I’ve been meaning to comment on Ed Felten’s fantastic four–part discussion of the nuts and bolts of network discrimination for a while now. In particular, I think his second installment hints at a strong argument against network neutrality legislation:
If a network provider is using minimal delay discrimination, and the high-priority traffic is bursty, then low-priority traffic will usually sail through the network with little delay, but will experience noticeable delay whenever there is a burst of high-priority traffic. The technical term for this kind of on-again, off-again delay is “jitter”.
Some applications can handle jitter with no problem. If you’re downloading a big file, you care more about the average packet arrival rate than about when any particular packet arrives. If you’re browsing the web, modest jitter will cause, at worst, a slight delay in downloading some pages. If you’re watching a streaming video, your player will buffer the stream so jitter won’t bother you much.
But applications like voice conferencing or Internet telephony, which rely on steady streaming of interactive, realtime communication, can suffer a lot if there is jitter. Users report that VoIP services like Vonage and Skype can behave poorly when subjected to network jitter.
And we know that residential ISPs are often phone companies or offer home phone service, so they may have a special incentive to discriminate against competing Internet phone services. Causing jitter for such services, whether by minimal or non-minimal delay discrimination, could be an effective tactic for an ISP that wants to drive customers away from independent Internet telephone services.
Here’s the problem: let’s say Congress has passed a strong network neutrality rule and charged the FCC with enforcing it. Comcast installs some new network equipment that happens to increase the jitter on its networks. Some Vonage user gets annoyed and files a complaint with the FCC.
The FCC investigates. Comcast says that they installed the new routers for reasons that were unrelated to impeding VoIP traffic. Perhaps the new router offers improved performance in other respects, such as increased throughput or better network-maintenance features. Although they suspect Comcast’s executives chose the routers deliberately to increase jitter, they have no way to prove it. The FCC will be forced to make a judgment call: did Comcast violate network neutrality or not?
That puts tremendous discretion in the hands of the commissioners. Because they can plausibly come down on either side of the issue, their decision could easily be swayed by inappropriate considerations. Outright corruption and influence-peddling are one possible set of inappropriate considerations. There are others as well. One can easily imagine a commissioner who happens to be crusading against smut on Internet-based video services dropping hints that if Comcast stops showing so many naked ladies via their video-on-demand service, he’ll be more inclined to go easy on Comcast’s potential network neutrality violations.
Another possible problem is that the FCC might have to start inspecting routers and certifying some as being non-discriminatory. I doubt I need to belabor the reasons that would be a bad outcome.
A third problem is the possibility of rent-seeking by Comcast’s competitors. Filing a network neutrality complaint could be an effective way to slow the introduction of new technologies that would otherwise give Comcast a competitive advantage. This would be exacerbated by the fact that the FCC commissioners aren’t likely to be experts, so they won’t necessarily have the expertise to quickly and decisively reject frivolous complaints.
A final problem is that once the FCC starts getting involved in vetting the design of networks, it’s likely to slow the evolution of the network. If the FCC is going to penalize ISPs who introduce unnecessary jitter to their networks, that means that network administrators will have to look over their shoulders every time they make major changes to their network design. Comcast might get in the habit of seeking advisory opinions from the FCC to make sure that any changes will pass muster with the agency. That, in turn, would cause administrators to tune their networks less frequently in order to avoid the hassle of additional paperwork.
I’m belaboring this point because I think proponents of network neutrality regulations are seduced by the seeming simplicity of the concept when it’s stated in the abstract. But the regulators charged with enforcing the rule won’t be able to retreat to generalities. It will have to rule on the legality of particular changes to network configurations. And in doing so, it’s likely to have a host of unintended consequences on the decentralized, rapidly evolving character of the Internet that we’ve come to take for granted. Once bureaucrats start saying yea or nay to network management decisions, ossification is all but inevitable.
Supporters of network neutrality like to draw favorable comparisons to the 20th-century regulation of the telephone industry as a common carrier, as though it were a model to be emulated. What they seem to forget is how incredibly slowly the 20th century infrastructure evolved, and how resistant to change it became. Yes, the network had some positive characteristics, notably excellent reliability, but it also hardly changed during the second half of the 20th century. Modems were a crude hack to work around the Baby Bells’ refusal to offer their customers a reasonably priced data service prior to the late 1990s. I don’t think it should be our goal to make the 21st-century Internet look like the 20th-century phone network.
While I’m on the subject of jitter, here’s another fact that Felten doesn’t discuss: jitter affects a lot more than just VoIP users. Another important class of users who are likely to be strongly affected by increased jitter is gamers. For someone playing a Quake 3 death match, intermittent 2-second latencies just aren’t acceptable. If Comcast started introducing jitter into its cable modem connections in order to discourage the use of VoIP applications, they would simultaneously piss off the millions of people who play realtime networked games. And thanks to XBox live and a thriving PC gaming industry, they easily number in the millions.