The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Volokh Conspiracy

Does the FCC really not get it about the Internet?


[I am posting below a short essay by my friend and colleague Brett Frischmann of Cardozo Law School concerning the "net neutrality" rules now being considered by the FCC. I've stayed largely away from the whole net neutrality debate over the years—too much inside [FCC] baseball for my taste—but Brett's been in the middle of it for some time, and I think he draws attention below to a simple, but very profound, problem at the heart of the approach the agency is taking: the distinction it is drawing between "edge providers" (suppliers of content) and "end users" (recipients of content). One can hardly imagine a more inappropriate distinction one could draw on the Internet that we now have, so much of whose power has come from its "end-to-end" design: all senders are recipients, all recipients are senders, and all IP addresses are equal. But I'll let Brett fill in more of the details./DGP]

The key to an open Internet is nondiscrimination and in particular, a prohibition on discrimination or prioritization based on the identity of the user (sender/receiver) or use (application/content). I explain why at length in my book, Infrastructure: The Social Value of Shared Resources (2012). Unfortunately, the rules now being considered by the FCC don't come close to implementing this simple and important benchmark. There are many reasons for this, but perhaps the most fundamental is a simple misconception, one that persists in the work of the FCC but also of proponents and opponents of network neutrality. It is the false distinction between what they call "edge providers" (YouTube) and "end users" (people who watch videos on YouTube).

The currently proposed FCC rules for an open Internet are based on the following definitions:

Edge Provider. Any individual or entity that provides any content, application, or service over the Internet, and any individual or entity that provides a device used for accessing any content, application, or service over the Internet.

End User. Any individual or entity that uses a broadband Internet access service.

The basic distinction suggested by these definitions is that some distinguishable subset of end users use broadband Internet access to provide content, applications, or services (or devices to access content, applications, or services), and these end users are edge providers.

Who exactly are the end users that are not edge providers? In other words, who uses the Internet but does not provide any content, application, or service? The answer is no one. All end users provide content as they engage in communications with other end users, individually or collectively. YouTube content, for example, comes from end users uploading it. But even passive "consumers" communicate and exchange data. That is the beauty of the Internet architecture; on the Internet all end-points on the network are equally capable of exchanging content/applications/services with all other end-points. Think of all the startups and small businesses run from people's homes on home Internet connections, using WordPress tools or Amazon hosting services. Are they "end users" when they email their friends but "edge providers" when they switch windows to check their business metrics?

What is the point of the provider/user distinction? It appears to draw a line between (commercial) entities that generate a lot of traffic (and revenue) and those who don't, but that is confused. In truth, traffic is generated jointly by many actors communicating with each other. Besides, that line ignores the fact that the Internet provides a smooth, scalable path that allows all end users to grow businesses and become future YouTubes, Amazons and Googles.

The provider/user distinction is not only false, but misleading and not a helpful foundation upon which to build a regulatory regime for Internet communications. Everything that occurs on the Internet can be understood as a conversation between end users. When I interact with a website I like, whether,, or, or when I send an email to the firms that own those websites or to a friend, or when I engage in a teleconference using Skype, I am having a conversation with other end-users. The contents of our conversations and nature of our interactions and relationships may vary, but those are all higher-layer considerations that have nothing to do with the infrastructural facilities we use to connect and have our conversations. All of us end users necessarily rely on access networks to have our conversations.

But when you use the Internet, do you have a conversation with your access network? It makes sense to frame my interactions with,, or as conversations, but no one besides the access networks themselves believes the access networks are part of these conversations. The access networks simply provide what the telecommunications act defines as a telecommunications service, which is one good reason why the FCC should reclassify broadband Internet access service as a telecommunications service.

The FCC should cast aside the false distinction between edge providers and end users. There is a better way for the FCC to implement a nondiscrimination principle to keep the Internet open and innovative. Management of the Internet should be based on quantity and timing of traffic and other related cost factors, not on a false distinction. A nondiscrimination rule that precludes discrimination or prioritization based on the identity of the user or use leaves more than sufficient room for traffic management that is based on quantity of traffic generated, timing of traffic generation, and other related cost factors; in other words, usage-based pricing, congestion pricing, and other cost-based management techniques are not precluded, despite persistent misrepresentations by opponents of nondiscrimination rules.

Some people find content delivery networks like Akamai confusing. I don't understand the confusion, to be honest. Akamai is an end user. Like the rest of us end user / edge providers, Akamai has relationships and conversations with various end users, and to do so, it relies on various access providers. Though we might not describe ourselves as such, many of us who use peer-to-peer file sharing applications also act as content delivery networks, although perhaps on a smaller scale and in a more distributed fashion. Neither Akamai nor those of us who use peer-to-peer applications is thereby transformed into a broadband Internet access service; we are simply end users.

In these confused discussions of "edge providers" and "end users," some folks at the FCC also point to "two-sided market" theory. This theory, expressed by the FCC Chairman once as involving Netflix and a cable company, oversimplifies the webs of relationships that exist among end users to create the Internet ecosystem and overstates the role of access providers in facilitating those relationships. The idea of "two-sides" does not account for the many (many) sides of the markets and non-markets on the Internet, and it also largely ignores the spillovers from user generated public and social goods that are perhaps the most important reasons that an open Internet is socially valuable (see my book, Infrastructure). When we use WordPress or Wikipedia to have extended conversations with many different end users distributed around the world, the access providers that we rely on (Comcast, Cogent, etc.) are not mediating a two-sided market; nor is WordPress or Wikipedia or Tumblr. To frame our conversations and relationships as though there are two parties in a market-mediated transaction distorts and cheapens them.

Embedding the false distinction between edge providers and end users in law not only distorts the law and the Internet environment we share, but it also, over time, solidifies the distinction and becomes our reality. It shapes the environment, and we tend to fit the mold provided. The danger of accepting the false and misleading distinction between edge providers and end users today is that it will help the distinction to become reality tomorrow.

I realize that my argument is at odds with the 2010 Open Internet Order, the 2014 NPRM, and the proposals of some open Internet advocates (e.g., Mozilla and others who have proposed creative ideas for regulating only the "edge provider" side of things). So be it. Those same advocates admit that their proposals might not be the best path forward. The FCC has a track record of losing unanimously in court with creative legal gymnastics, and should abandon the entire framework of distinguishing between giant companies and the little guys. All of us pay for Internet access and the FCC needs the authority-whether it uses that authority or not-to ensure that access remains open and free.

The wonderful thing about the Internet and its end-to-end architecture is that we all are capable of being providers, having conversations at the edge without having to include access providers in our conversations.