 Today, our great pleasure is to introduce you to the right network coordinator center. And we have some kind of crash course with a duration of two and a half hours. And we will try to discuss two main topics, internet and data. Bahar will talk about the team and also about our rights. Thank you. It's a great pleasure to be with you. And we have a really good team that came here today to introduce you to some topics that might be of your interest. So, first of all, about our team. My name is Bahar Nogsetian. I'm a external relations officer from Friar LNCC. We are an ER team of government relations team and I'm now responsible for the center of Asia and folks. We have here Christian Deschel. And this one. Bahar Christian. Christian is assistant architect in a drive-in CC. So, he will be the main speaker. We have here Gabriel Ajavahia. He is Armenian. He is also from drive-in CC and he is working as an engineer at drive-in. Alex Semiaka is also external relations officer technical advisor responsible for Eastern Europe and Central Asia. We also have our local counterparts here. People that were engaged in drive-in CC activities and they are working with us. And some of them are our fellows. Some of them have participated in drive-in CC project. And we will talk about that later, how we can participate in our activities in the region and in Europe. So, Christian. Yeah, can I start? Perfect. So, first of all, thank you very much for taking the time to come here with so many people and listen to this presentation. Just some words about this presentation. We started already since last week to give this workshop with the idea to provide an interest and education in what the internet is about. And also with a bit of a relationship to data science, which is a very popular topic at the moment. And the goal for this presentation is that it is not only a monologue from my side, so I don't want to talk all the time. But I also want to get some feedback from you. And in order to do that, I would like to know a bit more about you. And that brings us to the first topic, which is the introduction. So, as I understand it correctly, we have here a lot of students from the American University of Armenia. And from different fields, right? Is that true? Yes. All right. So we did the same workshop on Saturday. It was Saturday, right? Yes. Yes. And no, it was better before, yeah. Then I'm not so... Yeah, try it again, what you just did. Perfect one. Perfect. So what I would like to hear a bit is your background, because the design goal of this workshop is that we're going to bring someone with basically zero knowledge about the internet up to speed for the second part, which is more data science driven. So to make that for you a good experience, I would like to understand what your background is. So if we have a lot of technical people in the audience, then of course we can go through the first part a bit faster and then focus on the second part. And to be honest, the second part is also more interesting, especially for me. So let me ask you, who of you is technical, has a technical background? So show your hands. Okay. With technical, is it correct that we are talking about computer science? Yeah. Okay. Do we have some math students here? Okay. Hold on. Did you raise your hands before? I mean, I consider that technical. No? No. All right. Okay. That's good. Well, sort of. So we have more than 60% non-technical people here, right? And can you just quickly tell me what your backgrounds are? So I mean, right now, okay, I'm going to put you on the spot. I know that's not comfortable. But if the people that didn't raise their hands before could just, you know, raise their hands and then quickly say what their field of study is. So what I would expect is economics, law, political science, philosophy. Exactly. Yeah. Something like that. So you're done. Okay. Thank you. Do we have more people who would like to contribute? And it's only for your benefit, right? Because if I assume that you're very technical, you will not like the presentation. Political science. Perfect. Okay. Okay. Cool. Do we have something else here? Design? Law. Law? Okay. Do we have someone from design or? No. Industrial engineering. Industrial engineering. That's sort of the technical, right? All right. Let's not get hung up on definitions. So to finish the introduction, let me quickly talk a bit about myself. So I started to work in 2010 at the RIPE NCC. I started in a department that was called Information Science Systems. And the main focus was that we provide information to our members. We will talk about our members a bit later. But after a year, I joined research and development. And there I do a lot with data. So my daily job is usually not giving presentations, but doing data. And I'm holding a – so I studied computer science. That was my first study. And last year, I just finished my MBA in big data and business analytics. And I tried to get some of this knowledge also in this presentation to make it more interesting for you. Okay. If I understand it correctly, we also have two and a half hours. Who of you would mind if we, you know, go a little bit longer? So that would be possible. Because I mean, I would like to put a 10-minute break after the first hour. Then we're going to go into the second part. And then we also have local contributions. Yes. Okay. Good. Then let's see. So we did the introduction. Then the first part consists of a brief history of the Internet. Then we're going to learn how TCP, IP networks are working. Then we're going to talk a bit about the DNS system. Then we cover Internet governance. I'm going to talk about IoT, which is a very big password at the moment. And then we're going to have a short interactive session. And then we will learn a bit about RaySea and the right fellowship. And then we are ready for the second part after the break. So let's start with the brief introduction to the history of the Internet. So this is not a full-blown course about the history of the Internet. For that, you should go to YouTube. It's much a better source. But I just want to highlight a couple of things. Before we could develop the Internet or before the Internet was developed, I think there are three fundamental inventions that took place. First of all, it's the electronic telegraph and the telephone, of course. These two things are connected over line, right? They are also line-switched in a way. And then we have the radio. And the radio is specifically interesting because it's for the first time that data got transmitted wireless, right? And nowadays everything is wireless, right? And then it took roughly more than 100 years until the beginning of the Internet started in 1969. That's kind of the cornerstone when we talk about the development of the Internet. This is a very condensed overview of the different development steps and also applications on the Internet. So you know that the Internet was based on a research project which was called ARPANET. And that was created out of a military project which was DARPA, the Defense Advanced Research Project. Something like that. You can look that up. Then a very important step in the development of the Internet was the invention of FTP and TCP IP and Ethernet. Ethernet in that sense is very important because when you're going to connect your laptop at home, then the cable that you're going to use is to 99% Ethernet. And that simplified the connectivity between networks significantly. And just for your understanding, at that time there were actually a lot of network technologies around. But only TCP IP became the dominant standard and I think that's why we have such a growing or the growth that we saw over the years of the Internet because of its simplicity. But we're going to talk about that a little bit later. Then I'm not going to go into details about the different applications or the different companies that got created over the years and some of them don't exist anymore. For example, Netscape. I think it should just show that we are living in a very high-paced environment and the tech sector is moving very fast. When we're going to talk about the Internet and the impact on our world then I think there are four aspects very important. The first one is the technology. The second one is operation and management of the Internet. So that's separate of the technology. And the RIPE NCC, so our organization that would fall in the operation and management but we're also going to talk about that a bit later. Then we have the impact on society and the commercial aspect of the Internet. I would think that the Internet would not be at a stage as it is right now without the commercial aspect of the Internet. I think if the Internet would not have such a commercial effect then I think by now we would still browse on a speed of a modem from the old days. This is also reflected in the growth of the number of Internet users and I think it's important to note that these numbers they can never be accurate, right? These are all estimations. So don't got too hung up on the numbers but I think the trend is important. And the trend is definitely that the number of Internet users is growing and according to this statistics at the moment we have 3.5 billion Internet users which if we're going to compare that with the total number of people in this world then it's roughly around 50% of the world population. That brings us to the next topic which is TCPIP and let me just ask you how many of you and right now I'm pointing more to the technical students know about TCPIP. All right, three people. So as I said before the invention of TCPIP enabled this huge growth of the Internet. In a way the Internet in itself is nothing else but smaller networks connected, right? They together form the so-called Internet and TCPIP is the language that these networks talk with each other and that makes it the Internet because you can imagine if they are not talking the same language then it would not be possible to build such a huge network. Okay, so my drawing skills are not the best one. That's why I ask if we have some design students in the audience so what's important here is that I try to sketch computers on the Internet and there are obviously more computers on the Internet but what I wanted to show with that is that to be able to communicate with each other every machine needs to have a unique number and I added that here. That also means that every device that is on the Internet for example your mobile phone also has a unique number. If it does not have a unique number then you will not be able to participate on the Internet because then you see a lot of errors. So Internet is about sharing data, right? Either you fetch data or you're going to send data somewhere and to show a simple transmission and in this case I picked that node number one assume that that is you wants to send or request a movie from Netflix and in order to do that it is necessary that every node in this network knows all the other connections, right? And this is visualized in, well, I draw a book for that which is called the Rounding Table and I think right now I'm going to use the mouse to show you what I mean with that. So compare the Rounding Table with a phone book and the phone book basically contains all the participants on the Internet. If someone is not in this phone book then for this node, so for node number one they are not existing, hence they can't be reached and this is something that in the second part of the course we will see in more details but in this example we're going to see how the data would flow from node number one to node number three because in the Rounding Table we have this entry saying like if you want to send something to node number three then they are connected via node number two, five and four. Is that clear so far? Are there any questions so far? No? Okay, don't be shy. I mean if there is something that's not clear please raise your hand and then we try to solve it. So in that sense node number one would send the data to node number two because what's also in the Rounding Table is that node number two is on the right side. And if we're going to assume that all of these nodes have something like the Rounding Table then you can assume that the data transfer will reach the destination. Then to make this a bit more realistic I want to introduce so-called IP addresses because these simple numbers, they are in reality not used because if we would have so simple numbers that it would just not scale with the number of nodes that we have on the internet. So smart engineers back then introduced what's called IPv4. I guess some of you already heard about IPv4. Yes? So what's the interesting thing about IPv4 is that it is an address space of 32 bits and what you can address with 32 bits is 4.3 billion IP addresses. So what does that mean? That means that for every node on the internet if it has to be unique we have 4.3 possibilities. And this is a huge limiting factor because if you're going to go back to the slide where I showed you the graph of internet users how many internet users have you had there? It was 3.5 billion. And if you want to compare it with this number then this is already very tight. Now I assume that you don't have only one device that is on the internet like you have your mobile phone or your laptop, you have your TV and probably other things that are connected to the internet so that means that if we're going to have 3.5 billion internet users then you have to multiply it with the number of devices they have and then 4.3 billion is definitely not enough. And this is at the moment a challenge that the entire internet community is trying to resolve. So all the internet organizations and it is