 while we need End-End security, NAT does not provide that security. I think it's a funny sentence. NAT is not security. IPv6 is about discovering the Internet principles, removing tunnels. We want to have End-End all the time. It's about making life easier for users. It's about who can play. They don't read manuals. Even ourselves most of the time don't read the manuals. So how we can ask the users to read the manuals to connect the devices. They must work through the network. We need efficient mobility. That's not possible with IPv4. If I have mobility in my laptop and that means I need to have some kind of device in my home network to provide that mobility functionality I cannot provide every network I visit to provide me with the functionality. With IPv6 I need to have a device providing my mobility functionality, my home network and I don't care about the rest of the networks around the world. I just get mobility all around the world. In addition - this is one of the most important things in my personal opinion - IPv6 is an extensible protocol. We can have whatever new we need with IPv6 without the need to integrate all the network. Just to integrate the two end notes willing to use that protocol. That's what we can't do with IPv4. Clearly IPv6 has a very, very nice advantage here. Probably it is the most important one. What about the transition. We started a long time ago to work in IPng, the original scheme or name for IPv6. During all this time IPv6 has been designed with coexistence in mind. During this time there have been a lot of doubts about the success and the real deployment, if it will happen or not, of IPv6. But I believe since a couple of years the picture is clearly changed for everyone. If we just look to 24/12 six months ago, I think it's clear not just in Europe, in Japan, in Asia-Pacific, in North America, it's here, it's coming. The question maybe is when. I prepared this picture for a meeting in March in Montevideo. They asked me to have a crystal ball. While my conclusion is probably when is not the right question, but it is when and where. Obviously even if IPv6 is a global thing and will not happen if it doesn't happen all the world around, I will not see IPv6 just in Europe or just in Japan. It really needs to be a global thing. But it will not happen at the same time at every place. It will change depending on which types of networks, which types of geographies, which types of business sectors. So my prediction - I hope not to be too much wrong - if we look to how much traffic, between 20 per cent and 30 per cent of the traffic could be IPv6. The IPv6 standard of traffic is important. I guess the Asia-Pacific will have a lot of traffic in the next couple of years and then Europe, North America, Latin America and the rest of the world - maybe even regions like Latin America will give us some surprises. I would really like to see that. Because I think for the development of these regions it is really important, the same as for the development of Asia-Pacific. We need also to consider a lot of issues when looking at the when and the where. For example, the analysis of the cost, the maintenance costs, the operation costs of the networks - comparing IPv4 and IPv6 - we need to look about how much IPv6 is going to be there and if it's needed. We need to look at what applications will take advantage of this type of IPv6 traffic. I think a killer application happens - we don't have a crystal ball to say it will be the killer application. Sometimes the killer applications that afterwards will be recognised are those applications we have already today but we are not using because let's say it's not possible to use them in a good way with IP connectivity, plenty of NAT boxes - so we need to look into it all to see which would be interesting to deploy IPv6. So maybe to help everyone answer this question about when and where, it's important to say we need to start reducing the cost starting now to deploy IPv6. Today we know the cost is not any more a problem in general. There is not additional cost for the networks. We maintain and upgrade them every so often. Today you get IPv6 out of the box without asking for it. It's already in the breaking system. It's already on all the devices, on all that you upgrade. Maybe the cost is education and training - that's an important cost. But it's not so difficult to get knowledge about IPv6 if you have already the knowledge of IPv4. So I Will say just start now. It's up to you to decide your own when. In the other way it's a question also about saving costs. NAT is expensive. For home users we are not considering how much - that's a cost. What about application developers - they will tell you they spend 80 per cent of their development time to make the applications working in every network with any kind of NAT device. One of the problems we have is we don't have a single kind of NAT. There are a lot of uses of NAT. The developers need to take care of all these different types of NATs. It creates a problem for innovation - because instead of making applications better, developers spend time making them work with any kind of NAT. Some operators have started to install IPv6 on a larger scale. They reported that the transition for them was almost no cost considering that they did that in several months or years - not just overnight, of course. They even consider that they are going to start having big savings in managing IPv6 networks compared to the same network as IPv4. About the applications. Well, I already started to mention this. I think IPv6 is an opportunity for new advanced applications. We have today a lot of technology. NAT devices are talking to the server to decide how they talk to their peers. We are not exploring the possibilities of peer to peer. For example End-End gaming - we demonstrated a few months ago at an event in Brussels how you can use IPv6 and quality of service measurement to see they can have a fair game, a peer-to-peer game with good quality network and not depending on how much bandwidth each user has and so on. We can start doing a lot of things here. Today End-End or peer to peer gaming is not really peer to peer and not fair. We have development of GRID. We have, for example, a very well known project that is not really peer to peer. It's using servers to make the communication