The following are the outputs of the real-time captioning taken during the Thirteenth Annual Meeting of the Internet Governance Forum (IGF) in Paris, France, from 12 to 14 November 2018. Although it is largely accurate, in some cases it may be incomplete or inaccurate due to inaudible passages or transcription errors. It is posted as an aid to understanding the proceedings at the event, but should not be treated as an authoritative record.
***
Good morning, everyone. Welcome to Room III of the ICT. I'm Félix Blanc from the Internet Without Borders and we're very happy to have you here. And I would like first to thank all the participants and who accepted to our invitation to come to Paris to discuss this topic, submarine cables and sustainable development goals. I would like also to thank Access Now, our partner, and IGF and UNESCO for hosting us. The topic of this session is, as I said, is the link between submarine cables and sustainable development goals. It's maybe it's not obvious at first but we will investigate and (no audio ‑ technical difficulties).
The first is conflict prevention, conflict prevention because, as you know, we are celebrating the century of the first World War, the first World War, one of the first things done by the different axis was to cut telegraph cables, it was a consequence of the war. In Cameroon, the Cameroonian government cut the cables, to cut off access to the region, maybe, of course, it was to be studied but maybe some link with the conflict which is now going on and degenerating into a kind of a cyber war. This is the first point. The other one, of course, is the digital divide and the access to backbone and to reduce the digital divide access to countries, landlocked countries such as Chad, for instance, in central Africa, solution who don't have access to submarine cables also at the global level, implications of new generation cables like the (?) cables or other cables that we will discuss today that are very high capacity up to sometimes 100 ‑‑ 100 times higher capacities than the previous generations so what would be the implication of these new cables for digital divide. Last point, we are UNESCO, UNESCO and the ITU had an initiative in 2010 about monitoring climate change and geo(?) and so on and they proposed, the community proposed cable industries to implement sensors to prevent or to monitor climate change in submarine cables to have a network of sensors. So this initiative might not be relevant from a business perspective or for security issues but it's ‑‑ we wanted to do this kind of ‑‑ to discuss also this initiative even if ‑‑ there are many other initiatives that can help prevent climate change and this kind of infrastructure can be helpful to reach this goal which is, of course, very important in SDG and it's important and more important nowadays. How submarine cables meet the SDG to justify the title of this panel, and the second one was what a reasonable success shall we promote to success to regional gateways and (?), what kind is needed to reach these goals and the last question which are the backbone regulation, not only international submarine cables but also back bones in ‑‑ continental back bones, for instance, which regulation can provide conditions for a more peaceful and inclusive Internet.
So we ‑‑ each participant will have up to let's say seven, eight minutes, maximum 10, and then the red light and just a suggestion that each of you in your remarks make some ‑‑ speak about the good practices that you have noticed in your research or in your business activities because we have business sector here and the economic sector and also the Civil Society sector so 10 things are here. I will give the floor to Florence who will moderate this panel.
>> MODERATOR: Good morning, everybody, bonjour, my name is Florence Poznanski, I'm with Internet Without Borders. I'm going to introduce each panelist. We have a big panel with lots of people coming from a diversity of continents in the world. And the idea that after each presentation, we will give you the possibility of make some questions, and before that our colleague, Peter Micek who is from Access Now, he's codirector, he's going to make a small provocation and after we are going to be able to talk about that together.
Our first panelist is Robert Pepper, he wants that we call him pepper. He has a connectivity of planning at Facebook. Pepper, what is the strategy of Facebook about Internet cable and how it can promote human rights and network influence.
>> ROBERT PEPPER: Thank you, Florence, and thank you very much for inviting me to be in the workshop when we made the proposal, but also I want to thank the French hosts because it's not just about the free flow of information, this morning we've seen the free flow of coffee, and this has been really important.
So I'll come to ‑‑ in a moment to where Facebook fits into this and you're probably wondering why is Facebook talking about submarine cables. Actually, let me ask, how many people here already know that we are coinvesting and working with ‑‑ we're actually building out globally submarine cables, about a third of the people, OK, that's good. That's a higher percentage than usual when I asked the question.
So when I think about submarine cables for the SDGs particularly for all the things that Félix mentioned but also connecting the unconnected which is closing the digital divide and having people to benefit from the access and the use of the Internet. I think about submarine cables, I mean, literally they're invisible because they're subsea but as part of a global ecosystem of increasing global band width. It's the submarine cables that are global and regional and the regional sun sea cables are extremely important because you want to keep traffic within regions because it lowers cost, improves performance and latency, it includes opening up landing stations so you can have a competitive subsea market coming into countries that actually attracts the types of investment that are needed to increase the global capacity. You need open Internet exchange points, particularly when we're talking about regions such as subSaharan Africa so when you have landlocked countries the question is how do you keep the traffic within the region without having to route out. You need national back bones and back haul. One of the things that we have found is ‑‑ in especially the emerging markets, the way people access the Internet is through the mobile Internet, and operators, mobile operators, cannot scale, cannot provide 3G or 4G which is needed, too, actually have mobile broadband without having fiber back haul, so that's another component. You cannot do real mobile Internet on 2G. It doesn't work. And what prevents these operators from 2G to 3G or 4G is the lack of fiber back to their cell site, their tower, you can't do it in narrow band microwave. It also includes having points of presence distributed globally that reduces the, again, cost of global traffic and also lowers latency and then there's caching.
So if you take a look at all of these pieces, right, the sub cables are the necessary first global piece of this, but it's not sufficient by itself. So it's the necessary but not sufficient first step. And that's ‑‑ and in order to be able to have a global Internet and it's not just improving the performance for the 3.8 billion people on the Internet, it's also extending the networks so that the 3.8 billion people who are not yet connected can be connected. So it is about closing digital divide. And that actually then, leads to ‑‑ I'll just ‑‑ two other things: There have been some key decisions that opened up the global subsea cable market. It used to be if you go back 30 years, the subsea cable market ‑‑ maybe 35 ‑‑ was a closed club that was just the incumbent legacy telcos and they had bilateral agreements in which they invested in subsea cables for voice and a teeny bit of data. There were a couple of regulatory decisions that opened that up and completely changed the market and changed the world, that led to huge investment in competitive fiber global backbone subsea cables which changed the nature of global band width. The first was when the FCC in the U.S. opened up the ability for any company to obtain a cable license and opened up landing stations, right? That was a dramatic and radical change, that was opposed by all of the incumbent operators. That led to a huge amount of investment.
