SPEAKER: Rob Blokzjil: Good afternoon, everybody. I think most of us are here now, if you can find your seat or a seat, then we can start the last session today.
Before we start the last session, I have two announcements. One is that some of you might have expected the address Working Group to have an extra slot here, that is tomorrow morning at 11. So we do not discuss Address Policy this afternoon.
Secondly, you all have an invitation for the RIPE dinner tonight and it carefully explains where it is. It explains that it is today but it does not say when it starts, well it starts at 7:30 tonight. That is the arrival time you should be aiming for.
20 years of RIPE. So we got together with a few people and thought we can not let that pass unnoticed. So what we have done is we invited a couple of special guests and it's always difficult to pick special guests because there is a rich choice. So we thought the people who were involved in the very first RIPE meeting 20 years ago, we will invite those, we will invite them. And we invited a few more people who were very, very instrumental in the first couple of years of RIPE to get this whole effort on the road. And I have a presentation somewhere here.
In the old days we did this with equipment like you see here in front, and I managed to master that, and with this modern stuff sometimes a little bit of help is appreciated.
Thank you.
So the theme is RIPE 20 years young, because I had the choice between young and old and old seems like the end of a lifetime, of a lifecycle, and I after this almost one week now of a RIPE meeting, we are definitely not at the end of our discussions, so that's why I chose 20 years young because we still have a lot to do.
History of RIPE, this is one of the slides I used in my Monday morning newcomer presentation. It started in May 1989 with a meeting where 14 people showed up. It was half a day meeting, and this was a meeting of people who in their own environment, most of it research environment, universities, had local networks which were using the TCP/IP protocol stack and who started to make connections and found out you better start thinking about a little bit of coordination like routing or maybe it's not a good idea to offend your own address space, but try to use coordinated address space. So let's get together for half a day and sort this out and then we're done. Well, I think that was very optimistic, if I look at this crowded room 20 years later. And a little bit later I will highlight some of the issues which we identified 20 years ago and which we have been discussing this week.
We have made progress, but some issues are still there.
I want to spend not too much walking through 20 years of RIPE, RIPE meetings and RIPE documents. It's all on the RIPE website. I want to go a little bit more in detail in describing the networking environment in Europe in the late 80s, because when I went through some of the old documents, I was surprised to see that some remarks made 20 years ago are still extremely valid today, and I will show you examples of this.
So what was the environment? To summarize it in a few lines, some people have called this the period of the protocol wars. OSI was good, open systems interconnect, invented by mainly European phone companies which in those days all preferably owned by national governments and TCP/IP was considered bad because it was not invented by these phone companies but came out of the US research environment. It had two aspects which were highly suspicious in the first place: It worked; in the second place, it came for free, whereas the OSI stuff was famous for the fact it didn't work and was very expensive.
So 14 people came together and I think I'm very pleased to give you a list of ‑‑ a subset of those people because they're all here, all sitting here in the front row and later I'll introduce them more properly at the end of this session. Francois Fluckiger, Arnold Nipper, Ruediger Volk, Enzo Valente, Antonio Blasco Benito, Daniel and myself were among those 14 who were trying to fix the internet in an afternoon in May 1989. So I think it's appropriate to give them a round of applause.
(Applause.) There were many networking organisations in those days, though most of the networking was still in ‑‑ done in the academic and research communities. One important organisation there was an organisation that doesn't exist anymore, changed it's name to another one, and that was RARE. RARE is important here, especially Kees Neggers, because they provided umbrella, protection against these activities which were started and considered to be very suspicious by many, many other organisationsment so in order to give us some space to grow, to exist, RARE played, if we now look back, an extremely important role, and I think we can all agree, if you remember those days, that specially Kees Neggers who was on the executive board, that's what it was called those days, has been very actively providing the environment where we were not too much hindered by all this European politics. Most of you know him because later he was for many, many years the chairman of the executive board of the RIPE NCC till May last year, I think. But he's here in this capacity as one of the old boys who made this all possible. And Kees Neggers thank you for that role as well.
(Applause.) So this first meeting we had was quickly followed by a second and a third and a fourth and I can go on like that. But if you meet you need room to meet and if you meet for more than a few hours, do you want coffee, tea, refreshments, something to eat? I think we're all grateful to NIKHEF, they have hosted 19 RIPE meetings in those early days. Hosted in those days meant and we feed you and you get coffee and tea and there's your meeting room and a copy machine and you can use our computers. I'm not quite sure whether this was a decision by the director to do this, but they never complained when I informed them after a meeting that there would be a bill coming from the central canteen and things like that and they always signed for that. Those directors are not here today but one of the current directors is, Arjen van Rijn and through the whole community we thank Arjen van Rijn. Thank you.
(Applause.)
And last but not least, TERENA, ‑‑ this was the new organisation from RARE and that still exists in the lifetime of RIPE, specially in the creation of the first years of operation of the RIPE NCC. TERENA has played a commendable role as has supported this until the day when that child was five times bigger than the parent. And RIPE NCC became really independent. But I think Karel Vietsch, the secretary general, I think, of TERENA, he's not here this afternoon, but will be at the dinner tonight, and through him we thank this organisation as well.
(Applause.)
Okay, back to networking, OSI. You might think, oh, some invention, lots of nice and expensive books but it really worked and these are the first three verses of a long, long song composed by Vint Cerf and this is one of the nicest illustrations of the mood at the times in the 80s. This is '85, so this is a couple of years before we started thinking about OSI.
This OSI thing was, of course, very much promoted by governments and by the European Union which went by another name in those years, same people, same organisation. And some of them, though they were paid to promote this OSI and fight against TCP/IP got worried already in quite early days. Horst Huenke was a high level official in those days and DGXIII was under a different name but he was director of general Ford Telecommunications. This was European Commission which had a different name ‑‑ was getting worried at an early stage. Well, you can read it here. But the interesting thing is you can replace the name and the organisation in the title and you can replace OSI with things like ‑‑ pick your choice, DNSSEC ‑‑ IPv6, and soon in this theatre digital certificates. I find some of these statements, if you generalize them or translate them to 20 years later, I think still extremely valid.
This was a fight not only in smoke‑filled meeting rooms ‑‑ in those days everybody smoked, specially in meeting rooms ‑‑ but it also made the news from time to time. X.25 was the network of this whole OSI model, and this is how the press summarized it, which I think was quite amusing as well.
