1 2 3 4 5 6

organization and management of the Internet and concluded that 99 percent of the Internet's management could be--was designed to be--handled by automated means.[2] The remaining 1 percent includes tasks such as coordinating routing when there are software errors, creating standards for protocols and service quality, and coordinating the assignment of domain names and addresses. Gillett and Kapor argued that any proposals for change should distribute authority to avoid creating situations where resources are scarce (as in defining protocols that limit the number of available new addresses, for example) and provide a path for technical evolution, which they see as key to keeping the Net alive.[3]

Although Metcalfe was made to eat his words publicly in early 1997, lots of complaints about the Internet did surface in 1996. But they didn't come from users, who continued to flock online even though it meant learning why the World-Wide Web is so often called the World-Wide Wait. Rather, they came from the telephone companies, who complained that the combination of flat-rate pricing for Internet access and free--or at least, flat-rate--local calling encouraged users to stay online for lengths of time far beyond those the network was designed to manage. Just how big a difference was explained to the Federal Communications Commission (FCC) Bandwidth Forum, a January 23, 1997, meeting that assembled the points of view of a variety of telephone and online industry figures as well as consumer advocates.

Lee Bauman, vice president of local competition for Pacific Telesis, summarized the problems his company faced in grappling with California's 1 million active Internet users, who averaged sixty-two minutes a day connected to their ISPs whereas average residential customers typically originated only twenty-two minutes' worth of calls. "We last year had 23 billion minutes of use going to Internet access providers in California," he told the meeting, estimating that the company would have to spend $130 million in 1997 to beef up its network to handle the increased traffic. "We believe that a good portion of that money could be much better spent for the benefit of the country in building the new forms of data networks."[4]

The telephone companies' preferred solution is to charge ISPs, or "enhanced service providers," usage charges for access to the phone network, charges that presumably would be passed on to Internet users in the form of per-minute access fees. This would reverse five years of falling prices for online services of all types, and the concern is that higher costs might mean inhibiting the growth of the Internet.

This is not just a theoretical assumption: Internet use has grown much more slowly both in rural America, where users frequently have to pay long-distance charges for access, and the world beyond North America, where local calls are metered and charged by the minute. Conversely, when AOL switched from metered, per-minute charging to a flat-rate plan at the beginning of December 1996, the service immediately got blocked up by users who went online and stayed there. Discovering that it could be difficult to get through to the busied-out service, some users apparently began camping out online for even longer periods. AOL's CEO, Steve Case, begged AOLers to be responsible and log off as quickly as possible until the service could be upgraded; within weeks, junk email was going out advertising a program to keep your AOL connection open continuously.

Commenting on the Bandwidth Forum discussion, Shabbir Safdar, co-founder of the lobbying organization Voter Telecom Watch, wrote in the organization's newsletter, "At risk is the actual design of the Internet. Will it continue to be open and available to anyone with a computer and modem, or will it be stratified by pricing such that it is available to fewer people than are [sic] available today?"[5] Of the telephone companies' specific suggestion for usage charges, he wrote,

This solution places the majority of the cost on today's Internet users and businesses without adequately defining what the money will go to pay for. Worse, it runs the risk of focusing costs on a single group for a benefit that will be appreciated by everyone today and in the future.