Is the Internet Doomed?

Before doing business on-line, consider the problems that could make cyberspace go the way of CB radio

Sooner or later, everyone who connects to on-line services wonders what he or she is getting into. My own anxieties began about eight years ago, soon after I joined CompuServe, then the largest computer network available to people who didn't work in government or science. Grazing happily through the service, I discovered many games that would be of interest to my stepson. I was pulling one down into my computer late one evening when the incident occurred.

Being a novice user, I did not know that when I was on-line my presence could become known to other CompuServe users. I was astonished when my computer beeped to inform me that Dave from Texas wanted to know if I wanted to chat with him. Taken aback to learn that strangers could jump into my personal space, I had no desire for a tête-à-tête. "No thank you," I typed.

Dave from Texas cursed me out. Vehemently. I cut the connection quickly, of course, but Dave from Texas had given me the willies. Seeing the four-letter words pop onto the screen had been like opening the study door to discover a vandal ripping out my desk drawers. I felt violated, invaded, soiled.

With more experience, I now realize that Dave was probably a harmless 13-year-old reveling in his newfound ability to tell off a stranger. Still, he taught me something. Even in those days, computer visionaries were foretelling a future in which the whole planet would be wired together -- a global village, in the argot of the day. With a few keystrokes, I would be able to reach anyone, anywhere, anytime. Dave from Texas made me realize that the opposite was true as well. Anyone would be able to reach me.

Today the buzzwords have changed. The global village is now the Internet, the World Wide Web, the information superhighway. But the problem still remains. As long as the superhighway is filled with people like Dave from Texas, other people are going to get sideswiped when they take a spin. In response, they may get off the road and not do their shopping there after all. The problems will only grow worse when millions more clamber aboard the Internet. As the tales of outlaw hackers demonstrate, not all the Daves are harmless loudmouths. Some will treat the Internet as a vehicle for crime and its access to millions as a huge opportunity for vandalism and theft.

The specter of Dave from Texas, though, is only one facet of a greater, more general difficulty: the openness and flexibility that give the Internet its dazzling promise as a universal means of communication and commerce also work to frustrate that potential. Indeed, they may even lead to the network's eventual collapse.

The Internet is often described as a "functional anarchy," in the phrase popularized by science-fiction writer Bruce Sterling. The description is accurate. The network resembles a group of kids hanging out at the mall. No one member tells the others what to do or how to behave, but they tend to stick together anyway. They sheepishly lumber from the video arcade to the CD store to the shop with $130 basketball shoes. Things work out fine, despite occasional personality conflicts. Note, though, what makes this mode of existence possible: mom and dad, who furnish the money. The Internet is like that. It is a sociotechnical entity wholly evolved in the giddy freedom that comes from not having to worry about the bills. In the case of the Internet, the parental unit was the Advanced Research Projects Agency (ARPA) of the Department of Defense. (It was later called DARPA, the initial D standing for "Defense.") True to its name, DARPA was concerned with what Vinton Cerf, the former Stanford engineer and one of the principal designers of the Internet, calls "highly flexible, dynamic communication for military application." Its researchers were investigating how to link officers at headquarters with soldiers on battlefields thousands of miles away. Because wartime bombing might destroy any individual computer and its operators, the network needed to function no matter what pieces remained; it couldn't have a single brain.

To achieve that, Cerf and others designed the TCP/IP protocols. When one part of the Net requests information from another, the Transmission Control Protocol software breaks up the data into chunks and numbers them for reassembly like so many model-plane parts. The Internet Protocol then shoots each chunk zigzag style from computer to computer to its destination. The process is inefficient but highly reliable; pieces may travel by different paths, take longer than necessary, and arrive out of order, but they show up at the proper destination.

TCP/IP and the other innovations behind the Internet were remarkable technical achievements. But they were designed to link together a few thousand people, most of them on the government payroll; they were not intended to become a medium for millions of ordinary people, most of them paying their own way. Still less were they conceived as vehicles for commerce. As a result, the innovations that made the network possible have also created its worst problems.

For instance, the TCP/IP protocols allow the Internet to be expanded with wonderful ease. Anyone can join. Instead of straining the system, additional sites mainly represent new transmission routes. When the National Center for Supercomputing Applications in Illinois released its Mosaic software in 1993, tens of thousands of new individual users climbed onto the Internet from their homes.

Unfortunately, the TCP/IP protocols have another, less desirable feature, according to Bridger Mitchell, an economics consultant who is the coauthor of Telecommunications Pricing: Theory and Practice (Cambridge University Press, 1991): they make it difficult to allocate services. DARPA didn't care about how much the Internet was used. But because the costs of use are disconnected from the price paid by users, Mitchell says, the TCP/IP protocols represent "a recipe for gridlock."