termed as IPv6, IPv4 depletion that basically means that we're running out of IPv4 addresses. So what can we do to solve that? Any ideas? Come again? Perfect, yeah. Yes, by the way, Wahan do we have a presence that we can give away for very good students? All right, Gabriel, can you do me the favor and note down the special students? Okay, perfect. Yeah, well, that's what happened. Around 25 years ago engineers started to create a new version which is IPv6 and right now you all ask what happened to IPv5, right? You do that, right? Okay, yeah. Anyway, so that was the experiment and that's why we have this jump from IPv4 to IPv6 and that has 128 bits. So right now if you're good in, well, if you can do math in your head very well then the top number was 2 to the power of 32. The lower one is 2 to the power of 128 and as we all know, that is that one, right? It's a huge number and I think with this number it is possible to address every cubic centimeter of the earth and that's a very big address space and I got a question in the previous workshops that if we run risk to run out again we don't really know. So there's one difference between IPv4 and IPv6. At home you have one IPv4 address so if you're going to talk with your internet service provider and that could be, for example, Beeline or what is another internet service provider in... I mean, Yukon. Yes. And with IPv6, so if your internet service provider has IPv6 then you don't get only one IP address but you get a slash 64. That means you get a address space of 2 to the power of 56 or 64 which is bigger than the internet on IPv4 at this moment. So theoretically, if you're really crazy and geeky then you can have more than 4.3 billion internet devices just in your home, right? I mean, this sounds really awesome, I would say. There is one thing that we have to note to that. We can't switch to IPv6 before everyone on the internet move to IPv6. So the problem with these two protocols, they are not compatible. So that means that we first need to switch every device that's right now on IPv4 to IPv6 and then we can remove IPv4, right? And I show you right now a graph and this is the traffic analysis of Google and it shows how many percentage of users coming to Google have IPv6. And still remember, if that's not 100%, we will still have to maintain IPv4. And at this moment, we have 22%. I mean, you see that the trend is growing but I also said that IPv6 is 25 years old, right? And so the internet community is very unhappy with the uptake of IPv6. And I hope that in the next few years we will be able to switch to IPv6 because if we don't do that then we might run into many problems because I said before 50% of the world population is using the internet. That means that there's still a lot of space to grow, right? Another 50% and then we are completely running out of IPv4 addresses. And just to give you an idea what that means if we don't have enough IP addresses, that means that usually a company is coming to the right-wing CC and ask for IPv4 addresses, then we're going to give them IPv4 addresses if we still have one, and then they can connect you at home to the internet. If we run out of IPv4 addresses then the company can come to us and we're going to say, yeah, sorry, we don't have any more. That means that you don't have internet at home, right? So this is a very important point. And the last term that I want to introduce is autonomous systems. As I said before, the internet is a network of smaller networks interconnected and we usually have a name for these smaller networks and they are called autonomous systems. And to be able to refer to them, we're going to give them a number, so AS1, AS2, AS3, and at this moment on the internet there are 52,000 of these smaller networks. Okay, then this is just a graph that I showed before with the proper terms. So I put IPv4 addresses in there and I gave them ASM numbers. Then let me just check how we are with time because I'm willing to sacrifice that. Yeah. Yeah, okay. So I tried to mark the slides that are a bit more technical with a T+. So if you don't get that, for the first time that you hear the presentation, at the third time you should get it. Then it doesn't matter, right? Because it could be a bit more complex. So TCPIP, I said that's the language that these computers on the internet talk and what's very interesting is that whenever you're going to send data from one machine to another, so basically you're at home from your mobile device to Netflix, then the data travels horizontally but on your mobile phone and on the server of Netflix it travels vertical. And these are the so-called IP stack and the interesting thing about that is that every layer abstracts a certain problem to solve. So first of all we have the link layer that takes care of the physical element, how the data travels. Because the internet is a network of interconnected networks and how these networks are connected with each other is basically irrelevant, right? That could be copper, that could be fiber, that could be satellite and you might laugh but it could even be pigeons or, you know, it could be a truck that just carries data from one location to the next. Related to the standard it does not matter as long as it fulfills certain requirements. Then on the second layer we have the internet layer that makes sure that data travels from one IP address to the next one. Then we have the transport layer that makes sure that the data arrives safe or not. I'm going to get to that in the next slide. And then last but not least we have the application layer where basically it could be your browser, it could be the client of Netflix that takes the data that's coming from the lower layers and either shows it as a web page or a movie or a document or a blog or whatever or music, right? And then on the other side this data travels these layers up again and then the application in this sense the Netflix application knows what you want to watch, right? And the interesting thing about that is that because of these layers do we have any software developers in the audience? Software developers? The computer scientists they wrote something, right? Yeah, basic program, it doesn't matter. So you can sit here and you can create the application and when you write code you can assume that you don't have to take care of these things because someone else took care of that and that's the huge benefit. So that means that you can write internet applications without going into details about that stuff and that of course accelerates innovation. Okay, as I said that on the client goes down and on the server it goes up. Then the next one is the transport layer. I said that there are two different ways on how to transfer data. One is TCP which is the transfer control protocol and the other one is UDP is the universal data protocol and basically it just means the difference between TCP and UDP is the one is connection oriented and the other one is connection less. What that means is if you choose as a software developer to connect to a service via TCP IP then it means that the data that you send will arrive on the other side and you get a confirmation on that or you get a confirmation that it does not arrive. And the other one is UDP. That is fire and forget. You're going to send something and you don't know if it is received or not. And that's the price. Gabriel, listen, listen. Can you give me an example of an application where UDP is being used? Yes, perfect. Gabriel, remember, over there. Okay, good. So the thing is that TCP is very verbose and I'm going to get in that on the next slide and UDP is used for something that needs to work very fast but you don't have a guarantee of the data if the data arrives or not. Every one of you had a Skype call where either the sound goes very robotic or the picture distorts. That is because of UDP because there is no guarantee that the data arrives and if half of the data only arrives, then Skype or the application needs to make something out of that and that could be sometimes very funny visualizations. Then, and this is the last technical slide for the first part. No, maybe not. I don't know. This is called the TCP 3-way hand check. So whenever you're going to do a connection via TCP to another machine, then the client and the server go through these three steps before they're going to send data. And to visualize that a bit, I included a joke and you know the joke already, right? No? Okay, all right. Then it's worth to repeat it. So, as I said, TCP, IP is very verbose and right now I would like to have someone from the audience to help me read it. Yeah? Gabriel Price, okay? Wait a second. Yes. Would you help me? Okay, no problem. Okay, good. I'm going to do the intro and you're going to do the other guy, okay? Exactly. So I'm going to say, and remember, I'm the client and you're the server, right? I'm going to talk about TCP, IP. Are you ready to hear a joke about TCP, IP? Here is the joke about TCP, IP. Did you receive the joke about TCP, IP? Excellent, you have received the joke about TCP, IP. Goodbye. So it basically means that it's very verbose and when that person posted this joke, he said like, I don't care if you get it or not. And that's a hint for UDP because as we said before, fire and forget. So you don't care if the data is received. All right. If you're going to study computer science, those jokes are funny. Okay, then the next one is the domain name system. Who of you know about the domain name system, DNS? Okay, good, good. So we talked before about these numbers and we said IPv4 is dead long and IPv6 is dead long, right? If you want to connect to any of these services, you would have to know these numbers, right? But since that's not very feasible unless you have autism or something like that, we have something that helps you to remember these numbers. Basically, you don't have to remember these numbers, but you can use the so-called domain names, right? And the domain name system just resolves or translates a name to a number, right? As simple as that. So if you're going to type in www.netflix.com under the hood, it will be resolved to either an IPv4 address or an IPv6 address. And that basically makes it a very important part of the internet. Right now, are there any questions? No? Okay, good. From a technical perspective, this one is decentralized and hierarchical, which I will explain on the next slides. So here again, we're going to go back to the drawing and we added the domain names. And each of these domain names will be translated by the DNS system to the appropriate number. Then I said that it's hierarchical and it's decentralized, and that means that the operation of all these different parts of DNS is done by someone else, right? And that makes it possible that if you're going to register a domain, has anyone ever registered a domain? Okay, that's already more. So if you want to have your own web page, then you need to create something that is around here, right? And the tree of DNS is structured like that, that you're going to start from the front and then you're going to go back. And www.is just the part that refers to the web server that hosts your web page, right? And this is kind of standard, right? You could use basically everything there. You could name it Gabriel1, right? Gabriel1.ripe.net. It does not matter. But I think most of us got used to www. And we expect a web page behind that, right? So let's look how a resolution, so basically a name to a number, how that would work systematically. And again, we have you here as a client. Let's say that's your mobile device. If you're going to type in www.yahoo.com, then first the domain name system will ask your local cache. So there's a small thing in your laptop or your mobile phone that will cache all these resolutions, so from name to number for a certain time. And if it's not in there or if it's the first time that you're going to go to this web page, then the following will happen. First of all, you're going to connect to the DNS server of your internet service provider. And right now that's the university. When you're connected to the network of the university, the IT department, they are running a DNS server, among other things. And this DNS server would do the same, right? It looks into the cache. If it's there, then it will serve it directly to you. And then you can basically skip this part. And that could for example be if another student went to the same web page. So that means that if a lot of students go to the same web page, then the entire thing loads much faster, right? But let's assume that nobody went to this web page. And then the resolution starts from the back to the front. And the first one is COM, right? So first you would contact the root DNS server system that basically knows where all the top level domains are. And the last ending of a domain is called the top level domain. And then they would say, oh, you want to go to yahoo.com. Then they're going to give you the IP address of this server, which knows about yahoo.com. And then you connect to them. And they will contact you and will tell you what's the domain name server of yahoo. And then finally that server will reply you with the answer, and that's the IP address. So basically the DNS system from a helicopter point of view is very simple. You have a name and you want to have an IP address. How that's being done, that could vary in sophistication. Okay, that brings me to the next part, which is Internet Governance. And I wonder if someone of you wants to talk about that a bit. Just to be reminded, let's keep that short because I really want to do the second part. So as I promised, we'll have here Alex Semenyaka that will talk about some of the