possible. So a lot of good development of new applications will come in this area. Ambient intelligence sounds like fiction. We talk about it during a lot of years when we see the fiction films and we are in the rooms talking with the lights or talking with the ambient noise and getting information without a screen or keyboard. That's going to happen. That means we need to employ a lot of sensors, a lot of devices. They all need to have connectivity. Can we make that with private networks. Do you really believe it will be possible to have connectivity between different networks End-End and those devices will be able to be small if they need to handle different kinds of NAT boxes and be able to work in any environment? No, it's impossible. That's one of the reasons the people for BGP started to think about IPv6. They cannot really make sure to do very small scale reforms to deploy a lot of new applications if they need to spend cycles in computing to go through every kind of NAT device. So that's a very important thing. While the people will say we have again the chicken and egg problem - we cannot deploy or develop new IPv6 applications if IPv6 is not natively deployed. It will take some time. But maybe this is one of the deployment mistakes that we are doing. We should - we must - take advantage of IPv6 within the existing IPv4 infrastructure. We have the transition mechanisms to do that. There are a lot of applications that didn't succeed with IPv4, but if we have these applications running across IPv6 in IPv4 tunnels, while we have already some examples - I will show one of them at the end of my presentation - that we take advantage of all the applications and even we increase the possibilities of doing things with all these new applications. I think some of the application developers are already starting to realise that. So what happens with IPv4? Remember it has been designed at IPv6 to work with IPv4. I said we can use IPv6 with the existing IPv4 infrastructure. Of course it's not the best way to use it, but it's a starting point. So IPv4 is not the show stopper for IPv6. Until when will we have IPv4? It's difficult to say. Some people say 20-25 years probably. I don't think it's that important. IPv4 is actually surviving because we took measures - technical and policy - that we can change. So what is going around. Internet clearly is the victim of its own success, but with restrictions. That is the point. Innovation with IPv4 is not possible anymore. We cannot say IPv4 will finish in not just two years, five years, 10 years, 100 years - we don't care. Possibly never. There is not year 2000 effect with IPv6 - I mean with the transition. But not doing the transition to IPv6, it's clearly an opportunity lost if we are hindering the innovation, it means we are damaging our business, because at the end we owe it to business. So if we cannot innovate, we cannot provide new services to all customers. And the most important thing - depending on your knowledge, your expertise, the size of your network, obviously the transition is not overnight. It could take several months - from six to 24 months. So you need to start now. That's the important question. The most important one again insists on that. Take advantage of already existing applications that probably they didn't succeed because they were not able to perform as expected with IPv4 but with IPv6 we're changing that. My last slide is a kind of conclusion and personal thing - I hope most of the people here agree with that. Are we promoting IPv6 as a community? I think this community has a responsibility. Maybe this responsibility needs to start with asking ourselves should we promote IPv6? We have policy. We auto-regulate. We do the policy. Users don't know about IP. We need or we should help to make Internetting easier for everyone, maybe decreasing the costs, scaling the Internet, providing freedom of movement, making the Internet much more efficient, increasing the security. Is everybody deploying today IPv4 and IPv6? Then maybe we should provide IPv6 together with IPv4 blocks. That could require a policy change. Again it's my personal opinion, but I think it's something interesting to think about. It's not just for our own benefit. It's also for the community benefit. I don't think I have too much time. But just explain quickly before I finish my presentation. One of the applications that we move to support IPv6 is a home enabled application. This is a picture of my own home. I have several blinds, lights, different types of devices, IPv6 cameras. All these devices are connected by different means, including cellular interfaces, Internet, wireless LAN. While I am not going to do the demo now, if anyone is interesting, I can show is later, especially when it is daytime in Madrid. I have three docks and I connect the applications. These applications allow me to connect to every device in my home and open the blinds, see my dogs, even talk to them. While of course they don't bark at me anymore because they realise that I am not really there. But at least when I open the blind they still come to the window to try to find me, because they heard the noise of the blind opening and they believe I am there. But I can even give them additional food or water or whatever. These are the kinds of applications that maybe exist already today and we want to take advantage to say, "OK, until now, this application - home automation, industrial automation - has not been possible in a remote way unless you have a dial-up network. Now with IPv6 with End-End connectivity and security, we can take advantage of that and come back to this principle. With that I finish my presentation. Thank you. PAUL WILSON: The next presenter is Sharil Tarmizi from the government advisory committee. If you have any questions for the previous speaker, I will ask you to hold on to those until the end of the session. We'll try to reserve some time for questions at the end of the session. I'm sure you may well have some questions for Jordi during the meeting or afterwards. SHARIL TARMIZI: Thank you, Paul. Bula! Morning and as we say in Malaysia... my name is Sharil Tarmizi. I have several hats that I wear. I work for a regulatory agency in Malaysia so I'm technically a government person. That's not always been the case. I used for be in the