It has been two whole weeks since the FCC issued its Internet openness rules, and with holiday celebrations out of the way there has been some time for the details to start to sink in. Some observers seem to be perpetuating a high-level debate about whether the FCC went too far or not far enough, ignoring the extensive amount of detail to be analyzed in the actual text of the rules. But understanding the nuance is important to understanding what the impact of the rules is likely to be on ISPs’ behaviors. Studying the impact of Internet neutrality regimes in other countries might also provide clues about the likely upshot of the FCC’s decisions.
Traffic management provides one place to start on both of these endeavors. Since the FCC made its original Internet policy principles “subject to reasonable network management” in 2005, the issue of whether traffic management practices that discriminate between different applications, types of applications, or applications protocols are acceptable has been central to the Internet neutrality debate both in the US and abroad. More than a year ago, the Canadian telecom regulator, the CRTC, issued its own framework for evaluating ISPs’ traffic management practices. The framework expresses preferences for technical traffic management practices that:
• address the need and achieve the purpose and effect in question, and nothing else;
• result in discrimination or preference as little as reasonably possible;
• minimize any harm to a secondary ISP, end-user, or any other person; and
• achieve a purpose that cannot be achieved through network investment or economic approaches (e.g., pricing) alone.
This framework presents a strong preference for non-discriminatory traffic management, but stops short of disallowing discrimination altogether. Perhaps as a result, more than a year after the framework’s adoption, a substantial number of the largest Canadian ISPs, including Bell Canada (whose traffic management aimed at peer-to-peer applications originally spurred the CRTC to action), continue to state in their traffic management policies that they discriminate against particular applications or applications protocols. For those of us hoping for a better (i.e., less- or non-discriminatory) result in the US, how does the FCC’s new definition of reasonable network management compare? Take a look:
A network management practice is reasonable if it is appropriate and tailored to achieving a legitimate network management purpose, taking into account the particular network architecture and technology of the broadband Internet access service. (FCC Order para. 82)
The Commission goes on to clarify a number of points about this definition. While traffic management practices need to be “tailored,” they need not be “the most narrowly tailored practice theoretically available” (perhaps stopping a bit shy of the CRTC’s requirement that a practice may address “nothing else” other than its singular purpose). The FCC also explains that its evaluation of the reasonableness of traffic management practices will be based on the same principles it will use to evaluate potential violations of its broader “no unreasonable discrimination” rule (laid out in an earlier section of the order), specifically calling out the principles of “transparency, end-user control, and use- (or application-)agnostic treatment.” With particular relevance to discrimination, the use-agnostic treatment principle states the FCC’s presumption that “differential treatment of traffic that does not discriminate among specific uses of the network or classes of uses is likely reasonable.” Thus, as a baseline, it seems that traffic management practices that are blind to application types or protocols are likely to be deemed reasonable.
Although it is not explicitly listed as a “principle” that will guide the Commission’s evaluations of traffic management practices, the order does raise concerns about discriminatory practices (in general) that harm ISPs’ actual or potential competitors, end users, or free speech. If such concerns were to factor into traffic management evaluations, they would be not unlike the third prong of the CRTC’s framework, and they may close the door on traffic management that, for example, targets peer-to-peer file-sharing (which could potentially be construed as causing any of the harms listed).
In sum, the CRTC goes a bit further in its explicit language about narrowness and harm while the FCC is more explicit about its presumption of reasonableness for non-discriminatory practices. The impact of the FCC order on traffic management is of course not solely dependent on the text of the rules; if case adjudications begin to materialize, we will gain even more insight as to the future of non-discriminatory traffic management. Case adjudication may well have contributed to the current state of affairs in Canada — the fact that the CRTC allowed Bell Canada to continue its discriminatory practices at the same time it adopted the traffic management framework may well have sent a signal to other Canadian ISPs about which practices would be considered reasonable. In any event, drilling down on the FCC order provides meaningful guidance about how traffic management may evolve in the US. While a handful of deeper dives on this and other aspects of the rules have appeared, more are certainly in order.