How Comcast and Huge Telecom Players' Latest Gambit Could Destroy the Internet as We Know It
On December 21st, the Federal Communications Commission issued new rules governing Internet transmission. In essence, they effectively divide broadband content distribution between wireline and wireless transmission, providing nominal “net neutrality” protection over content distributed over telephone and cable lines, but deregulating such content sent through the airwaves.
As Timothy Karr of FreePress wrote in the Huffington Post: “The rule is so riddled with loopholes that it's become clear that this FCC chairman crafted it with the sole purpose of winning the endorsement of AT&T and cable lobbyists, and not defending the interests of the tens of millions of Internet users.”
While Karr and others are focusing on the front story and the FCC’s capitulation to corporate interests, there is a more compelling back story as to how this regulatory farce came about. It is this story that maps out the underlying fictions that provide the rationale for the FCC’s actions and the likely long-term consequences for telecommunications in America.
* * *
As the telecom wars heat up, the American Internet user is the proverbial pig on the spit. A series of recent developments have drawn public attention to major challenges that are redefining the Internet. And it doesn’t look good for those championing Internet freedom, meaningful competition, improved quality of service or an end to conglomerate integration. Sadly, the fix is in.
In May, the DC Court of Appeals handed down a decision to what is popularly know as the BitTorrent case, Comcast v FCC. Originally, Comcast had blocked BitTorrent transmissions over its network and BitTorrent complained. The FCC investigated Comcast's traffic management of BitTorrent and ordered it to end its "discriminatory" practices. However, the Court ruled that Comcast has the right to limit the Internet connections of its customers who were using BitTorrent’s peer-to-peer (p2p) services on the grounds that such limiting falls under the role of network maintenance. More critical, the Court raised serious questions as to whether the FCC has the ability to regulate broadband Internet access altogether.
A few months later, Level 3 Communications drew the public’s attention to an effort by Comcast to impose a new pricing tier or “toll booth” on its transport of Netflix’s streaming video. This comes amidst deliberations by the FCC and Justice Department over Comcast’s bid to acquire NBC-Universal and a string of other questionable actions, including censorship charges, leveled against the company.
The BitTorrent and Level 3 actions were trial balloons pointing to deeper issues related to the future of broadband Internet. On one level, they reveal how key players in “content” – and especially video content – delivery are lining up against one another over, what else, money, or how the fees they charge are to be allocated.
More significantly, the battle puts into stark relief the issues of who controls the wires and wireless spectrum, the backbone of America’s communications infrastructure. Internet traffic, whether that of BitTorrent, Netflix or your email, is transmitted over the primary wires and spectrum controlled by AT&T, Comcast, Verizon and other networks and wireless services. The Internet rides over these wires and airwaves, and these companies control the access points.
The battle lines are being drawn over three key issues. First, will the current distribution model known for an open Internet, with “net neutrality” safeguards, persist – in other words, will all data continue to travel over telecom networks at the same rate and remain uncensored by Internet Service Providers (ISPs), excluding “illegal” content like child pornography?
Second, will the current consolidation of ISPs persist with giant, integrated conglomerates like Comcast and Time Warner controlling both the network and the content -- or will there be a renewed commitment to “open architecture,” to an Internet promoting meaningful competition among multiple access and content providers on the basis of services offered and prices charged?
Third, will the FCC further extend its 2002 reclassification of broadband service as an “information service,” thus further removing data transport from traditional “common carriage” obligations? Will it further collapse the difference between the Internet as a distribution network from the content or applications it supports?
FCC chairman Julius Genachowski December 21st announcement is a fictions compromise, like Obama’s plan to extend the Bush-era tax give-a-ways, and will only serve to further commercialize Internet traffic. His support for "usage-based pricing" will permit wireline ISPs to charge extra fees to heavy Internet users, like Level 3, who transport lots of video or video games. And, in keeping with the Obama spirit of compromise, he opposes ISPs from “throttling” or slowing data traffic, thus protecting one aspect of net neutrality.