Francois Fluckiger was not only known as a key figure in European networking in those days, he was involved in the cert team, who at one point in time in the '80s, probably managed the largest concentration of international networks in Europe, everybody wanted to connect to search because everybody else was connected to CEC. Francois Fluckiger has been very active and sometimes his sense of humor, he could not contain it any longer, and this is the OSI, X.25 observation or whatever. And I don't know why the ‑‑ how good it is readable at the back but the conversation goes like this: And you, how many packet on your net? Six last week. Well lucky you are. Yes, that was my test program.
Now, replace X.25 here by IPv6 and suddenly it becomes very familiar. We have seen many examples of that in presentations. Thank you, Francois Fluckiger.
Applause.)
IPv6 has made it in some other parts of the world as we can see here. Dental chewing gum, IPv6 ready, wow. IPv4 never made it that far. And you get a two‑pack. Two packet is.
Horst Huenke again, two years later, 1991, at a conference of the annual networking conference of this organisation, RARE, which was still very, very much officially involved in implementing this OSI networking, which did not exist, but some people thought that is not a good enough reason not to do it because politically it was correct but I think he said if we ignore RIPE we ignore the real requirement ‑‑ how refreshing ‑‑ but if we accept RIPE we sin against the Holly principles of standarization. I think it is a good example of standarization but he was still on the right track. After a good dinner sensible things came out of the Commission. Maybe there's a lesson there sometimes.
This was more or less translated in another cartoon by Francois Fluckiger. I'll read it again. There's a man about to be burned. What did he do? Well, he said we like IP. Part two of the story. And this one he said we dislike IP.
These are nice illustrations of the environment of the early years of RIPE those of you who have been around for 15 or 20 years might have met or heard of Arne Moi he was the representative of the Norwegian government in various networking committee organisations and projects. I think his official title was state secretary for higher education and church. An extremely nice friend, extremely cultural and with a good sense for taking three steps back and put some interesting thoughts on paper. And I ‑‑ in the beginning of 1992 he got deeply worried about all these political discussions and no progress, so he wrote a white paper we would call it today. And to make clear that this was not the Norwegian government talking, he signed it by Arne Moi, a citizen of Norway. On TCP/IP: From a dumb end user point of view like myself it stinks to high Heaven. This was the politician. On the other hand it seems to work. This was the useer speaking, because Norway was one of the fore runners of TCP networks all over the country and of course the e‑mail left his office over TCP/IP.
How big was this network in the early days? Well, one of my colleagues who is no longer with us, wrote this small program to try and just measure how big is this internet in Europe. In those days, this was the days before commercial interests knew about domain names, so a domain name was for real and second level domain name belonged to one organisation. So in October ‑‑ October the 2nd 1990, one of the first counts of how big is this internet thing, 258 organisations could be traced in the DNS that were using the internet. And close to 32,000 end hosts.
It's a bit bigger but in 1990 this was considered ‑‑ it was a major result but all things are relative over time as we all know but I just wanted to give this historical point.
Arne Moi again: Lesson number one: Never let the prophets run a project because they have the answer instead of asking what must be done in order to reach the target." I'm very tempted to say digital certificate. Listening to some of the discussions we had earlier this week, but we can find other examples. The lesson is: First define what you want to do and then start doing it instead of giving some nice tools to some technical, extremely educated people and let them build something what they like. But don't lose your site on what you want to do when you start a major thing.
One of the last ones, which I think contains a very important lesson: "Part of the success of the TCP/IP on a pan‑European scale seems to be that there are no regulations on it." This is when they told you what telephone you hooked up to your line and they owned the telephone as well in many countries. "It has so far been outside the realm of the PTTs and the users themselves have had to manage it." Almost right. Central management of things like the internet is a nightmare. And even today there are people not in the room here, of course, but out there, who are having large international meetings discussing how they can get their hands on the internet because it needs central management. I'm talking about governments, ITUs, Commissions, internet governments for ‑‑ so not much has changed. They lost control over what the PTTs used to control and new generation of civil servants think that it's only natural for governments to control this resource again.
So what was this all about? Lots and lots of acronyms, which we couldn't explain now. Horst Huenke: "Acronyms is an area where we in Europe have profound experience. Our acronym technology is leading the world."
(Applause.)
Meanwhile, back on the farm, there was not only the early beginings of the internet in Europe, there were lots and lots of networks, we had deck net phase IV, deck net phase V, that was there was SNA from IBM. There was EARN/BIT net, again protocols used to connect universities in Europe and the United States. There were the colored books, a set upon application protocols developed in the UK academic world to run over X.25 and basically providing things like file transfer, e‑mail, remote log‑on and simplifyed version of what we now know as DNS. There was CERNET, when was inside the Lucern scientific institute which was one of the largest networks in Europe, a hell of a lot of computers have always been there. In those days you had more than one or two makes of computers so there was a wide range of incompatible computer equipment which made use of a common protocol set to communicate, which in itself made it very interesting. But there was TCP of course and there was X.25 because that is the way forward according to the PTTs, the telephone companies (PTT) and that was the only way you were supposed to do networking. It worked sort of, sometimes, and it was very expensive.
This was ‑‑ I came across this and I could not resist. I'll show it to you. All these network technologies had their own addressing problems and policies. One of the largest global networks was deck net, digital equipment computers were extremely popular in the research world and high energy physics world in Europe and in the United States, European space agency and NASA. They all had their DECNETs and they were all interconnected. That was not by design. I mean, you have university, which is involved in high energy physics and in space research and there are these machines and there's one department and somebody connected everything. So address coordination, Address Policy, was needed. Each of the four communities delegated one person to a small committee that sorted out global policy or deck net IV. The Europe, Enzo Valente, who explained once to us that the way this whole thing worked was very simple. It was a democratic dictator ship which means you elect me, that's the democratic part, and I tell you what to do. That's the second part. Thank you Enzo. Because you know what, it worked, it worked for many many years, and it did much more than digital equipment corporation ever imagined it could do. I think it was one of the largest global networks in its days.
Yeah, back to TCPIP, so we were having problems with building something on a European scale. There were many organisations, and some of them are listed here, who all had sort of an island networking, how to make that into a European wide back bone. There are two ways to do that. Either you're politically correct and you get ‑‑ after filling in a few forms, load of money out of Brussels and then you have to scratch your head because what you promised to do doesn't exist or you don't have the money and you put all your resources together, you sit around a table and say I can contribute this, I can contribute that. That was the birth of ebone which I think this was the first original design of eborn. This was never implemented because some of the players didn't want to play. If you want to know more details, if you ask very politely tonight, Kees Neggers might give more information tonight. He played an important role in getting these people together and keeping them together. But we had to be politically correct, so the surface definition of ebone is: We will provide a plug with two outlets and one you use for your internet IP stuff and the other for the ISO stuff. And this has been implemented. Can you imagine which plug was more popular than the other? But this was implemented, it worked, and this was the only way, political correct way to go forward, because politicians could point to the right hand of this picture and say we made great progress, the users clustered around the last side and said at least we have a decent service.
