These are unedited transcripts and may contain errors.

Plenary session

Tuesday, 13 May, 2014, at 4 p.m.:

ROB BLOKZIJL: If people can find a seat, please, then we can start in a minute or two. Good afternoon. I welcome you to this special session on the occasion of the 25th anniversary of RIPE. I would like to ask people in the back of the room to find a seat, and then we can start this session.

25 years of RIPE. We have a special programme this afternoon, three speakers; the first one is Vint Cerf, who I am sure will need no further introduction to you. I will give an overview of 25 years of RIPE; and the last speaker is Geoff Huston, who also has some reminiscences of RIPE.

So without further ado, I would like to welcome Vint Cerf, and invite him to give his presentation.

VINT SERF: Thank you very much, from western Virginia where I am enjoying very pretty spring weather and I hope you are as well. This seems to be a year for anniversaries and celebrations. First let me wish you a happy 25th, it's a little hard to believe that two?and?a?half decades have gone by and I am sure those of you who were around when RIPE was first started must be wondering where those two?and?a?half decades went. I can sure that view with you because we have just celebrated the 40th anniversary of the publication of the first paper on TCP and the design of the Internet in May of 1974. We had a very nice picnic celebration and we hung a plaque in the hotel where we wrote the first paper that was published in May of 74 in the IEEE transactions and communications and I am sure awful you are aware that the world wild web is celebrating its 25th anniversary along with RIPE.

Is RIPE right by 25 years and I think the answer is yes but I think, also, RIPE continues to mature as the Internet itself continues to evolve.

I must say that I was glad to hear that Geoff Huston is with you. This is one man with a remarkable variety of data to tell us what has happened over the past 25 years and what is happening in the present. Which brings me, by the way, to a very important point which I know all of us share, in particular the run out of IPv4 address space. I personally am a little embarrassed that we didn't recognise in 1973 that we would need more than 4.3 billion terminations in order to make the Internet successful, but frankly, at the time it seemed like more than enough for an experiment, so we just had the problem of the experiment got loose into the commercial world, so now we need IPv6 and I know you are fully aware of that.

I have been encouraging anyone who will listen to please ask their Internet service providers whether and when they will be getting their IPv6 allocations. They are saying, some of the ISPs are saying well no one is asking for it so I would like to ask each of you go ask your ISP when am I going to get my IPv6 address, if I don't have it already.

The second point, of course, is that because the IPv4 address space is pretty much run out, certainly at IANA, and ICANN, by the way I saw lease on camera earlier. Now that we have run out basically of IPv4 as many of, you know, there is a lot of tension associated with the address space and people who need it wanting to buy it from people who don't need any more of it. So that is going to create complications for all of the Regional Internet Registries, RIPE included and I speak with a certain amount of awareness since I now serve as chairman of ARIN over here in North America.

So we have work to do to make sure that this transition period is reasonably stable, especially for the users of the Internet and for the providers, who offer access to it and those like Google and others who offer services through it. So the v4 transition to period is really tricky and something that we all need to be thoughtfully and to work collaboratively to make sure this transition takes place in a reasonable way.

One thing that I think we also were going to have to pay a good deal of attention to is the proposal coming from the national telecommunications and information agency, to end its specific contract with ICANN, with regard to the IANA functions and I am sure every person in the room is well aware of that, and well aware of the many meetings that will be undertaken to in the course of 2014. The most recent of which is NETMondial meeting in San Paulo which produced a statement concerning principles and roadmap and I think tell you as the vice chairman of the high level panel, which is instituted by ICANN, which just met recently in Estonia, that this group also was ?? in Dubai, I am sorry, that this group also very strongly endorsed NETMondial principles but we have more meetings to come as you know. The Internet governance forum will come up in Istanbul and there will be the ITU and my guess is that these activities and discussions will continue on into 2015. As you all know, the official current contract expires at the end of September 2015, and ICANN has been charged by NTIA to come up with a multi?stakeholder process who are replacing the oversight role which NTIA has undertaken for the last 15 years and here I think it's absolutely vital that all of the Internet?related organisations, all the RIRs, ICANN, ISOC and so on work very closely together to fashion this multi?stakeholder process so that in the end we have a multi?stakeholder and not a purely multi?lateral or inter?governmental oversight for the Internet, which leads me to the next point, which is that security and safety and openness on the Internet must be considered at the top of the list of priorities.