The FCC proposal is another example of how Washington politicians and their dutiful bureaucrats capitulate to big capital to the determinant of ordinary Americans and the long-term erosion of the U.S. economy. The FCC’s plan will set the stage for another round of federal give-a-ways to the giant telecoms, accompanied by an increase in customer charges and a further erosion of service.
* * *
At the heart of the FCC proposed new pricing models for the Internet is the shibboleth of network data traffic congestion. It is an alarm being raised by many within the telecommunications industry and is based on the false assumption that, as video becomes an increasing larger proportion of data traffic on the Web, the network is in jeopardy of collapse. The myth of congestion provides the rationale for a new pricing model as well as the need to end net neutrality and restrict ISP competition.
The Internet and Web consists of four interlinked components: (i) the phone or cable company that provides the “last mile” facilitating the consumer’s broadband connectivity through the residential telephone company’s digital subscriber line (DSL) or cable modem fiber line; (ii) the ISP that connects the customer to the PoP (point-of-presence) on the Internet and World Wide Web, (iii) the “middle mile” provider that links the customer to wider network; and (iv) the provider of long distance, high-speed connections to the network “backbone.”
While the middle mile and backbone utilize high-capacity fiber networks, the crunch comes at the last mile. And it is this piece of the grid that reflects the telecom trust’s failure to upgrade its networks. They have pocketed billions of dollars in subsides and tax breaks and have little to improve their networks. As with most American industrial sectors, the demands for short-term profits makes telecommunications providers unable to meet long-term market demands. No wonder the U.S. ranks 15th among developed counties in terms of broadband utilization.
According to a recent report from Cisco Systems, by the end of 2010, worldwide global online video users are projected to surpass 1 billion and video is projected to account for 40 percent of consumer Internet traffic. The traffic includes Internet video from YouTube, TV programs from Hulu, video-on-demand like Netflix movies and p2p sharing like BitTorrent. Enhanced video quality like 3D and HD only further compounds the video signal.
Online traffic grew 45 percent during 2009 to 176 exabytes per year (an exabyte consists of a million trillion bytes) and is projected to reach 767 exabytes per year by 2014. According to Cisco, this means: “The average monthly traffic in 2014 will be equivalent to 32 million people streaming Avatar in 3D, continuously for the entire month.”
Determining U.S. Internet video traffic is much trickier. Sandvine, an Ontario-based network services company, found that in the August-September 2010 period, North Americans gobbled up only one-third as much broadband video as users in the Asia-Pacific region; North Americans consumed an estimated 4 Gigabytes per month of Internet bandwidth whereas those in Asia-Pacific region used 12 Gigabytes. Sandvine also found that Netflix, with nearly 17 million subscribers, accounted for more than 20 percent of downstream Internet traffic during the primetime TV viewing hours of 8 to 10 p.m.
ComScore, a Reston, VA, market research firm, offers a different take on Internet video usage. It found, in May 2010, that 183 million U.S. Internet users watched nearly 34 billion online videos. Google sites, especially YouTube, account for 14.6 billion videos, representing 43.1 percent of all videos viewed online. Hulu, a joint venture of NBC and FOX and aggregating videos from nearly 200 content providers, came in second with 1.2 billion videos (3.5%), with other sites trailing behind.
Champions of the telecom trust rallied to the Sandvine findings as proof of network congestion and the basis to impose a new pricing model. This model goes by a variety of names, "usage-based pricing,” "paid prioritization" and “pay-as-you-go Internet access,” among others. Kyle McSlarrow, president, Nation Cable and Telecommunications Association, a trade group, recently wrote that the cable industry backs the new pricing model: "A usage-based pricing model, for instance, might help spur adoption by price-sensitive consumers at the lower end of the socioeconomic ladder," he wrote. Craig Moffett, an analyst at Bernstein Research, chimed in with an investor’s glee: "Usage-based pricing will preserve, and even enhance, the economics of cable's infrastructure . . . even if consumers eventually get some, or even all, of their video content over the Web."
