After reading over some of the postings from the few weeks and exchanging emails with Richard Bennett at Technology Liberation Front, I am coming to see how disastrous a decision it was for the FCC to pursue sanctions against Comcast over its throttling of BitTorrent files.
True, the case, and the court decision has allowed activists to foam at the mouth about a “crisis” in Internet service. Yet despite the breathless warnings, none of this resonates with the public. The results of a recent Rasmussen Reports poll, posted here by Adam Thierer, that found that 53% of Americans oppose FCC regulation of the Internet. Perhaps Americans are sanguine because there is no Internet censorship problem.
Even though the issues in the BitTorrent case are a bit technical, the public groks on some level that claims by proponents of regulation that the recent U.S. Court of Appeals decision in favor of Comcast would lead to rampant Internet censorship don’t ring true. That’s because first and foremost, the BitTorrent case was not about blocking or “censorship.” In fact, in the more than four years of debate, the only real instance of a network neutrality violation, that is, an outright flouting of the guidelines set up by former Chairman Michael Powell, came in 2005 when Madison River Communications blocked Vonage’s VoIP service. And Madison River got caught and fined.
Since the rather short-sighted actions of Madison River Communications, a small North Carolina competitive local exchange carrier, hardly generate the emotions or headlines that an AT&T or Comcast would, network neutrality activists were left scrounging for any “censorship” tidbit they could find, even if it had little to do with the Internet access (Jim Harper’s recent podcast critiques recent claims of neutrality violations).
Aside from being a bad example of a network neutrality violation, the Comcast-BitTorrent case brought out specific problems with the neutrality concept, including the idea of whether the Internet is neutral at all. But in plain language, the Court of Appeals asserted an age-old principle about freedom and boundaries: Your right to swing ends where my nose begins.
Now let’s look at what BitTorrent is and what Comcast actually did.
BitTorrent is a file transfer protocol. That is, it is a set of instructions for encoding and decoding content for transmission. It is one of many protocols the Internet uses, which include the basic Internet Protocol (IP) itself. Other examples are the Hypertext Transfer Protocol, which is long form of the “http” abbreviation used in web site URLs. There are also other protocols that users rarely encounter but govern the way computers on the Internet exchange information and execute programming. The Session Initiation Protocol (SIP) and the Simple Object Access Protocol (SOAP) are two examples, for those who want to get granular.
IP, HTTP, SIP, SOAP and BitTorrent are all open protocols. That is, the coding languages are available to anyone who wants to use them. They are also intentionally built to be compatible with the underlying Internet protocols. But while a protocol may open, it may not always be license-free.
This is the first major misunderstanding in the BitTorrent case. The BitTorrent protocol takes its name from BitTorrent Inc., the company that developed it. BitTorrent Inc. itself is not a content or applications provider. BitTorrent Inc., rather, develops content delivery technology, of which the BitTorrent protocol is an example. In order to create a market for the BitTorrent protocol, BitTorrent supplies clients and plug-ins—pieces of software that tie into your browser—to consumers for free. With millions of BitTorrent clients attached to consumer browsers, there is an incentive for larger content providers to pay BitTorrent Inc. for the right to use the BitTorrent protocol to encode their content. Trouble is, because of the way it is designed, the BitTorrent protocol tends to increase congestion on networks owned and operated by Internet service providers like Comcast. BitTorrent files could move easily across the network, but Comcast said they were decreasing quality of service for the majority its customers and applied countermeasures that slowed the rate of BitTorrent transfers.
This is why Comcast v. FCC will be remembered as a bad test case for network neutrality. The court exposed the FCC’s blunder of framing what was a legitimate dispute over network management as an issue of Web censorship. The Comcast-BitTorrent dispute was not about a giant service provider squelching consumer access to the Web, it was a business conflict between two companies with competing agendas. BitTorrent’s interest in maximizing the use of the BitTorrent protocol ran up against Comcast’s interest in preserving high quality of service for the overwhelming majority of its customers. Pointedly, the two companies worked out this conflict within weeks, mostly because as businesses, they understood each other’s motivations.
The BitTorrent decision raises more questions going forward, especially when it comes to the idea of “reasonable network management,” which the FCC’s current notice of proposed rulemaking on net neutrality would permit. But if Comcast’s throttling of BitTorrent was not reasonable, what is?
Comcast’s initial attempt to manage BitTorrent files was ham-handed, but its tactic, it could be argued, was content-neutral. Comcast targeted files that used the BitTorrent protocol, regardless of what they contained. It certainly made good headline when AP decided to use BitTorrent transmit the King James Version of the Bible, only to have it throttled. But the same thing would have happened if AP had used instead chosen the entire issue library of Playboy magazine, only it would not sparked the dramatics from the Christian Coalition (as reported here).
Finally, the BitTorrent protocol itself is a neutrality workaround. The protocol was developed to accelerate the transmission of rich media, in other words, to help providers of hefty content like TV shows, movies and other hefty content, carve out a “fast lane” for their data—an idea that runs directly counter to the so-called network neutrality principle. Here’s how BitTorrent describes its software, branded BitTorrent DNA, on its website:
BitTorrent DNA is a disruptively effective content delivery technology. It significantly reduces bandwidth costs for popular files while dramatically improving the performance and scalability of websites. BitTorrent DNA enables websites to seamlessly add the speed and efficiency of patented BitTorrent technology to their current content delivery infrastructure, requiring no changes to their current Content Delivery Network (CDN) or hardware in the origin infrastructure. Businesses can benefit from the efficiencies of peer-assisted content delivery while improving the end-user experience.
As an interesting thought-experiment, let’s suppose Comcast, or any other Internet service provider, was to acquire BitTorrent. Now, in the above quote, substitute the name of the ISP for BitTorrent. What you end up with is something like this: “Comcast DNA enables websites to seamlessly add the speed and efficiency of patented Comcast technology to their current content delivery infrastructure.” Under the current guidelines, as well as the proposed rules, this would be a network neutrality violation.
But wait, the FCC already attempted to sanction Comcast for throttling the BitTorrent protocol. The FCC’s underlying assumption, then, is that the BitTorrent protocol is beneficial and that an ISP should not be allowed to impede it. Under a neutrality regime, however, the FCC could turn around and sanction any ISP that chose to market it, or, more realistically, another protocol or technique like it. Here’s where network neutrality breaks down and presents an inherent contradiction, not to mention all sorts of equal protection issues.
Yet regulation proponents dismiss the idea that network neutrality would slow network investment. Simply put, no ISP, software developer, or venture capitalist would know in any given instance how the FCC would treat any innovative technology that would add “speed and efficiency” to “content delivery infrastructure.” On Monday, one technique could be acceptable. On Tuesday, another technique might not. The D.C. Court understood all these problems, and they were among the reasons it reined in the Commission.