The Internet has enabled incredible information-sharing capabilities, but it has at least one Achilles heel: centralization. Beyond simply being subject to constant surveillance by government agencies such as the NSA (and, likely, many others), access to the Internet is controlled by a few mega-companies: AT&T, Comcast, Verizon, and so on. Furthermore, these companies are hardly independent of the government, as the cooperation between AT&T and the Drug Enforcement Administration (DEA) on maintaining decades of phone records illustrates.
Combined with a centralization of resources and services in the hands of IT giants like Amazon, Apple and Google, the result is that an Internet “kill switch” may be closer than it might appear. Perhaps the president can’t press a red button to shut down the Internet, but he could likely do it with half a dozen well-placed calls to heads of major telecommunication companies. So what’s the difference?
But the tipping of the scales toward centralization, government control and excessive corporate power isn’t just the musings of conspiracy theorists. The macroeconomy is another indicator of this trend: corporate profits are rising while personal incomes are dropping. Yet the economic recovery mantra continues—largely because of a relatively pricey stock market and ostensibly growing real-estate market.
An engineer might look at a technological analog and say that the problem with the system is that it is quickly headed toward a single point of failure—although the failure in this case might just as soon be intentional or malicious as it might be random. Decentralized currencies like Bitcoin have made headlines owing to their lack of central governing authorities and their (theoretical, at least) security and privacy. Could a decentralized network (or set of networks) arise to either complement or even replace the Internet?
Threats to a Free Internet
Every year, some new (and ill-founded) legislation is proposed to increase taxes, restrict freedom or otherwise burden the Internet with new regulations—most of which inordinately favor large companies (if for no other reason than the little guy seldom has any practical legal recourse against corporations). If it’s not SOPA, PIPA or the euphemistically named “Marketplace Fairness Act,” it’s spying by some alphabet-soup agency (NSA, DEA, etc.)—whether warrantless or with the all-too-easy-to-obtain say-so of some possibly secret court. These laws and agencies have a fairly simple task, owing to the centralization of Internet resources and the complicity of major IT and telecommunication companies.
The alternative is, of course, decentralization. The challenge is implementing it. The MeshNet (formerly DarkNet) project is one effort in this direction: essentially, instead of relying on centralized infrastructure provided by, say, Comcast or another provider, each user serves as a node in the network—a potential “hop.” Data is routed to its destination over these hops (users) in a manner similar to how it would be routed over servers. The MeshNet concept relies on wireless rather than wired networking, so it is more localized when implemented at small scales.
Regardless of the particulars of the project, a decentralized network faces a number of challenges, but over time, technology (even current technology) could be adapted to meet them.
The first challenge of an Internet alternative is to establish network links. The near ubiquity of Wi-Fi creates a strong foundation for local connections. Assuming you live relatively close to others, you can probably check your computer or device and find a number of Wi-Fi networks in range. Imagine interconnecting these networks: you have created a simple mesh network and taken the first step toward a decentralized Internet replacement. Of course, working out the software details may take some time, but in its infancy, the Internet had to deal with similar challenges. Certainly, any Internet replacement will start small and take years—probably decades—to reach a large scale.
What about a broader reach? Long-haul communication may be the biggest snag for mesh networks. Wi-Fi capabilities could be expanded to increase the range of local networks, but interconnecting communities/cities and even countries is another story. Radio communication is one possibility, although it certainly lacks the bandwidth of fiber. Installing fiber is expensive and almost necessitates some kind of centralization, which defeats one of the main purposes of establishing an Internet alternative. For low-bandwidth communication, however, it at least enables a connection.
Implementing decentralized infrastructure may be somewhat problematic if one holds onto the notion of large amounts of traffic flowing through single points, but if the mesh network implements traffic control that reroutes data in a manner that prevents bottlenecks, individuals need not maintain racks of servers. Given that desktop computers are overkill for most users, their spare processing and networking capabilities could act as node-like “mini-servers” in the network. And Moore’s Law—although perhaps nearing its end—still has a ways to go, even if desktops have already reached the “good enough” level for most users.
Such a network could also enable cellular-like communications without the heavy fees of cell-service providers. In most cities and suburbs, you’re almost always in range of some Wi-Fi network. Were those networks meshed and accessible, phone calls could be made in a manner that emulates cellular. The “carrier” is the mesh network—a group of independent users.
Do Internet Alternatives Have a Future?
One impediment of such a network is, of course, stinginess. Individual users might be unwilling to let others use their networking resources (they do, after all, consume power, and someone has to pay the electric bill). But the monthly cost of participating in a properly implemented mesh network could be much less than a typical high-speed Internet bill. And it could be thought of as insurance (something everyone loves these days, albeit for mysterious reasons): an ISP outage doesn’t mean you’re completely out of touch with the world.
The Internet currently has momentum, but that momentum has been hampered by revelations of NSA spying and other efforts to restrict electronic freedom. But mesh networks and other alternatives may still seem like curiosities with no real benefit. Should the government ever “kill” the Internet—as happened in Egypt—even for a short duration, the benefits would be quickly illuminated.
Furthermore, a mesh network need not be a complete replacement for the Internet: it could act as a local complement that operates alongside the Internet and serves as a replacement only when needed. Unfortunately, however, necessity may be the only realistic driver of significant progress in mesh networking; pending an emergency that leads to Internet outage or some other negative result, alternative networks face a long upward climb to visibility and relevance in the eyes of most users.
The Internet grew as a model of freedom and opportunity, but its increasing control by a few corporate giants and the expanding voyeuristic intrusion by a violent U.S. government threaten this heritage. Moving all at once to a new, decentralized network model would be nearly impossible, but parallel investigation of alternatives could prove valuable in emergency situations or if government controls become onerous. The MeshNet project illustrates one possibility.
Dealing with long-haul, large-bandwidth communications is likely the main technical sticking point for mesh networks. The expense and need to traverse multiple properties with physical infrastructure (fiber, for instance) virtually demands the work of large organizations like corporations and/or governments. Unfortunately, this approach defeats one of the key goals of decentralized networks.
Like most problems, there is no perfect solution in the case of the Internet. Presently, the benefits of using the Internet outweigh the costs (in lack of privacy and so on) for most users, and given the vast array of Internet resources, alternative networks offer insufficient value to drive widespread interest. In the event that individuals and small companies begin to see a need for such an alternative, however, these networks could quickly gain interest and investment.
Image courtesy of OLPC