Back to RIPE. What I would recommend everybody to read is the minutes of the first RIPE meeting and while you're at it, take the second and the third as well. In those days we were with 20, 30 people around a table, meeting a day, day and a half, and we wrote proper minutes of what was said, we still keep a record but in a slightly different format and you have to search a little more.
So what were we talking about? This was the very first RIPE meeting and we said there's a name for root name server in Europe. I couldn't believe my eyes. Yes we do have root name server now but it took many many meetings and years before we got that far.
We were talking about connectivity. That was a very hot item because connectivity was extremely expensive in the '80s, late '80s, if it was available at all. This was a major obstacle towards international networking. This we don't discuss anymore today but the next one, we talked about routing and that is still a very active RIPE Working Group today. And the last one I really like, we identified the need for registration data quality. We identified the need to have a database and Daniel proposed to use the who is neck mechanism where everybody would put information on the existing ‑‑ that was on the first RIPE meeting already identified that it was for operational reasons extremely important to have high quality registration data. And we had one or two presentations earlier this week, one from the RIPE NCC and another one where I was very glad to see we have taken up this requirement again because it still is ‑‑ and I think in the near future will be even more important than in the past.
Another one, these minutes of the very first meeting, and with these two sentences: All agree on the fact that there should be someone to monitor the progress of RIPE. Rob Blokzijl is appointed as volunteer. Someone may wonder how I got into this position of what we call chairman of RIPE. It's all their fault because they're the ones who volunteered me. We don't have a procedure in place for the election of a chairman and one of these days we may want to put one in place. I don't think I'll do the next 20 years.
One option, of course, is that if you thought that this action of these gentleman was rather successful, tonight at the dinner, you may ask them very politely to think about appointing another volunteer.
Okay, I will think about plan B.
So I think I'm coming near the end. I could not let this one pass: We are now 1994, and this whole confusion about what TCP/IP, now it had penetrated so far that the old established telephone companies woke up and said, if so many people are using it there must be money in it, so British Telecom was one of the first ones to come with an internet access connectivity range of products. So what do you do? You make this beautiful, colorful flyer, and of course this is a whole new thing and nobody knows how it works, so you poor potential customers should get a little bit explanation of what it is all about. So the domain name system or DNS is the method used to name your computer or its associated host ‑‑ well, that's all right ‑‑ by giving it a unique IP address. Um. It is also the system used to translate unfriendly names to user‑friendly numbers.
(Applause.) And I don't want to change a single word when I quote other people, ‑‑ some people say, Rob, you make this up. I said no, I don't, I still have a copy of this in my study. This was distributed at an INET conference in Prague in 1994, but it ‑‑ yes, it's highly entertaining, but the message is TCP/IP had really made it in Europe if large telephone companies are taking it serious.
So I would say thank you to your valuable contribution to the discussion which, according to a small old dictionary published by RARE, it's a translation of pointing out to someone he is talking nonsense. And the other one I think for some of you might remember really black and that was the end of this little dictionary, and I think it's the end of my presentation. Thank you all for your attention. I don't take questions. Questions you can approach me tonight at dinner. Thank you.
(Applause) Next will be a contribution by Daniel Karrenberg ‑‑ actually it's his fault more than mine because in those days he worked for the company housed by CWI and he said CWI director might be reluctant, a bit political and sensitive, you're a physical activity, you have a reputation you don't care, can you play host to this first meeting. So it was my fault because I said yes.
Daniel Karrenberg: Thank you. I'm trying to use this one. Yes, I'm trying to be a little bit faster because we have dinner in front of us and I want to try to be focused.
The talk is about RIPE, and I've been asked many a time how the French text is here. Finally, I can tell the story. It's not 20 years now. After a while all the most secret documents get declassified but this was a story that could only be told in private and smoke‑filled rooms before. Now I will tell it in public. How did we get the name? Well, actually, the name was suggested by a US American. So the press in the room, your secret plot revealed after 20 years. And I'll tell you the story and it's not so short.
Actually, it was a Texan, and I already had a picture of the other Texan on the slide but we should get ‑‑ and before you get the wrong idea, this is not John Crain. So long and involved story.
Actually, Rob already told you about the historical background. There was an abundance of acronyms and it was very fashionable, actually, to have English‑sounding acronyms with French expansions and it was not only (French word) also known as RARE, some say RARE is not well done ‑‑ it was not only that, there was also zillions of others. The interesting thing here is that ‑‑ two anecdotes, one, if you look carefully on the office sign that they put on their 20‑year anniversary website, the room is H 222 and there was Nick H and room 222 was the first room that RIPE NCC used as an office and we placed them. So that's another anecdote here. For the story you only need to know that at the time it was very fashionable to have these English‑sounding acronyms with French expansions. Actually RARE, I have a really sort of mixed feeling about that because at some point RIPE NCC was part of RARE and RARE has played a pivotal role here, but not immediately willing. I think we took a long time to convince them to do this. But what you have to know, RARE, if you spell it or pronounce it in a Dutch way it comes out at [] RAR and all the Dutch people would know that means ‑‑ what does it mean? Someone strange usually. So when we were setting up this RIPE NCC and we had to deal with external contractors and we had to spell our address, I spelled the address and they said it comes out as RAR and they said are you sure it comes out that? EU So at the time RIPE didn't exist, RIPE NCC didn't exist, and I was sort of spending my time promoting something that's called ‑‑ was called EUnet and Antonio Blasco Benito, who's here, was involved and this was a UUCP network and that's Unix to Unix copy, it's a storage forward mechanism that could transfer mail and something called User Net News which I think survived to this day. In Europe, this was organized by the European Unix user group and I was on the board of the Unix user group and I was one of the operations people first in Germany, and then at the mathematical center here in Amsterdam. So it was very nice because I was on the board and also someone working on it. That was very good. I liked it, having both perspectives. And I would go to these EUUG conferences, explain how the network was growing, how we were changing the telephone connections and how then we would have first nine 900 thousand bits per second and then faster bits, 64 kilobits per second and since we had them we could turn on TCP and also use all the other things. And I actually think ‑‑ at first I thought what I'm telling you about ‑‑ what happened at the EUUG conference in Florence, when I looked through my attic and found the cover of the proceedings, it was in 1986 and that's probably a little bit early, maybe I was misled by the network idea on the cover. So I was going around these conferences, also the national ones, and I was telling them about the network and I still have somewhere in my house this original slide and that's why the slide projector is there. I hoped I could find it, but I couldn't. I had one slide in my presentation and they were really plastic sheets and you wrote on them and if you didn't use the right pen they would get all smeary, and things like that, but you didn't make a new set of slides for each presentation because that meant sit down and write it again. So I had this slide pack telling about the eunit that I would update and there was one slide that looked like this. This is a reconstruction. It showed the European IP networks and this was most at the time of the IP stuff that was going that was internationally connected. So it really fitted on one sheet of paper. This particular one is actually ‑‑ that I used for the reconstruction is document RIPE 05 so it was made from switch and it says it on the bottom here it were.
