Invisible Links
by Bob Nardini, Senior Vice President & Head Bibliographer
Michael Bugeja
"I literally lost 10 percent of the book," recounts Michael Bugeja, describing what happened to him about a year ago when checking the footnotes in his manuscript for Interpersonal Divide: The Search for Community in a Technological Age. Dr. Bugeja, director of the Greenlee School of Journalism and Communication at Iowa State University, found that a third of the online sources he'd cited could no longer be accessed through his original links. "That really made me shiver," he says. "As a journalist, that could look like fabrication."
Bugeja (the name is pronounced Boo-SHAY-ah) pursued each of the lost links, finding valid ones when he could, composing new citations to fresh supporting material when he had to, and dropping some points altogether. He was forced to change 10 percent of the manuscript, he estimates, before submitting a final version to Oxford University Press, which published his book in January 2005. Bugeja's most remarkable chase, perhaps, was a search for the mission statement of Microsoft he'd cited in his introduction. This was the very first link that didn't work, and a portent. Microsoft's own version of the firm's mission statement had changed between the time of Bugeja's research in late 2001 and his fact-checking in 2003. He eventually found the original, but had to pursue it to an obscure Australian telecommunications website. In the end, Bugeja made a screenprint of each online citation in the manuscript, gathered them up in a thick binder, and sent it to his editor at Oxford.
"Like everyone," says Bugeja, "I was aware of the phenomenon of 'linkrot.'" But, he associated that with dead hotlinks, not with footnotes, and the extent of deterioration in his own notes was an unpleasant surprise, leading Bugeja to research the problem. Collaborating with his Greenlee colleague Dr. Daniela V. Dimitrova, Bugeja found that 33 percent of the online citations in issues of 5 leading communications journals from 2000 to 2003 failed to work in the summer of 2004. Bugeja and Dimitrova estimate the half-life of the citations they've studied to be under four years. They formally presented their findings at a May meeting in New York City of the International Communication Association, which gave their research "top paper" status with others in the technology division. The Chronicle of Higher Education and Washington Post have run interviews. The Serials Librarian will publish their work later in 2005. "How to use the Web," as Bugeja recaps the project, "to show how unreliable the Web is." Bugeja and Dimitrova summarize and update their research on a website, www.halfnotes.org.
Bugeja believes that helping to address the decay of online footnotes needs to be at the top of concerns for academic libraries, whose core mission should be "serving researchers as ultimate repositories of unassailable fact." He fears that academic librarians have strayed by embracing computer technology before fully understanding the risks. "We are aswim in manipulated information posing as fact, all accessible via the Web. Academic libraries," he says, "should not solely be 'dissemination' points for so-called information; they should be the final word on matters of reference and citation."
Greenlee students are taught to print out the online sources they've used or cited, as Bugeja did for Oxford, to show retrieval dates, and to cite printed sources whenever possible. "The closer you are to a piece of paper," notes Bugeja, advising citation to pdf files over html, whenever there is a choice, "the more stable a text will be." Journal editors, he says, have an obligation to educate authors about the problem and to maintain the integrity of their own online material. Footnotes lapse for a variety of reasons. By changing the file structure of their online material without considering the impact on existing links, for example, editors can easily invalidate all of the citations researchers have made to their articles.
In science and medicine, this could literally be a life and death matter. The inability to replicate the research of others would undermine the foundations of scientific research as it has been known. In the humanities and social sciences, where the stakes might not be life and death, the entire research enterprise is nonetheless threatened in the same way when citations and even texts turn to vapor. Bugeja, who earned his Ph.D. in English and once aspired to be a Shakespearean scholar, draws comparison to the 16th-century printers who would corrupt the texts with which they worked. Today's scholars can thwart these printers through close comparison of variant versions, enabling at least an approach to the original texts. But on the Internet, "print's opposite," as Bugeja puts it, a domain where impermanence is the rule, how is that possible?
"I'm at a loss at what to do," he says. "I don't have the ability to conceive of a new research paradigm." Perhaps younger people, scholars less grounded in the world of print than academics of his own generation, will be able to reinvent research for the digital era. As for himself, "I'm a product of the printing press."
Bugeja, 52, grew up in the New York City area, "in the shadows of Manhattan," where an enduring memory for him is the sight of newspapermen, notebooks in pocket, running in and out of the Daily News Building. Beyond New York, Bugeja recalls, the entire country was for him "a map of newspapers."
"The New York Times," he lists papers in rapid-fire, "Daily News, Washington Post, St. Petersburg Times, Atlanta Constitution, Cleveland Plain Dealer, Minneapolis Tribune, Des Moines Register," he pauses-"that Mecca of so many Pulitzers." Bugeja began his own career as a newspaperman in New Jersey, but something about the Midwestern work ethic as displayed in Des Moines drew him to the center of the country, where he has lived most of his life since. He earned an M.A. in Mass Communication at South Dakota State University and his Ph.D. at Oklahoma State. Before joining the Greenlee School in 2003, he taught at Oklahoma State and Ohio University, and has worked for United Press International. Beyond contributions to numerous magazines, journals, and newspapers, Bugeja, whose own vita could be the very definition of work ethic, has written 19 books, among them fiction, poetry, scholarly non-fiction, and popular non-fiction.
In a writers' magazine he once wrote about the importance of choosing the right name for a poem. Today a book's title is important, says Bugeja, in ways he wouldn't have imagined then. Not only, as was always true, should a title capture a work's identity, but today a book title needs also to do service as a domain name, for its own booksite. At www.interpersonal-divide.org, which Bugeja maintains with Oxford University Press, readers and potential readers can see the book jacket and table of contents, and inspect even the index of Bugeja's latest book; they will find blurbs written by Hodding Carter III, Theodore Roszak, and others; they can read press releases, news stories, and reviews; can read about Bugeja himself; can of course place an order for the book; and teachers can access suggested lecture material, course syllabi, sample assignments, even sample midterm and final exams, "all you need to adopt the book," he points out, "for free."
Bugeja, despite the decay of footnotes, despite the erosion of face-to-face contact he bears witness against in Interpersonal Divide, despite iPod-carrying teenagers asking him what he was doing-they really didn't know-when set up in a shopping mall for a book-signing, "like an exhibit you might find at the Iowa State Fair," is hardly anti-technology-as his own booksite demonstrates. In fact Bugeja himself bought the domain name, "weavingtheweb," which is the title of a book by Sir Tim Berners-Lee, inventor of the World Wide Web. He thinks Berners-Lee should have it, for a booksite. "And I'll give it to him," says Bugeja, "if he visits ISU."
Published by YBP Library Services
999 Maple St., Contoocook, NH 03229 USA
v: 800.258.3774 f: 603.746.5628
w: www.ybp.com
e: academia@ybp.com