slides. So we'll try to keep it short. How do you do this? It is complicated that you can erase. Actually, the usual question is how to internet who runs an important fashion. But unfortunately, there is no one quality who can run internet. You know, who will decide everything. What are the different stakeholders? Each and every stakeholders play its own role. And actually the stakeholder is important role because for the processes in internet, we usually use the term Mark is a holder and Mark is the holder of George. We have a lot of different players who are involved in this process. And they act together to gain some decisions. So we talked about IP addresses, about DNS systems. So I make names. And there are two basic resources in internet. It might not be different. But in all cases, we have one very simple demand. Same name, same address. Should be used only by one part. That's why postal address. If you have your postal address, you don't want it to be duplicated in some other place. So it should be unique on the whole road. Same demand we have for IP address. Same demand we have for the maintenance. So it's important to ensure that these things are unique. So we have to have some top level organization who is responsible for these movements. And yes, this organization was formed about 30 years ago. And it's all I can. Here we go. On this slide, I'm really not sure that any of you can agree with that. But it's I can. Actually, I can just add a bit. This slide is, especially here, to understand that it is complicated. You cannot read all of this. And you cannot, if you see the here, a lot of people and organizations. We cannot go through all these people and organizations and nobody knows the internet. Everybody knows that. That's the issue. But I can, this is not the organization which makes, you know, some specific decision who will be used to be suppressed, who will be used to be smitten, or something like that. No. It's very, you know, specific. Okay, okay, okay. Yes, this is the organization who, you know, make very top level job. And it just distributes names and hardware services for large and it is called registers. Both for my names. You just get it. I will not describe too much about my names. Because I don't concentrate on IP addresses. So basically the system is like this. I can have a global IP address of free IP addresses. That's important. They have only free IP addresses. And they distribute the most free IP addresses or five parts on the road. Can you switch to the map? Yes. This one? Yeah. And there are five so-called regional internet registers in the road. Corresponsible each for its own geographical region. It's every American register of internet numbers. Latin. That's Latin America. Afrinik. The youngest one. Epinik. Which is responsible for Asia Pacific region and Australia. And Rai PCC. Which is responsible for this large geographical region. Consists of 706 countries. So this is the largest area. And we represent exactly this organization. How things go further? Yes. Rai PCC obtains addresses for ACAM. And then this is over so-called ACAM. ACAM is regional. Regional internet registers. And it distributes addresses for local internet registers. And those parts are members of Rai PCC. And this is important. That's how it works. I mean that it's not Rai PCC who managed those organizations who create some artificial leaves and tell what's right and what's wrong. No. Our members and local internet registers are our members. And from the point of view of the law 17 or 18 already. They They are calling us. They decide what the rules will be. This is very important. All rules concerning organization, administration and money issues. All those rules are created by our members. We are organization who deploy those rules. We do not invent them. We just deploy them. This is very important. So this large community and this number is audited now it's 18,000. 80,000 members. They decide which rules will be applied for our region's 76 countries. And of course we have members from Armenia as well. And Armenian members also decide as well. But this is the only one part. I told you if you were careful you noticed that I talked about organization other than this. I asked for a presentation to put that there on material. Say a word about technical questions. And that's one the most appropriate one. Okay, that's one. Non-profit organization. Here we have this group. What does it mean? In parallel to that organization brightness is non-governmental association registered in Netherlands. And so on, so on, so on. It's back in town with a global management director which is a father, great mom and so on. We also have an old Arab community. An Arab community is formed from people, not from organization. It's formed from persons from from youths. Alright, this is with us. And threat immunity is huge. Every one of you can join the Arab community just by registering in one mailing list and it's free. It's simply to open no obstacle known exams or something. Anybody who is interested in the topic can join me on the list and start working immediately. And this is right community in form of several working groups which decides what technical rules will be. And we will be in my position to deploy that's how it works. So, there are two parts due to kinds of community. One community is community of organization LARs. Basically, half of them are internet operators but about 30% right now that's large large companies but I speak as whole bunch of stakeholders and second kind of community that's community of personal persons of individuals who are deciding which technical rules how they should be developed and who develop those technical policies. And that's why we have this in our official registration documents. And that means that we are more than just organization who is responsible for distribution. That's community who told us to do some additional resources for some additional projects. We want to have some pro's of the group. It was a decision of community and Christian will cover that. I will cover that. But it was decision by community not by us or we do want resource certification. Okay? We started a project community so now community has this certification that was decision by community and so on all those things are performed according to the rule of the community. So talking about Internet Governance it's very important to understand that it's not ruled by some special party, special organization or something. It's ruled by basically everybody and there are a lot of interests and there is a system to take it down all of those interests and people produce the best results for all of them. So probably that's it for now. Thank you. Thank you Alex. Thank you. So actually I just told us about right by NCC and right by community itself a lot of other organizations as you have seen from that first slide and some of these organizations just not go and describe all of them some of these organizations are called some of the I-STAR organizations and they are implementing different functions and all of you depending on what is your interest and if you're involved in these I-STAR organizations you can take this presentation and go through all these organizations occur. It's about I-Q and AM and O etc. I-STAR also has a very important function here in our community. I-ATF Engineering Task Force also has meetings which is implementing and designing the standards. So you can take it upwards and just go through all this presentation all these I-STAR organizations but we will be more we will be more concentrated on technical issues so how I-OT changed the internet I guess you have heard about I-OT internet of things how it will change Yes, okay by the way, do you hear me? The microphone is not perfect, very good thank you very much for this contribution so right now how do I get your attention back we were talking about I-OT so you all maybe you don't know about I-OT but I-OT is as with many buzzwords like cloud and what else traffic data, machine learning and all that stuff it's nothing new but it describes a certain situation so since we already said that we have a lot of people on the internet and everyone has mobile devices we get more and more of these devices on the internet and that is kind of going in the direction of what internet of things is so it kind of describes that the internet is not just on our laptop but it's around us and I think we're going to have later on a presentation that goes directly into I-OT and describes that in more details but what I would like to talk about right now is some use cases and for that I picked out a company and this company provides sensors is kind of the foundation of internet of things I mean imagine that you have sensors all around you and all of these sensors to make it possible that we have these sensors I mean you would not take a laptop and install it on on your wall just to measure the temperature they have four parts in common first of all they are small and the computers are getting smaller and smaller then they are getting cheaper and then they don't use a lot of battery and that basically enables a lot of different use cases and I just want to read you the list of use cases that this company Libelium has on their portfolio and with a lot of buzzwords you need to also work on the PR side so the people studying business they know that and then you don't have agriculture but you have smart agriculture you don't have animal farming but you have smart animal farming you have smart cities, you have smart environment, you have smart water, you have smart metering, you have smart security and so on and pick out one of the use cases then I think last time I talked about wildfires, I think I am going to do that again because wildfires are a problem in Armenia as I heard is that true? no? thank you very much or otherwise just provide me with another problem and we try to solve it with IoT so the thing is that right now you have a huge forest and the problem with wildfires is that you notice them when almost half of the forest is already burned down right? so what you want to achieve is you want to identify that there is a fire starting as soon as possible so what can you do? you can develop these sensors smart sensors that are small that don't use a lot of energy and they are super cheap right? you can fly with a helicopter or with a plane over the forest and you just distribute a lot of these sensors what would happen next? so it is summer either someone forgets a cigarette in the forest or you know it gets very warm there and then suddenly a flame starts and then it starts to grow bigger and bigger and then a sensor notices that there is the increase in temperature and this information could be sent to the fire fighter immediately and they know there is a fire starting and then the fire is still small and they can fight it with two people right? I mean this is much easier as if you have already half of the forest burning and it is going in this direction so you will see in the next few years that a lot of things around you will get smarter could be the fridge, could be the toaster could be the coffee machine and so on yes? yes so it is basically a micro-internet and as we said the internet is decentralized that is from the history that it was created to send a nuclear strike so you are going to take out one note and the system will reconfigure itself and everything will be still fine it is with IoT the same so it is decentralized you can destroy one sensor in a way but there is one thing that is important with IoT or at least at this moment and I would say that it is as important as the move to IPv6 because if we don't move to IPv6 you will not have internet this problem is similar important and what we can do is we can give out a price for the person that guesses it I mean you are of course exempt so we said that IoT has these three traits it is small it is cheap and it doesn't use a lot of battery what could be a fourth one so what do you think is come again time yes you study philosophy that is why you are going to say time it is a very interesting aspect but it is not what we were looking for so you have 50% chance for the next price so what do you think could it be now of course you can't say it I mean that would destroy the entire fun right so please absolutely you know sometimes I am surprised you study computer science right yes data science interesting cool data science data science cool then the next part of the presentation will be very interesting for you anyway the problem with IoT is security because they are produced very cheap security is not really on their list from the producers so you might have heard about that last year there was a DDoS I know you know what a DDoS is but the rest can you shortly describe what a DDoS is yes perfect so basically in simpler words you have a service like Netflix and Netflix sees per day 1000 users and suddenly 100,000 users try to access this service and then Netflix can't handle that anymore and then nobody can access Netflix anymore and that basically happened so I need some activity anyway so that happened last year with Facebook and Twitter yes with Facebook and Twitter so there was a day when on the east coast first Facebook and Twitter didn't work anymore and then on the west coast and that was for a couple of hours and you can imagine that Facebook had a lot of money with so many people on Facebook with advertisement and if they are not available for two or three hours then we are talking about millions of money dollars that are being lost and that DDoS attack was originated by these home smart home devices so your surveillance camera if you are going to buy it they have the same password on