I am a huge advocate of openness and freedom of expression on the network and innovation and all the other things it's capable of providing but at the same time it's vital that we provide people with safety, freedom from harm as well as freedom to express and that requires security and safety measures, some of them technical and some of them already certainly legal, without doubt require cooperation among governments as well as the other interested parties, those of us here in the room and elsewhere around the world who care about the Internet and the way in which it is governed.

I don't want to go on and on but I want to thank each and every one of you and the founders of RIPE for their initiative so many years ago, and you really brought the Internet to Europe and at a time when there was a great deal of controversy and a great deal of pressure to adopt a very different protocol suite and so the people who braved that conflict and insisted on bringing IT into the European context deserve an enormous amount of credit not only for that early initiative but for the way you have supported the system for the last 25 years. I am sure that you will do the same for the next 25 as the Internet continues to grow. So again, happy anniversary, thank you for letting me join you this afternoon and I return the microphone now to the chairman. Thank you.

ROB BLOKZIJL: Thank you, Vint, for your nice and kind words. Right. Next on the agenda is some thoughts I developed around 25 years of RIPE.

We have just seen one end of how far the technology has brought us. Vint sitting quietly in sunny western Virginia talking to us. I see here in overhead projector, an essential ?? the essential tool for at least the first 25 or 30 RIPE meetings. I bet there are people in the room who have never seen such a contraption in operation, but I have been told that we will give a demonstration of this technology.

But all this stuff is even too modern for me, so I made notes on this.

For those of you who were not born when these things were in use, this is called a punch card, an essential tool for statistics and computing for, I think about 100 years, and these are special ones because on the front it says "CDC" a world famous computer manufacturer who doesn't exist any more, but was extremely popular in scientific world, and then it says "CDC6400" that was one of their models, and was one of the first computers I encountered as a physicist. It says "scope binary". Scope was a very low level part of the operating system and you had to programme it in binary code via punch cards and then it says "Ziemann Laboratory Amsterdam" that was the institute for nuclear physics where I studied and later worked.

That contained my notes. All technology is being used today.

RIPE 25 years, you may have found already a tool to remind and help you and if you missed it, I am being told that after this session there will be more of this outside. 25 years. We started in May 1989 with 14 people and a meeting which lasted half a day. Because we thought that would be sufficient to sort out some problems. Networking in Europe 25 years ago, some people described it as chaotic, that is one way; other people were more polite or diplomatic and says there was a rich choice. We had ?? just on top of my head I am absolutely sure it's not a full list but we had things like DECNET. No? And that came in two flavours, phase 4 and phase 5. And DECNET had an address field of 16 bits. And that should be enough to have a global network. Well, we did it ? well, we being the physics community in ?? mainly in Europe, North America and bits and pieces of Asia, and the space research and exploration community, ESA, NASA and the likes, among ourselves we managed one DECNET based on 16?bit addresses, that was quite a challenge.

Then we had things like SNA coming from IBM, RJE coming from IBM. We had things like UUCP, which was popular and coming out of the UNIX world. I am glad to see I am not the only oldie one here. Then we had newfangled things like X 400, a great tool to lose your mail. Of course, that was running over a fascinating technology called X25, yes. And then there was this newfangled thing called TCP IP.

So, what to do with all that. Well, one thing was clear: Was a rich choice far too rich and we should try to put our minds together and come preferably with one ultimate network technology. That is what everybody agreed and that was the last bit before the split in the networking world. And there are two banners here which one could march behind: One was TCP IP; and the other went under the name Open System Interconnect, OSI, a set of formally?developed protocols by basically the ?? under the umbrella of the International Standardisation Organisation, so TCP IP and ISO or OSI.

What was the problem? Well, OSI was good, especially in Europe, because it was invented by European PDPs. It had a few main characteristics: It was very expensive, and it didn't work. But apart from that, it had the full support of national governments, not surprisingly because in the late '80s many of the national governments were the happy owner of the national PTT. So this was good, it was European and it was coming out of a respectable organisation and it had the full support of everybody, and then this TCP IP thing, that was bad in Europe, in Europe.

