Dangerous Ideas

Some critics think it's time to reinvent the wheel.

By Joel Snyder

Recently, The Economist ran an editorial about the Internet. If you haven't seen this British publication, it's a weekly newsmagazine, like Time or Newsweek, that's well respected all over the world for the quality of its reporting. At any rate, the editor decided it was time for The Economist to take a strong position regarding the way the Internet is being run. The editorial made it clear that the magazine is fervently opposed to technicians and academics making up the rules. Instead, it proposes that these "amateurs" of the old guard be cleared away in favor of a broader, more businesslike approach managed by qualified professionals. The author drew specific attention to Jon Postel, suggesting that he be put out to pasture. Before I tell you why this position is both misguided and reprehensible, let me explain who Jon Postel is.

Postel Addresses

Every system on the Internet has a unique serial number attached to it, called an IP address. Simply put, Jon Postel is the person who allocates those addresses. Without a unique IP address, you can't participate in the Internet. Thus, in many ways, Postel is in a position of considerable power on the Net. Forget domain names--that's just a convenience. IP ad-dresses are essential.

Of course, Postel doesn't personally assign you an address every time you dial in to the Internet. Your address comes from your local Internet service provider (ISP). Your ISP probably acquired a block of addresses from a larger provider. And the biggest of these (at least in the United States) get their addresses from the InterNIC. The InterNIC has peers in Asia and Europe, each of which has huge blocks of addresses that they allocate to Internet users. But where did the InterNIC get its addresses?

Technically, these addresses--millions of them--come from something called the Internet Assigned Number Authority (IANA). And the IANA, if you examine it closely, is mostly Jon Postel.

Who is this awesome individual? Is he a high-level Washington bureaucrat? Was he personally appointed by the president? Well, no. He's a researcher at the University of Southern California. Postel has been involved with the Internet since the beginning. Passing out numbers isn't a full-time job; like many of the scientists who led the development of the Internet, this is just one of the many things he does.

Minority Rules

This is typical of the way that the Internet operates. As a loose confederation of networks interconnected with each other, the Internet has no central authority. However, there are things which must be centrally managed for it to work: such as the assignment of names and IP addresses. So far, these tasks have been assigned either to individuals or very small groups.

And it's this awesome concentration of power in such a small area that The Economist finds so infuriating. It believes that Postel should step down, to be replaced by a governing board of businesspeople who could manage the Internet in the same way as, say, the CCITT/ITU and ISO. Unfortunately, this invariably means making compromise decisions by consensus that make little sense and are carelessly implemented for the sake of expediency. One of the main reasons the current system for managing the Internet works is because it is free of the bureaucracy typically associated with large governing bodies.

What The Economist (and many critics of the Internet's apparently casual way of decision making) does not understand is that the IANA only rules with the tacit consent of the major ISPs. The operational side of the Internet is run by a select group of large ISPs (such as Sprint, MCI, and Alternet). The IANA has no way to enforce its allocations; if the big ISPs were to decide against Postel, they have the power and the means to change how the Internet works.

But, surprisingly enough, they haven't. Well, it's not really that surprising, except to the technical illiterates who find it galling that the control of this otherwise democratic medium is in the hands of a few self-appointed guardians. In fact, the system works quite well. For example, when it was predicted that the Internet would run out of IP addresses by 1996, forward-thinking technical advisors were quick to come up with a strategy to extend the life of the Internet.

The strategy, proposed by the Internet Engineering Task Force (IETF), approved by the IANA, and implemented by the InterNIC, has three elements which no politically astute leader could have accepted. It was not egalitarian, since late-comers to the Internet world would shoulder a greater and more expensive burden than early adopters. It was very expensive, requiring many companies to upgrade software and hardware, completely without subsidy. And it limited the choices of businesses accustomed to having complete freedom.