every device that is being bought the default password and as a responsible user of these devices you are not going to change the password so you just install it that was right now a joke you should change the password but a lot of people did not do that so what hackers did they just gained the internet for these devices and then they tried to hack it so they were in control of that I mean it is bad that they are going to see what you are going to do at home without you knowing that but it was also very bad that they could control a network of millions of these devices and they could orchestrate an attack against Dyn which is a DNS provider which provides Facebook and Twitter with DNS resolutions as we had before so for a couple of hours so that means that DNS, what we learned before was not working for Facebook and Twitter and hence it could not be reached because nobody remembered the IP addresses so that was the problem and then I think we can go to the interactive session we are running already 10 minutes over I would suggest that we are going to do that 10 minutes and Vahan will tell us about Rayse and the right fellowship in the next block because then we have an easy start so we do that right now I expect a bit of feedback from you so participation and then we deserve all a coffee break so as you all know the internet changed a lot of things and I give you some examples Google changed the way we search information before if you want to look up things that you didn't know you had to go to a library and with Google right now you can do that at home and much better than at a library then we have WhatsApp which changed the way we are going to communicate so group chats that was not possible before WhatsApp was there we were exploring images or coordinating meetings another thing is Google Maps that definitely changed the way we navigate right you are going to get to a new city you have no clue where to go you just use Google Maps and it will suggest you which restaurant you can go to we did that also yesterday when we tried to explore the night life apparently Sunday is not a good day exploring the night life anyway relevant to the university Coursera is changing the way we learn and we study and I also experienced that last year when I was studying because many of the content that we learned was via Coursera back then we had well the best machine learning expert of the Netherlands who was our teacher and he is a brilliant researcher but as a teacher he was really bad so what we did instead following his lecture we just went to Coursera and we followed the machine learning course from Andrew and G which is a really I can recommend that if you want to learn about machine learning and I think that in future every university needs to think about how they gonna deal with this challenge of online courses because otherwise I think they will go over time because the benefits of online courses is much higher then we have two local examples menu.almenia and Gigi so now I would like to know from you as a starter what kind of applications do you use regularly and what applications you can't live without anymore and then maybe if you wanna if we wanna challenge that I would like to hear from you what you think will be the next big change on the internet or technology in general and we gonna give out prices for that of course so you can be very active in that yes please oh you already got a price you are greedy no I'm kidding please oh do you want a microphone no are you sure it feels very good I think I heard that from the other oh yeah in Ciamut Ciamut is that something where you can put a math formula and then it resolves it because I would call that cheating but it gives you like some problems that you can solve like it's better okay then it's going in the direction of Coursera right so that it will help you learning okay interesting yeah tonight we are completely bored then we gonna check out that app yeah this was again related to math it enables you to plot graphs not only two dimensional also three four dimensional graphs it is quite amazing yeah okay alright a lot of you are using those apps right okay this is all related to education is there something that you know would not fit in this yes please mobile banking yes that is true yeah do you think do you think that banks so brick and mortar banks will be obsolete in the future no why not ooh that's that's that's a very hot topic we gonna talk about that in the interactive session in the second part and yeah blockchain is I personally think the blockchain would you think that blockchain is a threat for banking some maybe some some disease will be why do you think that do you think because bitcoin is much cheaper it's more secure and many other reasons there is no third party for transactions for paints or other where do we have the economic students in the room actually I am a financier okay after graduating I now study I am an auto-didact I now study lead sense I will continue directly it will get cheaper that's for sure because blockchain technology is a decentralized public lecture you know what a lecture is you know that right so you have a business and you need to keep what's coming in what's going out because otherwise the text office will have a talk with you and that's a lecture book that keeps the transactions both sides in a way and blockchain technology is considered as an electronic lecture safe I mean it's safe against tempering so nobody can change something without someone not seeing it right that's the benefit of the blockchain and I think that banks will benefit from the blockchain instead of seeing as a threat because at this moment doing all these transactions between various customers is very cost intensive right you have systems that need to be keeping copies of what happened because I mean data loss is possible with computer systems and losing this data is not a good idea at all because I mean if you're going to look at your savings account and sorry we don't really know what you had on your saving account so let's start from zero again right that is not good but with the blockchain they can have that in a public space so they have a lecture so a database where they're going to write down oh you have that amount in your saving account but the interesting thing about the blockchain compared to other database technologies you can just share this data with everyone because nobody can change it without noticing right and that will make the operation of banks much cheaper and you might have heard about Ripple I mean there are more than 800 of these cryptocurrencies every computer scientist or student that you know start on a weekend they're going to create a new cryptocurrency and then find someone who is stupid enough to buy it so Ripple is being backed by Google Google is very much interested in that and the idea is that Ripple will help banks to keep all these transactions coordinated it could of course be that Google will enter the market of fintech companies and in the long run you will not go to your bank anymore but Google will manage what you're going to have in a way ok the price goes to Christian no that's not true so let me ask you should we go into a break yeah ok so Vahan if I understand it correctly we're going to meet in the cafeteria and for the ones that won a prize they get coffee for free and the rest as well right yes we can do it but we can also have some prizes so we can just award them with the prizes after the launch ok so Vahan what I suggested it right now we're going to have a break and then you're going to start with the fellowship and with the racing program ok alright good do you want guys now to have a rest 15 minutes 10 minutes 10 minutes ok 10 minutes and we're coming back here thank you very much I'm presenting Shraab Technology we are a local Armenian company and we are developing IoT systems current active project that we are working is an IoT systems don't do with physics institutions do not this is the stabilization system for CERN Switzerland so I will talk about very quickly because most of you know what is IoT what is the driving forces for IoT because its sensor technology was developed very quickly for the last few years this is because of the self-driving cars the self-driving cars was boost the technology the cellular technology the industry and a lot of investments were done in this area so this is the first thing as a driving force the second is cheap devices and miniaturizations of the devices then a low-power design and low-power devices which is the traditional that's a TTL logic 3.3 volt logic replaced with 1.2 now there are devices which are working on 0.1 volt so low voltage, low consumption and very small batteries allow this technology to grow the capability of mobile devices this is the next generation 5G that is coming which is creating a new environment because IoT systems will not work on 3G or 4G one tower can serve only a limited number of clients let's say 1,000 clients dedicated 16 channels for mobile devices but for mass sensor technologies mass connectivity you need millions of connections and one tower should support 10,000 100,000 parallel connectivity this is coming with 5G then the latency 1 millisecond of latency this is a requirement for these kind of systems and another thing is the bandwidth one gigabit per second bandwidth is necessary for 5G so these are all technologies that are driving forces and also the power of the cloud that is growing very light these are some examples of the tiny and very cheap 1.2 dollars you can get 1 dollar even if you buy 100 these devices and the mobile capability all these things Google platform, Amazon, web services this cloud services are allowing you to boost this technology very quickly and connectivity communication is bringing a new way of communicating the devices are becoming smart not only the traditional devices that we know the shoes, the table chair, pad anything can be connected the first question could be why to connect this you can have a homework to think about anything surrounding us why to connect to the internet what kind of capability, what kind of sensors what kind of application to develop to make it connectivity and the system is required the scalability because you need to develop a system not to customize everything not to develop some system that is particular for this kind of product you need to develop a system that is scalable easily you can scale up if you change the application another thing is the security and privacy because the traditional way of security is not working you are opening holes into the network because these cheap devices you cannot the traditional way to put firewalls or expensive devices or UTM devices threat management systems to protect this system this is the advantage of this thing is that you are using a very cheap technology and you cannot go for the traditional method to protect from one single point of entry of your network these are holes that you are opening the security holes into your network and you need to protect differently and your local network which was a trusted zone in terms of networking this is becoming untrusted so anything is untrusted the trusted zone GMZ zone untrusted zone all are untrusted and this is the hype by Gartner that you can see internet of things is in top this is 64 billion devices will be registered in 2020 and silicon industry machine to machine learning communication and big data all these things are investing billions of dollars and where is the IoT is used? Everyone any area that you may think of the facility utility, energy sector safety, security retail, business, banking industry, everything so it's everywhere and this is a nice slide about the segmentation of different types of industrial sectors that IoT is entering into this is the system flow data flow in IoT system where we are having let's say controllers, microcontrollers for example in our case we've been using our Cortex family products let's say Linux based small microcontrollers where you are connecting the sensors and different types of sensors and actuators you are connecting to this embedded controller this is all devices like the computer and then it is collecting the data or providing the API commands and then collecting the data into the gateway and through the gateway you are sending this data into central servers which we call cloud server and then getting the commands from there like controlling the actuators that you are having so this is the overall system and this kind of system if you deploy this kind of system if you have the firmware here, the sensors here the sensors can be very different they may work in different protocols like IPA CSPI VEA and then you have all the API commands then this infrastructure you may deploy to any type of application so that's what we are calling the scalability so once you develop this kind of system you can adopt this for your own needs these are some of the community different types of workshops and conferences that are dealing with this kind of system and this is different types of machine to machine communication, wireless and servers IPv6, Bluetooth, ZP Wi-Fi, LTE all these things and this is the technology that you are using with the PCPIP layer then transport layer then application layer where you are using the REST API to communicate with there are also on the HTTP level you may have now there is a new protocol which is called NIPP on HTTP based protocol you are communicating with the central server directly from the system without having any intermediary device or gateway and then IoT application visualization systems etc and also read data on top of that these are devices and platforms and read data