private sector until the last four years when I joined the government. I have that different perspective. I'm also chairman of the government advisory committee in ICANN. For those of you wondering why I'm not using my own laptop. I have a problem with this laptop. If I plug in something it has to be rebooted - rebooted again so rather than waste time I am using Paul's laptop. I would like to thank Paul and APNIC for the invitation to come here and the honour they've given me to be able to share with you some of my thoughts. I see many friends in the audience so I'm quite comfortable. A word from the sponsors - the organisation I work for is a regulator and Co-developer of the converged sector. We look after the telecoms, Internet, broadcast spectrum management, number allocation including IP addressing, digital signatures and postal. We have many perspectives on many things. That's the organisation I work for. I'd like to introduce you to my friend Elmer the elephant. I don't know how many of you have heard of the three blind men story with the elephant but I'd like to tell this story simply because depending on whose looking glass you use you will have a particular picture of the thing that you look at. In this case the elephant. Blind man No. 1 stands in front of the elephant and says the trunk, the elephant is a very strong creature. A tough, muscular thing. Blind man No. 2 touches the elephant from the side and says, ah, the elephant is like a wall, blind man No. 3 touches the elephant from the tail and says no, the elephant is a stick, not a wall. All three are right. But all three can not be - are also very, very wrong. So the point I'm trying to make is all three blind men need to get together so that they can form a better picture of the elephant. That brings me to the next thing. This is a telecoms map. Many of you who work in the IP address space, the network space don't even think about this. The applications that you do right on top of this are physical connectivity layer, but governments look at this, they look at how many cable ending points we have, in a given place, the kind of price you pay for this kind of connectivity and so on and so forth. That may not apply to but I'm trying to bring another perspective. The Internet we all know, nobody really owns it. There have been a lot of discussion in a lot of newspapers, a lot of articles written whether it's owned by the US government, or a group of RIRs, whether it's owned by - all sorts of people. But really it's a connected network of networks. You know this and this is the Internet perspective. It looks like the human brain. We know this, we know what keeps the Internet ticking, you guys are the ones who make it work. In trying to make the Internet work I think there have been various co-ordinating bodies set up, one of the organisations on the international level is ICANN. The key reason why governments are interested in the Internet, convergence and ICT. I think we heard in the opening speech this morning many developing countries are looking at ICT as in opportunity to reduce digital divide, as an opportunity to level the playing field, remember the two maps earlier, governments understand telecoms, they're trying to understand the web, called the Internet but that's the main reasons why they're interested. Which is exemplified by this picture. These are the kind of people who have interests now in IP addressing, in DNS, in Internet generally. You have the intellectual property. You guys are likely to be here in the RIRs. Some of you may be in universities, some of you may be somewhere here. But there's a whole other bunch of people who are interested in this. In those days - many of you know this also, there used to be the one man that used to do this and that wasn't working any more because look at how many people are interested in this - in the Internet. So, ICANN was created to try and address some of these things. You can read this, but what is interesting at least from my perspective in ICANN is it gives an opportunity for collaborative structure for people to work together. Slightly different from some of the things you might have seen. I consider this a more accurate reflection of what the ICANN framework is all about. Surrounding this you can have the public in put in the public forum. What does ICANN do? Just a quick question how many of you here have heard of ICANN? Oh... this side. OK. I recognise some faces so I know you guys know ICANN because I've seen you guys. For those who don't, they coordinate largely these things and also the DNS root names of the system. They don't do - I think I've just lost the microphone. Sorry. (Pause) hello. OK. It doesn't do a lot of other things, now for those of you who are fro governments - I know some of you are, or from telecoms companies, ICANN doesn't do a lot of these things. So this is where the governments tend to have a role to play. Now, I can coordinate unique identifiers but they're not cyber governments. One of the things that I - I - (crackling on the microphone) Can you hear me? Sorry, I apologise for that. One of the things that people in the technical community may not always recall or remember is that non-technical policy issues, technical issues sometimes have non-technical policy attributes. I'll speak about this more later, but what I'm trying to say is that when you guys talk about policy you talk about technical policy, you talk about operational policy. When governments talk about policy they get very nervous. Because to the government, policy is a 15,000-foot level thing and to you guys it's an every day operational thing. What's the big deal? And this is what I mean about the blind man and the elephants. Now, all these identifications and some of the things you guys do are areas of interest of governments because of these kinds of implications. So public policy, technical policy, operational policy, that's my closest definition to separating these areas of interest. I was in several forums yesterday and the day before, I hear people talking about policy, I'm intrigued to learn more simply because I want to understand a little bit more about the technical policy and the operational policy so that maybe I can contribute in the public policy arena. WSIS is government's attempt to try and catch up with Internet