The second, which was similar but really important, probably 15 years later, was when Singapore, which had been adding licenses for additional landing rights to the landing station in Singapore, basically said we're opening it up. Right? We're just going to open it, we're not going to limit the number of cables coming into Singapore and by opening up the cable landing station, that led to huge investment in the region and it kept a lot of the traffic in the region, lowered the costs, improved performance and also Singapore became a hub.
So why is Facebook interested? Well, two reasons, right? One is the mission that we have from Mark Zuckerberg who is passionate and personally involved in the fact that the world needs to be connected. This is part of the mission in addition to creating community and it's not about connecting the next billion, it's about connecting the next three and a half billion so if you get the last billion the others get pulled through. That cannot happen unless there is global connectivity, right?
The second is we have data centers globally, and so we're building out, in partnership ‑‑ always in partnership ‑‑ cables that are open for others as well. And there are seven that we've ‑‑ that we've announced public ‑‑ we're already announcing publicly and working out. And we do this and it increases band width for others as well because it's shared.
About a year ago we turned on a cable from the U.S. and Europe in Spain which was in partnership with Microsoft and Telefonica. We each have our own dark fibers and we like to use them for our own purposes, we use them for our data centers but the increasing capacity with Telefonica is opening it up and increasing capacity and driving the cost down. Others include a cable with the U.S. and Japan, a partnership with KDI, AT&T and Amazon, significantly increasing transPacific band width. Another to the Philippines, another that's to Hong Kong, that's in partnership with tell extra, India and China, again, dramatically increasing transPacific band width. And then ‑‑ and then another one ‑‑ then we have another one we're building to Europe in partnership with others. But then we also do regional. We have a partnership we're bringing ‑‑ we're building a cable from Brazil to Argentina and there's been almost no new investment in international capacity to Argentina I don't know in how many years but it's been a very long time. Prices are high, band width is low. And so this one cable that we've announced that we're building in partnership there will increase the international band width into Argentina by 6X, right? There are other cables coming after that so it will even increase that more.
In just ‑‑ in terms of connecting the unconnected, going to the first point about scaling the backbone, we have an open fiber that we had a coinvestment with Airtel and a small USP in Africa, in Uganda we built a 770‑kilometer fiber core network that has enabled Airtel to convert its cell sites to 2G, 3G and 4G. That is providing mobile broad band as part of that ecosystem I described in northwest rural Uganda that was not available and I'll stop here and we can come back for questions.
>> AUDIENCE: (Off mic).
>> MODERATOR: Now we're going to talk with Camille Morel. She's a Ph.D. student at Centre Lyonnais with the Center of Strategy on the International Security in France. With Cami we're going to understand a little bit what is the geostrategy issue of the submarine cables and also understand how is the international norms context, OK.
>> CAMILLE MOREL: Good morning, thank you. So first to international organizers for organizing this table and the physical aspects of the Internet, I think it's something good for the general (?) here and it's with a lot of community, of course, I will present my research here as a Ph.D. student. So the undersea network has two complementary aspects that gives its (?). The first is the physical aspects of the infrastructure and the second one is role in conveying information. Submarine cables link two or more countries or even continents together and thus international dimension involve interstate relations which leads to some political considerations, of course.
Firstly, the geographic distribution of submarine cables around the world is (?) which relations of dependence between states, indeed as some states are less connected than others, to the industry web data need to pass through arteries. And, of course, that was the case, sorry, for South America, for example, before the submarine cables Atlantic two, international and Internet data from South America and to Africa had to transit the American continent to reach Europe. This has encouraged some states to allow or launch new projects of cables like Brazil with their latest cable section, for example, I think we're talking about that later.
In addition to the willingness of dependence new projects are initiated by less connected countries to ensure data transmission in case of (?) cable system. In particular Algeria experience and important on its single international cables to Asia in 2015, significantly its connectivity and the access of its citizens to Internet.
As a vector of information, so the undersea network has helped the expansion and the keeping of powers in the Internet arena, of course, it's played a major role of the growth of information between financial centers as well as in the development of economic powers. We can also highlight its important role in the control of population at very long distance by admitting communication, for example, between the empires and their colonies. The information transmitted through the network has been always ‑‑ has always been a matter of strategic interests for states. This explains why they have always conducted and still conduct straightening actions on infrastructure and its flow. For example, cuts in times of war or cuts of intelligence through cables. One famous example concerns the disclosures about 9 international security on cables through extreme. More recently we can also mention contents of U.K. and U.S. officials about Russian vessels near submarine cables suspect censorship of traffic during the Arab spring. The brief overview show that submarine cables are subject of geopolitical issues which, of course, can be contemplated by a matter of perspectives like cyberspace perspectives. And in this context. So international norms appear insufficient to protect human rights related to submarine cables. Among them, of course, the freedom of information which is an extension of the freedom of speech implies freedoms of seeking, receiving and spreading information without any consideration of borders or media. And another important right is the right to privacy well‑known in the context of Internet neutral. These two fundamental rights are (?) and are still (?) today. Submarine cables in the context of big data allows states and companies to have access to personal data. Traffic interruptions submarine cable systems can lead to Internet ‑‑ into Internet down times, freedom of information during the time of (?) of cables ‑‑ maintenance of cables. In this context protection of submarine cables mainly conferred by the 8084 international convention for the protection of submarine cables and the United Nations convention of the law of the sea seems to be insufficient. In particular, the parity of maritime areas crossed by submarine cable impede this legal form of protection. Some of them are under territorial (?) whereas others are not. In areas of international protection and the level of protection depends on the goodwill of states to take sufficient measures to protect cables.
So in order to be effective the legal regime of submarine cables has to be (?) and I think redesigned, in particular, it has to include new types of (no audio) international conventions also has to be expanded. For example, the current status of international cables lets belligerent states to use submarine cables in times of war. It does nothing to improve the cooperation between neighbors and the navies to improve the coordination in case of emergency in the network. Furthermore the international norms have to take into account the current growth of service providers on the market to ensure the diversity of operators and honest to protect the net neutrality.
And finally ‑‑ I don't know if it's OK, if I'm good? Submarine cables are also considered critical infrastructure and they're often regulated by specific norms, national (?). Nothing is planned at the international level, even though a definite flow can produce in case of maybe force majeure, the effects on the global society.
