Comcast’s New ToS: Company Admits to Tampering with P2P Traffic

In Comcast’s new Terms of Service, the company explicitly admits that it degrades peer-to-peer traffic as a means of reducing their network load.

The company also admits that they will kick off end users who use (what they determine to be) too much bandwidth:

The Service is for personal and non-commercial residential use only. Therefore, Comcast reserves the right to suspend or terminate Service accounts where bandwidth consumption is not characteristic of a typical residential user of the Service as determined by the company in its sole discretion.

Harold Feld has a great explanation of why Comcast blocks p2p and boots the heaviest users: they don’t want to invest in the infrastructure to improve capacity, and they don’t want to use tiered pricing.

If only we could get network neutrality legislation…

(Link via Marvin Ammori (who ROCKS), guest blogging for

Supercapitalism Really Is Super

Robert Reich’s latest book, Supercapitalism, is a fantastic analysis of the current relationship between corporations, citizens, and politics.

I put Supercapitalism on my wish list after Prof. Lawrence Lessig’s glowing recommendation. While I make no pretense of being such a gifted writer as either of these scholars, here I attempt to summarize the book and follow with a few minor points.

Reich, the former Labor Secretary and current Professor of Public Policy at Berkeley, describes us all as being of two minds. On one hand, we are all consumers and (most of us are also) investors. As such, we’re always seeking to minimize our costs and maximize our profits. This leads to lower costs and higher profits; companies that cannot deliver lose customers and investors.

On the other hand, we are also all citizens and employees. In that capacity, we are generally frustrated by the effects of our growing collective power as consumers and investors. Those low prices and high profits squeeze employees, main street family-run stores, and the environment.

Our civic selves object to these negative effects, but we know that our individual purchasing and investing power cannot reverse these trends. Even were we to make the sacrifices of paying higher prices and earning lower returns by supporting more “socially responsible” businesses, we cannot make a difference with our dollars alone. Even social movements calling for corporate responsibility fail because, even if the companies comply, they leave an economic vacuum to be filled by other companies; otherwise, companies just revert to their old ways once the heat is off.

In the “Not Quite Golden Age” of postwar America, companies could pay high wages and CEOs could act on what they saw as the public interest. Most major industries were composed of cozy oligopolies with little product variation. The high cost of industrial production set high barriers to entry, leaving companies with plenty of room to negotiate relatively good deals for employees and the public.

Thanks in large part to new information technologies, as well as the growth of worldwide shipping infrastructure, we have entered what Reich calls supercapitalism over the past 30 years. Companies can design a product on a computer in Denver, buy parts from Brazil, Egypt, and Hungary, and subcontract with a factory in Korea to follow the computerized assembly instructions.

The increasingly fierce competition between companies has led to the squeezing along every part of the supply chain. Main Street retailers can’t sell refrigerators for $1200 when the same icebox is $799 at the big box store 2 miles away. Ford can’t stay profitable by paying its workers $70 per hour in salary and benefits when comparably skilled Koreans will do the same job for half. Suppliers get squeezed, too; ask any of WalMart’s suppliers about this process.

In the era of supercapitalism, companies have little choice but to minimize prices and maximize profits. In the Not Quite Golden Age, a system of cozy oligopolies gave consumers and investors little choice; both groups had mediocre but predictable deals all around. Now, consumers and investors who do not get the best possible deals will take their money elsewhere. Companies that do not ruthlessly squeeze their costs go bankrupt or get bought out.

This process has also led to the corruption of the democratic process. In the ever-accelerating contest for strategic advantage within and between industries, companies have begun to game the system to a degree that was generally not necessary 30 to 50 years ago. Reich’s own experience in government illustrates the impact of the rapid influx of money into the DC area:

Even by the mid-1970s, when I worked there as a political appointee at the Federal Trade Commission, much of the downtown was still run-down. I’d take any lobbyist who insisted on a lunch to a cockroach-infested sandwich shop on the other side of Pennsylvania Avenue, after which I would never see the lobbyist again. But when I returned to Washington in the 1990s, the town had been transformed. … The flow of money had inflated everything in its path. (p. 132)

Corporations are the primary folks funding this influx of capital. NGOs and labor make just a drop in this rapidly growing bucket of lobbyists, PR firms, campaign donations, “expert” consultants, hotels, and fancy restaurants with leather menus and $75 steaks.

