Internet Access Should Be Application-Agnostic

Julius Genachowski, the Chairman of the FCC, recently announced that he would ask the FCC to adopt rules to protect the open Internet at its open meeting on December 21st. We applaud the Chairman’s effort, but we worry the proposed framework, as it is currently drafted, will not result in the free and open Internet that is his goal.

The proposed rule has several problems:

  1. it prohibits only unjust and unreasonable discrimination but does not clearly define those terms,

  2. broadband access providers are not prohibited from charging web services like Google, Facebook or Twitter a fee to reach consumers or to get faster access to consumers, and

  3. users who access the Internet over wireless networks have few protections.

If these concerns are not addressed, access providers could use their ability to control access to the Internet to control the market for Internet applications and services.

I remember too well, the experience of investing in cable television programming start-ups back in the 90s when there was limited channel capacity on cable networks and the companies that controlled access to consumers made it very clear that they would need to own 20% of your company before they would agree to carry your programming on their network. The Internet we know today exists only because, until now, there have been no gatekeepers between consumers and service providers. We need to keep it that way.

The good news is that the FCC can balance the interests of web services innovators and consumers with those of telephone and cable companies without changing the substance of the proposed rule simply by defining application-specific discrimination as unreasonable.

Barbara van Schewick, a professor at the Stanford Law School, describes this approach here. She says the correct approach is:

“A non-discrimination rule that would ban all application-specific discrimination (i.e. discrimination based on applications or classes of applications), but would allow application-agnostic discrimination.”

The brilliance of this approach is that it offers cable and telephone companies great flexibility to package and price their services and to manage their networks without harming investment and innovation in web services.

If a user wants more packets or less latency, an access provider should be able to sell that to them. But for that access service to meet the test of being application-agnostic, the choice of when to use these services and for which applications must be left to the user.

Similarly, if a user consumes a disproportionate share of packets at certain times of day, a network provider should be able to temporarily reduce that user’s throughput to avoid degrading the experience of others. These actions would not threaten a free and open Internet because they are targeted at a consumer’s use of network capacity, not a specific application.

On the other hand, if access providers throttled only the bandwidth available to BitTorrent to deal with congestion, that would clearly be application-specific discrimination. Blocking or throttling video would be discrimination against a class of applications.

This approach works equally well for wireless.

If an older wireless network does not have the capacity to handle lots of packets at peak times, it can reasonably limit the number of packets available to users. When congestion is eased it can open up the pipe again.

This is reasonable network management that does not distort the competitive market for web services. Blocking or discriminating against a specific web service like Skype or against a whole class of web services like streaming video would be prohibited under this framework.

If it is not possible to solve all network management problems on older wireless networks in an application-agnostic way, there could be an exception; but the presumption should be that network management would be as application-agnostic as possible.

If cable and telephone companies intend to use their control over consumer’s access to the Internet to extract outsize profits from the innovative companies working in the dynamic and competitive market for Internet services, it should be pretty clear to the FCC that they cannot reconcile their interests with those of consumers and innovators. If, on the other hand, access providers are, as they say, concerned only about their ability to invest in their network and manage it responsibly, they will support this application-agnostic regulatory framework.

This is not just a problem for venture capital investors. There is a great post here that summarizes all of these issues from an entrepreneur’s perspective.

We believe it is in everyone’s interest to improve the current proposal by:

  • defining any application-specific discrimination as unreasonable,

  • extending that reasonableness test to include wireless Internet access, and

  • making it clear that pay-to-play access fees (whether for access to users or faster access to users) are prohibited.

If you agree, we encourage you to write to Chairman Genachowski, Commissioner Copps, and, Commissioner Clyburn, Commisioner McDowell, and Commissioner Baker and urge them to work together to make this modest but important change before bringing the rule to a vote later this month.

We also encourage your you to exercise your own authority and influence, using the services that you use everyday to let the FCC know you understand the problem and will support their effort to create an application-agnostic regulatory framework:

1. Post your thoughts on your own blog.
2. Tell your friends on Facebook
3. Post this on Tumblr, and
4. Tweet this to your followers on Twitter.