is that there are alternative modes for developing new infrastructure in communications, and these modes are treated asymmetrically by present legal structures. Second, the costs and benefits of restructuring communications industries are difficult to observe; therefore, revised policies for these industries will be difficult to formulate. Third, some core and long-established principles for the regulation of communication activities are being upended by current changes in technology, which raises new regulatory challenges.
He approached these three themes by looking backward five years and forward to the next five. In the last five years, we have gone through a period that has almost no precedent. Researchers have shown that the Internet emerged from a two-decade incubation period as a government-managed technology program and became a refined and malleable commercial operation whose features are relatively advanced.
The early Internet had some peculiar features. It was optimized to a non-commercial environment, with end applications generally assumed to be independent of transport mode. It also presumed a lack of gateways, which is now being questioned. The Internet accommodated the needs of the Academy as well as a number of other organizations that had helped to set its standards.
When this technology—in other words, TCP/IP-based applications—was thrown into a commercial environment, it forced an extensive retrofit of the existing communications infrastructure. The experience was not unlike the displacement of the telegraph by telephony, when two independent communications structures arose independently. The Internet retrofit was straightforward because it was expedient to link dial-up modems with the existing public telephone switch network. This fostered rapid entry of customers and a huge demonstration effect. People quickly saw how easy it was to use and wanted to join. It also stimulated unprecedented and almost spontaneous growth of the communications infrastructure. The growth of the Internet represents a massively decentralized, unorchestrated investment activity. There is no precedent in communications for such a pervasive application of a relatively refined technology, although there is precedent in other realms of electronics.
The business aspect of this new technology, however, was not refined, which resulted from its incubation in academic and government environments. That, said Dr. Greenstein, presents three problems. The first is the last-mile problem: how to wire or otherwise connect individual users and businesses from their homes or offices to the nearest Internet backbone. The earliest and still the most common way to do this is by dial-up Internet access through the telephone sys-