Corporations spend this cash in the search for competitive advantages within or between industries. They may not even want to play, but they have to in an attempt to counterbalance other companies’ or industries’ efforts. Competition for customers and investors is too fierce, and one bill can kill a company’s bottom line.

The result is the increasingly impenetrable Beltway we all know and love. Corporate cash has purchased such a cacophany that citizens’ voices are drowned out.

Reich does offer some hope for a cure. Some of the usual suspects are here, from policy changes such as stronger labor protections to procedural reforms such as publicly funded campaigns. The really interesting recommendations, though, center on Reich’s argument against the anthropomorphic view of corporations as people.

Corporations are nothing more than bundles of contracts, so he insists we should neither give them standing to sue to overturn duly enacted laws nor find them criminally liable nor tax their income as though it is the company that owes. Their shareholders and employees would still retain all their rights and responsibilities, which is proper, since a corporation is just a collection of shareholders and employees.

He makes a compelling case for the feasibility and benefits of taxing shareholders instead of companies; corporations would withhold taxes on shareholders’ behalf and give them something like a W-2 form at the end of the year. This would be feasible in the era of computer-processed financial transactions, and it would be progressive, since the wealthiest would pay a higher rate on this income. It would also eliminate corporate inefficiencies caused by some wrinkles in the tax code.

The jaded may initially blow these off as politically impossible (I certainly did), but Reich points out that most companies would rather not be shaken down. A coalition of likeminded corporations helped leverage McCain-Feingold into law, and combined with public pressure, a similar coalition could create even greater reforms.

The prospect for procedural reform in particular is not impossible, but it is quite optimistic. A few industries with a history of winning backdoor negotiations with little effective opposition would fight tooth-and-nail against anything that would reduce their unique power position: oil, telecom, and the entertainment industry all come to mind.

For instance, he perpetuates the mistaken notion that the net neutrality debate was just another contest between corporate interests. In fact, tech companies were seriously outmatched on The Hill, and it was only due to the outstanding work by NGOs such as Free Press and the mobilization of over 1 million citizens that Sen. Ted Stevens’ (R-AK) 2006 telecom bill died as a net neutrality hostage.

Additionally, Reich regrettably fails to consider the potentially obstructionist role of the corporate media in blocking political reforms. The media have an obvious economic incentive to keep campaign funding the way it is: teeming with corporate cash that winds up buying tons of ads for several months every two years.

Any attempt to tie campaigns’ spending to taxpayers’ willingness to pay would generate substantial media opposition, and a bill mandating free airtime would drive media companies to break out every political tool they have. Congress speaks to its constituents through the media; these same media will turn on them (even if not as overtly as, say, Fox) in a heartbeat, and politicians know that.

Finally, Supercapitalism could better integrate theory generally and political economy more specifically. This is ironic; the book is itself an excellent introduction to political economic analysis. But theories about the flow of political information (such as Oscar Gandy’s theory of information subsidies) and the policymaking process (perhaps Baumgartner and Jones’ theory of punctuated equilibriums) could add some heft to Reich’s analysis. This is clearly a trade press book, but it is not impossible to drag a little theory into a book with wide appeal. Paul Krugman’s highly readable book, The Age of Diminished Expectations, is a fine example, and he was using economic theory.

All told, though, Reich’s book is nothing less than a beacon of hope in a world of dark political realities. This should be on your must-read list, and I’m already thinking about how I could use it in the classroom.

700 MHz Auction Starts Today

While the process will drag through late February or early March, today marks the kickoff of the FCC’s auction of the 700 MHz spectrum, in which TV channels 50-69 will be sold to the highest bidders in four blocks.