development, essentially. It was something that the UN started, through ITU, and it was supposed to cooperate with interested organisations and partners. Now, however, I think Paul Wilson, who I regularly meet at the WSIS meetings will find that cooperation is somewhat lopsided at sometimes with private sector but they've - they've been given 5 minutes to speak out of the entire day or something at those meetings. That's the framework of governments, which makes a bit of a mockery of cooperation, I think personally, but then again, that's my view. The intention is that WSIS is to be a high level gathering. This is the blind man called the government on one side, trying to have its own perspective of how to deal with ICT, and when you're talking about ICT Internet is key in the middle. They want to bring heads of state, heads of government, they want to define the roles of participants. Many of you in your daily work don't even think about some of these things. But this is what governments talk about all the time. Control, management, and who's on top of the food chain. That's what this is about. So the goals are noble - economic, social development, digital divide. The challenge is that ICT telecoms - it's impacting how people do business. So now governments say "We have to have some rule", that's what the WSIS program is, the governments trying to coordinate between themselves as to what their role might be. For you guys who started these things, many, many years ago are saying, you could be saying one of two things - you could be saying "Finally the governments have caught up" or "Why are you guys even interested in this space because we've been doing it fine all these years?" Well, I don't have the answer to that, unfortunately. It's just that I think - I'll come to one slightly later where governments I think see this as a public resource, see IP addressing, see the Internet as a public resource and therefore they feel they have a role. The intent of the WSIS guiding principle is to benefit the information society and so on and so forth and you can read about this in the WSIS Web site, but the key thing, the key messages which I think have been lost in the debates in WSIS are these things - ICT as a tool for development, universal access to information, importance of ICTs to disadvantaged groups and the importance of linguistic and cultural diversity. Those of you who have been following WSIS see a lot of the debate centres only around two things - funding, Internet governance. I personally think that a lot of people are now missing the point because the point of WSIS was to find a way to use ICT to preserve these things and have these things. So, whilst the WSIS process is ongoing and continues to go on to 2005 with no-one knowing quite how much it's going to go away, it's going to be, and - there is a group of governments in the ICANN framework that tries to play a small role in providing public policy advice and that's the group called the GAC of which I'm the current chairman. What's unique about the GAC - governmental advisory committee, those of you who may not realise the government system might find what's the big deal, again. The big deal is this - for governments, they work in silos, you're from ministry of foreign affairs you stay there. You're from the ministry of science you stay there. You're from the ministry of telecoms you stay there. Very often if you go to an ITU meeting, with telecoms people you go to a UN meeting it's the foreign affairs people, you go to UNESCO it's the science people. Technically they're from the same governments but they don't always talk together very well. We don't decide at the GAC who comes to the meetings from governments. The government decides itself who it should send to the GAC. Because the only criteria is that we look at DNS and IP addressing public policy issues, i.e. Whatever is in the ICANN framework. I can share with you - there's a web site address of the GAC, if you look at the members, we have a couple of ambassadors there, which means they're from foreign affairs a couple of regulators there, a couple of telecoms people, a couple of science people, a couple of professors, so it really depends on where that country places that responsibility. We provide advice on public policy to ICANN. Some of the - I'll come to things like Whois, IPv6, this is the third blind man trying to do his contribution to make the picture whole. Our mission is we want to see every country represented in the GAC in the ICANN system. We want to be able to be the conduit to liaise with other groups in the ICANN framework. We are an advisory body, a forum for discussion, we work by email quite a lot. Again, those of you who deal with government people find that this is not quite common. Not many government people even bother reading their emails at times. I've heard comments from sending email to a government guy three weeks, no response. Well, we try not to, we try to work by email. It's open to all national governments, government organisations. Happy to say in this region we have quite a few. The government of Fiji has fairly recently joined. And a few other South Pacific governments are coming in as well. Because we work online you don't necessarily have to be physically in a physical meeting to make your contribution. We have more than 95 national governments, and the vice chairs from Sweden, Nigeria and Mexico. We hope to see more. These are the areas we currently are working on. Some of you are working on these things as well. In the ICANN scheme. I'll leave that for you to consider. We can have a discussion later on this, but these are the issues we concentrate on. TLDs, IDN, IPv6. We just heard Jordi talk about IPv6. What's IPv6 to governments? IPv6 is about resource allocation, unfortunately. IPv6 is about making sure there's a fair allocation and distribution because there is a perception that address allocation has not been fair, not structured, not coordinated. Remember the telecoms map I showed you, remember the Internet map I showed you after that. Many governments don't understand the way in which IP addressing works. Many governments have only a reference to telecoms, which means in telecoms every country has a country network number. In IP addressing that's not the case. Again, you know this but your governments may not. The