Missing the point of the Internet

Network neutrality advocates have been preening and cooing since the election as they expect the Obama FCC and the new Democratic Party-dominated Congress to enact new laws and regulations advancing their pet cause. They got some support today from an unexpected quarter when the Cato Institute published a paper by graduate student Tim Lee echoing and supporting their main argument:

An important reason for the Internet’s remarkable growth over the last quarter century is the “end-to-end” principle that networks should confine themselves to transmitting generic packets without worrying about their contents. Not only has this made deployment of internet infrastructure cheap and efficient, but it has created fertile ground for entrepreneurship. On a network that respects the end-to-end principle, prior approval from network owners is not needed to launch new applications, services, or content.

Tim Lee, bless his heart, is wrong about the importance of the Internet’s end-to-end architecture. While the Internet, along with all other computer-based networks, certainly does have such an architecture, it’s not the only architecture or even the most important one in the mix. The most important part of the Internet is its “network-to-network” architecture, because that’s the part that makes it what it is. The Internet is only an internet because network operators have agreed to exchange traffic with each other according to terms that they develop among themselves without government interference. This exchange of traffic is what makes it an interesting place.

Internetwork packet exchange is not as simplistic as network neutrality advocates make it out to be. Network operators do not simply forward packets first-come-first-served to anybody and everybody for the end-to-end layer to sort out; they discriminate in all sorts of ways to provide good service to as many people as possible at a reasonable price. Some network operators offer different tiers of service to different customers, and exchange traffic with other networks accordingly. This is good, but it’s not the “stupid network” that our regulators want to see.

Network neutrality is an attempt to shackle the Internet with regulations that mirror a failed model of network architecture, to give a victory to a failed vision by government fiat that it could not achieve in the market. The government should not be picking winners and losers in the competition among network architectures.

Even if you don’t accept that argument, there’s another reason that the proposed regulations should be rejected: the Internet is a technology, and technologies can always be expected to improve over time as parts to build them become cheaper and faster. Net neutrality is a backward-looking agenda that seeks to freeze the Internet core at a particular level of technology. This can only have the effect of hastening its obsolescence, and make no mistake about it, it will be obsolete some day. Nostalgia has no place in technology regulation.

Indeed, Tim’s argument against net neutrality regulations is weak and non-specific. It’s a good reminder that advocates only make arguments about unintended consequences, slippery slopes, and camel’s noses when they’ve lost the argument.

Any attempt to add new regulations to the Internet should be examined from a bias against regulation. If a case can be made that new regulations will make things better, well and good. But arguments about restoring a once golden status quo should be rejected out of hand as incoherent and reactionary.

Technorati Tags: , ,

4 thoughts on “Missing the point of the Internet”

  1. Richard, it’s good to see that SOMEONE understands the actual, original intent of the Internet. The “end to endian” religion was not fundamental to the Net’s design, nor was it embraced by many of the original designers. Rather, the purpose of IP — the “internetworking protocol” — was to connect networks. Networks which were under different administrative control and different ownership, with different business models and different constraints on permissible uses. (To this day, the academic institutions and government contractors which comprised the original Internet each have their own, unique acceptable use policies.) And one of the reasons why it was created was to allow this freedom and independence — unlike the Bell System (which was under centralized control).

    Alas, many people who are not knowledgeable about the architecture of the Internet want to treat it like the Bell System of old (or perhaps even turn it into such a system). But this would kill the opportunities to innovate that the Internet has afforded to date. We must therefore expose the “network neutrality” agenda for what it is: a drive to regulate the Internet in such a way as to serve the interests of various special interest groups. As I’ve said many times, the only regulation which should apply to the Internet or Internet service providers should be prohibition of monopolistic or anticompetitive practices (just as these practices should be prohibited in any industry).

  2. Richard,

    What do you think the chances are that any one is going to focus on technical issues seeing as Susan Crawford will be heading up Obama’s transition team?

  3. The selection of Crawford and Kevin Werbach to head the transition team’s FCC work is certainly an attention grabber. It’s a break from tradition in the sense that both are Internet people rather than telecom people, which is something I like.

    The FCC needs a new direction, and these are good people to provide it. Werbach in particular has done some very intriguing academic work on network formation theory that shows him to be an original thinker. It think this was a good choice.

Leave a Reply

Your email address will not be published. Required fields are marked *