First place, it was not invented by the PTT, so how could it ever work? It was for free, not expensive, not cheap; it was just for free. And the mid?'80s saw the advent of scientific workstations like Sun Microsystems, Apollo and a whole load of different makes, and they all came with UNIX and UNIX came whether you liked it or not, with a working implementation of a full set TCP IP and services like mail and transfer and remote login and a whole load of other things and all for free. And it worked. So people started using it. That is no good.

So, the war was declared and I am sure I see faces in the room of people who were present in these dark days of protocol wars. If you think current discussions about Internet governance are convoluted, complicated and boring, go to your local history department in the computer science bit of your university and if you are lucky you can find tens of metres of documents of all sorts of organisations who were involved in developing the next generation of networking, i.e. nice documentation of these protocol wars. And TCP IP people, people who were playing with TCP IP, usually computer scientists, physicists, people who really needed networking in the very early days of networking, and they were considered to be some sort of guerillas. At one meeting in ?? of the European Community, whatever they were called in the late '80s or early '90s, people like Daniel Karrenberg and me were characterised as the Bondite and Bond aus Amsterdam, which we took as a compliment.

So, some of us thought action was needed, for two reasons: In the first place, together you stand stronger in these wars than being fragmented over individual small groups, deep inside universities; and secondly, we were all happily playing mainly inside the university or a campus with TCP IP on a local network but then you have a friend who is doing the same thing at another university, and you say, mm?hmm, let me find a way to connect these two networks. Well, with some filling with the analog phone system and right kind of modems you could get along way and you find out, oops, suddenly you have to start thinking about routing, addressing, because if you use the same set of addresses, things don't work any more.

So we need to coordinate our he was, European?wide. Well that resulted in 14 people sitting around the table for half a day, because we thought that is enough, and we identified some issues which we would resolve when we went back home. And then a couple of months later we thought, mm?hmm, maybe one meeting was not enough; let's have another meeting. And then, mm?hmm, maybe we had better write down what we are doing, because otherwise we might forget it. And then every body in networking was hiding behind fanciful names of fanciful organisations, so we need a name. And this is where the mystery starts but a name came up and one theory is Daniel went to gave presentation in Paris and, hence, he came back with a name for this group of people, that happened to be French, which we thought that distinguishes from the rest who all speak some flavour of American. And then, oh, we need a chairman, and I must have not paid full attention because that is what I ended up with, and that was 25 years ago. But that will last until the end of this week and then we will have a new chairman, as announced at the last RIPE meeting.

So, that is how, basically, we started. Very lightweight, very bottom up. You organised the things that really need organising, and you don't organise because you want to organise things.

So, from the start of RIPE, coordination is the key word that describes our activities, 25 years ago, and today. 25 years ago we were not in the numbers game, you needed numbers, you wrote an e?mail to John post he will. If you used the right words and phrases you got numbers, and you could get a few or if you wanted more, you got a lot.

But we were concerned with routing. Routing in those days was mainly a manual affair. Handy tools like BGP had not been invented yet. So, we had the Working Group in RIPE in the early days that has its charter to make the complete map of all the interconnections in Europe which basically meant all the routers in Europe and their interconnections and that nicely fitted on one page of A4. And I remember that we have spent many, many hours studying such a map and decide how we could improve routing across Europe. And then people scribbled their notes, went home and typed things into their routers and most of the times, most of that worked.

So we had all these routers scattered around Europe and we had people scattered around Europe so let's put a simple database together so that you can easily find people and routers, so that is the start of a large RIPE database today.

And last but certainly not least, a changing knowledge and experience has always been, from day one, a very important part of this RIPE community.

So ?? and then everybody lived happily ever after. No, not really. If we jump to what RIPE is today, in a sense, nothing has changed because we still talk about routing, we still talk about DNS operations, we still talk about what a database should look like and how to use it, and we still exchange knowledge and experience. We do a few things more that came up over this 25 years. So, in a sense, nothing has changed, but everything is different. Routing, today, is in no way, can in no way be compared as how we did it studying, staring at one page of A4 with a view boxes and lines on it. Another major thing that has changed is in 1992, we concluded that there was too much day?to?day administrative and related work for a bunch of volunteers to commit themselves in doing it, so the idea was born to have a professional secretariat, professional in the way that a handful of people would do this work as a job on a full?time basis. And that is how the RIPE NCC was born, as the secretariat, for the RIPE community, and that is what it still is today, basically.

