1 2 3 4 5 6 7

A certain amount of this information gets stored in a file called a "cookie." A lot has been written about this, but in fact, cookies aren't as big an invasion of privacy as they sound at first, especially because they're actually stored on your system, not on the Web site. If you've used the Web since about the middle of 1996, you probably have a cookie file in the directory on your hard drive where your Web browser is stored; it's called COOKIES.TXT and is just a text file full of arcane information that won't mean much to you but helps the Web site give you a somewhat personalized service. The idea is that, for example, a site's "What's New" page might reflect what's new on the site since you last visited, rather than since the last time the programmers updated it.

But precisely how much people dislike giving up personal information can be shown by the fact that, in GVU's Sixth WWW User Survey,[17] roughly a third of respondents reported lying on Web site registration forms. (It also raises questions about direct marketers' glib statements that people really like to receive unsolicited product information.)

These GVU surveys take a different approach: instead of counting the size of the Internet population, they are interested in studying the values and attitudes of those already on the Net. Because their surveys depend on voluntary response from a self-selecting sample, rather than from a scientifically selected and weighted sample, there are inherent biases. People who dislike answering surveys, never use the Web, or use it only in highly rigid ways (such as looking only at a company's preprogrammed sites) won't appear, and those paying timed access charges may be deterred from answering the lengthy sets of questions. GVU did try to counteract this effect by advertising the surveys in a carefully chosen set of newsgroups, announcing them in non-Net media such as newspapers and magazines, and using advertising banners on high-traffic sites such as the main search engines. Nonetheless, it's well known in demographic research that a self-selecting sample of people interested in answering surveys is unlikely to be representative of the general population.

In her book Tainted Truth, an examination of the effects of corporate sponsorship on the results reported by scientific research, journalist Cynthia Crossen looked at the complex interplay between question design and sample selection:

To show how much difference poor questions and a non-random sample can make, consider this question posed by Ross Perot. In a mail-in questionnaire published in TV Guide, the question was "Should the President have the Line Item Veto to eliminate waste?"; 97 percent said yes. The same question was later asked of a sample that was scientifically selected rather than self-selected, and 71 percent said yes. The question was rewritten in a more neutral way--"Should the President have the Line Item Veto, or not?"--and asked of a scientifically selected sample. This time only 57 percent said yes.[18]

Unrepresentative or not, the sixth GVU survey still contains surprises indicating that the make-up of the Net might be different than we tend to think. One such example is the political leanings of Net users. A lot of expectations derive from the writing that appears in Wired magazine or from net.prophets such as John Perry Barlow, who has said a number of times that the Net's infective culture has spawned 30 million libertarians. GVU's sixth survey tells a different story: the largest group of Web users, 38.4 percent, identified themselves as centrist, with the next largest groups being left-liberal (27.3 percent) and libertarian (25.1 percent).[19] Other findings are less surprising, though they should sound a warning note for commercial interests: only 9.88 percent report reading junk email messages, nearly 4 percent retaliate, and, most significantly for the future, younger users are less tolerant of junk email than older ones.

As another twist on what is commonly believed, the Net may not recapture younger consumers for news media. The GVU survey reports that the younger generation consumes less news online, just as it does in traditional media. Further, only about 20 percent of those questioned in a survey commissioned by the Radio-Television