key interest of governments right now - and this particular group is spear headed by the government of Japan - is to try and promote the use of IPv6 from a policy perspective to see what needs to be put in to swell businesses, to develop IPv6, which may include things like tax incentives, promotion schemes, what have you. Whois. Some of you I think are familiar with this. Anyone using the DNS to look at who is running a particular sight. Privacy. Some people are debating why should we allow those kinds of things - information to appear publicly because that puts me in danger or that puts my private information out. Before I go there - do all of you - I think most of you appreciate the choice whois as an issue, right? From your perspective, what's the big deal - again, what's the big deal about putting personal information, contact address on a web site? The whole reason why the technical community put Whois data available is so that if that particular site is downloaded there's a problem you can pick up the phone, call the guy, say, hey, check your servers because something's gone wrong. That was the intent, but the public policy guys are saying "Hang on, there's public policy issue here, it's an issue of privacy." IDNs. Big issue of governments in the WSIS process. That's simply because again, language issues are closely related to cultural issues which are both very sensitive issues for governments. So if you - up until fairly recently you could only access stuff on the Internet on the URL in English. You couldn't do it in any other character set. That was one of the reasons also why governments got involved to say, we need to push this because if not our people will all have to learn the alphabets A to Z in order to find something on the Internet. Again, that's where governments come in. Most of our contributions, because it's in the ICANN context, relates to some of these things, which are new TLDs, multilingual domains, and country name.info. Some of you will say, "What's the big deal about securing country name.info". Governments have been known to sue for the loss of their country name in dot-com. I won't name the names but there are some countries in this region and some in another region, some governments have gone occupy to sue country name dot-com because they say this is a governmental right. I personally don't think so, but that's me, but many governments actually do think so. The kind of request I get from governments on a daily basis is - "dear Mr Sharil, chairman, GAC, why are you allowing so and so domain operator to register my president's name in a dot-com domain? " I didn't even do that. But that's the kind of questions we get from the government site. It's very difficult to explain some colleagues in government that this is part of the whole thing called the Internet. It's an open process. And it's again, the blind man and the blind man's perspective of things. So what we did was we worked with affiliates to secure country name.info so that we would be able to at least offer governments something back. Don't go and sue on dot-com if you want, here's your name and.info, use it. Quite a few governments have done so. Spain have done very well, they've used it as a portal to their country. Fiji might want to do that. Sweden has done that. If you go to Sweden.info or in the case of Spain,espana.info you'll find there's a gateway to Spain. Many other countries have done that. On the last couple of points, this is my take on governments on the Internet. You have a house, I have this structure! In many cases people have been operating in silos, much like governments. Businesses are businesses, IETF, IAB, the engineering community there, then the intergovernmental agencies and civil society. Sit in a political silo structure. In my view, the ICANN structure - and this is based on the Malaysian regulatory framework which is splitting the acilities from the service from the applications to the content. I use this because it's a structure I'm familiar with. It's my perspective as one blind man. I think, if you remember the blind man and Elmer, I would categorise things in the Internet governance space, as this. As in infrastructure layer, key applications, web companies and all that. That's my thing. Some people would turn it on its head, some people will agree, disagree but this is a matrix where we can start the discussion. This is something I'm sharing with the WSIS people as well because right now the debate on Internet governance personally scares me because governments want to control everything. I'm trying to suggest to them, no, no, some things are working fine, leave it to the private sector to do. The things you need to do let's find a spot somewhere in that space. I think most of the people here work in that box, although I could be wrong, but with this box, or these boxes we have an opportunity to discuss and point as to where you think you are in this scheme of things in the space of Internet governance. My key wishes - I think the Internet belongs to everyone. And governments should include all parties and international multistakeholder partnership. It is both a resource and identifier as part of the infrastructure. From your perspective I think you guys see it as part of the underlying infrastructure but from the government perspective it's a resource. I think it should be inclusive, multilingual is a key thing I've always been in support of. Privacy is something we think about when you guys do your addressing. Education I think is key. And this is where if I can make a plea to all of you in this region, those of you who are present here - please help educate your governments. 