Information technology - the network of shadows / Julian Dibble

Governments and corporations now control the Internet more than ever. Activists in the digital field are interested in creating an alternative network that will never be able to be blocked, filtered or closed

Just after midnight, on January 28, 2011, after three consecutive days of mass protests against the government organized in part through Facebook and other online social networks, the Egyptian government did something unheard of in the history of communication in the 21st century: it turned off the Internet. Exactly how this was done is not yet known, but the evidence suggests that all it took was five phone calls to the right places - one call to each of the country's major Internet Service Providers (ISPs). Network routing records show that at 12 minutes after midnight, Cairo time, Telecom Egypt, the leading Internet provider, began disconnecting its customers' Internet connections, and the other four providers followed in the next 13 minutes. The operation was completed 40 minutes after midnight: according to the estimate, about 93% of the Egyptian Internet was no longer accessible. In the morning, at sunrise, the demonstrators made their way to Tahrir square in almost total digital darkness.

From a strategic and tactical point of view, the impact of the internet blackout was not great. The number of protesters that day was the largest, and in the end they had the upper hand. However, the shutdown was a startling example and an important lesson we should have learned a long time ago about the Internet's vulnerability to takeover from above.

We have heard a lot about the Internet's ability to resist such control. The technological foundations of the network, so we are sometimes told, have their origins in the search made during the Cold War for a communications infrastructure so robust that even a nuclear attack could not paralyze it. Although this story is only partially true, it describes a little of the power that is built into the network's elegant decentralized design. The TCP/IP protocol, which defines the Internet through the multiple and redundant paths it creates between every two nodes in the network, and through its ability to adopt new nodes at any moment, should ensure the flow of information regardless of the number of blocked nodes or the cause of the blockage, whether it is a nuclear bomb or a dictatorial regime. And according to the famous saying of the digital rights activist, John Gilmore: "The Internet interprets censorship as physical harm and creates paths to circumvent it."

That, anyway, was the plan. In practice, if five phone calls can cut off 80 million Egyptians' access to the Internet, something is not working as planned. The outage in Egypt was just the most prominent example in a growing list of examples of how the Internet can be very vulnerable to takeover from above. During the revolution in Tunis, a month before, the authorities took a more targeted approach and blocked only a few websites from the national internet. In the demonstrations held after the elections in Iran, in 2009, the Iranian government slowed down the traffic on the domestic network instead of stopping it completely. Also, for years the "Great Firewall" gave the Chinese government the ability to block any website it chose. In Western democracies, ISP mergers have led to a small number of corporate entities, which control larger and larger portions of Internet traffic. This gave companies like Comcast and ATT the drive and ability to drive their media partners' traffic at the expense of their competitors.

So what happened, and can it be fixed? Is it possible to restore the durability and dynamism of the Internet as they appear in Gilmore's idyllic description, qualities that allow the network to overcome governmental and corporate barriers and delays? A small but dedicated community of digital activists is working on a solution, and this is what it might look like.

It is a bright afternoon at the Wien-Zemering power station, in Vienna, Austria. Aaron Kaplan spent the last seven minutes locked inside a dark and crowded service elevator, on his way to the end of the factory's exhaust chimney, which is 200 meters high - the tallest building in the city. When Kaplan eventually emerges onto the surface at the summit, the view around him stretches from the foothills of the Alps in the west to the green lands of Slovakia in the east, and the Danube River shimmers straight below. But Kaplan didn't come here for the view. He went straight to the edge of the pad to take a look at four small Wi-Fi routers that were attached to the safety rail and that the weather was giving off its signals.

These routers form one of the nodes in a non-profit community network called FunkFeuer, of which Kaplan is one of the founders and chief developer. The signals that the routers transmit and receive link them, directly or indirectly, to about 200 similar nodes on rooftops all over Vienna. Each of them is owned by a private user who installed and maintains it, and each of them contributes its bandwidth to a high-speed, community Internet connection that serves an area almost as wide as the one Kaplan sees around him from the end of the chimney.