But as unacceptable a strategy as the conservation of IP addresses might appear to some, the bottom line is that it worked, it continues to work, and it was the least expensive and most rational solution available. The Economist, I suspect, would have hated this strategy, had its staff been technically literate enough to understand it five years ago.

For the most part, the people who manage the Internet and make the major decisions do so with the best intentions. That is, to make the Net run more smoothly and efficiently. They continually strive for the most equitable solution to the ongoing challenges of this ever-evolving technology.

Dissension in the Ranks

Needless to say, those who agree with the opinions expressed by The Economist and similar critics can't abide by this. There are many users who are dissatisfied with the way key decisions about the Internet are being made. They believe that somehow the people who designed, built, and nurtured the Internet and have guided it to its present state are suddenly no longer capable of taking it further. They want to sweep aside the existing system and start over again with something new.

In some cases the source of discontent stems from their own lack of power and the feeling that those who run the Internet are not accountable for their actions. The Internet has two faces: one of strict control in administration, and one of complete freedom in content. When those accustomed to the free and open communications suddenly discover that their vast playground actually has a rock-hard border, frustration can result.

To some companies, the point of contention is the perceived financial inequity of the Net. They look at the supposed millions floating around for the few organizations that register domain names and allocate IP addresses and want a piece of that pie. These are the ones most likely to file suit--and several have--to change all that. They'd rather see an ignorant court step in and order some inane behavior, possibly destabilizing the entire Internet, than be cut out of the cash flow.

Under the guise of fair play, what these companies really want to do is break up what they see as a monopoly so they can create one of their own. This may result in some of the power being spread around a little, but not enough to really change things. And because of their inexperience and impatience to get on the Net, vital technical and operational issues often are neglected, leading to major problems later on. For example, follow the antitrust suit PGP Media filed against the InterNIC and you'll get an idea of how this works. (Visit,4,9525,00.html for more details.)

Despite all these concerns and criticisms, the changes made by the controlling elite have been mostly moderate adjustments that, instead of imposing unfair restrictions, may actually account for much of the freedom we enjoy in our use of the Internet.

Dumb and Dumber

There is a third group of discontented cyber citizens that appear to have too much time on their hands and too little sense or sanity to go with it. These are the solicitous folks who barge in on any online discussion, flooding the Net with absurd blather in an attempt to promulgate some ill-conceived notion that they believe is the only correct solution.

The combined force of these soapbox savants is harmful as well. The danger comes from the noise level of their disruptive prattle rising so high that all useful discussion and debate is squelched. On the one hand, the Internet is a revolutionary medium for unfettered communication, even at the highest levels. Anyone with something to contribute is welcome to participate. Unfortunately, this privilege has been flagrantly abused, and this misuse could drive the real movers and shakers behind closed doors--which would be a serious loss.

The other danger is one of confusion. With dozens of idiotic proposals for restructuring the DNS, IP addresses, and the Internet protocols themselves being disseminated by a crowd of self-styled pundits, the few good and rational proposals could get lost in the noise.

The Self-Repairing Net

And this brings us to the fourth group of vocal Internauts: those who have something constructive to say. I believe that the Internet should embrace, not resist, change. There are technical problems facing its continued growth, and there are business problems because of the wild and informal way in which it was assembled. All of these need good thinkers to help solve them, and the Internet itself could be the principal instrument of that change. After all, the Internet was designed for the rapid exchange of information, making it a most powerful forum for collaboration. Used wisely, the solution to many of the problems with the Internet, plus a working plan for its future, might well be found on the Internet.

What the Internet doesn't need are more profiteers, power mongers, and uninformed polemicists. What it does need are more people like those who built it: technically adept, generous with their time and energy, and driven by the goal of a better, stronger, and faster network. In short, people like Jon Postel.

Joel Snyder is a senior partner at Opus One in Tucson, Ariz.
Reprinted from Internet World magazine Vol. 8 No. 8, (c) 1997 Mecklermedia Corporation. All rights reserved. home