'cause very often the technical community does not any way engage - I know APNIC doesn't, but many others, the technical community does not engage with governments now. APNIC has been doing a lot of program was APT and started doing programs with PITA, both governmental organisations and I find that's helped bring the different perspectives of the different blind men together. Some thought pieces for your consideration - information society, ICT, Internet, is all part of the knowledge economy. International governance of the Internet is one part of it. If you're talking to your government colleagues on this please advise them that many of the - there are many other parts. The one thing that's clear to me at least is all three blind men need to cooperate, coordinate and collaborate because if you don't each blind man's gonna go away and talk about his own thing in his own way and not realising he may not have the full perspectives or the picture. With that thank you very much ladies and gentlemen for your attention. (APPLAUSE) PAUL WILSON: Thanks. This is Sharil's new office number, if you want - office address if you want to visit or call him or look at his web site. I don't think Sharil would mind me relating a story that one of the ICANN board members traditionally tells at this time, Betty McKowski from Bulgaria who relates a story about the three blind elephants and the one man. The blind elephants are trying to work out what this man is like. They stomp around for a little while and feel, trying to work out what this is. They all decide after a little while that they all agree, he's sort of flat and squishy and sort of mushy. So no disagreement there at all. I think the point of that story is that there are a few blind elephants running around in this space and there could be a victim that doesn't quite look the same as it did at the start. That's sort of - sort of tongue in cheek. I think we're working a bit better now. The elephants maybe have dark glasses on these days but aren't necessarily completely blind. Without further ado the next presentation is Adiel Akplogan. There is a bit of formality going on here that is quite important in the history of APNIC. Let's welcome Adiel who's come from South Africa for the meeting. (APPLAUSE) ADIEL AKPLOGAN: Thank you, Paul. I'm really happy to be here today. And taking advantage of being in this special location, which is Fiji. To move one step ahead in our process as regional registry. And giving first of all, updating our activity progress and also use this opportunity to officially give our application. I will quickly go through the - this presentation. A few years ago the African Internet community decided to set up a regional registry to be set in one location and have a uniform policy applied to the continent. After a few months, years of discussion on what will be the best way of doing this, in 2003 it was decided to incorporate AfriNIC as a not-for-profit organisation in Mauritius, from where the company would be managed, and in addition to that location three other locations have been chosen to incorporate AfriNIC operation during the first two years. South Africa, equipment for technical replication centre, Ghana for training coordination. Since that year, 2003 then, AfriNIC has started working closely with all the local and international communities to set up a strong base for the new registry. For example, by participating in various international number resource organisations, like ASO where AfriNIC has been an observer since 2000 and also the NRO where AfriNIC is participating as observer. We also participate in various member meetings of all the area to explain, to give updates on our activity. And our project too. Moreover, we organised and participated in many local Outreach and member meetings in Africa, in - we organised a meeting with ISP from the region. In May this year we organised the first member meeting of AfriNIC which was in May 2004. Where the first set of policies was discussed and approved by the board. We attend ARIN and the RIPE NCC regional meeting organised in Africa to update our future member which now ARIN and RIPE NCC member on what we are doing. We also used this progress and this formal organisation to sign an MoU with the NRO to support, financially support AfriNIC startup. So now after all of that, we think it's time for AfriNIC as an organisation to move one step ahead by achieving a formal status as the regional Internet registry. That means the new organisation must comply with the criteria stated by IANA under ICANN, which is stated in the document called ICP-2. I will quickly go through all the points in the ICP document and give a brief overview of how we think we are complying with those criteria. The first criteria is the region of coverage. The region of coverage has been defined for AfriNIC corporation. It includes 54 countries, the continental and also Indian ocean islands. The information is available on the AfriNIC web site. 2: AfriNIC needs to demonstrate a strong support from the community. Since 2000 AfriNIC has received a lot of support from local ISP and international organisations. We received about 25 formal letters of support from different organisations in ISP in Africa. We also have today 13 registered members from 11 different countries and have about 68 organisations that sign, formally signed the startup charter of AfriNIC. You can visit our web site and look at all of those country and organisations too. Self governance. AfriNIC needs to show that it is structured to be self governed. For that, we sent up a policy, development process and also a working group to work on AfriNIC policy on the organisation, the way the organisation will work. This policy working group has developed a policy development process which has been passed through the community for discussion. And this first policy development process was used for our first policy adoption process, meaning the IPv4, IPv6, ISN number was going through this policy development process and submitted at our last AfriNIC development policy meeting held in Dakar in May this year. The process is also available on the AfriNIC web site. And a report from AfriNIC, one meeting is also available with all the documents. The fourth point in the ICP document is the impartiality. It is registered as not for profit organisation, a non-government organisation. It is located in Mauritius. The structure of the organisation is such that it's governed by its member which has control on its operation by defining policy and also approving activities and activity plan. AfriNIC has also an independent board of trustees composed of representatives from 6 sub-regions in Africa. The continent is subdivided in six different sub-regions, so the member of the board are elected based on this subdivision of the continent to make more people involved in the process and show that the way the organisation is working is impartial, no matter the region. Technical expertise. The