>> MODERATOR: Thank you, Cami. Now we are staying in the research approach, going back to America, in Canada with Dwayne Winseck. Dwayne is professor at the university and also director of the Canadian media concentration project and with him we are going to understand better the trend of the concentration around the cables, submarine cables and maybe try to understand also what is the public role in this situation.
>> DWAYNE WINSECK: OK, well, thank you very much. Thank you very much for attending this panel session and thank you very, very much to Félix and his partner for organizing it and I'm very happy to be here. And my paper really follows up quite nicely on what Pepper and Camille just talked about and in particular if you think what we're seeing after Pepper talked here it's a vertical integration between the apps and the services that we're all so familiar with on our devices and on our desktops, the Googles, the Facebooks, the fangs of the world, so to speak, and this idea that the fangs have come to rule the Internet and we have a very strong case for a kind of hegemony of the Internet the worldwide over and this would seem this kind of vertical integration from our devices and desktops from the apps and services down into the cables to kind of provide further evidence and kind of proof of the extent to which GAFAM has come to rule the world. I'm going to argue, actually, that this is a superficially ‑‑ superficial case but also a misguided case because if we look down the stack, so to speak, while we see the dominance at the level of service and apps and operating systems, once we get down to things like the submarine cables, the Internet exchange points, the ASNs upon which the Internet is built, the traffic and its distribution and shifts in its distribution over time we see a very different world, we see, actually, a kind of a post‑American Internet emerging, a kind of multi‑polar world and we can see that in the nature of the consortia that own the submarine cables and, in fact, the consortia have changed from the traditional telecoms carriers to a much more heterogenous telecom players which are significant players here but new relatively new two‑decade‑old now class of competitive carriers that have emerged, the level threes, the CATAs of the world, the cogents and so on and then we have the CDNs. So we have a much more complex and heterogenous system and when we look there we do not see the levels of American dominance that we see in our devices and in the desktops and in our services so we have a split between the content where I think it's safe to say that the American companies dominate the world over in many places except for China and Russia and so on but not at the level of the material infrastructure and instead what we're seeing is a tilt in the kind of geography and geopolitical economy of the global Internet infrastructure more towards Europe, towards China in particular, the Asia‑Pacific region, the application countries and the global south. One of the truly remarkable things in the last couple years is the extent to which Africa is finally being wired up with fiber optic cables and that's driving the Internet much deeper into Africa.
There's some really interesting things, I've been looking at the development of the global information infrastructures, basically, since the late 19th century. The first global information infrastructure was, of course, built on copper, the telegraph system. And it was really kind of laid down in the mid 1870s, about a decade after the technology became well understood and investment started to flow, but there was a major financial boom in the early 1870s that ended up in a spectacular bust a few years later, the great financial crisis of 1873. And there was right on the leading edge of that was the copper wires that built up the worldwide telegraph system and so there's a tough of investment plowed into the copper wires, they wired up the world, the market collapsed, many of the companies went bust, but still the wires were left behind and the world was wired.
So there's something interesting about these global financial bubbles and busts because when did we see this occur again? We saw it occur roughly a century later with the dot‑com boom, during the dot‑com boom in the final three years of the 20th century and into the first kind of three years of the 21st century, we saw a massive explosion of investment in the submarine cable system. It was very ‑‑ there were 60 new cables laid across the transatlantic at that time, between Europe and North America. There's a very specific geography to this, though, and it was very much northern focused and focused on the European and North American economies. There's so much new capacity laid at that time that 90% of it was kept aside as dark fiber, I love that phrase, "dark fiber." It sounds mysterious, we can write a spy novel around it. The dark fiber was unlit, they didn't put the optics to turn the fire on and make it available. There was so much capacity there was a glut on the market and they were concerned to basically preserve pricing power, so they kept the cable unlit. And they've only been bringing it on in the last half decade or so, we're still around 50% capacity today.
Now if we go, though, there's been a major boom after 2008 but the geography has completely changed. And if we could go to the first slide there we could see this and Pepper talked a couple more recent projects that I haven't yet included but the geography now has clearly shifted from the northern transatlantic region over to the west coast, so from the west coast of North America over to Asia and particularly in the Asia‑Pacific region, we've got the six major projects that were developed over the last decade and while there are some who are starting to talk about the fangs' takeover of the world in this vertical integration as if Facebook and Google and Amazon and Microsoft are going it alone and build their own fiber to basically move the telecom carriers aside and that competitive class aside the fact of the matter is that we're seeing them joining these consortium and they are sitting cheek with joule stake‑based incumbents like China telecom, and so on, I don't think they've dominated whatsoever. I think they've dominated some of the fibers within the cable and the capacity for their own uses. All this is really important because where you build up the submarine cables you build out at the ends of those things a couple things like Internet exchange points and the data centers and this has been particularly important with respect to the Asia‑Pacific region but also Africa. Eight new cables laid around Africa within the last five years or so, finally the IXPs that the OCD and the OIT has been promoting in terms of Africa are now going there. In terms of these cables, these cables are governed through the lens of the landing licenses and the basic rule is the stronger the state so like the U.S. or Britain the stronger their authority over the cable and the greater they can exercise capacity. The weaker the state, the less capacity the state has to effectively govern these cables and I'll leave it there. So it is a post‑American Internet is my ‑‑ at the infrastructure level is my basic argument.
>> MODERATOR: Thank you, Dwayne.
>> DWAYNE WINSECK: Can you go to the next two slides very quickly?
>> MODERATOR: You still have a few minutes.
>> DWAYNE WINSECK: Go to the next two. What I wanted to show is the ‑‑ not working? It's frozen. Basically what it shows is the American share of ASNs dropping from 60% to 30% since 1997. This rapid decline, the second from the top is the American share of ASNs, basically their share cut in half over a 20‑year period.
Go to the next one. Internet traffic. Same kind of thing. From over 50% of all Internet traffic in early 2000s down to around 25%. And then the last one, of course, is the geographical distribution of users so America used to account for over 50% in 1997, they now account for 5%, China 20%. OK. So all these as indicators of a move away from American‑centric Internet. Thank you for that indulgence.