analysis and it is very simple that when you are dealing with a big amount of data like in Wi-Fi sensors how to analyze that and also the technology challenge you can see that the environment is different the sensors are different the variety of different types of devices and technologies and it brings up to the main question of convergence what I think is convert convert means you have sensor technology embedded controllers software development website development big data analysis machine learning all these things are coming together as a convert technology to deal with IoT to deal with IoT so many different types of expertise so that you can have all this together to bring up the system in a working condition and this technology also brings some challenges this violates single point of entry opens three walls in the meantime you don't want to bring up the price of your system so that's why you need to rely on low power of computation low power of devices in terms of power consumption so that this system will be very scalable and you will easily distribute thousands of sensors anywhere you can put because they are cheap and then convergence technology brings all together because everything is about man humanity right man currently if you calculate the bandwidth and what each of us bandwidth will consume the bandwidth is our consumption of sky video, our telephone, our voice anything that we communicate these are intermediary devices that is fulfilling our desire to communicate so the final consumer is not this device but the man itself so we are a convergence technology we are consuming totality of some kind of data bandwidth that is necessary for that for the person depending on what kind of all communication surrounds us and this is IoT big data and cloud also is getting converged because these are all distinct separate technologies there are many companies that are working only in particular cloud area big data area in further development area but now to break up the system you need to deal with all of that so this is the overall convergence and I will skip this case study because Christian has gone through some cases this is smart city case Madrid city environment intelligence this is Germany smart village this is Sahara and education we are having this presentation because one purpose there are a lot of opportunities in this area research interest and if anyone will be interested in research in IoT systems we are welcome you can contact us and there are many opportunities because it is a so different variety of interest one could be an electrical engineer work only in the area of sensor development or sensor technology or embedded controller design etc another one could be a computer science specialist working on software development another one red design another one data analysis etc so it is a growing area it will blow all over the world because this already a big blow has happened because of this the self-driving car and this technology accumulated so much resources that it will be used many many other areas as well and lastly I will talk about security there are several things the confidentiality that you need to maintain when you are developing the system these are possible attacks the effect of the service and the little attack the integrity when your data is corrupting or not complete and the tiny devices can be served as zombies to easily develop those attacks and this is the cost of hacking anyone who is hacking the system they can work in different areas one is the communication attack this is the most cheapest one because you will need to write a code etc software attack is more complicated where you are hacking the software system and the hardware attack where you have to go into the hardware itself and the hardware attack is most complicated where you can access let's say through JTAC interface through the debugging interfaces and do something in the hardware so in conclusion we just have an opportunity about IoT system development you can contact us thank you very much I think you just leave that ok thank you very much that means that we are going back to where we stopped before and right now it's already 10 past 6 all of you I think know that it's two and a half hour workshop started at 4 theoretically we have 20 minutes left I think we will push it a little bit further so it will take a little bit longer but we do all our best to make it as short as possible and also as interesting as possible so you are going to have 5 minutes for presenting Racy and the Ripe Fellowship actually this part is quite interesting for you because this is the place where there are possibilities we get involved in our Ripe system so Racy is the Ripe technical corporation we should do that if we are around with academic institutions so the participants from these institutions but they can be students or searchers, academics they can apply if they have any research for any work of interest in these areas but not only limited so it's network measurement and analysis IPv6 deployment network security, internet governance security and internet of things but it is not limited remember each and every topic that might be interesting for our members and our community can be represented and can be applied so if you apply with your search and with your topic and with Racy problem on your complimentary tickets travel, accommodation and support your participation in our events different events it is Ripe Meeting that is twice a year it is ENOG, Eurasian Network Co-operators Group ENOG, Middle East Network Co-operators Group and SEE Southeast Europe so and Ripe Fellowship is for anybody else if you are an academic it can be something like a police officer that has good ideas and good research in this area so go ahead and the deadlines and the applications so we have ENOG this year it will be in Terran it will be 25, 26 and application deadline is 25 February 2018 you are welcome to apply it will be in Marcel for March deadline and it will be in May ENOG Meeting will be in Moscow by the way we have ENOG Meeting in Armenia who knows that great ENOG Meeting is quite large and this time it will be in Moscow in June, good time and SEE in June again what can you what is a good thing that you have and others might not have these you can apply with your ideas with your topics with each and every thing you want and I will advise you because sometimes I am based in Armenia so you can just find me on Facebook you can write me I have such an idea and I want to participate in the Pride Meeting or ENOG Meeting etc and I will guide you all through the process and will help you to understand how to fill the application form and how to be successful so basically that means that whatever field you are in if it is interesting for the internet community and a lot of fields are interesting for the internet community you are going to put some work into that you are going to present it at one of these Pride meetings the presentation is around 20 minutes with 5 minutes of questions and answers and then you get the rest of the time completely paid the flight is paid, the accommodation is paid the entire stay is paid and you will get to know a lot of very interesting people so I highly recommend that great companies, great people fun, we had one question in general internet everything that is related to the internet could be interesting, I give you some examples Alex just a moment I give you some examples what we had as a focus point in the last Pride meetings that was legal implications of the internet you might have heard about the data protection law so the GDPR so that means that the European Union is really stepping on the toes of big companies that they should not use personal data in the wrong way because fundamentally there is a legislative difference between Europe and the US in Europe personal data so your data belongs to you whereas in the US this data that is being collected by big companies belongs to the company that is a very big challenge not only for the technology sector but also for the lawyers that are the only thing you need to understand it is not only for tech guys it is also for lawyers, it is also for even journalists etc. if you find because internet is everywhere now and it can be very interesting also for our members if we have a good presentation and good understanding what do you want to do for our members and without our members you might not be awarded a possibility to make a presentation because it is a competitive process but it can be awarded with the possibility to be at this meeting and be represented there without presentation and we want to encourage other fields than technology right because I mean imagine the right meeting is a meeting with 600 highly technical people and we want to get new ideas into this process and I think it's a win-win situation for both parties but we know meetings we have founded so you can see that there are lawyers there are feminists we have such cases so you are welcome to apply and we will go to the next yeah, it is okay I would say it provides the basic assistance for successful candidates it's about right fellowship it is for all other people that are not academics that are not students and that are not academics and lectures etc each and every person can apply so that's good perfect, alright so welcome to the second part of the presentation and I'm just remembering or help me to remember Ellen, did we present that on Saturday? okay alright and this is a premiere for you for you as well and I'm completely aware that I'm standing between the end of your evening and the end of the presentation so what I'm going to do, what I suggest to you is I'm going to quickly explain you what you can expect in these topics and then I let you choose in which field we should go in more depth and the other one that's less interesting for you we just skip so the first one is internet measurement data sets this is something that I highly recommend because it's about data sets that help you to understand the internet the next one is measurement networks so the right BNCC runs networks that collect certain data sets this is something where I could say that we can go a little bit quicker through because in the long run you can all look that up if you're really interested then we have big data at the right BNCC also a thing that I would say if you are not specifically interested in that then I'm going to go over that very quickly then the last one and that is a system that allows you to get all of the data that we're going to talk about in the first item this is also something that we can do rather quick examples of data analytics this is also something that I highly recommend because it's very interesting and then we have again an interactive session will the internet fundamentally change in the future we mentioned blockchain this is something that I would reserve 10 minutes so that we can talk about that a little bit so I would say that the first one we're going to do with the second one who wants to hear about that in more details I'm interested to hear about these things please okay Alex come again about the main topic yes, about the sub topics so right now internet measurement data sets we do definitely because otherwise the other ones don't make sense right BNCC measurement networks RISC and RIPE Atlas okay we'll get there patience yes, this is something that also in the light of presentation about RIPE Atlas I keep the second part short, big data at the RIPE NCC is anyone really interested in how we're going to deal with big data at the RIPE NCC yes, okay one person I'm going to mail you the slides if you're going to are you going to take notes of that okay perfect, well we quickly go through the slides if you're going to have a burning question then just you know, just get it somewhere to me, then the RIPE NCC RIPE is that also something we do quickly, then we're going to do data analytics and then to the last part, okay cool perfect so in general why are by the way do we have a clicker no that's not a clicker because otherwise I always have to walk back and forth anyway okay, I'm going to do it for me so why is it important that we're going to talk about data, measurement data in general and there's a simple reason and that is that you can't improve anything if you don't know what it is about I mean this is a fundamental principle in engineering and I think in basically every aspect in life and there's a simple method how to improve any system that you have and that is usually build measure and then learn and that is with internet measurement data it's completely the same as I said before you can't improve what you can't measure could anyone give me an example of that think of your field I mean philosophy is I think a little bit a tricky one but I think it's also related because we have the different philosophers I mean there was Hegel who came after Marx before Marx was inheriting the knowledge from Hegel alright and that was also kind of an improvement and I think you can measure that with the happiness of people and what is going on in the world or in the area of the universe first of all we should know what there is and then to see what type of problems there are we don't want to be able to overcome the problems because without knowing this exactly yeah I mean it's not possible that you measure the happiness of people how should you do that with IoT we will be able to do that because we all have this thing in our body that measures how many endorphins we have in our body right perfect directly correlated with happiness so then there are three different types of data sets when we are talking about internet measurement data sets if you're gonna know what they are about so