Kill My ISP
We trust Internet Service Providers with some of the most intimate details of our lives, but why should we? Around the world, people are repurposing inexpensive routers and antennas to build their own, community-controlled internet.
Every morning when I go online to access my favorite counter culture publications and email my boss, my computer requests and receives data from web servers far away from my house, and this internet traffic whizzes through cables owned by strangers. I rely on these strangers to relay my page requests to the proper destination without alteration. I also trust them to treat me fairly and not slow down my traffic if I choose to stream music from Bandcamp.com instead of Apple Music, for example.
The strangers that own and control most internet traffic are Internet Service Providers (ISPs). When using the internet, I trust them with some of the most intimate details of my life, but why should I?
Over the past 40 years, the internet has gradually passed from being controlled by scientists, hobbyists, and defense department contractors to multinational corporations like AT&T and Verizon. Decades before the invention of the World Wide Web in 1989, the US Department of Defense and later the National Science Foundation funded early development projects to connect computers within a network. The technical and infrastructural groundwork of the internet was designed by scientists who used the network to link their universities and rapidly share research. In other words, the people who were using the early internet the most were also writing its rules and protocols, giving the users a strong sense of network ownership and autonomy. This dwindled rapidly in the 1990s when the US government sold off infrastructure and technology funded by tax dollars to a handful of private companies, which facilitated both the popularization and privatization of the internet.
The internet then (and now) relied on a method of transmitting information called “packet switching.” Messages are broken up into packets that each have the destination’s numerical address. The packets are sent from one node in the network to the next, moving closer to the destination until it is finally reached. Packet switching differed from the circuit switching of telephones because instead of having a central point where actual people (and later machines) would create a dedicated route connecting the phone lines of two people, you could have a distributed network where different portions of your message could take different paths across the network and still reach the same destination in one piece. If I was at Stanford in the 70s and wanted to send a message to someone at MIT, my message would be broken up into digital packets each with MIT’s address on it and passed along the fastest route to MIT. We can imagine the packets would go from Stanford to University of Southern California to the University of Utah and perhaps a few other points until they made their way to MIT. If that path was congested with other traffic, some portion of my message may be sent via another route, say from UCLA to Illinois to Rutgers. No matter the route they took, once all my packets reached MIT, the recipient’s computer would be able to reassemble them into the complete message. This required trust of other members of the network to properly relay packets towards their destination and to refrain from reading messages not addressed to them.
In the early days of the net, this trust wasn’t a problem since the other members of the network were fellow researchers and part of a tightly knit community, but as the network expanded to more casual users, the division between users and administrators got sharper and demanded more and more trust of strangers.
While the early web had a bottom-up ethos, it was used by a small and fairly homogenous group of people. In the twenty-first century we have seen significant deregulation that has allowed telecoms to further centralize their control over the web.
Last year we saw a strong blow to our digital rights when the FCC undid several Obama-era protections. In the spring, a rule which barred ISPs from selling customers’ traffic data was repealed, giving ISPs free reign to sell our private data to marketers and inject ads into our web traffic. If my web traffic suggests that I’m in a pinch and looking to get rich quick, my ISP might guess that I’m susceptible to gambling advertising and subtly insert it into my traffic, or it may infer what sort of ailment I have based on my webMD searches and sell that information to insurance or pharmaceutical companies. By the end of the year, the FCC repealed the widely supported net neutrality rules which makes it possible for ISPs to prioritize traffic they profit off of and slow down access to any website that doesn’t pay their tolls.
This recent wave of internet deregulation exposed many people to the willingness of politicians and legislators to please powerful telecommunication giants at the individual’s expense, and this has reinvigorated pushes for locally controlled broadband. Digital rights advocates on the left and some tech journalists have highlighted the need for municipal government-run projects that treat broadband like a public utility in the same way many cities already treat services like electricity, water, and sewage. A popular example is Chattanooga where the city’s Electric Power Board invested tax dollars into laying fiber-optic cables that provide high speed internet to all of those already served by the city’s electric utility company. A study by Harvard’s Berkman Klein Center this year showed the economic viability of community-run fiber-to-home ISPs. These public utilities rely on infrastructure identical to private ISPs, but they are generally less expensive and more invested in protecting the individual than their private competitors.
Fiber undeniably provides the best and fastest connection to consumers, but the cost of burying cables can be prohibitively expensive for some local governments. Even in cities with the necessary funds, public fiber is unlikely a feasible option since one in three states have laws that ban public competition with incumbent ISPs. This is a sign that telecommunication companies have been able to influence politicians even at the municipal and state level.
Thankfully, there exists another approach to building community connectivity that has much lower barriers to entry and involves the entire community in its very design. Across the US and internationally, there are a variety of community-led projects that are repurposing inexpensive routers and antennas to create mesh networks.
Mesh network architecture is radically different than that of fiber optic or copper cables used by commercial ISPs. Commercial ISPs use underground cables, which require permission from local governments and lots of money to install and manage. Mesh networks, on the other hand, are made up of routers owned by participants in the network, and are therefore decentralized and nonhierarchical by design. Routers are usually used to route internet traffic between an ISP and different computers and phones in a home, but in a mesh network, routers are reconfigured to communicate with other nearby routers as well. These routers serve as nodes that relay traffic from one nearby router to the next until it reaches its destination in the network, rendering the ISP obsolete.