>> MODERATOR: Perfect. After this presentation, we stay in America but go to the south, in Brazil, to understand what is the reality there. Just before that maybe say that we have just organized two weeks ago a research (?) in Brazil to talk about that. And Barbara, as the representative of IDEC, was a partner of this organization, and she's going to talk for us a little bit what is the point of view of Civil Society and the context of connectivity in Brazil.
>> BARBARA SIMAO: Hi, everyone. Thank you, Florence and Félix for the invitation to speak here. I'm Barbara, as Florence already said, and I'm a researcher in telecom and digital rights in IDEC. It works with consumer defense basically claiming for justice and balance within consumer relations. And in telecom we specifically work to increase quality and access in the Brazilian network.
I think to answer these questions, it's necessary to use some background on the connectivity reality in Brazil. According to the most recent research on Internet access there, which is called (Speaking non‑English language), we have 61% of the population connected, but that number varies when we consider regional and socioeconomic inequalities. The rural areas have only 34% of the residences with access and the lower classes have 30% of residences connected. And the price, the research shows, is the main challenge of universalization. So this data represents all inequalities we have there and the urgency of public policies to tackle that. And how does submarine cables can help on that. And as everyone said here before, they're an essential part of Internet infrastructure that connects data all over the world. So there is a huge impact of the submarine cables on global connectivity and how the Internet is distributed. But the question is what's the impact on local connectivity? And IDEC was invited by (saying name) to promote the submarine cables and its possible impacts. The seminar took place, as Florence said, two weeks ago in Rio de Janeiro, and it was really, it was a huge success and it was important to help to clear some thoughts we had in mind. And I would like to emphasize a few lessons of what we learned there. And the first is, yes, submarine cables do have an impact on local connectivity and quality and prices. They help to reduce the cost to ISPs and therefore they can help to reduce the cost to the final customer and they have the impact also on quality, reducing the latency of connection and to increase beat which is really welcome when you think of the future of Internet of things and ICTs, et cetera. So they should not be seen just as private infrastructures that no one knows nothing about and by no one I mean Civil Society in general. And they are a really important part of policies directed to connectivity issues.
And the second lesson I think is, OK, they are important, but they are not the magic missing elements that will suddenly help us out and solve all of our problems. Yes, they are important but they will only have impact if they are thought as aggregated in a way in a larger policy that brings to the table different aspects such as taxes, local distribution of Internet access points, for example. So speaking more of ‑‑ whoops ‑‑ speaking more specifically about the new cables that are now being built in Brazil, they have Ellalink which is a spotlight because it links Brazil to Europe and it has a project Bella (phonetic) which is a consortium of different research organizations such as Jean here at Europe, (saying name) at Brazil, (saying name) in Latin America in general and they bought a part of the capacity of the Ellalink, to connect the research organizations around these regions. So it will definitely have a public impact. Sax which is a cable already put in place between (saying name) and Brazil will also have an impact because it's the first south transatlantic cable that links these two continents. So already an impact on local, regional corporation and connectivity. And Monet is a cable financed also by Google which is a content provider, we all know about it, and a part of the cable will be used to link astronomers who are building ‑‑ like, building a telescope in Chile. So another example of how this have an impact on research and global and regional cooperation.
I think that's all I had to say here and IDEC and Civil Society organization will watch out because I think now we really know the impact that these cables have on connectivity and they should be seen as a part of policies there directed to tackle these problems and I think our mission now is that. Thank you.
>> MODERATOR: Thank you, Barbara. After this point of view of Civil Society it's also interesting to understand technically what is happening inside the cables and sometimes what are the traffic issues and if we are sure that they are used at full capacity where sometimes it's not happening and to help us to understand that, we are now hearing Doug Madory. He's director of Internet analysis of (?) in the United States.
>> DOUG MADORY: Thanks. Is my light on? There you go. Thanks. Thanks for having me. I do have a couple slides here. So I am not in the submarine cable industry per se but I have done ‑‑ been a speaker at a number of submarine cable events over the years and part of what we do is we are basically a team of computer scientists that were originally a company called Renaissance, later we were acquired by DNS service provider dine and rebranded as Dine Research and Dine got acquired by Oracle, got rebranded as Oracle Intelligence, so there are a number of names to keep track of which is why my slides keep changing as far as the background. We do a lot of Internet measurement and we build supply tools to folks in the Internet community, telecoms and we try to be fairly knowledgeable about what's going on in especially Internet routing and submarine cables became a little pet area ‑‑ interest of mine to try to see if I could spot, understand the impacts of submarine cable cuts and also spot cable activations prior to their public announcement. And I'll just give a couple examples of these, if you go to the next slide so the famous one was a submarine cable in Cuba. Prior to this cable becoming active Cuba was entirely dependent on satellite service. In fact, the state that Cuba's in is not unlike a number of nation states around the world, if you don't have a submarine cable you cannot conduct a modern economy, saint Helena is one that people are trying to advocate to get a submarine cable to help the people that live on that island in the Atlantic, in Cuba, I have one story, the Venezuelans put up the money to build the cable, it was constructed and a couple years there was no notice of any change. And I would ‑‑ I got familiar with a story just seeing people write about it, there's a cottage industry of people who watch ‑‑ especially in the United States watch the Internet connectivity in Cuba and figured I'd set something up within our tools to spot the cable activation as soon as it came up and sure enough about two years later after I wrote my little script I got my ‑‑ my thing woke up gave me an e‑mail and said there's a new connection into Cuba so if you go to the next slide, a couple of graphs like these, these scatter plots, the one on the left the dots are the latency, the Y axis RTT latencies and X axis is over time. It's hard to see from the distance but the upper band is symmetric geostationary satellite so there's a lower bound on how fast ‑‑ or how low latencies can be when you're on a satellite, they can't be lower than 480 milliseconds and that's due to the speed of light so when we spotted a lower band that was not consistent with the speed of light, then it had to be some surface connection. I wrote this up, we got a lot of press for it. I did notice that the latency was still kind of high for surface and theorized that it was probably an asymmetric connection where traffic was coming in on the submarine cable but then still going out satellite and that's how this round trip time was still above 300 milliseconds from anywhere in the United States and then a couple days after we wrote that up then it dropped down and it became symmetrical in the ‑‑ going over the cable.