if you grasp what they are about you basically yeah you are very far in terms of data science related to internet data we'll quickly go through them the first one is registry data registry data is something that is usually manually maintained and it's being collected to measure right it is administrative data but Gabriel will talk about that a bit more in details the next data set is passive measurement data and that is observed to measure um let me think of an example in the real world that um is comparable to that for example right now I observe to measure I observe how many people are in this room right um we're gonna go into more details about that and then the other one is active measurement data and for active measurement data you need to the name implies you need to actively do something to get the measurement results and that for example would mean that if you want to measure the speed, the average speed on a highway you need to send a policeman there and he's standing with the radar to measure it right the analogy might be not completely accurate but you get it in a way so the important thing is that you can't just observe it you really need to actively do something to measure the data and that gets us to the first one which is the registry data and Gabriel as a software developer of the right database will give us the pleasure to introduce this data set hello anybody know who is who is who who is who is the database of the and all the who is so my colleagues spoke a lot about the computers with numbers and ASNs in them and also Alex spoke about them who is the database saves all this data it has multiple types of data for the customers and we maintain that in our region and other regions have their own databases and they ask who owns these resources so there is some ISPs I assume in here there is also the university here so everybody who has an IP is registered to someone and when you search it you can go online and search on our website you can search that IP address and inside this resource there is multiple types of data one of them is the technical so for example if a network engineer want to speak with another network engineer of this IP he will search it up and contact if we have the technical we have the abuse the abuse is very important in case the IP address is doing something crazy then you contact the owner of this IP address and we have multiple use cases for it and one of them is Christian was explaining about the internet book which is the routing and that exists also in the database so the right database is one of the core applications that other applications like right start and other measurements comes to us to take the country information and more information of really who owns this resource I think you can give a really concrete example when you set up use I mean internet abuse could be hacking in a way because all of these people that are on the internet they are identified by an IP address but as a policeman you can't arrest the IP address right so you're going to arrest the person behind that and this data is exactly giving this information or someone who is providing software that provides the copyright of who owns this data set and this IP address and via that you can contact the ISPs and the ISPs as in each country regulations but that's for example Unicom does come to us and to the movies and everybody you can do it yourself if you like and you can also go to check what's your IP and then go check who owns this IP and how I'm here and that will make sense for you how you connect to the browser and through this kind of like a phone book of the internet we call it do you have any questions about the OSD no so you can yeah OSD is open source it's also an no you cannot the personal data is protected under the European law but you can query the personal data to a certain amount but you can query unlimitedly to know the routing of each of those data sets but you have this data set yes we do have it and the code is posted on github so if any of you is a Java software engineer who likes to help us I mean we accept any putting words or any other is it like secretaries secretary you mean like a secretary we have an API that you can many people build their own systems and they connect to us and grab information and one of them is the routers so also routers comes and says ok give me routing information and we give the routing information and there is more of this type of things so it's also it has a web interface that you can write in our web interface and you will actually see who owns it and you will see if it's right or other yes Evo the data of Europe is set we do merge also from the rest of the layers and we keep all the data so you can search vri r2 but you have to add only r so if you look at the interface but you could also use RIBSTAR which brings you all those information plus from when it is announced which has a bit more even deeper in the data so we are more our LIRs the colleagues from the LIRs come and maintain their data and there is multiple applications one of them RIBSTAR who grabs data from us we usually speak about those data sources but we are more like record for our clients and yes it is simple and when let's say if a company or two companies in Armenia merge together and they want to take the IPs to one those companies they contact us and then the RIBSTAR will merge them and make them under one IP let's say one LIR sorry I mean this data set is I mean from a data science perspective because I work in R&D very interesting because it is the only record that shows how many resources and we are talking about IP addresses and these ASMs numbers are being used in a country and I mean just think about that and if and that makes it I think very unique as well here we have some examples that is that's the country list for Georgia so there is an interface, I am going to show you where you can find it but right now just what you are going to see on this screenshot is all the ASMs that are registered in Georgia who do you think is very interested in that? programmers exactly, the government because that was two years ago I met some people from the government the cyber security department I mean I can't tell you which country but they were very delighted to see this data because finally they know what they have to monitor right sometimes it's very trivial but if you don't have this data you can't do anything right you want to create laws and that makes it very specific that legislation and technology need to work together because otherwise it's just chaos there are some countries in which this coordination between technology and law is just not there and they can't do anything about that some countries like child pornography is being hosted in countries where we don't have anywhere we as the internet community we have only small ways to change that and take that down right and I think a lot of people in the internet community are very busy to change this situation but of course it requires also stable government anyway then let's go to the next dataset which is the passive measurement dataset and you remember when we had this drawing with the numbers and these routing tables and what the routing data, the bgb routing data basically is all these phone books collected together because I told you before that if you're node number one and a certain machine is not in this book you can't reach it as an example Netflix is not there so you're at home you try to go to Netflix and if it's not in this book then you can't reach it this is a problem and that's why a lot of people, so the internet service provider is very busy of keeping this s up to date as possible and to measure that we have the routing data and the way we're going to measure it, we're going to collect these routing tables from as many people that provide this data to us and I'm going to show you some applications that we can do based on this data what is important is that in this data you would see if a cable connection between two machines just breaks then you would see a change in the routing table so right now we have a broken cable between 5 and 4 that means that all the connections that go from 5 to 4, well that was before here right 2, 5, 4, 2, 5, 2, 5 will and that is the characteristic of the internet that it will reconfigure itself it will change that the data flows this way and the routing protocol just makes sure that that will happen so nobody has to do manually something about that but the importance is that this would be reflected in a routing table so imagine that you have one month ago this cable was still active then you have in the routing data not the line that goes via 7 but as soon as this cable breaks then the data in the routing table would reflect that and then you are going to see all the data that flows from 5 to 3 that goes over 7 ok and the bgb routing data really ok alright we are going to make a mark see you so the routing data is all these phone books collected together and you can also visualize that then you would see something like that and let me explain that so in the middle lets assume it is facebook or netflix then netflix is connected to everyone else on the internet and if you are going to have access to netflix then you are sitting at one of these blue dots and if you are not sitting at one of these blue dots then you don't have access to netflix and this is a very important thing to know right because it enables you to draw results from this routing data and there are a couple of applications that were enabled with that then lets quickly talk about active measurement data in comparison to the passive measurement data I said before what do we have to do anyone? it's basically in the name so it's a super easy question for what is that? for this one you have to yeah it's all measurement data yeah yeah yeah but more I mean this passive measurement data is also collected measure serve it no that was passive measurement data what do we have to do with this one? hold on yeah was that what you wanted to say? yes hold on before you answer because I'm willing to give out two so was that what you wanted to say? and what would your answer be? for active one active one there should be kind of like in the case of passive measurement you just collect the data or just observe it in the case of active measurement you should do kind of actions also perfect that's enough okay perfect can I try? perfect good wonderful so I draw this graph there and what we have there is the following situation between one and two there is a broken cable and between six and five it's very long cable right? and what do we want to do? we want to do active measurements so we have to set an action let's say that every hour we want to send data from number four to one and six I think you can already assume what will happen these are the results so basically it says the data that we get out of that or the information or knowledge that the line between four and five works fine the line between five and six takes hs because it's a very long one right? do we have any questions there? yeah I think I heard something no? is that clear so far? okay perfect I liked when you're not synchronously anyway then we have five and two five and two is also okay and then the line between two and one is not working right? this is the thing that you get out of active measurement data what would we observe so just going back to slides what was something that we also observed in passive measurement data was it this one or this one? or both? yeah so the broken connection so this is a I don't know what that is for oh you are collecting it already yeah so it is important that each of these data sets can solve a different problem the routing data could tell me that the line between one and two is broken so if you are collecting in one of these blue dots then you don't have connection to Netflix whereas if you are going to sit on one of these blue dots you have connection to Netflix but then there could be another problem it could be something like that so that means that the data flows very slowly so you can't watch you can't enjoy the movie that's why we have active measurement data to solve different problems there are different examples of active measurement data just we are not going to go into details and there is an inherent problem with active measurement data because you constantly have to do it if you miss the chance to do something like that then it's gone it's over with passive measurement data we have a very good picture of what's going on on the internet so we know what's up and what's not but for example you represent Netflix and someone does a DDoS against Netflix we did not measure the connection to you every hour or so and something happened but I didn't measure that then I have no chance to see what's going on that's also important we need to have a lot of actions on the internet later on when we quickly talk about Atlas I show you the impact what that means for us running these networks sure yeah it's basically a DDoS you are absolutely right the way we are going to do active measurements is still small enough so that we are not affecting the system too much but you are absolutely right when you do active measurements you influence the system and the worst thing of active measurement is this one broadband data so I guess you have internet connection at home and your internet service provider tells you you buy internet with 20 megabits per second I mean just a number and you don't believe them because you said internet is slow, I pay a little bit more and then they give you 40 megabits per second and you think like no there is not really a change then you could do a broadband check where you are going