The FunkFeuer network is what is known as a wireless mesh network. You don't have to pay connection fees to this network. All you need is $150 worth of hardware ("No more than a Linksys router in a plastic lunch box," Kaplan says), a roof to put the equipment on, and line of sight to at least one additional node. There is no need for direct radio communication with more than a few nodes, since each node relies on its nearest neighbors to pass on information intended for nodes that it itself is unable to reach. In the first months of the network, shortly after Kaplan and his friend Michael Bauer founded it in 2003, there were a little more than 10 nodes in the network, and the "side-to-side" transmission method was a dubious matter: if only one node stopped working, most likely the others would disconnect From that, or worse than that, they would also disconnect from the only router that connected the entire network to the external Internet. In those days, Kaplan recalls, "we had to struggle" to keep the network active around the clock. He and Bauer visited the houses many times to help repair broken junctions - among other things, working on a roof at two in the morning, in the middle of a snowstorm and minus 15 degrees, which was only possible thanks to the mugs of hot wine that Kaplan's wife gave them.

However, when the DIY tech community in Vienna realized what FunkFeuer had to offer, the network began to grow. At a certain point, when it had between 30 and 40 nodes, it became stable. The topology of the network was made rich enough to ensure that if any node failed, the others that relied on it were able to find another path. The network has reached a critical density where, as Kaplan puts it: "The magic of the tangle network begins to work."

Mesh networking is a relatively young technology, but the "magic" that Kaplan talks about is nothing new: it is the same principle that underpinned the reputation of the durability of the internet infrastructure. Packet-branded routing using the "store-and-forward" method - where every computer connected to the network is able not only to send and receive information, but also to transmit it for other connected computers - has been a central feature of the Internet architecture since its inception. This is what creates the variety of transmission paths available, which allow it to simply "bypass vulnerabilities", and this is what makes the Internet, at least in theory, so difficult to eradicate.

If the reality of the Internet today were more like the theory, there would be no need for tangled networks. But in the twenty years when the Internet grew beyond its academic roots and became the common commercial service we know, the role of the "store-and-forward" principle was less and less significant. The vast majority of new nodes added to the network during this period were home and business computers, connected through Internet providers. In the providers' connection model, the customer's computer never serves as a relay station: it is always just an endpoint, a terminal that sends and receives information exclusively through the provider's computers. In other words, the tremendous growth of the Internet has added almost no new paths to the network map, but mostly dead ends. Service providers and other traffic hubs have become focal points for controlling the hundreds of millions of nodes they serve. If the service provider goes down or cuts them off, these nodes have no way around the damage. Instead of keeping the Internet immune from vulnerabilities, the service providers have become, in effect, the weak points of the network.

In contrast, mesh networks do exactly what service providers do not: they allow the user's computer to act as a relay station for data. In less technical terms: they allow the user to stop being just a customer, and start being an internet provider himself. To better understand the meaning of this, imagine what could have happened on January 28 in Egypt, if the communication between the citizens had not gone through a limited number of providers, but through a tangled network. At the very least, it would have taken a lot more than five calls to close the net. Since each user of the tangle network controls their own small piece of infrastructure, it may have required as many calls as there were users - and much more persuasive effort would have been required for most of them than was required of service provider managers.

Sasha Meinert, 37, has been a major player in the community tangle chain scene since it first appeared. While a graduate student at the University of Illinois, he helped create the Champaign-Urbana Community Wireless Network (CuWiN), one of the first networks of its kind in the US. After that, he co-organized a volunteer response team that set up a 60-kilometer-diameter makeshift grid in the Hurricane Katrina disaster area and restored communications to the area in the first weeks after the impact. In those years, he moved to live in the capital, Washington, with the aim of establishing a business for community networks, but along the way he was "hunted", as he says, by the New America Foundation - an influential think tank, which hired Meinert to create and supervise technological initiatives. There he launched, in 2011, the wireless project Commotion - an open source wireless tangle networking initiative, backed by a two million dollar grant from the US State Department.

According to Meinert, the goal of the project in the short term is to develop technology that will "bypass any barrier and any centralized eavesdropping attempt." To illustrate the idea, he and the other prominent Commotion developers created a prototype that can be called "Internet in a suitcase": a small, integrated case of wireless communication hardware, suitable for smuggling into the territory of a tyrannical government. From there, rebels and activists will be able to provide unblockable internet coverage. In fact, the suitcase system is nothing more than a basic, ready-to-use collection of technologies well known to tangle mesh enthusiasts. Any geek with sufficient motivation can create one and run it.

The question facing Meinert and his colleagues in the long term is: "How to make the suitcase simple enough, so that even 99.9% of the world's population that is not made up of geeks can operate it?" Because the more people use a tangle network, the harder it will be to eliminate it.