Funny story, though, a couple months after we wrote this up, I was at a conference in med dell lean Colombia and got addressed by the director of Cuba and I introduced myself and I said I wrote this thing about your cable and he looked down at my badge and looked up at me and said, yeah, I know who you are. Great, I'm not invested in the politics, it's great, I hope it works well for you, he said you were right about the asymmetric routing thing, I said, oh, great I figured. This tweet at the bottom is (saying name) he's a prominent dissident that kind of became our best friend around that time where we were just putting out information that the government was not ‑‑ not going to be forthcoming ‑‑ had never answered any questions about the status of the cable. And either the government or the state telecom. And so they were forced to ‑‑ this thing the grandma the state newspaper were forced to confirm our report within a couple days and so that's kind of a neat outcome of Internet measurement, that's a highlight from the past. But if you go to the next slide, this is a more recent activation, and so this has been mentioned a couple times, this is a south Atlantic cable system going from Brazil to Angola. So this, too, we spotted ahead of the public announcement, Angola cables for the guys that put this on, actually, Angola cable and myself we did a joint talk, LATNIK, what month was it, September, two months ago, I guess, and, yeah, first they were happy because they were so excited about the cable being up, then they were annoyed that I didn't let them call their own cable up and then they got happy again. So we made up and we did a little joint talk about just the improvement of service for, you know, going across the south Atlantic. Last measure transoceanic route that really wasn't covered by a submarine cable so that's a neat milestone of the Internet to see this thing come live.
If you go to the next slide, one thing that's kind of neat about this cable is, you know, aside from the fact that this is one of the rare cables that connects to developing parts of the world, you know, most cables that get put on are connecting Africa to Europe or South America to the United States, they're connecting a developing part of the world to one of the hubs of the Internet. This is going from South America to Africa and that makes it doubly unique, but also in conjunction with the cable that goes from Brazil on up to the United States latencies from the United States to subSaharan Africa are actually faster because it's a shorter geography path to go down and bounce off Brazil to get to Angola than it is to go across Europe and down the west coast of Africa so it's kind of an interesting ‑‑ it's marginally better, it's not, like, dramatically different but it's just the fact that that's possible ‑‑ these are latency measurements from some of our servers in Tokyo and Singapore improving their latencies to get to Angola and you think of the geography here they're crossing the Pacific crossing the United States and there's still benefit in going and bouncing off of Brazil from the United States as opposed to going through western Europe to ‑‑ nerds like me who look at this all day that's super neat to see not just from Brazil but from Tokyo to Angola it got better. It will be interesting to see, there's some debate around the financial viability of this cable as a business, but setting that part aside it will be interesting to see if there is benefit to the disintermediation to the United States and Europe from Internet paths, how that affects these connections.
So in our work documenting developments about the Internet we do a lot of documenting of Internet shutdowns, that kind of puts us trying to help out ‑‑ we're often in a position to try to assist to provide technical data to Access Now or Internet Sans Frontières ‑‑ OK, I'll wrap it up ‑‑ for example, the ace cable going down this year, it was kind of like two events back to back, there was an event on the cable that was visible because you could see impacts on all the countries receiving service from the cable at the same time and so when that occurs you can be ‑‑ that's a reasonable ‑‑ you can make a reasonable assessment that this was cable‑related as opposed to something happening in a single consider where I think it was Sierra Leone had an Internet shutdown and then they blamed it on the cable we dug into the data there was two events there was a cable cut and I want shutdown, that's where we feel our role as far as being Internet measurement trying to apply facts to these and make sure everybody's got the right facts and we know what's going on. With that if you go to the next slide, two more slides here, we launched this website this year called Internet intelligence map and this is to try to democratize some of the data we've been putting out over the years, if you want to see routes going on or DNS queries drying up from a country due to some kind of event, then you can go see this for yourself, you don't have to wait for me to write this up and generate a graph. If you go to the next slide, this is just ‑‑ let's see using an example, if you advance to the next slide I'll wrap this up. Is it hung up? Can you go ahead? Thanks. So we're just looking at Syria in I think it was June, was having high school exams, they take their country down for high school exams. It shows up here, bubbles up at the top as the country most affected, if you go to the next slide, oops, sorry, go ahead. One more. So this is the last slide. So then you can kind of see on these three different dimension that we use to analyze just, you know, is a country communicating or not, the routes up are active measurements responding or are we receiving queries coming out of the country, then each of those shows some sort of perturb bans, you don't need me to interpret that to see something is going on there. One is a surge of DNS queries, there's an outage, there are more queries coming out because they don't get response, there's a lot of retries. That's kind of a gear head thing. But anyway check out the site and you may find that interesting with some of the work you guys do. That's it.
>> MODERATOR: Great, Doug, thank you. I think this is directly related with what our last panelist is going to talk about, (saying name) he's the Internet Without Borders and he's going to talk to us what is the relativity of the connectivity there. You know about Internet shutdowns and he can talk about that.
>> First of all, I would like to thank you, Florence and Félix, for the invitation. I am responsible for the coordination in West Africa, I come from Benin, in West Africa who is connected to cable, we have two Internet cables because we have a good geographical position in the Gulf of Guinea, but the thing is we were expecting to have a real impact with those cables on the quality and the price of Internet. So basically we have a new population coming to Internet, like more and more people are using the mobile telephones to connect, to have access to Internet, but at the same time the digital divide is still present in the population, so nowadays we were able to get information through several reports saying that the cable, the submarine cables are now used undercapacity. So basically we have two cables to help transmit the flow, an impressive, as massive flow of information, cables are used 40%. We have nearby countries that are also connected to our cables so we are still facing issues, though we have ‑‑ we've seen investment in the infrastructures. But I'm glad to have here Robert Pepper from Facebook, because we saw that new ‑‑ a newcomers are interested, I mean, about investment in cables, submarine cables, and because we already have infrastructures in Benin and we're still not able to use them to the full potential but we would like to know what guarantee, for example, you at Facebook, for example, as a new ‑‑ like a new player in this market could offer to, for example, a population like users, like basic users of Internet regarding the quality, regarding the price because in Benin, for example, recently there was, like, a huge debate in the population because the government wanted to increase the price on the Internet usage, so through the mobilization, we managed, like, to cancel this tax. So I'd like to know what kind of guarantee you can offer as a new player in this market. Also based on the recent (?) scandal I would like to know also how you could approach like the data protection on new markets where we have very weak regulations or non‑existent regulations at all so that would be my question for you, thank you.
>> MODERATOR: If you want to talk a little bit more, you have still left a few minutes.