For background on the process, see the Public Knowledge page on spectrum reform. As always, Harold Feld has another awesome post on the subject.

CNet has a fiery critique of the FCC’s handling of the D Block auction by Gregory L. Rosston.

Finally, for the least informative page of all, here’s the FCC Spectrum Auctions Portal.

Why Time Warner’s Bandwidth Pricing is Good Short-Term Solution

David Isenberg has a great post on why Time Warner’s bandwidth-sensitive pricing plan is good.

The very short version is that it’s honest and a reasonable substitute for non-neutral packet discrimination, but the details will matter tremendously.

( link via Berkman Buzz)

UPDATE: I failed to mention the important caveat that bandwidth-sensitive pricing will not solve our bandwidth problems. We still need more capacity to keep up with growing demand and to fuel new digital innovations.

If we think end users are an important source of innovation (and they are), higher capacity for homes and small businesses is a brutally important part of our economic and cultural future.

The bandwidth-sensitive pricing will reduce congestion, but that problem is radically overstated. The continued popularity and usability of VoIP (despite it being continually invoked as the fragile traffic in need of protection) shows that the tubes are not exactly full.

Rather, the real problem is our general lack of growth in capacity. Other countries have really beaten us to the punch in subsidizing, requiring interconnection, and making other policy moves that have resulted in much faster growth in broadband adoption and network capacity for end users.

We may have untold dark fiber connecting major transit centers, but the last mile is still a serious bottleneck.

As bandwidth demand keeps growing, we still have a huge need to grow our capacity. Discriminating based on bandwidth, though, is a good interim solution for the reasons discussed above.

On a personal level, this is a wildly gratifying rebuff of Christopher Yoo’s suggestion that metering would just be too expensive to pull off. See Opening Bottlenecks, p. 146, for more.

Best Blog Post Ever: Why Google’s Serious About Spectrum

Many folks have wondered how serious Google is about winning a slice of the 700 MHz spectrum in the upcoming auction.

Harold Feld answers this for us in the best blog post ever. Specifically, they’re bent on destroying the current business model of the mobile industry and preserving the last vestiges of an open internet. In other words, pretty serious.

So go read it already.

Retraction: Net Neutrality Probably Stops Comcast

In my last report on this story, I put little thought into a minilink to an article by CNet’s Anne Broache, which frets that proposed net neutrality bills probably would not prevent Comcast’s ongoing peer-to-peer blockade. I no longer agree; I think even the weaker of the two bills on this count, if properly enforced (big caveat), would stand a good chance to stop these shenanigans.

This further investigation was sparked when I learned a good deal more from this awesome EFF article about the Comcast affair. This looks like a tight case: Comcast is almost completely obstructing Gnutella and BitTorrent traffic, and when confronted with effects on other programs (specifically, Lotus Notes and Windows Remote Desktop) has changed their algorithms to let that traffic through.

Thus inspired, I re-read the bill now in the Senate, Byron Dorgan and Olympia Snowe’s S. 215. I also re-examined Ed Markey’s bill from last year. Having done so, I’ve changed my mind: I think the plain text of the Senate bill and any reasonable reading of the House bill (pending FCC interpretation) makes it pretty hard for Comcast to continue blocking specific programs.

I start with the Senate bill which, unlike Markey’s proposal, has already been re-introduced in this Congress. Section 12(a) states, “With respect to any broadband service offered to the public, each broadband service provider shall (1) not block, interfere with, discriminate against, impair, or degrade the ability of any person to use a broadband service to access, use, send, post, receive, or offer any lawful content, application, or service made available via the Internet.”

This seems straightforward: BitTorrent and Gnutella have legal uses, thus Comcast would be prohibited from impairing or even degrading their subscribers’ uses of these programs. They could slow down the traffic to and from their heaviest users; EFF points out that they’re already doing this. But they couldn’t block specific programs just because some of their users are bandwidth hogs and some of their content is illegal. (A large share of HTTP is illegal, too, but Firefox is both legal and unencumbered.)