And he actually made some of these. If you look at the RIPE archives you can trace the development of the European IP networks.
There's also one that shows the institutions, you can look it up yourself and I had a slide like this that said European IP networks or something and then I had to go to France to the AFUU conference and my French is not at all that good, but when I give talks in France, I tell people it's okay to ask questions in French and I will respond in English to encourage them to ask questions. In order to do that in this particular time I changed this slide a little bit, it said European IP networks and I changed it to a French title, Reseaux IP Europeans. To you, it's familiar; to me, it wasn't, I just wrote it down. Next, I had to go to this conference, and this conference I have the proceedings here, and that was the EUUG of that idea and being my lazy self, I didn't make a new slide pack and I didn't even change the French back to English, it's a title, everybody may not read that. At this conference, it was a bit like this, the stage was about twice as wide and the speakers were ‑‑ the microphone was on one side and the equipment was exactly on the other side, so I went, oh, my god, I have to give this talk in this session and people will see me running from one side to the other. Wireless mikes didn't exist, or they existed but we couldn't afford them. So I asked the speaker who was speaking after me to actually turn my slides and I offered a reciprocal deal and he was very happy about that. And after I was finished with my talk, he handed back this slide to me, my slide pack to me, and he said I heard that you're looking for a name for this thing and I said yes, we've thought very hard but haven't really found one, and he said it's there, and I go like where? It's on this slide, he says, and I go yeah, European networks, that doesn't have any ring. And then he points out RIPE. I think I dropped the slide pack, laughed myself silly and then called Rob and said I have this great idea.
(Laughter.)
And it's, of course, Rob liked it immediately and everybody liked it because it was kind of a pun on RARE and it was a bit of an expression of the mood at the time, that the time was RIPE to now do it and get out from under the U‑boat and doing it all in secret so it also expressed the spirit.
Actually, these proceedings actually reveal who the Texan was because he was speaking right after me, this guy, John Quarterman, and some of you may have read his book, The Matrix, which appeared at that time where he meticulously documented all sorts of networking activities around the world because he also likes to travel. So finally this story is now out in the open. It's really a conspiracy, it's really been an American who discovered this name or suggested it, but I wrote it first.
So my last slide, the answer is often right before one's eyes. Looking carefully will reveal it.
(Applause.)
Rob Blokzijl: Thank you, Daniel. Yeah, 20 years of RIPE, that has been made possible by basically you, of course, you are RIPE. Nobody and nothing else. So we thought we will ‑‑ in a little bit symbolic way, emphasize this that RIPE is you and not some mysterious office somewhere. And the way that we thought about this and we thought, well, continuity, 20 years points to some continuity, well, continuity in this environment means ‑‑ we did something very silly, some people would say, we started counting, going through a whole pile of documents, covering all the RIPE meetings and counting who has been to the most RIPE meetings and took the top ‑‑ don't worry Daniel, don't worry, you and I are one and two, but we don't count.
So I would like to invite Ruediger Volk, Wilfried [], Keith Mitchell, and to the stage
(Applause.)
In the good old tradition of these silly little prize contests in the RIPE framework, I'm sure there are other people who might have beaten you in attending even more RIPE meetings, but they're not here today. So if you're not here, you don't count. I would like to thank all of you as representing a great community. Thank you.
(Applause.)
We ‑‑ so far we have been looking back, but I think if there's one thing that this week showed, there's a lot of work in front of us, and I think we are all very glad to realise that at this RIPE meeting we had 149 people registered who came for the first time to a RIPE meeting.
(Applause.)
Because it might well be that we need at least another 20 years of RIPE, and that is then in the hands of new people.
So we did something similar and we had to count a lot ‑‑ kind of lottery of all these newcomers who registered and actually are here, so I would like to invite Geoffroy Rivat, excuse me if I don't pronounce your name correctly.
If you're here, you are lucky. (Applause.)
So this brings us to basically the last bit I had planned. I would like to invite all of you special invited guests here on the stage so the people can see the front part of your face as well.
Please come up because we have a little gift for you as well.
We have a big stage. Don't hide behind each other.
So, I think, once again, thank you for participating in the first RIPE meeting, for initiating this great work and some of you, some of you couldn't stop, like Ruediger Volk, he got his prize for the first RIPE meeting after having attending most of the RIPE meetings.
(Applause.) Okay
(Applause.) Tonight, you can all admire this in greater detail, but it says RIPE chair. See I hope there's only one copy. No, I hope you make more copies.
So once again, thank you all very well. Thank you all very much. I think we are almost done. Please take your seat, it's awkward standing here maybe.
We are almost done. So we have done the 20 years of RIPE and my closing words really are: Are we ready for the next 20 years? Well, I don't know about 20 years, but I think we should be ready for the next couple of years because if this week has shown one thing, it is ‑‑ there are a couple of hectic years in front of us, running out of IPv4, fixing the global IPv4 database, introducing digital certificates, and I have not mentioned all the outside influences like governments thinking about governance over the internet and other disturbing disturbances coming from the environment. I think it's ‑‑ we have celebrated and tonight we'll have a very fine dinner, 20 years of successful operations, that is your operations, but I hope we are ready for a couple of more years, especially the coming few years where we will see some interesting things coming in to us.
So without further ado, I thank you all for sitting so patiently and listening to us. I thank our guests for coming and for other, I think except P ‑‑ IET who has other obligations later today, we'll all meet again at the dinner tonight.
See you at the dinner. The dinner starts at 7:30, so you have ample time to prepare yourself for that. On the back side of the invitation is a little map. I think it's four minutes walk, maybe three, from here. See you all tonight.