But a few things have changed there, as well. I don't remember exactly when the RIPE NCC started, how many members did we have. I think in the order of 30 or 35. Today, we have passed the 10,000 members. Well, you can say it's just a number. That is true. But there is a fundamental difference in having a membership association with 35 members where everybody knows everybody else quite intimately and having 10,000 members, so new challenges are there so the fundamentals have not changed.

The future of RIPE: Well, I don't know. If you look back at the last 25 years, nobody 25 years ago could have forecasted where we are today or where we would be ten years ago or even 15 years ago so I think one thing is evident: This community develops by evolution and not by revolution. So, we take things with little steps. But the core values still hold and that is coordination and cooperation in the field of operation of IP networks, not just distribution of IPv4 addresses. And basically, as Vint Cerf also said, most of the number games are over. IPv4 is almost over. There is a little bit left but the great bulk of this 4.2 billion addresses are out there in the field.

IPv6: Well, luckily, we have learned from IPv4 and so IPv6 has been defined with one characteristic which is very important: we should not create another scarce resource. So there is enough of IPv6. Well, there is never enough, some people say, but there is enough for our lifetime, once you get your first IPv6 allocation, it's probably also your last one because what you get is enormous. So, all the effort that in the last 10 or 15 years has been spent by this community, driven partly by the scarcity of IPv4 address space, the effort into developing fair address distribution rules or policies, as we call it, is no longer needed. So we can go back and put more of our resources and efforts in the other things that makes up RIPE. And I think we have seen in the last two or three years that the other parts of the work of this community has gained in weight and importance.

We have a very well organised and this, as a result, high level presentations in the first two days of the RIPE meeting, so that is the education, knowledge exchange, experience?sharing part; and we have a group of well functioning Working Groups that cover a wide range of subjects and and that is the coordination part in action. So I think that the infrastructure is in place to have a healthy evolution and there is then, there is no need for a revolution. And there is still enough to do and what I have observed in the last few RIPE meetings is that the people who organise the plenary part of the programme, the Programme Committee, and the Working Group Chairs, have one complaint in common: I don't have enough time, I have too many interesting, high quality subjects that we have to deal with. Well, I think that is a problem which I love to see, because it's a sort of a luxury problem. The alternative would be very worrisome: How do we fill the time and it's all so boring. We are not there, luckily. No, we are certainly moving in the other direction, and as long as people keep complaining about not enough time and too many interesting presentations or subjects to discuss or work on, I think we are in a healthy situation.

A few loose remarks. Somewhere in the beginning I said we didn't do the numbers game because if you needed a number, there was only one place in the world where you could get your numbers, and that was John Postel. One of the hottest items that has been discussed on our mailing lists in the last six weeks or so, was the problem of the legacy space. Well, I never saw that as a problem, and I want to remind you that the numbers that John Postel handed out to Europe in the first three years, that is what we call the legacy space. So these are the people who were at the start of building the Internet in Europe and that sometimes seems to be lost, this knowledge or recognition in the discussions.

But I am sure we get out and over this non?existing problem. Which brings me to another observation:

We have a whole library of address policy, mainly in IPv4 address policies, which are all totally useless today. They are no longer important, they can go to the history department because with one or two exceptions, they were all designed as tools for a fair distribution of scarce IPv4 space. Well that has been extremely successful, we have distributed all, so these are policies are no longer needed. And out there in the wild we see IPv4 addresses and there is no collar bit saying that this packet or this address is a green one or a yellow one or a red one; they are all equal and they should be treated as such. So, where we still have left overs of our policy rules which hit us today if we try to remember what the fine distinction between PI, PA spaces and what you are supposed or allowed or forbidden to do with the one or the other in certain circumstances, I think we should get out of that. There are IP addresses floating over a network and we ?? it's our job as operators to see that these IP packets reach their destination in an efficient way and there are no collar bits hidden in the headers, so it's all artificial. Private observation.

