TechKnowledge No. 3

One Internet Is Not Enough

By Clyde Wayne Crews Jr.
April 11, 2001

Despite its Wild West reputation, the Internet is subject to substantial regulation. There are rules on children’s privacy, copyright, domain names and more. Congress is looking at new privacy and spam requirements, Internet taxation, gambling, and the digital divide. And the American Bar Association, the French government, and the UN are pushing global Internet governance plans.

A solution, however, is more Internets, not more regulations.

The Internet needs borders beyond which users can escape damaging political resolutions of these battles, which are rooted in the Internet’s nonowned, common-property status. Conflicting legislative visions in a cyberspace populated by exhibitionists at one extreme and would-be inhabitants of gated communities on the other, reveal the basic truth that not everybody wants or needs to be connected to everybody else.

To escape the regulation trap of the commons, the Internet should splinter and go private. One Internet is probably not enough. Instead, owned Internets-proprietary “Splinternets” where prespecified ground rules regarding privacy and other governance issues replace regulation and central planning-may be superior. What matters most is not necessarily the Internet as it exists today, but Internet technology.

Physically, parallel Internets could be up in short order. The Internet itself runs mainly on 13 root servers (computers). Other server and router hardware is obtainable. The fiber backbone is in place. And the audience and applications are growing. Already, dedicated video networks operating on Internet protocol are emerging.

The catch is securing widespread adoption, a horrendous marketing and money problem. Some consider today’s root server system to be a “natural monopoly.”

But as broadband infrastructure strategies and server warehouses expand, it is likely that we could surf distinct networks the way we surf Web sites today. At some point, the benefits of tailored, owned networks capable of harmonizing issues of privacy, values, access, and participation outweigh the costs of regulation, endless governance fights, and the costs of inherent insecurity on a nonowned Internet from which criminals and hackers can’t be excluded. One National Security Agency official recently even argued that real security is not an option on the Internet, and a more secure network should replace it.

Besides, diverse applications-streaming video, gaming, secure commerce and teleconferencing-may benefit from separation. The Wireless Application Protocol, a proprietary means of aggregating Web-based content on cell phones, already exists apart from the Internet. And AOL has been separate for many years.

While today’s Internet will forge ahead, companies may profit from a presence on several networks with correspondingly different relationships with users. Or they could provide content to newfangled middlemen who troll for information across networks and then filter it according to consumer preferences. To address widespread privacy fears, banks, hospitals, and insurance companies could combine with major Internet Service Providers, data hotels, and phone or cable companies to create an alternative Net. The purpose would be to keep confidential information off the “hackable” commodity Internet altogether.

Future networks could offer porn-free surfing-others, porn-only, perhaps with privacy guarantees. Some might allow kids to provide information online under specified conditions, others obviously would not. While networks paid for by ads could be frequented by people who don’t mind giving up personal information, those who don’t want to reveal personal information might use for-fee networks that ban mouse-click “profiling” by third-party advertisers. Since contracts govern private network property, companies that violate their promises to safeguard information they collect would be held accountable.

In a thousand ways, self-selecting users, through contracts, could sort themselves among competing networks with unique ground rules and policies. The same individual may present different faces to the world depending upon whether he is surfing a shopping, educational, adult, or medical network.

Certain overarching issues, such as encryption policy and Internet tax policy will apply to any network-and splintering won’t help. But even the issue of freely available copyrighted multimedia files exists largely because there has been no network owner to hold accountable in the Internet commons. The commons helped exacerbate the “Napster problem” in the first place.

It may be preferable for consumers and businesses to organize among assorted open networks rather than accept an increasingly regulated cyberspace cage. Warfare on the digital commons invites more regulation and adds to a deteriorating and antiquated Internet. Splintering, though it will be criticized as Balkanization, increases our options and wealth. It also protects our rights, which depend upon the institution of private property.

Clyde Wayne Crews Jr. is director of technology policy at the Cato Institute. A version of this article appeared in Forbes magazine.