1 2 3 4 5 6 7

Donna Hoffman and Tom Novak, associate professors at Vanderbilt University's Owen Graduate School of Management, in an article published in Wired in November 1994, in which they called for a net.census.[1] "It is time to act," they wrote. "The Internet has changed dramatically in size, character, and economic importance, but may not evolve further without careful measurement of its users. Until then, the lack of accurate and credible information about Internet users is likely to hinder the continued health and positive development of electronic commerce."

Hoffman and Novak are unusual fixtures on the Net's landscape, Hoffman in particular: she not only analyzes the Net but shows up every day on the WELL, where she sometimes styles herself "data geek." While others did the investigative footwork that was partly responsible for discrediting Martin Rimm's study of pornography (see chapter 9 ), it was Hoffman's knowledge of marketing research and statistical analysis that exposed the study's many technical flaws. While we browse, Hoffman and Novak are busy researching the marketing implications of commercializing the Internet, a specialty that didn't even exist a couple of years ago, as part of Project 2000, a five-year research effort sponsored by, among others, Daimler-Beng, HotWired Ventures, Sun, and the National Science Foundation.[2]

The immediate impetus behind Hoffman's and Novak's call to numbers was an article in the New York Times in which cyberspace correspondent Peter Lewis contrasted estimates of Net usership based on two competing surveys of Internet hosts.[3] Both surveys counted machines, rather than users, a process Hoffman and Novak compared to conducting a real-world census by counting buildings ("without regard to their function or contents") rather than people. One of the two surveys, Mark Lottor's July 1994 Internet Domain Survey,[4] put the number of hosts between 707,000 and 3.2 million; the other, John Quarterman's TIC/MIDS Internet Demographic Survey, administered by email in January 1994, put the number at 1 million to 1.4 million.[5] Depending whether you guessed 3.5, 5, 7.5, or 10 users per host (all common estimates at the time), you would get anywhere from 2.5 million to 32 million Internet users--a difference roughly equivalent to the population of Colombia or California.

In 1994, when the National Science Foundation pulled out of supplying the Net's backbone--the major arteries through which data is transmitted--in favor of commercial suppliers, some of these technical measurements became less easily obtainable because there was no longer a single main road through which all traffic flowed. At the same time, it became more difficult to count hosts using standard Internet utilities such as Ping (a network-testing routine that lets you determine if a system attached to the Internet is alive) because increased security on many sites blocked such inquiries. But this was also the moment when the Internet was being turned into a business, creating a new interest in such numerical data.

Noting that few Internet users at the time were interested in counting people rather than machines, except in the limited context of counting visits to individual Web sites, Hoffman and Novak complained, "It is foolhardy to be content with an 'adequate' number of visits to a site. In the explosively evolving Internet environment, we expect that the novelty of many commercial sites will soon fade, and then the real competition to attract visits to commercial sites will begin. In this competitive environment, accurate information on market potential and user needs will be critical." However, they added, "surveying the size of the Net will be difficult, complex, and costly."

Nonetheless, Hoffman and Novak proposed that such a survey should be attempted, taking into account why and how people actually use the Internet and how they react to its commercialization. They proposed the formation of an advisory panel and the development of a set of protocols and standards for measuring the Internet, with the key proviso that the information so developed should be made public. "Privatizing this information," they say, "flies in the face of the anarchic, yet democratic roots of the Net and may be the surest path to a monolithic, mass-market vision of a commercialized, yet sadly 'de-evolved' Internet." The point here is this: if the good demographic data are private, then the