to test what is the speed between me and the world and then you could get the result if it's what the internet service provider says or if it's not like that but on the other side if you are going to run a broadband test then you are trying to saturate full fill up your broadband and then every other connection that you do during that time will be super slow so there is a bit of a trade off because on the other side you also want to know it and for the people that are studying physics it's like the Heisenberg problem I mean you can't measure the location of a particle because you influence the system at the same time but do we have any physics here? it doesn't matter we have one here right? no? physics yeah okay cool I wanted to give you another example let me see what's the next one the next one is the measurement network maybe it helps if we're going to go back to this one in 2012 you remember that there was the Arab Spring yeah and the first country that it escalated was Libya imagine that's Libya and that's Egypt if you are a dictator you're in control of the communication systems now if you have our position so the people that don't like you they have no communication systems but they could use the internet and they did use that they used Facebook and WhatsApp to organize demonstrations and things like that and the dictator at that time was Gaddafi and what he did was he told the the only line that goes to the outside from the internet what he did was exactly that one so because he did that in that way that really showed up in the routing data so all the resources that remember the registry data belongs to Libya they suddenly disappeared from the routing table and then it was completely obvious something very dodgy and I can show you later on a tool where you can also see how the routing table for a country changes over time but the interesting point right now is completely a different one here we have Egypt and it was half a year later that the conflict swept over to Egypt and Mubarak at that time he had the same problem with Gaddafi but what he did was and I mean if it would not be that sad you could contribute that he was smarter he did something like that that means that on the routing table everything is fine we couldn't monitor that that there is something wrong and hence we didn't report it to anyone and but the problem was that because of the cable that goes to the outside he kind of throttled the bandwidth so basically that almost no data can flow through it's the same effect but very difficult to monitor for that you would have to have active measurement data which is very difficult to say like hey right now in Egypt there is a problem with the internet because for that you would have to set actions you would have to run these measurements to every network in Egypt and then it's coming back to the registry data right so you need to combine a lot of knowledge to really come to the conclusion right now someone is changing the internet in a country right anyway ok let's continue with so we do that very quickly because yeah we're still in time we run at the right BNCC so we already explained that no Alex, Alex explained that we as the right BNCC we are not only a registry our members decided that hey guys it's nice that you run the registry but we also want you to take more part in the development of the internet our mission statement is that we do something for the good of the internet so we're going to help to develop the internet and in this role we're going to run two measurement networks that collect exactly these two data sets that I described before I mean we already have the registry data we have RIS for the passive measurement data we have RIPE Atlas for the active measurement data RIS is very simple to collect what we're going to have is we have at the moment I think it's 20 servers all over the globe and they are usually at internet service providers at IXPs so where a lot of internet service providers come together and we ask them hey guys can you please share your routing table with us so the book right and they did it so we have more than 160 if full table doesn't say anything to you it doesn't matter but we have a lot of people that are sharing their view on the internet with us and this helped us to monitor situations like in Libya because it can be that one internet service provider doesn't have all the information right but as soon as 160 internet service providers don't see all the networks in Egypt or in Libya being on the internet anymore then it's for sure then we can be sure for sure and also any government we're going to share this information with they can also be sure that there's something going on right data science has a lot to do with trust right so how much do you trust the data it's about data quality what you're going to see here is ripe Atlas and I told you before that the problem with active measurement data is that if you don't do this action in time then you lose the chance to observe it right all of these small dots are measurement devices that we distributed to the internet community and recently we celebrated more than 10,000 active devices all over the internet and these devices can do couple of measurements I'm not going into the details of these measurements because I think right now it's not too important this network has more than 10,000 probes connected and we have almost 18,000 measurements parallel measurements running right this is against the problem that if we miss to measure something then it's gone this chance this number is still very low because imagine what I said before we have more than 52,000 networks on the internet but we are running only almost 18,000 measurements at a time right so that means there is a huge chunk of the internet that we just don't see right so everyone who uses this data also doesn't see this so it's important that we are going to grow this network and we are very active in doing that and if you wonder how these devices look like this is an example of how one of these probes look like and we are going to rely to connect that to the internet so this is about big data and I'm going to skip through that that's not so important I think right now to have access to this data because I mean it's one thing to collect this data but also to improve the internet and allow everyone to take benefit and advantage of this data we have RISC we have RIPE-STAT and RIPE-STAT is the open data platform of the RIPE-NCC and we provide not only data to the registry or the RISC or ATLAS there are a couple of other data sets as well so we have from MLAP these are broadband checks we have speed checker, we have geolocation data geolocation data is a very interesting data set it sounds very simple but it is in practice it is not simple at all so you have an IP address and you want to know where that's being used this is a very simple mapping but to get that right is I would say it is impossible to absolutely get it right but even to get it usable and accurate is also very difficult if you're going to come up with a solution that's accurate I guess you can easily sell that and be super rich then we have blacklist data that's I think not so important RIPE-STAT in itself is built in layers I think that's also not too important right now if you're going to go to stat.ripe.net then you would see something like that and you can enter so this is the input form you can enter an IP address an ASN, a domain name so right what we already went through and then you get additional information on that it's definitely outside of the scope of this workshop to go through all the different data sets that you would get in return but if you have a moment to look it up then please be my guest this might be interesting to see how many people are using that 2010 we started with RIPE-STAT and we had per day around 10,000 requests and the most recent number is that per day we get more than 55 million requests a day and that's coming from more than 1.5 million users this is an example if you're going to look up an IP range here you would see which ASN is announcing that IP range you're going to see the tier location of that so that IP range is predominantly used in Amsterdam then we also have registry information, Gabriel talked about that the widget API I'm not going into details about that it is so in a nutshell it is a user interface to all this data and we already saw the mapping of how these networks are connected so there is a tool where you can enter IP address and then you exactly see how it's being interconnected on the internet this is something that might be more difficult to understand but I just put it there because of the following one let me just quickly explain what the colors mean the colors they mean that if your IP space is being seen by a lot of these 160 peers so if your IP space is in a lot of these routing tables then it's green, if you're in less of these routing tables then it's more to the red so in essence it would mean that for that time in 2012 everything from Libya was red for a short period of time and we have someone in the community who is very geeky and what he did is so he controls certain IP addresses he played around with or they are announced on the internet so basically they are seen in the routing tables or they are not seen in the routing tables if you're an operator of IP space then you can easily do that and he was drawing the neon cat so that was rather funny and maybe also IPv4 space because we already talked about that IPv4 space is running out he was using IPv4 space for that to draw that and I mean imagine that there is a customer behind one of these IP addresses during the time when after their eyes are drawn there is no internet so I'm pretty sure there is no customer behind he was just playing around with that okay, as you see we talked about that so these Europol Interpol they constantly come to us and want to have information who owns what okay, this is the country statistics so what we did we just combined what we see in the registry so we're going to see like all these IP addresses belong to Armenia then what we're going to do is we're going to count how many IP addresses are being used in Armenia and that's what you're going to see there the blue one is I guess you know that, right? we talked about that IPv4 and IPv6, right? and it basically means this is the development of IPv6 prefixes in Armenia the green one is IPv6 and the blue one is IPv4 from this picture a quick interpretation that tells me already IPv6 is not relevant in Armenia Armenia is not alone with that we have many countries who just give a shit about but it's still 61% in comparison with other countries so it's still high I don't mean it there but they are announcing it in 61% of our colleagues here are announcing IPv6 so you're doing well Armenia is doing well and if you look at the wrapping which is also one of the leaders perfect wonderful well then you just have to work to get the blue line a little bit higher and the blue line a little bit lower yes a couple of explanations how technically are you? physics so in 2012 the world realized that we are running out of IPv4 addresses then how quick or not quick a country responds to that we're going to see that everyone panics we need IPv4 addresses and they are running to the registry to us to get IPv4 addresses you can see that they are in a way so you can see that a huge increase in there and then suddenly it dropped we as the registry we thought about ok if we're going to give out the last IPv4 space then in a month or two we don't have anything left and we decided well not we as Alex said the internet community decided that let's give companies that are not yet on the market a chance to use and participate on the internet so what they did is the RIPE NCC and all other areas they came up with policies and said like if someone wants to get IPv4 address space then we're not giving them a huge chunk of IP addresses but we can give everyone a little bit one right and this is what you're going to see right now there so the curve is slower because we give out chunks of 1000 IP addresses to everyone who registers so we make it much slower for someone to accumulate IPv4 addresses and on the other side you also see that there are more and more companies registering so that's why you still see an uptake on that and from an economic point of view it's very interesting because you know what happens to scarce resources right I mean just look at oil oil is not we don't have a lot of oil so what happens the price of oil goes up right same with IPv4 addresses so at around that time a second market developed so companies that have IPv4 addresses they said like hey guys we're going to sell it and we have a couple of examples of people that earned a lot of money by selling IPv4 addresses and I give you a kind of an idea of how much that is at this moment one IP address is being sold for 12 euros a piece right I mean kind of in this area right and now imagine that you would have you know basically in those days when you can get a lot of IP addresses if you would accumulate them right and then suddenly over here you're going to sell them right I mean it's like bitcoin right maybe as has anyone bitcoins no you have bitcoins I hope you already sold it no okay well then I hope you don't have too much alright is it late okay okay then quickly about examples of data analytics so with this huge amount of data that we have with artificial intelligence and with big data and machine learning we could create first of all the status of IPv4 in countries and what you're going to see there is this line is internet users per IP address so how many people are sharing an IP address and on the vertical axis you're