Network: Tell me another one

The story was too good to resist. In a row at Christmas, Hillary Clinton had punched Bill, claimed the Drudge Report (the Internet site that broke the Clinton/Lewinsky story). Editors at three UK dailies - the Daily Mail, The Express and The Times - snapped it up and printed it on Boxing Day. But had they known the origin of the story, they might have thought twice about running it. The source? None other than that notorious US supermarket rag the National Enquirer, best known for its exposes of alien abductions, Elvis sightings and monster babies.

It wasn't the first time newspapers had been duped by the Internet. Last June, an Associated Press reporter accidentally posted an obituary for Bob Hope on the wire service's website. During the 20 minutes it remained on the site, a US Congressman read it and eulogised Hope on the House floor, prompting other news organisations to pick up the story.

Had this happened just once, it would look like sloppy journalism. But so many instances of Internet-related misinformation suggest a larger problem: interpreting and evaluating Internet sources is difficult. Unlike print publications, online publications often don't list authors or date of publication. E-mails get sent under fictitious names and websites appear and disappear overnight. Web searches produce hundreds of thousands of sites related to a topic - without measuring the reliability of the sources. How much of what you read on the Internet should you believe?

It's a problem that two new books address. In the recently published Columbia Guide to Online Style (Columbia University Press, $17.50), Janice Walker and Todd Taylor recommend standard styles for students, scholars and writers to follow when they produce or cite online sources such as databases, discussion lists and professional sites. The Modern Language Association of America (MLA) added similar style guidelines to the fourth edition of its MLA Handbook For Writers of Research Papers (1995), and the fifth edition of the MLA Handbook, scheduled for publication this spring, includes more extensive guidelines that are summarised on the MLA's website (www.mla.org).

Both guides recommend that scholars give more information than they would when citing print publications. For instance, a citation for an article in an online journal would, like a print citation, include the author's name, the article title, the name and issue of the journal, and the date the article was written. But it goes further to include updates and details of recent accessing. The idea is to give enough information so that all sources can be tracked. As Joseph Gibaldi, director of book acquisitions for the MLA and author of the latest MLA Handbook, says: "For example, it's useful to have the name of a [site's] sponsoring organisation, so that if the site does change a researcher can find the site through a search engine." If Matt Drudge had documented his sources according to the guidelines in either of these books, Hillary may never have hit Bill in print.

Will better standards for citation fix the problem of Internet misinformation? Or is this an instance of academics with little online experience trying to impose archaic standards on a medium that thrives specifically because it is transient? Not so, say Walker and Taylor. "The pre-eminent goal of style is to support the continuous, communal, and cross-generational process of knowledge building."

The Columbia Guide grew out of scholars' need to cite research they were doing on the Internet. Its predecessor was a four-page style sheet that Walker developed in 1994. At the time Walker, a graduate student in composition and rhetoric at the University of South Florida and editor of the online Journal of Composition Theory, was doing research on MOOs (virtual environments). Very few of her sources existed in print. "I took existing forms that had been used to cite print and tried to translate those into electronic source citations," she says. The result was the Walker Style Sheet. In March 1995, the Alliance for Computers and Writing voted to endorse the style sheet. Soon after, Walker was bombarded with requests from libraries, universities and publishers for permission to use it.

Walker explains that the Columbia Guidelines do more than simply transfer print standards to the Web. Rather, new citation forms have developed to reflect the new medium's idiosyncrasies - like multiple authorship. "New forms are emerging on the Web that don't fit our preconceived forms," says Walker. "We have to come up with new forms for thinking about them."

But citation takes time, and Web sources may change, or even disappear. Is it worth the effort? Walker thinks so. "Documenting sources lends credibility. We shouldn't only document sources that are permanent. Aristotle documented sources that we no longer have. We should document whatever seems worthwhile, and if it disappears, too bad."