You may have seen Google’s “Take Action” page, which encourages netizens to voice their support for a “free and open Internet”. A number of organizations are offering similar petitions, and there is even a dedicated website with the domain, ProtectInternetFreedom.org. A good overview of the controversy by Ambassador David Gross and Ethan Lucarelli is available at WhosWhoLegal.
What lacks from these sites is the deeper context of the controversy: the history of the Internet. My paper on the early history of the Internet (paywalled, alas)* describes briefly how an organization now part of the ITU, the CCITT, tried in the 1970s to develop global computer networking standards. (Sorry – lots of acronyms here.) The standard they produced was called X.25, and was designed in response to a fear that IBM’s proprietary protocol, SNA, would dominate computer networking. TCP/IP was proposed by the US as a possible alternative standard, but rejected by the CCITT. Some networks were built using X.25, but many of these also allowed TCP/IP to run on top of X.25. Of these three protocols, X.25 (and SNA) are ancient history (by Internet standards) because pretty much everyone started using TCP/IP in the 1990s. Another way to look at this is to say that TCP/IP, X.25, and SNA competed against one another, and TCP/IP won because Internet users preferred it to the others.
To make sense of the previous paragraph, it may help to understand the various roles involved. At the top of the pyramid are users — namely you, the reader whose computer is connected to this page, but also WordPress, whose servers connect to the network as well. Even a company as massive as Google is technically a user of the Internet, because it does not own the network itself. A step beneath the users is an Internet service provider — a company or perhaps a government-owned network, which actually owns the connection to your house or workplace. Depending on your provider, that company may in turn connect to one of the corporations comprising the Internet backbone, or it might be one of those corporations. When you see an ad on a Google page, Google gets that revenue, but the provider and the backbone company do not; their revenue comes from what users pay for use of the network — your monthly bill for service. The way Internet service is priced (at least in the U.S.) assumes all data packets are the same, in part because TCP/IP treats all data the same, and this makes it relatively cheap.
All this to say: TCP/IP privileges network users over network owners. This fact was a deliberate decision built into the protocol from very early on in its development, and while the decision was made for sound technical reasons, it also has political consequences for the relationship between network user and owner. Especially in countries where the network is heavily regulated or government-owned, these consequences shift the balance between citizen and regime. This is the power of the Internet, and the value of a ‘free and open’ Internet. This is also what makes it possible for companies like Google, Facebook, and Amazon to thrive on the Internet as users. The technical decision to privilege users is the same feature underlying the net neutrality debate: network owners in the U.S. sought the ability to identify packets which are more valuable to users (advertisements, Facebook, Netflix streaming), and charge them more for those data. Users won that debate in 2011, when the FCC published rules prohibiting service providers from discriminating against lawful content.
The problem that TCP/IP creates for the ITU is that it profoundly disadvantages the telecommunications corporations (some private companies, some government-run), which make up the ITU’s membership. Companies like Verizon can be members of the ITU-T — as the CCITT is now called — but cannot vote on ITU business. Only governments can vote. The catch is that around 2/3rds of member countries have either a private or public monopoly on telecommunications — at least as of 2002, according to the ITU (PDF — see Figure 1), as of 2002 approximately 2/3rds. This fraction is somewhat lower now, but having a powerful national telecom monopoly creates a strong incentive for governments to act to reshape the Internet in favor of network owners. The U.S. — which hosts around 43% of the world’s top million websites, despite having only 10% of Internet users — gets a lot of traffic from other countries, and that translates to some degree into revenue flowing to the U.S., but also ideas, culture, and other information flowing out of the U.S. into these countries. Some governments and telecoms wish to capture a portion of that revenue, or to stem the influx of content, or both — and the ITU provides a venue for that debate.
So the ITU meeting is an opportunity to have the net neutrality debate all over again, only this time on the world stage, with odds against users. Because of the predominance of network-user corporations in the U.S., it likely has the strongest domestic lobby in favor of Internet users — yet even the net neutrality decision was not an easy win. We see evidence of this lobby’s influence in our work on free trade, where the U.S. has pushed for inclusion of ‘openness’ provisions for data flows into trade partners. Some trade partners have pushed back on these provisions, citing the interests of their citizens. So while the U.S. government may advocate for an Internet friendly to users, it is widely seen as advocating for the interests of particular U.S. users. It is not clear how this will play out in the ITU, where the U.S.’s best arguments for Internet freedom may be undercut by the government’s obvious interest in Internet commerce. In any case, most governments will be much more influenced by the interests of their telecoms.
To understand the debate in and around the ITU, it is important to remember that the telecoms’ vision of the Internet is very different from what users currently enjoy — and, in fact, that vision was tried in the 1970s and 1980s, but did not survive competition from TCP/IP and the Internet as it is. The irony is that as recent as ten years ago, the ITU could have likely implemented new regulations to revert the Internet to owner control — companies like Google, Facebook, and Amazon weren’t strong enough to mount any challenge. But in the last ten years the economic case presented by network-user companies, as well as the moral case from the Arab Spring, uprisings in Burma, and the color revolutions, have shown the world the power and potential of a user-centered Internet. This makes it more difficult — but not impossible — for the ITU to act for the network owners’ interests, but it also makes any such decision more devastating.
*a free but rough version, which covers most of this history and the relevant debates, is available on my website (PDF).