This principle is numerically self-evident: it is easier to silence a network of a hundred nodes, node by node, than to silence a network of a thousand. Even more important, perhaps, is the fact that a larger network tends to contain more connections to the general Internet. These connections, the rare transition points between the thicket and the rest of the Internet, become a less problematic weak point the larger the network. When there are more such data upload connections within the local network, day-to-day communication is less disrupted when some link to the global network goes down. Since every node in the network can, in principle, become a data upload connection through an external link to the Internet (by dialing the provider, using a cell phone, etc.), there is a greater likelihood of quickly restoring contact with the outside world.

In two words, size matters. Therefore, the open question of the ability to expand - to what size the networks can reach - is a burning question among those involved in the field. Are tangle networks capable, even theoretically, of containing a significant number of nodes without becoming too slow? Opinions are still divided, depending on the number of nodes that is defined as significant. Only a few years ago, some network engineers claimed that the size of the tangle could not exceed a few hundred nodes. Today, the largest pure mesh networks contain a few thousand, and there are dozens of thriving community networks, the largest of which use a combination of mesh and supporting infrastructure and reach five thousand nodes (the wireless metropolitan network of Athens in Greece) and even 15 thousand (the Guifi.net network in and around Barcelona). The doubt that still gnaws at is how much the tangle networks are able to grow further in terms of users, given that most humans are reluctant to mess with such complex technologies.

Unlike most open source technologies, which tend to downplay the importance of a user-friendly interface, the tangle movement is beginning to understand how essential it is that its equipment be simple. The Commotion project may not be the only one trying to simplify the use of mesh networks, but the main simplification it offers is unusual and revolutionary: instead of making it easier to install and operate a mesh node in the user's home or business, Commotion's goal is to dispense with this equipment altogether. "The concept is that cell phones, laptops, existing wireless routers and the like can be adapted for this purpose," Meinert explains, "and build a network from what is already in the pockets and bags of the users." He calls it the "devices-as-infrastructure" network, and according to his vision, adding a node to the tangle would only require flicking a switch. "Basically," he says, "you just press a button on your iPhone or Android phone, and confirm that you want to join this network. It should be that simple.”

So imagine a world where tangle networks have reached this level, crossed the barrier of mass accessibility and become, more or less, another application running in the background. What will happen now? Will the low cost of do-it-yourself Internet service crowd out the commercial possibilities, until the last of the centralized service providers disappears, giving way to one worldwide tangle?

Even the most ardent proponents of network decentralization don't believe it. "In my opinion, such a system will always be an 'Internet for the poor,'" says Jonathan Citrine, a professor at Harvard Law School and author of the book "The Future of the Internet: And How to Stop It." Citrin would love to see the tangle approach succeed, but acknowledges that it will probably never reach the efficiency levels of more centralized networks. "Centralization has real advantages," he says, "among them ease of use."

Ramon Rocca, the founder of Guifi.net, also doubts that tangled networks will cause Internet service providers to go bankrupt - and even that they will ever take more than 15% of the market from them. Nevertheless, even such a slight penetration can "clean up the market", Roca claims, opening it up to poor parents who cannot afford a normal connection, thereby causing the leading providers to lower prices for everyone else.

These economic effects are welcome, but the civilian effects—the inherent resistance of the tangle networks to censorship and eavesdropping—need a much larger market share than 15% to thrive, and if it's clear that market forces alone won't be able to greatly increase that number, what will?

Usually, when the market is unable to provide a "social" good, the first place to go is the government. In this particular case, it's not a terrible place: the same tangled web that is able to bypass censorship as if it were a physical injury, can just as effectively bypass real physical injuries, and this makes it a perfect communication channel in the event of a hurricane, earthquake, and other natural disasters of the kind that governments are supposed to protect. from them on the citizens. Therefore, Citrain argues, the wise policy on the part of governments would be to take an active part in spreading tangled networks, not only among revolutionaries in foreign countries but also among their own citizens. All we need is, for example, a requirement that every cell phone sold in a certain country be equipped with emergency mesh networking capabilities, so that the phones can turn themselves into hubs and relay stations at the push of a button when needed. In terms of public policy, says Citrain, "building something like this is very simple, and the national security and law enforcement authorities should, as a rule, welcome it."

The catch, of course, is that it is just as easy to imagine the law enforcement authorities going against some national tangled network, claiming that it is a means of communication for criminals and terrorists, which the telephone and Internet companies cannot allow listening to. This is the problem with relying on governments to support a type of network, which in many cases is intended to solve problems that the governments themselves created.

