Rumored network neutrality approach might break the Internet instead of saving it

If current leaks are correct, the Federal Communications Commission (FCC) may be on the brink of not only undermining hope for strong, enforceable and legally sound network neutrality rules, but may also be taking steps that would ultimately disrupt the very principles that have governed the way the Internet has always worked.

The press reports do not explain what the new proposal would fully entail (in part because the ideas from which the latest proposal appears to be derived are complicated and new), but it could mean that Internet users would be swept into new relationships with Internet service providers (ISPs) they've never interacted with in the past. Based on what we do know from the press reports, we are deeply concerned about this new proposal in several respects: its ability to actually protect against blocking, discrimination and fees for prioritized access; the relative strength of the legal underpinnings the FCC would use to implement those protections; and, importantly, the ripple effects that such a proposal might have across the Internet and around the globe.

For Internet users, this approach would place their ability to access the legal content of their choosing on the shakiest of legal grounds, using a legal theory that can only incidentally afford that protection. Moreover, every piece of traffic sent (an email to a friend, a response to a click on a website, a streaming video on YouTube) could trigger a new legal relationship with an ISP on the other side of the world.

Reports suggest that the proposal flows from one or more ideas in the record. One begins with the recognition that two separate and legally distinct relationships exist in the exchange of traffic - one flowing from the broadband provider's contract with "retail end users," and one flowing from the relationship between a broadband provider and other providers that seek to deliver their own services over the provider's network. This proposal would essentially recognize both a "call" from the end user and a "response" from a content provider.

But the FCC may also be looking toward a second, similar proposal in the record that would recognize a simultaneous and congruous relationship among "remote hosts" and end users, and would leave the relationship between broadband providers and end users unregulated, while imposing protections instead on the relationship between the broadband providers and these newly defined "remote hosts." In contrast to the relationship created by the first proposal, the much more esoteric one created here would not be directional, and would cover every interaction on the Internet where traffic is exchanged.

We aren't exactly sure where the FCC might be leaning among these approaches as it crafts its new proposal. And we don't know what rules the FCC would even adopt if a novel legal relationship among content "senders" and distant, last-mile broadband providers is recognized and classified as a Title II service. However, both approaches, and any combination of them, carry significant legal risk and are generally confusing to both laymen and lawyers alike. To survive the inevitable legal challenges that would follow, the FCC would be required to demonstrate that this new service between a sender of content and a broadband provider meets the legal definition of a telecommunications service. This statutory definition requires that there is a service, offered for a fee, directly to the public. A relationship between content senders and Internet providers fails this definition on its face - there is no fee (indeed, the fee is precisely what network neutrality advocates hope to prevent!); and there is no offering to the public — certainly not a direct one. Making this definition work would require significant legal gymnastics to recognize the new service, classify it as a telecommunications service under the statute and convince a court that this new approach makes fundamental sense. On any one of these points, the theory could fail, and the fact that the approaches themselves are both complex and enigmatic make the risk of failure even greater.

However, legal viability is only part of the concern. We are also fundamentally concerned about the collateral effects that a sender-side approach to network neutrality might have on the future of the Internet ecosystem.

The Internet, as Vint Cerf explains, has historically been "based on a layered, end-to-end model that allows people at each level of the network to innovate free of any central control. By placing intelligence at the edges rather than control in the middle of the network, the Internet has created a platform for innovation." This conception of the Internet ecosystem is bedrock, and would be undone by a network neutrality solution that, at its core, would entwine as a matter of law the very layers that must be kept distinct in order to preserve the open Internet as a platform for innovation and decentralized control.

Recognizing a relationship between senders of traffic and last-mile broadband providers for the sole purpose of imposing network neutrality rules would recognize new privity among all websites around the world and all retail Internet access providers in the United States, making every website and content "sender" around the world a customer of every broadband provider in the U.S.

There has never been this type of legally recognized relationship among all websites and broadband providers. Generally speaking, if someone is sending an email or making content available on a website somewhere in the world, that person is utterly unconcerned with how that content is accessed by other users. This is the very nature of the end-to-end principle — traffic is sent, routed across the Internet and someone on the other end accesses the information. Susan pays Comcast for access to the entirety of the Internet; Reddit similarly pays an ISP for access to the entirety of the Internet, in addition to servers to host content and perhaps a transit provider to send it. Reddit, however, does not pay Comcast to ensure that Comcast doesn't block Reddit traffic to Comcast's subscribers. That is precisely the kind of behavior that network neutrality rules are designed to prevent. The surest way to protect every user's ability to access the Internet is to simply afford all users the same protections, rather than create arbitrary distinctions among different types of users.

We also don't know what this new relationship really means, or how it could possibly scale. Does a new relationship among distant parties carry with it obligations beyond the network neutrality protections imposed? If other countries followed suit, what do those transaction costs mean for content senders? A website in Alabama could be forced to negotiate under this new proposal with broadband providers in any number of countries — with Comcast, with Time Warner Cable, with Charter, with Telenet in Belgium and HelloVision in South Korea — all broadband providers with which they have never previously had any interaction at all.

For these legal, practical and fundamental reasons, the FCC shouldn't go down this complicated, confusing and untested path. Reclassification of retail broadband access as a Title II service remains the surest, clearest path forward for strong network neutrality protections, and it's the solution that millions of Americans have called on the FCC to implement.

Morris is senior policy counsel for the Open Technology Institute at the New America Foundation.