>> I think for now I'm fine so maybe further discussion we could further, yeah.
>> CHAIRPERSON: Is responsible for central Africa, it's not because this country has no access to cables, you don't have a voice, so can you just say ...
>> Thank you. I just want to ask about our country, especially in Chad. It is a landlocked country and a country under non‑democratic governments. So how our organization like Facebook or Access Now or all organizations who are fighting for digital rights in the world, and fight to make Internet accessible for all people who have may have difficulty to access Internet in three ways, political ways because the government doesn't need people to have access to the Internet and to have information and to have, like, citizens right, find the citizens their rights, sorry for my English, I do my best. Secondly it is economics (?). The government do everything to make Internet very, very, very expensive so citizens can not access the Internet because they don't have economically power to access it. So how we can do ‑‑ make Internet better, cheaper and accessible for those people in this area. So ‑‑ and the third way is they shut down, the government make pressure on all telcos in our country and forces them to obey in order to shut down Internet or completely cut connection, telephone or (?) or everything like this. So we heard that many organization are thinking, like, making like satellites or (?) Internet, some kind of technology to help Civil Society of our country to be able to make connection and have access to the world. So in the central African countries, democracy or freedom of speech online, citizens online are not guaranteed. So we really want to know how you can help us to make better and better connections in the world because this area is the black hole of the Internet world. So we are like we said we are losing a very important rendezvous of the history, because if we are not at this time connected in the world, we will be like the second rendezvous of the world, which is the digital world. So if this area doesn't have access to the digital world what can be this world? We are the next (?) people of the next world, I'm very sorry for my English, I have a thousand things to tell you, as you are feeling now, as you really are seeing now all of this French country have not access to the technology, to the English world. So please help us to do your best to make diplomatic pressure, economic pressure, technological pressure, and institutional pressure to give us the chance to be part of the world's (?), if not we will be not visible at all. Thank you very much.
[APPLAUSE]
>> MODERATOR: Thank you. It's important to ask the question of Africa, it's a central topic and I think it's also a topic that has been treated by all the panelists here, it's very important that you can give us this testimony.
Before passing to the questions, now we have closed the first session of the panelists and I thank you, everybody.
We are going now to pass the microphone to Peter Micek who is general counsel at Access Now and you are going to help us to make a (?) of the situation, what are the hot questions that have been the result of our panel. And after we are going to invite our panelists to answer those two questions and also people who have questions and comments to do.
>> PETER MICEK: Thank you very much, I'm really privileged to speak as a member of Civil Society who we've learned knows nothing about cables, it's really exciting to be able to ask questions of your experts. And so I guess that's my first question is where do I go to learn. How do I learn about these cables that seem to be so important but don't always have the impact that they should have, that they could have on increasing our access and letting us enjoy the benefits of the Internet. Who's job is it to build capacity among Civil Society and the public sector to understand and exercise their rights and interests over these ‑‑ this essential critical infrastructure. Do cable authorities, whether they're public or private or both have a real interest in motivation to open up to public scrutiny and to invite us into our ‑‑ you know, what are probably still very closed discussions. And, you know, trust, I think, is the theme of this conference and so this seems like the ultimate, to me, trust ‑‑ trust us situation, where these cable operators have made back‑room deals and really stayed invisible. How do we know that they deserve our trust? We've had several crises that have come up, so the crisis over surveillance with Edward Snowden's revelations about the NSA's activities. I'd say we have crises over lack of access, lack of landing points are underused and dead landing points. We've had crises over disruptions. Is the cable infrastructure prepared for the next crisis which I'm guessing would ‑‑ let's say cyber security or terrorists, is it prepared and if it's not how can the Civil Society and the public help lay the groundwork in terms of both goodness and substantive measures and standards to prepare the sector for the next crisis? And then, yeah, using my prerogative as Access Now I would really promote using things like the guiding principles on business and human rights, putting human rights and the users at the center of this system and building outwards from there. If we carry out a human rights impact assessment on a new cable what would the criteria be? And who would be on the steering committee leading that, monitoring on an ongoing basis? I think I can leave it there. I think it's great that we started off listening ‑‑ hearing about the 1884, I think you said, convention was the first and so is a new treaty needed and if so what would be in that treaty guiding this essential infrastructure. Thanks.
>> MODERATOR: Thank you, Peter. So yet we have still 20 minutes. That's good. The proposal will be that if someone have a comment to do, a question, please present themselves, introduce themselves, herself, and maybe we can take a block of three, four questions and then go back to our panelists. OK for you?
>> AUDIENCE: Hello, I'm Marta (saying name) from Ecuador. I run an investigative journalist website. And we actually had a problem investigation ‑‑ investigating the owner of the submarine cable in Ecuador. It's a cable that goes from the United States, but the last part from Panama to Ecuador is (?) in ‑‑ a mogul in Ecuador. Some say he is really the partner of our last vice president that now is in jail because of the other cases and it seems that part of the money of the other BRIC bribes were used to build that part of the cable. And the problem is what happened, I think it's really good from the south point of view that we have new cable owners and all that stuff but it should be transparent who are the owners and what are the consequences of that, not just in a geopolitical sense but also for the citizens because in my country we think that we had a black (?) masked as a cable failure. Two days after the partner, the ex‑vice president uncle that was his briefcase man, something like that, I don't know how to translate that, went to jail, we made a report about his relationship with the cable owner viral and then two days after that I received some ‑‑ a telephone call telling me to take down the note, I said no, and we had an Internet practically shut down at least for social media, two days without social media on our computers and other stuff and they said that the submarine cable had suffered from a rupture between Panama and Ecuador. But we were looking for the outages report, and United States has a very complicated system that you can (?) even second outage is the same as a three‑day outage, at least we couldn't as journalists and Civil Society we couldn't understand that system. And we said, well, there should be an authority, a global authority, because our telecom authorities, they didn't give us any kind of reasonable explanation and we just have to believe what ‑‑ three days after the problem, the submarine cable owner told us. So we said there should be some kind of international war or act orb something that held these people accountable since a lot of people and some of them are not so straight in their behavior. So as users we have at least the right to know what's happening. Thank you.
>> MODERATOR: Thank you very much. Are you directing your commenter especially someone from the panel to facilitate ‑‑
>> AUDIENCE: I think to Camille maybe, I don't know.