Another one: Before the coffee/tea/Champagne break, we were talking about multi?stakeholderism. I love that word. When I still was a physicist and had to do complicated calculations or develop algebraic equations, there is a very popular scientific package which some of you might know and it's the wolfam alpha set of tools. And that do also fun things. One of the fun things they have is you can insert a word and I have algorithm and I say a word is a bunch of letters, how many English ?? true English words, according to the English Oxford Dictionary, can you make out of the letters of this word? Well, multi?stakeholderism is the right word for this Internet governance discussion because it is incredibly rich; you can make 4736 real English words out of that one word.

I think those of us who are sent to take part in these fora where all this is being discussed, should remember this and bring it up at the appropriate moment and see whether people have a sense of humour or not. Those who don't laugh, don't trust them.

So, I think this is all I wanted to say. I didn't want to go into too deep details, as I mentioned one of the first decisions we made 25 years ago was we should document what we are doing. So you can go to the RIPE documents tool and you can find the extensive minutes of the first, I don't remember, 40 meetings or so, and after that you can find the webcasting of all the meetings. So, history has been preserved and in our true tradition, open and transparent, it's easy to find and easy to play with it if you are so inclined.

So, I want to say, those of you who have saved a sip, congratulations to us all, and as Vint said, up to the next 25.

Next on the agenda is a live demonstration of this technology, and I ask Geoff ?? oh, I see somebody at the microphone.

AUDIENCE SPEAKER: I have been just pushed forward because there is a challenge for one of the more traditional sessions at RIPE. The longest pole constructed by only an grams of multi?stakeholders wins a prize.

ROB BLOKZIJL: If you go to the website and you find this tool it not only gives the number, you can also print the list of all the words, so that might help you to construct this poem. If you can make a Limerick out of it you get...

Geoff, you are next.

GEOFF HUSTON: Thanks, Rob. Thank you so much for inviting me to be part of your celebrations. I must admit I remember a cc IR N meeting in 1990, it was actually in Paris, and there were these two rather strange people from Amsterdam who were bashing the table saying, we want to do the addresses ourselves in Europe. And I remember scratching my head, thinking that sounds like a really good idea. And it just went on from there.

But what I want to do is look back and give you a taste of what life was like around 25 years ago. And certainly, it's the case that, you know, the past is very, very strange. They do things remarkably differently. And we live I think in an amazingly special point where we have taken what we thought was pretty ordinary stuff and turned it into brilliance. You know, this all came back from trying to automate mathematics, 160 years ago it was the lodge rim table and the first of these engines was built out of paper. And it got from there to Charles Babbage in 1837 wonderful analytical engine, a real computer. He never built it, they built it later but that is what it looked like. It's just wonderful stuff. Very real, very physical. And then you get to this ?? this is a reconstriction of the Colossis computer built in '43 and it's all valves, remember them. They couldn't cut it, when you try to make things big your problem was power and heat and that is power and heat at a massive scale. Ian yak in 46 was a remarkable problem of how much heat and copper you could what can into a building. Some bright spark said I can do it all on silicone and so they did. The revolution was amazing and it got to this: You had to wear a suit and it had to be checked. And he is wearing a tie because he is in front of a computer and formally addressing it. But the issue about this it was business all of a sudden, that it turned from the military into business and that was the really big evolution that happened across that ten?year period. But computers were still very, very special. But then something else happened around ten years later, that just ripped through this industry and rewrote it again completely. Ethernet was amazing. All of a sudden, this stuff was everywhere and it wasn't point?to?point or hierarchies, it was stuff that threaded through every office all of the time. This wonderful broadcast network that meant that could I talk to you and he could talk to she and everyone could talk to each other independently down the same piece of common high speed channel. Don't forget we were dealing with modems that run at 200 bits per second and the ethernet came out at 10 megabits. This was rocket science on steroids and so it matched the computing at the time. And, you know, you never came realised it until afterwards but the best things were always built at the end. And I always thought the CRAY was where main frames finished. This was the end, the water cooled monster built like that because it took so long to propagate the electrons, they built it into a circle to make it faster. The engineering was truly stunning, wonderful stuff. The very same year in California out came this, which was regarded as a hobbiest piece of shit and this was would cost you about a million dollars dare say if you tried to get it on ebay because the original Apple was indeed truly phenomenoninal, it wasn't business any more, it's just you and I ?? consumerism. And over there in mainframe land it was all becoming again commodity. If you were alive at that time in the late '80s you always had learned Vax VMS, you had to. UNIX might be around but real people knew Vax VMS.