To understand the difference between a traditional ISP and a mesh network it is important to understand the basics of how the internet actually works. The internet (a contraction of ‘interconnected network’) is not a single network but a network of networks. The commercial ISPs that most people begrudgingly pay every month (such as TWC, AT&T, Comcast) create the first network, a network of underground cables that connects to all the houses in a geographic area. This network connects to the “internet backbone,” relaying traffic to a series of larger networks such as the submarine cables that run along the bottom of the Atlantic Ocean and the cables that traverse the US connecting major cities. If I want to message my friend in Germany, my message first passes over the ISP’s local network, then through a series of larger intercontinental networks, and finally over my friend’s ISP’s network in Germany. When people raise concerns about threats to our privacy or net neutrality, they are primarily concerned with the first and last network our traffic passes through – the commercial ISPs who are the most likely to use our data for advertising or slow down our traffic for profit.
Commercial ISPs currently act as the gatekeepers of connecting us to the rest of the internet. Mesh network architecture uses routers in individual homes to create a decentralized network that can simultaneously keep local communication local – essentially off-the-grid – and connect users to the rest of the web without relying on greedy private ISPs. Every router/node communicates with other nearby routers to relay traffic within the mesh network until it reaches a node that has access to the Internet. These nodes are either sharing commercial ISP subscription with other members of the network, or the nodes are connected directly to the internet backbone and therefore completely cutting commercial ISPs out of the equation.
Regardless of how or whether the mesh network connects to the rest of the internet, it is first and foremost an independent network. Without being connected to the internet, users of a mesh network can locally host websites, share files, or message each other. This means that on a mesh network I can message my neighbor or visit a forum discussing community issues without this traffic passing over infrastructure in the hands of private companies and prying eyes. Mesh networks can have their own top level domain (things like .com or .org) that are only accessible to those nearby and connected to the mesh network. I can run MyWebsite.mesh and host discussions that I know can only be seen by my neighbors and other community members. This aspect of mesh networks has been put to use by different groups who need hyper-local, off-the-grid communication tools in times of ecological and political crises.
In 2012, when Hurricane Sandy left much of New York without internet service or electricity, FEMA workers and residents of the Brooklyn neighborhood Red Hook were able to communicate with each other over the mesh network launched earlier that year. In 2014, when thousands of pro-democracy protesters occupied Hong Kong’s central business district, organizers encouraged participants to communicate over the mesh networking app FireChat in order to avoid sending messages over the internet which was seen as prone to surveillance and blackouts. They were drawn to FireChat because it uses each participating phone’s’ bluetooth capabilities in place of routers to create an ad-hoc mesh network between users. Similar mesh network projects were launched in occupied Zuccotti park in 2011 to allow for local discussions.
Mesh networks create a digital network outside the state’s sight which is crucial for people involved in various political struggles, and the very process of building this network has the transformative effect of bringing together otherwise unconnected groups to collaborate on connecting and communicating with each other. Mesh networks radically alter communities’ relationship with connectivity. Instead of begging ISPs to refrain from throttling or spying on traffic, mesh networks put control directly into community hands. Around the country and around the world, communities are coming together and giving the boot to corporate ISPs.
A number of mesh networks already exist in major American cities. In addition to Red Hook Wifi, New York also has NYC Mesh. In Oakland there is the People’s Open Network and in Detroit there is the Equitable Internet Initiative (EII). In addition to building a cheaper, community-run network, these US-based projects also seek to tackle the problem of digital redlining by ISPs who overcharge and under-serve communities of color. Internationally notable examples include Altermundi in Argentina, Athens Wireless Metropolitan Network, and Spain’s Guifi which has over 34,000 active nodes.
All of these projects share the goal of building a community-operated, decentralized internet that provides free or inexpensive connectivity and promotes locally relevant content. While they all started off as mesh networks using inexpensive routers and antennas, some projects like Guifi have incorporated more expensive technologies – such as fiber-optic cable or more powerful antennas – to increase the speed and reach of the network. What is important about these initiatives is not so much the technology they use but their emphasis on horizontalizing control and knowledge creation. Of course, mesh networking is just one particularly inexpensive way to achieve this, but it is a powerful one that points to the efficacy of community technology amid what appears to be an irrevocable concentration of power and capital.
Community-led mesh networking presents us with a possibility to take back control of our communication networks. If you want to help build a more autonomous internet, find a mesh network project near you or start your own project. Share your bandwidth with your neighbors. Create a miniature intranet with nearby friends to share files and send off-the-grid messages. The internet is full of tutorials on how to repurpose cheap routers and start experimenting in the creation of local networks.
The first step in building autonomy is dismantling illegitimate hierarchies and recognizing the power we already have. The dominance of commercial ISPs rests only upon myths that legitimize hierarchies and mystify technical knowledge. Mesh networks show, however, that any of us can teach ourselves how to use inexpensive equipment to replace ISPs. Even if you have no technical background you can take part in building a network by simply copying and pasting code that others want to share with you. Big telecoms do not deserve their monopoly over our means of communication, we do.