>> MODERATOR: No, just in case of ‑‑ so, now OK, I'm going to try you first and then go back to another block but that's OK.
>> AUDIENCE: My name is Paul Nelson, I think it's pretty clear that we've moved into a world where the new cables are being build by the city ends and platforms like Facebook and Google and that gives the telcos something to worry about, not the top businesses but the bottom businesses, because they're being squeezed in both directions in this new environment. I think in light of this change, though, I'm interested to know what obligations the CDNs and the platforms feel interest open Internet to connect or maybe the countries they might go past in dropping a cable where the marginal cost of dropping a branch into a place which might ‑‑ who may have no other chance to get cable connectivity is where their cost is is relatively low and I ask because we're looking and very interested in new cable, cable in the Pacific and providing technical assistance to operators and to provide Internet services that are stable and secure and efficient for the people who rely on them. Because I think we all should know that just dropping a cable doesn't guarantee very much at all, whether it's price or security or quality of service. So there's a lot more to be done after a cable is dropped into a place in terms of making sure it actually produces the benefits that we expect. So, yeah, the question, again, is what are the obligations of the platforms which are no longer the traditional telcos laying cables, the open Internet and not just to their own services. Thanks.
>> MODERATOR: What is your name, just I didn't ‑‑
>> AUDIENCE: Paul Wilson from APNIC, yeah thanks.
>> AUDIENCE: I'm (saying name) from Brazil from NIC.BR. It seems to me for the comments, from the histories, that it shows that when we have less regulation from the governments, it happens to the ‑‑ that the market developed and we had more cables and things working better. It seems to me that the situation is improving. We have more cables, more cables in the suit. We have more players doing these cables. And my question is we really need a kind of regulation by the governments about this. It seems to me that no, but I would like to hear the people from the panel specifically on this point. Thank you.
>> MODERATOR: Could you just say your name, please.
>> AUDIENCE: Antonio Moretes (phonetic).
>> AUDIENCE: Hello, everybody this is (saying name) from Afghanistan. And stuck about the submarine cables but, as you know, Afghanistan is a dry country and so I want to talk about our fiber networks and (?). So first we have completed our fiber networks throughout the country and have provided our own fiber links to the people of our neighboring countries. Our basic problem is that with the exception of our two borders, the other neighbors, our other neighbors do not allow us to connect with their networks, fiber networks. So now my question is whether can we affirm to go or exchange the (?) through the fiber with other countries? If, yes, is there any policy for crossing fiber cables through the borders? I mean, do you have a clear policy for the fiber networks through the border, how can we cross our fiber network to other borders or from other borders?
>> AUDIENCE: (Off mic).
>> MODERATOR: This is green, and if this gets red, just we are ‑‑ because we have, like, 15 minutes, how many questions do we have still, just one? OK, maybe we can take this last question and after, go to the panel, it's OK for you, everybody?
>> AUDIENCE: Hello, my name's Christopher Foster from the University of Manchester U.K. First of all, Doug, interesting Internet measurements and things. These tools are democratized and open but it's a challenge for a layperson or Civil Society to use these tools to try to understand what's going on, some of these more geopolitical issues and cutoffs and so on so what kind of advice would you recommend for opening up these value data to the wider Civil Society and community. And a quick question to Dwayne or potentially also Pepper, just asking, I mean, in terms of cabling locating in certain regions, if certain region identify problems, there's a lack of cabling or potentially latency issues how might they attract Facebook or other conglomerates to come in and lay cables throughout the region. That's my question.
>> MODERATOR: Thank you. I would like to suggest someone have received some direct answer but I think all the panelists can make a suggestion, I would like to begin on this side starting with Camille and finishing with Robert if that's OK. Maybe three minutes, each one?
>> So briefly, thank you, Florence, once again. I will try to relate to the (?) Ecuador, because she mentioned the new mission, the issues related to your investigation, like ‑‑ because of lack of transparency and especially when there's ‑‑ there was this rupture of the cable. And I would say that we also have the same issues in Benin, so the cable was cut two times in 2007 and in 2011, actually. And in general, the stories related to those situations are quite close to the other situation, so basically we have less information, it's very like (?). In Benin now we have a local authority governing, like, all of the electronic activities so they are responsible for this, but the thing is that the members are chosen by the government, actually. So they are very close to the government. And when they have, like, to provide, like, information or opinion on these kinds of situations sometimes like they don't talk that much so we are putting pressure on the government in Benin, actually, to have like the possibility to have more information when this kind of situation happens. Last time, for example, we knew it was like a Chinese boat somewhere in the high sea who was unfortunately able to like broke the cable. And it's happened more and more in Africa, I know recently in (?) was the case, so for now we are putting more pressure to have the capability to get more information because for now it's like we are really facing the lack of transparency regarding this. Thank you.
>> MODERATOR: Thank you. Doug? Doug?
>> DOUG MADORY: All right. Let's see. I'll answer a couple questions. So the first question was directed to me as far as the ‑‑ how people can develop these similar skills or abilities, I guess there are ‑‑ there's quite a bit of data, public data out there, of active measurement routing data, I could talk to you afterwards maybe to point you to where that stuff is. I think, and we've got our own data that we would argue that it's better but I do a lot of what I do with public data and it's really the skill of trying to understand how do you get insight out of the data and how do you track with what's going on in the world and what would the activation of Russia's cable Crimea the data and go with a tool ‑‑ I tell reporters I set traps, I think I know what it's going to look like and write a code to do a thing and alert me when it becomes active. That I've just kind of made up on my own over many years of doing this every day over a decade. I'm happy to share insights with you, I can't do that in three minutes, though. Another thing that came up as far as some public requirement of reporting outages, there probably would be some benefit to that. I know that as someone who does some reporting on outages, it is a pretty tightly held ‑‑ it's tightly held information, cable operators do not generally publicly announce all of the different adverse events that are happening to their cables, for the big ones there's actually quite a lot going on of varying levels of severity. That is not public data. That's given to the people who buy service from them, but they also sign an NDA not to publish that publicly in general. I have a handful of contacts around the world that don't mind, you know, looking ‑‑ overlooking their NDA to confirm things for me, they won't tell me about it, not everybody agrees with this, I've got flak from the submarine industry, people are written to me in LinkedIn and stuff, saying you're ‑‑ even if you're not a signatory to the NDA you're violating the spirit of the NDA that these other guys signed, I'm making my own data, I'm making my own conclusion, you can't stop me, I never said I wouldn't. But that's the old‑school mentality of not sharing anything about outages and it ends up having real‑world effects. Last year there was a submarine cable cut, the easy cable had a cut near Madagascar, they had to take the cable down to do the maintenance. Unfortunately they didn't do public notices and when they took the cable down it's the one cable that serviced southern Somalia, Mogadishu, the cable went down as the candidates for the national election for president were starting their national debate, and as they began their debate the country's Internet went offline and everybody suspected foul play. The candidates stormed off, issued press releases saying the government's trying to throw the election and we could see this as the whole ‑‑ up and down the east coast of Africa all went down at the same time, it was within a couple days of the cut and I had, again, a couple contacts that would confirm this is maintenance. Unfortunately they didn't let somebody know and people were really, you know, thought the worst so it would be nice if aside from guys like me who like to write stuff up there's other people who would benefit from some public requirement of announcing at least major stuff, not all the little stuff, the big outages.