Thank you.
(Applause.)
Before we start the last session, I have two announcements. One is that some of you might have expected the address Working Group to have an extra slot here, that is tomorrow morning at 11. So we do not discuss Address Policy this afternoon.
Secondly, you all have an invitation for the RIPE dinner tonight and it carefully explains where it is. It explains that it is today but it does not say when it starts, well it starts at 7:30 tonight. That is the arrival time you should be aiming for.
20 years of RIPE. So we got together with a few people and thought we can not let that pass unnoticed. So what we have done is we invited a couple of special guests and it's always difficult to pick special guests because there is a rich choice. So we thought the people who were involved in the very first RIPE meeting 20 years ago, we will invite those, we will invite them. And we invited a few more people who were very, very instrumental in the first couple of years of RIPE to get this whole effort on the road. And I have a presentation somewhere here.
In the old days we did this with equipment like you see here in front, and I managed to master that, and with this modern stuff sometimes a little bit of help is appreciated.
Thank you.
So the theme is RIPE 20 years young, because I had the choice between young and old and old seems like the end of a lifetime, of a lifecycle, and I after this almost one week now of a RIPE meeting, we are definitely not at the end of our discussions, so that's why I chose 20 years young because we still have a lot to do.
History of RIPE, this is one of the slides I used in my Monday morning newcomer presentation. It started in May 1989 with a meeting where 14 people showed up. It was half a day meeting, and this was a meeting of people who in their own environment, most of it research environment, universities, had local networks which were using the TCP/IP protocol stack and who started to make connections and found out you better start thinking about a little bit of coordination like routing or maybe it's not a good idea to offend your own address space, but try to use coordinated address space. So let's get together for half a day and sort this out and then we're done. Well, I think that was very optimistic, if I look at this crowded room 20 years later. And a little bit later I will highlight some of the issues which we identified 20 years ago and which we have been discussing this week.
We have made progress, but some issues are still there.
I want to spend not too much walking through 20 years of RIPE, RIPE meetings and RIPE documents. It's all on the RIPE website. I want to go a little bit more in detail in describing the networking environment in Europe in the late 80s, because when I went through some of the old documents, I was surprised to see that some remarks made 20 years ago are still extremely valid today, and I will show you examples of this.
So what was the environment? To summarize it in a few lines, some people have called this the period of the protocol wars. OSI was good, open systems interconnect, invented by mainly European phone companies which in those days all preferably owned by national governments and TCP/IP was considered bad because it was not invented by these phone companies but came out of the US research environment. It had two aspects which were highly suspicious in the first place: It worked; in the second place, it came for free, whereas the OSI stuff was famous for the fact it didn't work and was very expensive.
So 14 people came together and I think I'm very pleased to give you a list of ‑‑ a subset of those people because they're all here, all sitting here in the front row and later I'll introduce them more properly at the end of this session. Francois Fluckiger, Arnold Nipper, Ruediger Volk, Enzo Valente, Antonio Blasco Benito, Daniel and myself were among those 14 who were trying to fix the internet in an afternoon in May 1989. So I think it's appropriate to give them a round of applause.
(Applause.) There were many networking organisations in those days, though most of the networking was still in ‑‑ done in the academic and research communities. One important organisation there was an organisation that doesn't exist anymore, changed it's name to another one, and that was RARE. RARE is important here, especially Kees Neggers, because they provided umbrella, protection against these activities which were started and considered to be very suspicious by many, many other organisationsment so in order to give us some space to grow, to exist, RARE played, if we now look back, an extremely important role, and I think we can all agree, if you remember those days, that specially Kees Neggers who was on the executive board, that's what it was called those days, has been very actively providing the environment where we were not too much hindered by all this European politics. Most of you know him because later he was for many, many years the chairman of the executive board of the RIPE NCC till May last year, I think. But he's here in this capacity as one of the old boys who made this all possible. And Kees Neggers thank you for that role as well.
(Applause.) So this first meeting we had was quickly followed by a second and a third and a fourth and I can go on like that. But if you meet you need room to meet and if you meet for more than a few hours, do you want coffee, tea, refreshments, something to eat? I think we're all grateful to NIKHEF, they have hosted 19 RIPE meetings in those early days. Hosted in those days meant and we feed you and you get coffee and tea and there's your meeting room and a copy machine and you can use our computers. I'm not quite sure whether this was a decision by the director to do this, but they never complained when I informed them after a meeting that there would be a bill coming from the central canteen and things like that and they always signed for that. Those directors are not here today but one of the current directors is, Arjen van Rijn and through the whole community we thank Arjen van Rijn. Thank you.
(Applause.)
And last but not least, TERENA, ‑‑ this was the new organisation from RARE and that still exists in the lifetime of RIPE, specially in the creation of the first years of operation of the RIPE NCC. TERENA has played a commendable role as has supported this until the day when that child was five times bigger than the parent. And RIPE NCC became really independent. But I think Karel Vietsch, the secretary general, I think, of TERENA, he's not here this afternoon, but will be at the dinner tonight, and through him we thank this organisation as well.
(Applause.)
Okay, back to networking, OSI. You might think, oh, some invention, lots of nice and expensive books but it really worked and these are the first three verses of a long, long song composed by Vint Cerf and this is one of the nicest illustrations of the mood at the times in the 80s. This is '85, so this is a couple of years before we started thinking about OSI.
This OSI thing was, of course, very much promoted by governments and by the European Union which went by another name in those years, same people, same organisation. And some of them, though they were paid to promote this OSI and fight against TCP/IP got worried already in quite early days. Horst Huenke was a high level official in those days and DGXIII was under a different name but he was director of general Ford Telecommunications. This was European Commission which had a different name ‑‑ was getting worried at an early stage. Well, you can read it here. But the interesting thing is you can replace the name and the organisation in the title and you can replace OSI with things like ‑‑ pick your choice, DNSSEC ‑‑ IPv6, and soon in this theatre digital certificates. I find some of these statements, if you generalize them or translate them to 20 years later, I think still extremely valid.
This was a fight not only in smoke‑filled meeting rooms ‑‑ in those days everybody smoked, specially in meeting rooms ‑‑ but it also made the news from time to time. X.25 was the network of this whole OSI model, and this is how the press summarized it, which I think was quite amusing as well.