team leading the AfriNIC project is composed of very experienced people that work and manage Internet services and are well armed to manage all required RIR services. Today AfriNIC has set up its own operational infrastructure in South Africa, web server, mail server, ticketing system for handling requests from Africa, all mailing lists and the DNS for domain numbers internally used by AfriNIC. The team of AfriNIC is composed as follows. We have the board, as I was saying it's composed of representatives from different regions in Africa. One as a primary and a second one as. The new ten isn't is from Benin. The new board member was elected this year. It's the first formal board elected by AfriNIC community. There is a CEO position that I'm heading now. We have a business assistant who joined the company one month ago in July. Which is Maria. We have also an engineer who is working as engineer and IP analyst. More to that we have a working group, dealing with specific area, like policy development, as I was saying previously, as meeting organisation, all the process of the recognition process. AfriNIC needs also to demonstrate it adheres to the global address policy, allocation policy. We fully adhere to the global address allocation policy, especially regarding address conservation, aggregation, and all of these principles are reflected in the IP resource policy that we adopted in Dakar in May 2004. Activity plan - AfriNIC is now in its transition process. It's main activity now is to follow up the transition and follow all the activity plans during this transition to make the transition from the other area to AfriNIC being smooth for LIR. In mean and long-term AfriNIC has planned many activities for information and training for the local community on IP resource management and request processes and all other issues linked to the Internet. AfriNIC is as an organisation, also has a mission statement. The goal of providing an interface environment for Internet development on the continent. Meaning to have ISP and organisations in Africa in their process to adopt Internet as a means of communication. Funding model. AfriNIC needs to demonstrate a very strong funding model showing that the organisation can step on its own feet. There is a funding model designed for AfriNIC and from that funding model the first two years of incubation, AfriNIC will be mostly formed by hosting country, who agree to host the different operations and contribute to make the company startup. But after two years of incubation AfriNIC will be 100% funded by membership fees and that's not for profit organisation. Our main goal is to ensure the sustainability of the operation. In our application we also attached a business plan that showed how we can achieve this goal of sustainability of the company. Record keeping - as a registry we are dealing with a lot of information, different kinds of information. We have to make sure that all of those information are kept in a secure place. We have developed a strong record keeping policy ensuring that all the physical and electronic documents are well saved, backed up and saved in a secure location. The public part of those data will be also available for the community on AfriNIC web site. And the last is the confidentiality. We need to make sure all the information we receive is kept confidential and inside the company. We have for that a non-disclosure agreement that will be signed with any external entity working in the company. Meaning employees, Internet company or any or company that will have access to information. AfriNIC also provide in registration service agreement a clause for non-disclosure of private data we will receive from ISPs during the request evaluation. Briefly, this is the key and main activity that we conduct and make - comply on some points on the ICP 2 criteria. So, what remains now? We have to move ahead. And now, I would like today, taking advantage of this APNIC 18th meeting, here to submit officially AfriNIC formal application to be recognised by ICANN as the fifth regional Internet registry to serve Africa's community and for that I would like if there is a representative from ICANN here, to come officially to receive our submission. (APPLAUSE) SPEAKER: Those of you who know me know that I've been in this field for a while. It's a great pleasure for me personally to be finally able to accept this from Adiel and the African people. Thanks to all the regional Internet registries that worked so hard with them to get them to this stage and hopefully we can move forward quickly with a registration. (APPLAUSE) ADIEL AKPLOGAN: Yes, it's an important moment for us as AfriNIC, a big step. We look forward to future cooperation with the Internet community as a whole to strengthen our organisation and to be fully integrated in the group of area. Thank you. Vinaka. (Pause) PAUL WILSON: I need to change hats for a second here. As you may know I'm the chair of the NRO executive committee, that's a rotating position which was assigned to APNIC in the first year of the NRO. I guess we won the lucky draw. In that capacity I've been asked by the executive committee to read a statement of support for AfriNIC, which is a sort of formal statement from the NRO representing the RIR. The statement is that the NRO was formed by the RIRs to formalise their cooperative efforts. One of the most important efforts undertaken by the NRO is to provide support for the emergence of AfriNIC. All of the RIRs have and will continue to provide support for AfriNIC. We are encouraged by the progress the submission of that application representation. We encourage provision - no later on the meeting in December 2004 in Cape Town South Africa. We pledge our continued support for AfriNIC and we look forward to welcoming AfriNIC into the community of regional Internet registries. Thank you very much and congratulations. (APPLAUSE) ADIEL AKPLOGAN: Thank you, every. PAUL WILSON: We have come to the end of the session. We haven't had time for the BGP movie so I'm afraid you'll need to wait until the opportunity to see that in the routing SIG. Thanks for that, Geoff, - Geoff, but you'll all be on the edge of your seats if you haven't seen that already. I would like to just open the opportunity for questions, which I promised at the beginning for any of the three