Saving Net Neutrality the BitTorrent Way
Eric Klinker, the chief executive of BitTorrent, the file-sharing software company, has proposed -- only half-seriously -- an interesting solution to the net neutrality vs. bandwidth-hogging problem. Instead of having heavy traffic generators such as Netflix pay Internet providers for "fast lanes" to their customers, Klinker suggests that providers pay other companies to get into a "slow lane", shifting traffic to off-peak hours.
Klinker's idea is typical of net neutrality advocates' arguments: It is largely based on the essentially Communist argument that Internet providers should shut up about any extra charges because they have plenty of money, anyway. There is, however, a curious and potentially useful technical side to the proposal.
BitTorrent knows all about slow lanes. In 2011, it accounted for 13 percent of peak hour Internet traffic in North America. Now, according to the broadband-equipment firm Sandvine, its share is just 5.96 percent (which still makes it the third-biggest traffic generator after Netflix and YouTube). One explanation is that BitTorrent technology is widely used to share pirated movies and music, and in a world of cheap or even free streaming, that is an increasingly obsolete way to get one's hands on content. The other one is that, after Comcast shut off BitTorrent traffic in 2007 and 2008, the company made a conscious decision to stop being a peak-traffic hog.
It moved to the so-called Micro Transport Protocol, which shifts traffic to less congested times. A BitTorrent download will slow down when the network is relatively crowded and speed up when there's less demand for bandwidth. That doesn't only affect the pirate downloads but also legitimate uses of BitTorrent, including its popular Sync service -- cloud storage without the need for enormous data centers or the danger of government control.
So now Klinker says Internet providers should pay people who, like him, voluntarily renounce bandwidth hogging, just like electricity companies encourage people to use more power in off-peak hours. "This would relieve pressure on the network, yield a better experience for users and would be worth real money to the ISPs," he argues. "Additionally, there would be no unnatural pressure for the ISP to deliberately degrade the base service in order to manufacture demand for the priority service, as some have suggested might happen."
In fact, in the old days Internet providers did charge less for off-peak use. One can still find such plans in places including Nepal. Introducing "night tariffs" is a variation on the idea of broadband providers going back to the old practice of charging only for traffic used, which would be fair but inconvenient to users who want their bill to be predictable.
Klinker's proposal is only a rhetorical device: He's arguing that the problem of net congestion doesn't exist, so there's no need for Netflix to pay the providers or for providers to reward "slow lane" customers. "It's time to stop engaging in this kind of zero-sum thinking, pitting one user of the network against another," Klinker writes. "There's no scarcity." As with many net neutrality advocates, he refers to Internet providers' fat gross margins to prove his point.
It's true that bandwidth is not really scarce. According to the Federal Communications Commission's 2014 reporton U.S. broadband performance, providers now routinely exceed their advertised speeds, which wasn't the case a year earlier. That kind of generosity would be impossible if there were a capacity shortage. The profits argument, however, remains ignorant or disingenuous. To make the increases in speed possible, providers have to maintain and upgrade their networks. The investment it requires does not affect gross margins. Comcast in 2013 reported capital expenditure of 8.5 billion, more than its net income for the year and about 13 percent of revenue.
It's wrong to expect Internet providers to keep paying for equipment that allows us to watch more video, play "heavy" games over the Internet or increase our use of the cloud. Now, they want heavy traffic generators to chip in, and the FCC appears to support them. Perhaps, however, Klinker is right about revisiting old traffic-based payment plans. I think it would be misleading to most consumers, but having a discussion about how we want to be charged would help figure out how much people really care about net neutrality.
If we truly want a neutral Internet in which all traffic is created equal, we should be happy to pay for our actual usage. That would be fair, and providers would not need to create fast or slow lanes. If, on the other hand, we want the convenience of a fixed payment, net neutrality is no more than a meaningless fetish.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Leonid Bershidsky at firstname.lastname@example.org
To contact the editor on this story:
Toby Harshaw at email@example.com