going to see the population growth right and right now you can easily see that all the countries so all of these thoughts they represent countries countries in this quarter they have a problem because they already have a lot of people using one IP address and then their population growth and I think there was today an article in a magazine I don't remember that Eastern European countries for example they have a shrinking they are leading with a shrinking population so I would assume that they are either here or here right any questions to that down there your development IPv6 that you have seen so many graphs another thing that we could tell based on the data is which country influenced IPv6 positively per year and which country influenced it negatively and there are a lot of surprises in there so first of all in 2017 you see Germany there and there you have Costa Rica that already tells us that it's not leading economies that you know are good with IPv6 right and this is very interesting for the governments again to know on the other side the red ones they should think about their strategies this is again population growth and a number of internet users and this is the last one before we are going to go into the discussion this is a tool that we developed which is I mean a lot of governments they are burning to see that just on Friday we had a presentation Tbilisi and representatives of the government were there and they were super happy to see that this is a map that shows how the data flows within a country let me explain that we have these probes distributed right in Armenia we have a certain amount of probes and what we are going to do we let one probe send a data to another probe to another probe in a country what would you assume one probe is in the country and the other probe is in the country then we would assume that the data stays within the country absolutely so what you would assume would it stay in the same country or not and this could go for banking also because if you are doing something with internet banking you think ok I am going to the Armenian bank am I in the same country coming to Russia coming back to Armenia so what was the question actually we are asking what do they assume if it is going outside the country if it is going outside the country no no it should stay in the country it should but it is not ok so first of all this is based on tracer and theoretically there is a high chance that the way that data travels is a little bit different right so it is not ground truth there is room for interpretation and actually I removed the special effect because the special effects was that there is a burning wall of fire and then I realized that these special effects in power point are not the best ones so what you can see there is on the left one IPv4 and on the right one IPv6 in many countries you don't see anything for IPv6 but we see at least a little bit for Armenia so that is good then Alex is coming and I wonder why yeah suspicious so here we have Armenia and again the source and the destination of these data packages are in Armenia but what we see is that the IP addresses that are found on that way are in Frankfurt for example then we have Sofia I think that Vienna that is the Vienna internet exchange so that means that and right now I am not talking about data that you know is connecting to Facebook in the US or something like that this is data that you are going to send with in the country so for politicians no for the engineers in the country that is bad because it is not efficient because it takes much longer to send the data first to Europe and then back if we too interchange data right then on the other side sending the data over there is not cheap because you have to pay the data and then especially interesting for the government is that it is national security right because as a government you can decide what is happening within a country you can basically say ok internet service provider they need to keep a log when when you are on the internet and you they should do it for two months or something like that right but as soon as the data goes outside of the country they have no chance to know what is happening with this data right could theoretically be that the FBI controls that data point and they are going to collect a lot of data so they suddenly know what is happening within the country that is definitely not good ok and then I think we are at the end future outlook forget about that interactive session right now we are going to come to the couple of words yes we have guest speakers and you want to add something couple of words couple of words actually Christian told that we do care about this probe distribution and increasing the number of probes and that is very simple and you personally you are interested in the internet analytics of something can help both the whole internet community and your country because the hosting of probe is very simple thing you saw the device it is a very simple device an internet cable and usb for power usb is just for power you can get this probe absolutely free of charge from right by usual post service you can ask even is your room you can get it from me yes you can ask if you provide you with the probe or with the url where you can order the probe very simple you just plug it in and that is it and after that you can participate in this in the research because hosting a probe gives you so called credits and you spend those credits to create your own research at any of 10,000 probes in the world so you can participate it is very simple really simple it is free of charge so please join the team we have later on a presentation about ok perfect then let's run into let's run into the interactive session how will the internet change our lives in the future and to start with that and give you an idea we have the blockchain we talked a little bit about the blockchain we have internet of things we talked about internet of things this represents artificial intelligence this is something the last images representing local caches and if you dare to go into a bit more technical discussion then let me explain you what that means so you know that facebook youtube they produce most of the data that is being sent around on the internet and google facebook know that so what they do as a strategy right now is that they are not sitting in the US and you need to connect from Armenia to the US to get the data but what they are using is local caches so they create a kind of complex system that they are going to into Armenia so whenever you contact facebook then you don't have to send data to the US but you stay in the country I mean they already got it it's not efficient if you are going to send data outside of the country that has some implications that for example on Friday we shared some images from the workshop and I was on the VPN there is nothing else then it hides your connection on the local internet and it will connect to a remote location and then you are going to access the internet from there can be used to bypass geo blocking or something like that and what happened was that someone was putting images on facebook from Tbilisi and I tried to download them and it took a long long time and that was because I was I was using a different local cache I mean ages it took longer than for someone who was directly connected on the local network and that is also not too important but what's important is that the internet became popular because it was this interconnection between all these networks because everyone was sharing information and the threat theoretically could be that right now if these local caches are in your network that the connections to the outside of your network are not that necessary anymore so that means that nobody is paying for that and that means that that's a bit related also with net neutrality you might have heard about that that in the long run let's say in 20 years that if you're going to go to google or facebook it's super fast but as soon as you want to go to Wikipedia it's super slow right that is a threat that at least the internet community is right now fighting against that at the moment it looks a bit like that you know that the internet organizations they put undersea cables between the continents to connect everything right and since a couple of time so since a few years it's not a telecom operator anymore that's laying out new cables it's facebook and google and that's already telling us something right so anyway let's start with a positive outlook into how the internet will change our future these large companies have separate cables yeah they do they have the money for that because that's a very business oriented question so imagine that facebook started out in the US and if you as a customer of facebook want to connect to them the question is who is paying for the internet connectivity in between right I mean for the last mile it's your internet service provider because you pay for that but there is also the discussion that the internet service providers they say like hey everyone is using google, facebook youtube nobody is giving me money for that right so let's see how we can change that right because these internet organization like facebook pool the big ones they are very smart they they know that people use it because of the content right and that is I think also a problem for the mobile operators because the mobile operators who is creating the mobile phones it's not the mobile operator right it's apple you know Samsung and that who is creating the reason why you have a smartphone why do you have a smartphone exactly so I mean in the long run why do you pay your mobile provider because there are more and more of these wifi networks right if you are in a city that's it's very dense where a lot of wifi hotspots are you don't need telecom or beeline or something like that so why do you have to pay them so that means that in the long run you will not pay them anymore and then they will die right and the same thing could happen with these interconnections between and that's the problem in a way right so I think at this moment we are in very interesting times what will happen in the future with the internet also net neutrality is a very interesting one so what do you think I mean I would like to pick up the question about does technology provide us with more time or with less time what do you think about that okay why is that okay I mean we would I would argue that if you have okay that means that so basically okay if if I'm gonna summarize that would mean that technology helps us to save time right because you said you can do things faster but on the other side we also do more and that's why we have less time in a way right what do you think about that is it a good or a bad thing okay it's called procrastination right so the thing is that we will have to learn more and more in the future because the economies they require that right 40 years ago if you didn't study okay that was fine meanwhile you have to study to be able to work what we also see and that goes a bit towards machine learning is that the barrier of you finding a job increases with the more knowledge you have right the more knowledge you have you're safer to not lose your job artificial intelligence right now is increasing that bar right so that means over the next few years we will see more and more jobs being replaced by robots and if the development goes on like now then this might be a wild guess but in 30 years we might not be required to work anymore because the machine can do it better than we do but this is quite a negative outlook right give me something positive how does the internet change our lives in a positive way the positive part of artificial intelligence is that we can concentrate on creative part and escape the routine so we can do better okay this have you seen the what is it the Van Gogh project where a machine was taking a random picture so you take a picture then you put it into that machine and then it creates an image that looks like if Van Gogh would have drawn it I don't know I don't know the exact name but there are a couple of these projects and at this moment they made a test between if human beings can distinguish the one that was created by a computer and one that was created by the real artist and you can assume that that was not a clear cut right so I would say that maybe we leave the creative part also to the computer I think that in 30 years we all have you know we can choose if we want to work or not but we don't have to right yes what should we do then and I think enjoy life okay I think that in the future when we solve the energy crisis right that will come with solar panels or something similar right because we produce solar panels at a certain moment we have enough solar panels that we have enough energy that we don't need to care about that and then energy will be for free and then we're going to add artificial intelligence and then we can be really lazy and then I think it's the time of philosophy because you need to find a reason why you you know stay alive exactly yeah yeah yeah you learn how to explore oh you have nothing to do use that a new app no something I think we're going to be quite late and that means that we're going to go to the last slide which is questions and answers well if you're going to have any questions then please contact me yes if you're going to have answers especially if you have answers then please contact me and then we're going to conclude it thank you very much for the patience that you brought up to still stay here we lost I think 40% which is still a good ratio and then we have two presentations still up and yeah let's just do that thank you very much