Francois Fluckiger was not only known as a key figure in European networking in those days, he was involved in the cert team, who at one point in time in the '80s, probably managed the largest concentration of international networks in Europe, everybody wanted to connect to search because everybody else was connected to CEC. Francois Fluckiger has been very active and sometimes his sense of humor, he could not contain it any longer, and this is the OSI, X.25 observation or whatever. And I don't know why the ‑‑ how good it is readable at the back but the conversation goes like this: And you, how many packet on your net? Six last week. Well lucky you are. Yes, that was my test program.
Now, replace X.25 here by IPv6 and suddenly it becomes very familiar. We have seen many examples of that in presentations. Thank you, Francois Fluckiger.
Applause.)
IPv6 has made it in some other parts of the world as we can see here. Dental chewing gum, IPv6 ready, wow. IPv4 never made it that far. And you get a two‑pack. Two packet is.
Horst Huenke again, two years later, 1991, at a conference of the annual networking conference of this organisation, RARE, which was still very, very much officially involved in implementing this OSI networking, which did not exist, but some people thought that is not a good enough reason not to do it because politically it was correct but I think he said if we ignore RIPE we ignore the real requirement ‑‑ how refreshing ‑‑ but if we accept RIPE we sin against the Holly principles of standarization. I think it is a good example of standarization but he was still on the right track. After a good dinner sensible things came out of the Commission. Maybe there's a lesson there sometimes.
This was more or less translated in another cartoon by Francois Fluckiger. I'll read it again. There's a man about to be burned. What did he do? Well, he said we like IP. Part two of the story. And this one he said we dislike IP.
These are nice illustrations of the environment of the early years of RIPE those of you who have been around for 15 or 20 years might have met or heard of Arne Moi he was the representative of the Norwegian government in various networking committee organisations and projects. I think his official title was state secretary for higher education and church. An extremely nice friend, extremely cultural and with a good sense for taking three steps back and put some interesting thoughts on paper. And I ‑‑ in the beginning of 1992 he got deeply worried about all these political discussions and no progress, so he wrote a white paper we would call it today. And to make clear that this was not the Norwegian government talking, he signed it by Arne Moi, a citizen of Norway. On TCP/IP: From a dumb end user point of view like myself it stinks to high Heaven. This was the politician. On the other hand it seems to work. This was the useer speaking, because Norway was one of the fore runners of TCP networks all over the country and of course the e‑mail left his office over TCP/IP.
How big was this network in the early days? Well, one of my colleagues who is no longer with us, wrote this small program to try and just measure how big is this internet in Europe. In those days, this was the days before commercial interests knew about domain names, so a domain name was for real and second level domain name belonged to one organisation. So in October ‑‑ October the 2nd 1990, one of the first counts of how big is this internet thing, 258 organisations could be traced in the DNS that were using the internet. And close to 32,000 end hosts.
It's a bit bigger but in 1990 this was considered ‑‑ it was a major result but all things are relative over time as we all know but I just wanted to give this historical point.
Arne Moi again: Lesson number one: Never let the prophets run a project because they have the answer instead of asking what must be done in order to reach the target." I'm very tempted to say digital certificate. Listening to some of the discussions we had earlier this week, but we can find other examples. The lesson is: First define what you want to do and then start doing it instead of giving some nice tools to some technical, extremely educated people and let them build something what they like. But don't lose your site on what you want to do when you start a major thing.
One of the last ones, which I think contains a very important lesson: "Part of the success of the TCP/IP on a pan‑European scale seems to be that there are no regulations on it." This is when they told you what telephone you hooked up to your line and they owned the telephone as well in many countries. "It has so far been outside the realm of the PTTs and the users themselves have had to manage it." Almost right. Central management of things like the internet is a nightmare. And even today there are people not in the room here, of course, but out there, who are having large international meetings discussing how they can get their hands on the internet because it needs central management. I'm talking about governments, ITUs, Commissions, internet governments for ‑‑ so not much has changed. They lost control over what the PTTs used to control and new generation of civil servants think that it's only natural for governments to control this resource again.
So what was this all about? Lots and lots of acronyms, which we couldn't explain now. Horst Huenke: "Acronyms is an area where we in Europe have profound experience. Our acronym technology is leading the world."
(Applause.)
Meanwhile, back on the farm, there was not only the early beginings of the internet in Europe, there were lots and lots of networks, we had deck net phase IV, deck net phase V, that was there was SNA from IBM. There was EARN/BIT net, again protocols used to connect universities in Europe and the United States. There were the colored books, a set upon application protocols developed in the UK academic world to run over X.25 and basically providing things like file transfer, e‑mail, remote log‑on and simplifyed version of what we now know as DNS. There was CERNET, when was inside the Lucern scientific institute which was one of the largest networks in Europe, a hell of a lot of computers have always been there. In those days you had more than one or two makes of computers so there was a wide range of incompatible computer equipment which made use of a common protocol set to communicate, which in itself made it very interesting. But there was TCP of course and there was X.25 because that is the way forward according to the PTTs, the telephone companies (PTT) and that was the only way you were supposed to do networking. It worked sort of, sometimes, and it was very expensive.
This was ‑‑ I came across this and I could not resist. I'll show it to you. All these network technologies had their own addressing problems and policies. One of the largest global networks was deck net, digital equipment computers were extremely popular in the research world and high energy physics world in Europe and in the United States, European space agency and NASA. They all had their DECNETs and they were all interconnected. That was not by design. I mean, you have university, which is involved in high energy physics and in space research and there are these machines and there's one department and somebody connected everything. So address coordination, Address Policy, was needed. Each of the four communities delegated one person to a small committee that sorted out global policy or deck net IV. The Europe, Enzo Valente, who explained once to us that the way this whole thing worked was very simple. It was a democratic dictator ship which means you elect me, that's the democratic part, and I tell you what to do. That's the second part. Thank you Enzo. Because you know what, it worked, it worked for many many years, and it did much more than digital equipment corporation ever imagined it could do. I think it was one of the largest global networks in its days.
Yeah, back to TCPIP, so we were having problems with building something on a European scale. There were many organisations, and some of them are listed here, who all had sort of an island networking, how to make that into a European wide back bone. There are two ways to do that. Either you're politically correct and you get ‑‑ after filling in a few forms, load of money out of Brussels and then you have to scratch your head because what you promised to do doesn't exist or you don't have the money and you put all your resources together, you sit around a table and say I can contribute this, I can contribute that. That was the birth of ebone which I think this was the first original design of eborn. This was never implemented because some of the players didn't want to play. If you want to know more details, if you ask very politely tonight, Kees Neggers might give more information tonight. He played an important role in getting these people together and keeping them together. But we had to be politically correct, so the surface definition of ebone is: We will provide a plug with two outlets and one you use for your internet IP stuff and the other for the ISO stuff. And this has been implemented. Can you imagine which plug was more popular than the other? But this was implemented, it worked, and this was the only way, political correct way to go forward, because politicians could point to the right hand of this picture and say we made great progress, the users clustered around the last side and said at least we have a decent service.