As Broache points out, the bill does allow an exemption for network management, as does Ed Markey’s House bill from last session (pdf) and his new bill expected soon in this session. The question is whether this exemption is big enough to allow Comcast’s new Berlin Wall impression to stand.

Broache and the experts she cites (including the very awesome Harold Feld of the Media Access Project) express concern that the Markey version in particular leaves a big enough gap to allow some potential for such a p2p blockade to stand. Let’s go to the tape.

In Section 4(b), we find:

EXCEPTIONS.—Nothing in this section shall prohibit a broadband network provider from implementing reasonable and nondiscriminatory measures to (1) manage the functioning of its network, on a systemwide basis, provided that any such management function does not result in discrimination between content, applications, or services offered by the provider and unaffiliated providers.

This is flawed and open to interpretation. If the period comes after “services” the exemption is airtight. You can manage bits however you want, but don’t pick and choose winners from losers. As it stands, though, I still think Comcast is almost without a leg to stand on.

First, Comcast is definitely in a business that competes with BitTorrent–namely, the cable television business. BT is largely a means of acquiring video; at least in terms of total bandwidth traffic (important when evaluating the rhetoric of network management), video is surely the lion’s share of BT usage.

Second, Comcast’s current efforts are neither reasonable nor nondiscriminatory. They’re effectively blocking all users from seeding any files regardless of current network congestion or usage patterns, which is unreasonable when they could provide much better overall traffic management by adjusting it to reflect congestion and each user’s bandwidth greed. (See the EFF link above.) And they’re blocking some bidirectional filesharing programs (BT, Gnutella) but not others (LotusNotes), which is clearly discriminatory.

Markey’s 2006 version is indeed not as tight as the Dorgan-Snowe version in the Senate. Dorgan’s network management exception requires that all decisions not be inconsistent with the mandated nondiscrimination provisions, including the requirement that the provider not pick and choose winners among applications and content. This is an important distinction. Under Dorgan, network management must be done without resorting to picking off specific applications. Under Markey, though, it simply must be “reasonable and nondiscriminatory,” which probably means something different to Michael Copps than Michael Powell.

Two more points are worth mentioning. First, any regulation is surely a substantial disincentive for Comcast to try something so boldly against specific applications. The marginal value of choosing this over choosing more sensible (and cost effective) traffic management is greatly diminished in the face of even a potential drawn-out, multi-venue legal proceeding. Thus, even if the bill does not force the FCC’s hand here, Comcast would never know how it would play out, especially in light of massive political pressure, so they would pick something safer, which means less discriminatory.

Second, both bills require providers to disclose all limitations on broadband networks. Comcast won’t want to publicly admit that they’re doing this; they’ve dragged their feet on doing so before finally chalking it up to “network management”. If they had to disclose it, this would dissuade them from doing it in the first place.

Of course, they could still do it and try to hide it, but they would surely realize this makes any necessary FCC proceeding even more likely to come out against them. They’d have a much harder time explaining it as mere network management if they were hiding it from their customers.

Either Markey’s or Dorgan and Snowe’s bill probably does the job, but the Senate version is the better of the two.

Markey Berates FCC Over Broadband Policy

In a congressional hearing this week, Rep. Ed Markey (D-MA), chairman of the House telecommunications subcommittee, took the FCC to task over its recent track record of deregulating broadband.

Especially since 2001, the FCC has rolled back many of the policies that require those who own the broadband infrastructure to share their lines. Instead, the Commission has let incumbent local broadband providers–the largest cable and telephone companies–dictate the terms and conditions for their competitors’ access. Markey laments:

Many other Nations took one look at our broadband situation, learned from our experience, and took the opposite approach. Japan and U.K. implemented the very policies that the FCC had gradually eliminated in recent years, such as local loop unbundling and broadband resale, which facilitate competition using the incumbent’s plant, regardless of technology. These foreign competitors are now enjoying broadband success stories.

If you think current broadband policy is remarkably defective, you should thank Ed Markey for his efforts to turn things around. You can read a transcript of Markey’s testimony here.