>> MODERATOR: Thank you, Doug. Barbara, do you want to answer something?
>> BARBARA SIMAO: I agree with him about this, this lack of accountability and transparency that cables have now. And there are infrastructures that have public interest involved so it should be really important to disclose some information about them, not all information, because there is a private agreement that, OK, we can't have access, but I think the ‑‑ some aspects are really important for the population in general to understand and ‑‑ and that's it, I mean I think.
>> I would suggest better research I've used theirs in the past, telegeology, they make a pretty rich, subforum publicly available in Africa, Steve Song is the God of good insights there. In terms of the question of trust, transparency, you know, what kind of standards or criteria could we use in terms of, you know, civil liberties and these kinds of things, this is a really, a big black hole. I mean, in the United States, cable landing licenses, their announcement are done entirely at the prerogative of the president who oversees and makes a final determination after a committee foreign investment in the United States which is made up of all the security agencies in the United States, vets an application, imposes whatever kind of back‑door lawful access requirements are on the cable and then the president decides whether he or she will deign to tell you if this cable's permitted but they won't tell you anything. So if this is the standard in the United States the idea of getting access in most other countries is very unlikely. So, you know, we can maybe tie in ‑‑ tie Bob up here or Pepper up here and try to force some info out of him but I don't think that would be in the spirit of the conference. So really difficult, but I think really, really important. You know, 1884, jeez, you know, this is why the last major agreement understanding on this is, you know, well over a century old.
One thing I'd like to push back on a little bit but I'm hearing from my colleague here the third time and it was raised by this gentleman here, the idea that the platforms are gobbling up the carriers. And I'm not convinced, but I want to be ‑‑ you know, I want to have an open mind and hear about this. My view is that still we have ‑‑ and I got a list from telegeography here, the purchase stuff of the cables that are being laid without, you know, any of the platforms involved. I also showed that the platforms are part of consortia and what I heard best was an interview I did with Telefonica is basically the platforms are their best customers, strategic rivals when they go to competitors and real competitors when they decide to lay their own so it's a multiple and conflictual thing.
>> MODERATOR: Thank you, Dwayne. Camille.
>> CAMILLE MOREL: To answer the need for transparency and the moral issues of public intervention in the public rights sector, I think first requirements on transparency can be had to maybe new conventions, in terms of public‑private partnership, and doesn't consider digital (?) some information that's today are missing. The second question of the public intervention in the public sector, in all issues, of course even in the context of, I don't know, common good or other related issues, my opinion is that in a security perspective, we can consider first that most ‑‑ more cables are landing in a territory, the residency is important, but the thing is that the lake of transparency in how the capacity is used and actually, yeah, used also can be a big problem of identity of which country made an improvement of residency or not. And, yeah, for the questions about new regulations on cables, I think that we have already problems on cyber perspectives on the station where our cables are landing on territory. And so we have to improve cyber gnomes on that especially on landing stations.
>> MODERATOR: Thank you. Robert Pepper?
>> ROBERT PEPPER: According to the clock there's 30 seconds. I'm happy to talk to people sort of offline and we can continue the conversation. I agree, Steve Song is one of the people who has a lot of data and he'll be in a session at 12:30 that we're doing on networks.
On the transparency question I'm a little confused actually, because every time there's an application, you must declare who the owners are, right, so there is a process, and this is a national process. So if in Ecuador they do not require ‑‑ actually ‑‑ I don't want ‑‑ there's a lot here going on, happy to take it offline but in the U.S. where we're part of cables when the application is made, you disclose who the owners are, right? And if a country is not doing that or does not require that, that's national legislation that should ‑‑ it should be required. We can talk about that.
The ‑‑ there's so much to talk about. The questions, I think, about ‑‑ Paul, I'm going to address your question, and it goes back to what Dwayne said. The reason we're investing in cables is because there's a lack of capacity, right? And the reason we also invest, we do not invest alone. So ‑‑ and this goes I think to the Telefonica point. I've had conversation with multiple traditional network operators who said if we had not been a coinvestor they would not have built the cable, right? So the cables they are building, right, as common carriers, right, they're providing connectivity. There is more band width available for public use versus private use because of the investment by a Facebook or Amazon or Google or Microsoft or SAP or IBM, the companies that have data centers that are essentially using capacity for private networks. They've always been private networks. But there's not enough capacity for that. So we see this as sort of mutually beneficial on ‑‑ with ‑‑ in terms of ‑‑ because of the mutely interdependent symbiotic relationship. We need public networks, we also need capacity. By coinvesting it reduce the cost for everybody. And if you look at very specifically all of those cables, what you're seeing is expanded capacity, not shrinking capacity. And, you know, again, we're out of time, but also ‑‑ it also includes things like ‑‑ which we're totally committed to which are IXPs which are open which help address your concerns. We have a project in which we are supporting in a very big way over multiple years ISOC to implement, it's an ISOC project that we're building, building out ISPs over Africa. Again, these are the types of things that makes sure and ensure that these networks are open even if there's capacity used for what have always been private networks for enterprise or corporate networks. I think we're out of time unfortunately but I'm happy ‑‑ I'm going to be around all day and I'm happy to talk to anybody about all of this.
>> CHAIRPERSON: Thank you for helping by promoting a more peaceful and inclusive system. We can continue with UNESCO so I'll see you soon.
>> MODERATOR: Thank you, everybody.