Yeah right. But at the same time, computing turned into fashion. The other thing that the MAC did, is no more computing language or JCL or crap. You just used your mouse and wiggled it and things happened. That whole par dime of turning all that computer power to do what my brain does naturally was amazing. It just transformed computers into something that was almost invisible, it was just me being creative and you being creative. What a thought. What a way to think about technology. It was about humans and people, not about the technology. What an evolution.

And these guys invented plug and play. Long before the rest of us tried to figure it out. You know this was just plug the MAC in and it just worked. We are still trying to figure out to do that with v6.

And there was one more thing that happened just before '89: Government said this is cool. Because that high performance computing initiative that happened in the US wasn't, if you will, something saying we want; it was actually government saying we are, we will, we are going there. And this was TCP IP, the NS F net, mainstream. All of a sudden, by then, there was no more OSI. It was over. We were mainstream even then.

But let's actually look at 1989. I have something that will require a little bit of booting. Oh, my God. Just before I go there, I do have a couple of photos here if I could just have it back. Someone remembers this. This was in a railway station in Japan around that time. At the time Peter Loftberg was employed as a luggage porter. Someone had to do it. Daniel Karrenberg was sporting a highly fashionable beard and the father of the DNS was there, wearing high fashion.


GEOFF HUSTON: It was very hot. What were we talking about then. I will look over, now I will flick over to real hardware. This is typical with slide wear isn't it. So I need to go back through the dim, dark tunnel of time and start looking at what we were actually talking about. Those people at the back, you are stuffed. And if this is out of focus, that is true, too. Now making the slides flash fast is always difficult. Dropping them on the floor is always necessary so let's just do that.

The other thing that is kind of necessary, too, is the gradual reveal. So, as you might recall ??

You had to start slowly. The anticipation is killing you, I know. And all of a sudden, you know, it's fully revealed, wasn't that a remarkable point, anti?climb axe. What were we talking about? Oddly enough, we were talking about OSI. We spent all of this time talking about OSI. So, let's talk about OSI, you are right. It was an open access something or other. Yeah, the dream was weird. It was all about the transition from vendor specific to open and end industry. The things that we take for granted today, that open technologies win and that proprietary technologies are a pain in the butt and if you can't see the source code don't use it. All of our world today is built around open technologies, you none of the world then was built around T and the issue was what do you after do after that? Vendor lock?in was not what we wanted and the industry approach was all about trying to create this top down. But, you know, the dream was just bizarre. The dream to the nightmare was that the folk you were trying to make this happen were telcos. How long does it take telcos to get something? To truly get it? And I can tell you what the answer is: More than 25 years. I don't know how much longer but it is going to be a lot longer than where we are.

Because they never understand it, they always thought they were in control. They need to own it. They need to hermetically seal the devices you use and not let you tamper with them. This is a network device, it's not my computer. And that is the attitude you find from telcos. So, you know, we were doomed. But, you know, part of it was, however, that there was a chink in the armour. The technology was crap. This is X 400 and X 500 that worked together to bring you the addresses from hell. And of course the whole thing about this was, that when you really looked hard in this, you really didn't understand if it was malice or incompetence. It sure seemed like malice, but by God it was weird. Because what we thought at the time or at least what those folk thought at the time, was that technology was something you could vote on, that you'd take an idea, put it through some wise folk of standards, jiggle around it a bit to make sure nobody had an advantage and it was equally bad for everyone. Do you remember heart beats in Ethernet? Why do we have them, no one remembers. It seemed like a good idea to level the field at the time. And that was going on there that, quite frankly, the good ideas, it was just impotent because their idea was no one can actually have a monopoly on the market, what we standardise must be equally inappropriate for everybody.