If so, it's hard to trust the government to help with this issue more than the market will, but Ibn Muglan has some alternative ideas. Muglan is a professor of law at Columbia University, and served for many years as the lawyer for the Free Software Foundation - a non-profit organization of activists in the digital field. In February 2011, partly following the news from Tunis, he announced a project called FreedomBox. He also announced, on the crowdfunding site Kickstarter, that he was looking for an initial investment for the project, and managed to raise $60 in five days.

The FreedomBox project is similar in some ways to Commotion and this is no coincidence as Meinert was a member of the FreedomBox Foundation's Technical Advisory Committee. Like Commotion, the project presented an attractive and ground-breaking prototype: a small white mesh device called the FreedomBox, which costs "$149, in initial quantities, and will eventually be replaced by hardware that costs half that or less," Moglan says. .

The project, again similar to Commotion, is not limited to any specific gadget. This is actually software code that can run through the networked processors that we accumulate in our homes and around our lives in ever-increasing numbers, like "accumulations of dust under the couches," in Muglan's language. All of these can become an Internet infrastructure that will "restore the balance to privacy" and restore the vision of a "decentralized peer network." There are IP addresses for smart TVs, refrigerators, and any of those, according to Moglan, could be a FreedomBox. This is not only about the decentralization of the infrastructure, but also the decentralization of the data. For example, Moglan considers the concentration of user data in a cloud service such as Facebook or Google to be no less a threat to privacy and freedom of expression than the concentration of traffic in the hands of Internet service providers. To combat this trend, the FreedomBox code will be optimized for the activation of alternative social networks, such as Diaspora, which store the personal data on the user's computer and only share it with selected people through peer-to-peer networks.

However, Muglan says that the main component of the project is "the political will displayed by a generation of young people who, due to their dependence on social networks, are becoming more and more aware of their online vulnerability and that of others." He relies on this awareness, among other things, to motivate the many programmers who contribute their work to the project. It is also the factor most likely to push users to adopt the technology. With the exception of a sustained campaign of techno-activism, Moglen argues, it is unclear what will succeed in making the average user understand the true costs of the erosion of freedom and privacy, in which we pay for convenience of use and other benefits, tangible in the immediate term.

According to Muglan, "People underestimate the damage caused by the loss of privacy, just as they underestimate the importance of the dire consequences of activities that harm the environment," such as the accumulation of garbage and pollution. "It is difficult for humans to think ecologically. It's something the primate mind didn't evolve for."

This means that reinventing the Internet can never be done just by adapting the technologies. A broad and long-term political movement like the environmental movement may be required. If both the market and the government are unable to get us there, perhaps only a collective change of consciousness will succeed, similar to the change created by the green movement through force of will. In the past, no one wooed. Today we do it. Nowadays no one uses a tangled infrastructure. In the future, it might happen.

Even in that case, there will not be a single technical measure that is sufficient to preserve the freedoms that the Internet promises and embodies. The reason for this is that even the most perfect and durable Internet cannot withstand, standing by itself, the social and economic forces that push in the direction of centralization. Mesh netting is just one way to push in the opposite direction. "Such tangle networks are good for communities, and the bigger the better," explains FunkFeuer's Kaplan. But even a single global tangled network would be at risk of repeating the evolutionary steps that created today's vulnerable Internet. "Rishot Sabach is not a substitute for the Internet, it is only a part of it," he says. "There is no place for utopia here."

______________________________________________________________________________________________________________________________________________________________________________________

in brief

The Internet was designed as a distributed system, where each node is able to connect to many others. This design helped make the system resistant to censorship or external attacks.

However, in practice most private users are on the edge of the network, connecting to others only through their Internet Service Providers (ISP). When this link is blocked, the Internet access also disappears.

An alternative option that is now starting to appear is wireless mesh networks - simple systems that connect end users to each other and find their own workarounds around blocking and censorship.

Mesh networks must reach a critical mass of users to function well. The developers must convince the potential users to give up some of the convenience of use in exchange for additional freedom and privacy.

About the author

Julian Dibble (Dibbell) has been writing about Internet and digital culture for about twenty years. He wrote the book Play Money: Or How I Quit My Day Job and Made Millions Trading Virtual Loot, and edited the book "The Best of Technological Writing, 2010".

And more on the subject

A Survey on Wireless Mesh Networks. IF Akyildiz and Xudong Wang. IEEE Communications Magazine, Vol. 43, no. 9, pages S23-S30; September 2005.

Freedombox: http://freedomboxfoundation.org

Funkfeuer: www.funkfeuer.at/index.php?L=1

Mesh Networks Research Group: www.mesh-networks.org