Back to RIPE. What I would recommend everybody to read is the minutes of the first RIPE meeting and while you're at it, take the second and the third as well. In those days we were with 20, 30 people around a table, meeting a day, day and a half, and we wrote proper minutes of what was said, we still keep a record but in a slightly different format and you have to search a little more.
So what were we talking about? This was the very first RIPE meeting and we said there's a name for root name server in Europe. I couldn't believe my eyes. Yes we do have root name server now but it took many many meetings and years before we got that far.
We were talking about connectivity. That was a very hot item because connectivity was extremely expensive in the '80s, late '80s, if it was available at all. This was a major obstacle towards international networking. This we don't discuss anymore today but the next one, we talked about routing and that is still a very active RIPE Working Group today. And the last one I really like, we identified the need for registration data quality. We identified the need to have a database and Daniel proposed to use the who is neck mechanism where everybody would put information on the existing ‑‑ that was on the first RIPE meeting already identified that it was for operational reasons extremely important to have high quality registration data. And we had one or two presentations earlier this week, one from the RIPE NCC and another one where I was very glad to see we have taken up this requirement again because it still is ‑‑ and I think in the near future will be even more important than in the past.
Another one, these minutes of the very first meeting, and with these two sentences: All agree on the fact that there should be someone to monitor the progress of RIPE. Rob Blokzijl is appointed as volunteer. Someone may wonder how I got into this position of what we call chairman of RIPE. It's all their fault because they're the ones who volunteered me. We don't have a procedure in place for the election of a chairman and one of these days we may want to put one in place. I don't think I'll do the next 20 years.
One option, of course, is that if you thought that this action of these gentleman was rather successful, tonight at the dinner, you may ask them very politely to think about appointing another volunteer.
Okay, I will think about plan B.
So I think I'm coming near the end. I could not let this one pass: We are now 1994, and this whole confusion about what TCP/IP, now it had penetrated so far that the old established telephone companies woke up and said, if so many people are using it there must be money in it, so British Telecom was one of the first ones to come with an internet access connectivity range of products. So what do you do? You make this beautiful, colorful flyer, and of course this is a whole new thing and nobody knows how it works, so you poor potential customers should get a little bit explanation of what it is all about. So the domain name system or DNS is the method used to name your computer or its associated host ‑‑ well, that's all right ‑‑ by giving it a unique IP address. Um. It is also the system used to translate unfriendly names to user‑friendly numbers.
(Applause.) And I don't want to change a single word when I quote other people, ‑‑ some people say, Rob, you make this up. I said no, I don't, I still have a copy of this in my study. This was distributed at an INET conference in Prague in 1994, but it ‑‑ yes, it's highly entertaining, but the message is TCP/IP had really made it in Europe if large telephone companies are taking it serious.
So I would say thank you to your valuable contribution to the discussion which, according to a small old dictionary published by RARE, it's a translation of pointing out to someone he is talking nonsense. And the other one I think for some of you might remember really black and that was the end of this little dictionary, and I think it's the end of my presentation. Thank you all for your attention. I don't take questions. Questions you can approach me tonight at dinner. Thank you.
(Applause) Next will be a contribution by Daniel Karrenberg ‑‑ actually it's his fault more than mine because in those days he worked for the company housed by CWI and he said CWI director might be reluctant, a bit political and sensitive, you're a physical activity, you have a reputation you don't care, can you play host to this first meeting. So it was my fault because I said yes.
Daniel Karrenberg: Thank you. I'm trying to use this one. Yes, I'm trying to be a little bit faster because we have dinner in front of us and I want to try to be focused.
The talk is about RIPE, and I've been asked many a time how the French text is here. Finally, I can tell the story. It's not 20 years now. After a while all the most secret documents get declassified but this was a story that could only be told in private and smoke‑filled rooms before. Now I will tell it in public. How did we get the name? Well, actually, the name was suggested by a US American. So the press in the room, your secret plot revealed after 20 years. And I'll tell you the story and it's not so short.
Actually, it was a Texan, and I already had a picture of the other Texan on the slide but we should get ‑‑ and before you get the wrong idea, this is not John Crain. So long and involved story.
Actually, Rob already told you about the historical background. There was an abundance of acronyms and it was very fashionable, actually, to have English‑sounding acronyms with French expansions and it was not only (French word) also known as RARE, some say RARE is not well done ‑‑ it was not only that, there was also zillions of others. The interesting thing here is that ‑‑ two anecdotes, one, if you look carefully on the office sign that they put on their 20‑year anniversary website, the room is H 222 and there was Nick H and room 222 was the first room that RIPE NCC used as an office and we placed them. So that's another anecdote here. For the story you only need to know that at the time it was very fashionable to have these English‑sounding acronyms with French expansions. Actually RARE, I have a really sort of mixed feeling about that because at some point RIPE NCC was part of RARE and RARE has played a pivotal role here, but not immediately willing. I think we took a long time to convince them to do this. But what you have to know, RARE, if you spell it or pronounce it in a Dutch way it comes out at [] RAR and all the Dutch people would know that means ‑‑ what does it mean? Someone strange usually. So when we were setting up this RIPE NCC and we had to deal with external contractors and we had to spell our address, I spelled the address and they said it comes out as RAR and they said are you sure it comes out that? EU So at the time RIPE didn't exist, RIPE NCC didn't exist, and I was sort of spending my time promoting something that's called ‑‑ was called EUnet and Antonio Blasco Benito, who's here, was involved and this was a UUCP network and that's Unix to Unix copy, it's a storage forward mechanism that could transfer mail and something called User Net News which I think survived to this day. In Europe, this was organized by the European Unix user group and I was on the board of the Unix user group and I was one of the operations people first in Germany, and then at the mathematical center here in Amsterdam. So it was very nice because I was on the board and also someone working on it. That was very good. I liked it, having both perspectives. And I would go to these EUUG conferences, explain how the network was growing, how we were changing the telephone connections and how then we would have first nine 900 thousand bits per second and then faster bits, 64 kilobits per second and since we had them we could turn on TCP and also use all the other things. And I actually think ‑‑ at first I thought what I'm telling you about ‑‑ what happened at the EUUG conference in Florence, when I looked through my attic and found the cover of the proceedings, it was in 1986 and that's probably a little bit early, maybe I was misled by the network idea on the cover. So I was going around these conferences, also the national ones, and I was telling them about the network and I still have somewhere in my house this original slide and that's why the slide projector is there. I hoped I could find it, but I couldn't. I had one slide in my presentation and they were really plastic sheets and you wrote on them and if you didn't use the right pen they would get all smeary, and things like that, but you didn't make a new set of slides for each presentation because that meant sit down and write it again. So I had this slide pack telling about the eunit that I would update and there was one slide that looked like this. This is a reconstruction. It showed the European IP networks and this was most at the time of the IP stuff that was going that was internationally connected. So it really fitted on one sheet of paper. This particular one is actually ‑‑ that I used for the reconstruction is document RIPE 05 so it was made from switch and it says it on the bottom here it were.