presenters, so please if there are any questions for Jordi, Sharil or Adiel. SPEAKER: Thank you very much. Axel, the managing director of the RIPE NCC and secretary of a number of resource organisations. I have a question to chairman Tarmizi if I may. A question regarding the ICANN meeting in Kuala Lumpur which I was unable to attend, myself. I was surprised to see that contrary to my expectation the... wasn't signed there. From what I have gathered, the governmental advisory committee was "Not happy" with the MoU, I would see it also as a bit of an elephant and me as a techie one of the blind men walking around and whoever you talk to it's difficult to find a unified vision or opinion of the GAC. My question to you is can you present us with the unified opinion of the GAC concerning the proposed ASO MoU. Thank you. SHARIL TARMIZI: I think one of the things I know it probably won't be clear but I think we're having a discussion later this afternoon so maybe more elaboration can be given there. One of the key concerns I guess, is the process and procedures that was put in place. It didn't seem to give enough opportunity and comments from a government perspective as part of the - I think there were certain time limits which we didn't think we would be able to meet. For example, we weren't able to comment during the comment period and that's how governments work, whether efficiently or inefficiently. we very much support the signing of the document but we want to be able to provide some input to improve that, at least from this blind man's perspective. SPEAKER: Thank you. PAUL WILSON: We have heard directly from Paul Toomey that on behalf of the ICANN board that the decision not to sign or the non-decision to sign in KL should not be interpreted by the RIRs as a... unwillingness to sign at some point. They are very supportive of having the agreement, they have assured us, and there's some more discussions apparently to take place. That's from someone who was at KL. Any other questions? Randy. SPEAKER: on a philosophical note, and I think it's shown in WSIS when on your slide 20 it shows the goals as being access to the Internet, et cetera and instead the WSIS spends its time on Internet governance. ICANN and the formation of the NRO, et cetera, et cetera. What we have done and what has happened is the ICANN process took what was an arbitrary graph and started a formation of a hierarchy. This is understandable, hierarchies, the instinctive organisational form, but it has done two damaging things. The first is a hierarchy is less resilient and less rigid and less reliable than a network. As we can see that we use routing instead of circuits and so on and so forth. Secondly, it has fooled people into concentrating on climbing up the hierarchy and ITU and WSIS trying to climb on top of ICANN, et cetera, et cetera, and losing sight of the goal of getting packets to people. And, I support AfriNIC, I've worked on it for I don't know how many years and I think the NRO, et cetera, et cetera but we've lost context and we're making a power structure and we're encouraging a culture of power instead of a culture of facility and access. PAUL WILSON: Thanks, very much, Randy. SHARIL TARMIZI: Thank you, sir. I would like to say, I personally very much support what you're saying. In fact, sometimes I get very worried. In some of the discussions in WSIS, only governments are permitted to sit in there. I know one of the colleagues I have on the ICANN board sometimes sits on behalf of Senegalese government and it worries us very much being people who have some, a little bit more maybe knowledge than some of our other colleagues in government to hear the kinds of things they want to do. For example, there was one guy who said I want a root server. The rest of us who knew looked at him in horror and said why do you want one? "Oh because I want control". This is the kind of talk that goes on amongst governments which is very scary. I personally have been part of this process in ICANN and all sorts of other groups of the Internet community for the last four years, every day I learn a new thing. Somebody said to me this morning, Sharil, why do you bother coming from the - for the newcomer meeting for APNIC? For me it's the first time I'm here at an APNIC meeting thank you very much and I'm here also to learn, to understand a different perspective. Hopefully be, my perspective as one blind man will improve after talking to a few more other people and listening to their perspective. PAUL WILSON: Thank you very much, Sharil. I think we'll finish with questions now, we're running into the break. Just a few brief announcements - please have a look at the onsite notice board for the meeting. That is easily available off the APNIC general meeting web site of the APNIC website, and that gives you a lot of details about the schedule, about locations, about arrangements for various things such as lunch today, the social event. Other administrative details. If you haven't collected your meeting kit there is a coupon in your registration, what you received at registration, you can collect a meeting kit including a bula shirt. There is the APNIC help desk here which is available for anyone to come and approach and use. There's a MyAPNIC demo, et cetera there. This is the opening plenary of the meeting. It's now finishing. For the rest of today and tomorrow there will be special interest groups, including the policy SIG, the IPv6 SIG, both of which are related in various ways to what we've heard about today and at the routing SIG you'll see the BGP movie. Thanks again to the sponsors of the day, Softbank BB and InternetNZ for their support. Also the meeting hosts Connect and Telecom Fiji. I've just heard that Telecom Fiji has prepared a lucky draw for the morning tea break, so if you are interested in putting your business card in the box at the meeting kit collection desk, you will enter into the lucky draw and who knows what you'll find or what you may win. I don't know, but I'm sure it'll be good. Let's break for morning tea. We are scheduled to start again at 11 with the SIGs. I expect we'll run a little bit late but let's try and pick up a bit of time on that. Thank you very much once again to all the speakers and yourself for coming to the meeting. Thank you. (APPLAUSE) Time: 10.44am morning tea break