So, you know, OSI was this wonderful dream of open technology, it's just that the Internet seemed to be living it. And the standards flowchart, by God this is what Ivor heads were like, this flurry of slides. Someone will clean this up, won't they? There is a flowchart about what we thought we were doing on the Internet. For a while the IETF really did this, for a while it was possible to get an RFC published in under a month. That was a long time ago. But the whole idea was, you had a good idea, you went to a few IETF meetings and it got published. That was just so brilliant. But what we really thought about this and I think the real thing was the standards were were free, the standards were free. Anyone could find them. They didn't have to pay for them, you could build whatever you wanted from it and it kind of worked because it was all about having good ideas, not about adhering to a particular theology. It wasn't about religious correctness, it wasn't about all of us doing the same thing at the same time; it was actually a debate about what worked better. And that was the difference. That it was actually, if you will, a discussion about what was really the right thing to do. And that worked. That worked. Because we started dreaming dreams then about the world. And for a while, that was just revolutionary, that this experiment was going to be the global communications enterprise. So, where does that leave us? Sometimes when you turn these off by the way the globe just blew.

Blah?blah?blah, if you missed the overheads too, you missed this because it was kind of different. Where I wanted to end up with was here, because like on that train station in Tokyo, we were a lot younger then. And we thought that this was a fight we needed yet to have. That the Telcos still dominated the world, that the Internet was still just just an experiment and the success of the Internet was not assured. We always thought ourselves as the under dog, that there were these nebulous others that we were constantly struggling against and say we are right and you are wrong. But, you know, things change. We won. By God, we won. We won so much. Even by '89 we won't those protocol wars. It was over. What about voice and telephony? It's over. If you have an SS7 switch, sell it now, ebay will give you ten bucks, no one else wants this shit. It's all about SIP isn't it? Voice is just another app, that is all it is. Television, it's over. Between Netflix and bit torrent, something is replacing it, but it's no longer television any more, is it? What is 80% of the Internet's content? Video streaming. We have won. Everything we do, we have won. It's everywhere. I love this set of photos, it's taken in Rome, the first one is in 2005 with a whole bunch of people looking forwards, right? And it's backs of heads, this is cool. Same thing in 2013 and what do you see? A sea of iPhones, iPads, Androids, whatever else, everyone has one in their pockets. Bring out your cameras now, bring out your devices, they are everywhere. And, you know, the Google glasses look, you are going to have to get used to because everyone is going to be doing it. Because it's just this constant presence now that they are literally everywhere, you know it's no longer looking at laptops, this is device, Candy Crush earned a billion dollars, it may well have been taken at a RIPE meeting, I found it on flicker. This constant obsession is all of the time, everywhere. So these days, you kind of look at this and go, who is the other, who is the them? Is this a fight with the ITU?T any more? Is this a fight about Internet governance any more? Is there an other that we have to worry about any more? Or is it us? Because I strongly think that even then, when we were fighting the OSI wars, it was over. It was us. Today, it's us. It's all over. Whatever happens next, is us. We are the problem. And the issue is, the dream is not always that good, because what we are doing to ourselves is bizarre. We are the product. Examining every little thing we do is now what we do out there on the Internet, and that is something that we really need to worry about. And that is the core, I think, of a lot of these issues coming up over the next 25 years, on how to be responsible with technology, how to give it creativity, enjoyment, to make lives better and not just make it an even more powerful microscope on every last thing we do.

But I think we are up to it. I think we can do it. We have gone so far to computers that now have just two buttons, of which only one works, to technology that is now so natural that we are going to forget about how it works and think about what we can do with it and that, to me, is just so liberating, so releasing in terms of it's us now, it's you and I and what we can do with this, that makes they think the next 25 years are going to be truly phenomenoninal. Thank you very much.


ROB BLOKZIJL: Right, then. This, I think, concludes our festivities for today, 25 years of RIPE. I think those of you who were not there at the beginning, got a good overview of how we came to be what we are here, how our environment has changed over the 25 years and thanks to Vint, you could also see that people can grow old on the Internet, sorry, Vint. But people from the very first days are still, some of them, still extremely active, which makes this such a rich environment. So without further ado, I think I close this session. There are drinks outside. There is a BoF later, but you can find that in your programme, and tonight there is a social and there are buses organised and all the details you can find somewhere in this marvellous badge book.

So without further ado, thank you for participating in this event. Have a joyful evening and see you tomorrow at 9 o'clock.