Peter Burrows, of Bloomberg Businessweek, takes the argument one step further. Drawing upon research from Juniper Networks, he insists that, based on a "revenue-per-bit" model, the big telecom conglomerates like AT&T and Comcast “will see [their] Internet revenues grow by 5 percent a year through 2020,” but traffic will “surge by 27 percent annually.” “By this math,” he insists, “the carriers' business models break down in 2014.” Or does it?
* * *
Freud once famously noted, “"Sometimes a pipe is just a pipe." And sometimes it isn’t. This is the same with Internet traffic congestion.
The Internet has evolved through three phases and each has been defined by “congestion.” The first phase, during the pre-1990s when the Internet had yet become a mass-market phenomenon, congestion was experienced in email delays, limits to bulk file transfers and low bit rate interactive sessions. The second phase of the 1990s saw rapid Internet adoption with dial-up accounts offering 56.6 kb/s rates and congestion taking the form of a slow system with only intermittent connectivity. In a famous 1995 case, the phone companies attempted to block companies offering the new service, Voice-over-Internet-Protocol (VoIP), on the base of, yes, congestion; companies like Vonage would not exist today had the telecoms had their way. The third phase began around 2000 and saw the wide-scale adoption of broadband. It eliminated the dial-up bottleneck with complementary advances in “upstream” capabilities (e.g., more powerful multimedia-capable PCs) and downstream content delivery (e.g., “rich” and interactive media content).
Internet traffic has grown due to four principal factors: (i) an increase in the number of online subscribers, (ii) the amount of time each subscriber spends online, (iii) the growing mix of wireline and wireless distribution options and (iv) the differences and complexity of applications (especially video-based apps) carried online.
However, as Andrew Odlyzko, a professor at the University of Minnesota and former AT&T Labs researcher, argues, “there is no evidence of wireline Internet traffic growing so fast as to require intrusive traffic interference to control it. … traffic growth rates have been declining, to levels slower than the rate of improvement of latest transmission equipment.”
Adding to this critique, three MIT scholars, Steven Bauer, David Clark and William Lehr, recently released an invaluable study on web traffic, “The Evolution of Internet Congestion,” and offer the following warning: “It is certainly possible that network operators, under the guise of managing congestion, may exploit their control over the network pipes in ways that are socially undesirable ….” Their warning should be the starting point for linking alleged congestion to changes in pricing and overall control of the Internet.
Comcast, like AT&T, Verizon or other dominant controller of distribution, has gained ever-increasing control over access by and to its subscribers. In order for a subscriber to reach the Internet, and for content providers to reach the consumer, they have to pass through an ISP's last-mile network, whether over the phone line or airwaves. The ISP is the gatekeeper. This situation gets more troubling when one recalls that in addition to being a last-mile gatekeeper serving residential, business and wholesale customers, Comcast, like the others dominant players in the telecom trust, also operates a powerful backbone network and is now moving in to content.
Comcast is seeking, like other dominant carriers, to be both “vertically” integrated, i.e., control connectivity from the last-mile to the backbone, as well as “horizontally” integrated, i.e., control available content or applications like NBC-U. Its two-dimensional system of integration pushes beyond the iPhone “walled garden” model by which Apple controls the applications available to the consumer. This two-dimensional integration suggests the real, long-term danger that the FCC and Congress refuse to recognize.
The real drama being played out in terms of Internet congestion and changing pricing models needs to be seen as part of a more profound and systemic change in the control of the nation’s telecommunications infrastructure, especially the Internet. Comcast and other conglomerates that make up the telecommunications trust, like the trusts that dominate the oil and gas sector, health care or financial services, are aggressively pushing to control all aspects of the market sector. Unless the debate over congestion and pricing is opened up, refocused to the larger question of industry consolidation, the FCC December 21st proposal will only make the problem worse.
Stay tuned – the worst is yet to come.