Combatants in the Net Neutrality wars often seem to talk past each other. Sometimes it’s legitimate miscommunication. More often, though, it arises from fundamental defects in the concept itself.
On December 2, Commissioner Ajit Pai wrote to Netflix, Inc., saying he “was surprised to learn of allegations that Netflix has been working to effectively secure ‘fast lanes’ for its own content on ISPs’ networks at the expense of its competitors.” Commissioner Pai noted press accounts that suggested Netflix’s Open Connect content delivery platform and its use of specialized video streaming protocols put video from non-Netflix sources at a disadvantage. Commissioner Pai concluded that “these allegations raise an apparent conflict with Netflix’s advocacy of strong net neutrality regulations” and thus asked for an explanation.
In its reply of December 11, Netflix made four basic points. Netflix (1) said it “designed Open Connect content delivery network (CDN) to provide consumers with a high-quality video experience”; (2) insisted “Open Connect is not a fast lane . . . . Open Connect helps ISPs reduce costs and better manage congestion, which results in a better Internet experience for all end users”; (3) said it “uses open-source software and readily-available hardware components”; and (4) applauded other firms for developing open video caching standards but “has focused” on its own proprietary system because it is more efficient and customer friendly than the collaborative industry efforts.
Three of Netflix’s four points are reasonable, as far as they go. The company is developing technologies and architectures to improve customer service and beat the competition. The firm, however, seems not to grasp Commissioner Pai’s central point: Netflix relishes aggressive competition on its own behalf but wants to outlaw similarly innovative behavior from the rest of the Internet economy.
When broadband firms enter into voluntary interconnection agreements that speed content to end-users, Netflix cries ominous “fast lanes” and “slow lanes.” When Netflix itself deploys infrastructure and protocols that speed content faster than its competitors, it’s in the service of consumers. Netflix’s use of “open-source software and readily-available hardware components” is irrelevant. Many firms build closed, proprietary systems using open-source and commodity components. But Netflix may think the casual reader will conflate “open-source” and Open Internet.
The righteous act is wearing thin. And so is the decade-old net neutrality effort.
In the beginning, net neutrality sought to banish any network tool or business model that amounted to “pay for performance.” The initial legislative vehicles were preposterously elementary and unworkable. They would have banned most of the Internet. When many of us pointed out that content delivery networks (CDNs), as just one example, were a ubiquitous “pay for performance” model used across the Internet, advocates said, “Well, we didn’t mean that. CDNs are ok.” The FCC’s 2010 rules explicitly exempted CDNs.
Over time, as the many shortcomings of their proposals were exposed, they narrowed their demands, at least when one could pin them down, which wasn’t often. The key, they finally settled on, is outlawing packet prioritization over the last mile. Although most broadband providers think such a blanket prohibition is bad idea, many BSPs last year nonetheless relented — “Fine, no paid prioritization.”
After paid prioritization was taken off the table, however, the activists still weren’t satisfied. They could not take yes for an answer. The net neutrality fans once again expanded their scope, proposing the regulation of network interconnection, which had never been considered a part of the debate or any proposed rules. They also, of course, made a radical new demand: imposition of the old Title II telephone regime on the Internet.
The FCC final rule will almost certainly impose Title II, and press reports say it could focus on interconnection and not last mile broadband. It would be a complete reversal from net neutrality history and Chairman Wheeler’s explicit statements that net neutrality is about the last mile, not interconnection.
Ironically, the new focus on interconnection brings CDNs back into the regulatory equation. CDNs were early adopters of successful interconnection relationships (known as “on-net transit” or “paid peering”) that may now be in jeopardy. Circling back to where we started 10 years ago would be hilarious if it weren’t so tragic. The focus on interconnection and not the last mile will, I imagine, also add to the FCC’s legal troubles. Among the innumerable legal infirmities of the forthcoming FCC order, a last-minute bait-and-switch, performed largely without public comment on Title II or interconnection, will help sink the FCC’s own legal ship.
Net Neutrality fundamentalists say they are worried about the Internet becoming another cable TV network. They draw stark distinctions between the two and want no overlap. Netflix, however, wants the benefits of both worlds. It wants the freedom, interactivity, choice, and low costs of the Internet. But it also wants the instant availability, big bandwidth, and quality-of-service of high-resolution cable TV channels.
Fortunately, the Internet has evolved to accommodate numerous content types, interconnection architectures, and business models, many of which overlap to varying degrees. Although the Internet and its chief protocol, TCP/IP, is not the most efficient way to deliver video, the ingenuity of network engineers and agility of the enterprises on all sides have made it work. It turns out Netflix can have the best of both worlds to a very large extent — its success in the marketplace proves today’s model of Internet freedom is working. Yet Netflix is not content with a healthy model in which all sides win. It insists on maximalist government rules that would radically alter the way we govern the Internet — and which also would subsidize its own business model at the expense of others.
We will know Netflix is serious about net neutrality when it offers to host its rivals’ content on its Open Connect platform for free — at the same zero price it has demanded from providers of transit bandwidth.
The technical incompetence, legal shenanigans, and refusal to take yes for an answer all expose what we said from the very beginning, a decade ago. Net neutrality is not really about high-minded Internet principles. Nor is it about paid prioritization or any other specific technical or business model. It is an open ended, omnibus vehicle to expand bureaucratic control over the Internet in particular and the digital economy in general. The latest Title II effort makes this crystal clear. The details of technology, economics, and law are quite beside the point. What matters is maximizing Washington’s ability to dictate, intervene, and harass without limit. If a small handful of firms find this arrangement helpful to their businesses, they will support the effort.
The inability to define net neutrality is also what causes the unproductive cross-talk. Ultimately, Netflix was unresponsive to Commissioner Pai’s letter, in part because the constantly shifting nature of net neutrality makes it difficult to have a useful conversation. The serial and often secret proposals lack substance at both ends of the spectrum — in both technical and legal detail and in broad, definable principles. When FCC Chairman Wheeler did propose a reasonable if imperfect solution last spring, the fundamentalists rebelled. When the FCC has actually put pen to paper and issued final orders, the courts have struck them down.
Commissioner Pai’s simple questions are useful not because they show Netflix misbehaving in the marketplace but because they expose the fundamental flaws and endless hypocrisies of the effort to regulate the Internet.