And he actually made some of these. If you look at the RIPE archives you can trace the development of the European IP networks.
There's also one that shows the institutions, you can look it up yourself and I had a slide like this that said European IP networks or something and then I had to go to France to the AFUU conference and my French is not at all that good, but when I give talks in France, I tell people it's okay to ask questions in French and I will respond in English to encourage them to ask questions. In order to do that in this particular time I changed this slide a little bit, it said European IP networks and I changed it to a French title, Reseaux IP Europeans. To you, it's familiar; to me, it wasn't, I just wrote it down. Next, I had to go to this conference, and this conference I have the proceedings here, and that was the EUUG of that idea and being my lazy self, I didn't make a new slide pack and I didn't even change the French back to English, it's a title, everybody may not read that. At this conference, it was a bit like this, the stage was about twice as wide and the speakers were ‑‑ the microphone was on one side and the equipment was exactly on the other side, so I went, oh, my god, I have to give this talk in this session and people will see me running from one side to the other. Wireless mikes didn't exist, or they existed but we couldn't afford them. So I asked the speaker who was speaking after me to actually turn my slides and I offered a reciprocal deal and he was very happy about that. And after I was finished with my talk, he handed back this slide to me, my slide pack to me, and he said I heard that you're looking for a name for this thing and I said yes, we've thought very hard but haven't really found one, and he said it's there, and I go like where? It's on this slide, he says, and I go yeah, European networks, that doesn't have any ring. And then he points out RIPE. I think I dropped the slide pack, laughed myself silly and then called Rob and said I have this great idea.
(Laughter.)
And it's, of course, Rob liked it immediately and everybody liked it because it was kind of a pun on RARE and it was a bit of an expression of the mood at the time, that the time was RIPE to now do it and get out from under the U‑boat and doing it all in secret so it also expressed the spirit.
Actually, these proceedings actually reveal who the Texan was because he was speaking right after me, this guy, John Quarterman, and some of you may have read his book, The Matrix, which appeared at that time where he meticulously documented all sorts of networking activities around the world because he also likes to travel. So finally this story is now out in the open. It's really a conspiracy, it's really been an American who discovered this name or suggested it, but I wrote it first.
So my last slide, the answer is often right before one's eyes. Looking carefully will reveal it.
(Applause.)
Rob Blokzijl: Thank you, Daniel. Yeah, 20 years of RIPE, that has been made possible by basically you, of course, you are RIPE. Nobody and nothing else. So we thought we will ‑‑ in a little bit symbolic way, emphasize this that RIPE is you and not some mysterious office somewhere. And the way that we thought about this and we thought, well, continuity, 20 years points to some continuity, well, continuity in this environment means ‑‑ we did something very silly, some people would say, we started counting, going through a whole pile of documents, covering all the RIPE meetings and counting who has been to the most RIPE meetings and took the top ‑‑ don't worry Daniel, don't worry, you and I are one and two, but we don't count.
So I would like to invite Ruediger Volk, Wilfried [], Keith Mitchell, and to the stage
(Applause.)
In the good old tradition of these silly little prize contests in the RIPE framework, I'm sure there are other people who might have beaten you in attending even more RIPE meetings, but they're not here today. So if you're not here, you don't count. I would like to thank all of you as representing a great community. Thank you.
(Applause.)
We ‑‑ so far we have been looking back, but I think if there's one thing that this week showed, there's a lot of work in front of us, and I think we are all very glad to realise that at this RIPE meeting we had 149 people registered who came for the first time to a RIPE meeting.
(Applause.)
Because it might well be that we need at least another 20 years of RIPE, and that is then in the hands of new people.
So we did something similar and we had to count a lot ‑‑ kind of lottery of all these newcomers who registered and actually are here, so I would like to invite Geoffroy Rivat, excuse me if I don't pronounce your name correctly.
If you're here, you are lucky. (Applause.)
So this brings us to basically the last bit I had planned. I would like to invite all of you special invited guests here on the stage so the people can see the front part of your face as well.
Please come up because we have a little gift for you as well.
We have a big stage. Don't hide behind each other.
So, I think, once again, thank you for participating in the first RIPE meeting, for initiating this great work and some of you, some of you couldn't stop, like Ruediger Volk, he got his prize for the first RIPE meeting after having attending most of the RIPE meetings.
(Applause.) Okay
(Applause.) Tonight, you can all admire this in greater detail, but it says RIPE chair. See I hope there's only one copy. No, I hope you make more copies.
So once again, thank you all very well. Thank you all very much. I think we are almost done. Please take your seat, it's awkward standing here maybe.
We are almost done. So we have done the 20 years of RIPE and my closing words really are: Are we ready for the next 20 years? Well, I don't know about 20 years, but I think we should be ready for the next couple of years because if this week has shown one thing, it is ‑‑ there are a couple of hectic years in front of us, running out of IPv4, fixing the global IPv4 database, introducing digital certificates, and I have not mentioned all the outside influences like governments thinking about governance over the internet and other disturbing disturbances coming from the environment. I think it's ‑‑ we have celebrated and tonight we'll have a very fine dinner, 20 years of successful operations, that is your operations, but I hope we are ready for a couple of more years, especially the coming few years where we will see some interesting things coming in to us.
So without further ado, I thank you all for sitting so patiently and listening to us. I thank our guests for coming and for other, I think except P ‑‑ IET who has other obligations later today, we'll all meet again at the dinner tonight.
See you at the dinner. The dinner starts at 7:30, so you have ample time to prepare yourself for that. On the back side of the invitation is a little map. I think it's four minutes walk, maybe three, from here. See you all tonight.
Thank you.
(Applause.)