Wikipedia in the Newsroom   

February/March » While the line “according to Wikipedia” pops up occasionally in news stories, it’s relatively rare to see the user-created online encyclopedia cited as a source. But some journalists find it very valuable as a road map to troves of valuable information.

When the Las Vegas Review-Journal published a story in September about construction cranes, it noted that they were invented by ancient Greeks and powered by men and donkeys.

Michigan's Flint Journal recently traced the origins of fantasy football to 1962, and to three people connected to the Oakland Raiders.

And when the Arizona Republic profiled a controversial local congressman in August, it concluded that his background was "unclear."

What all three had in common was one of the sources they cited: Wikipedia, the popular, reader-written and -edited online encyclopedia. Dismissed by traditional journalism as a gimmicky source of faux information almost since it debuted in 2001, Wikipedia may be gaining some cautious converts as it works its way into the mainstream, albeit more as a road map to information than as a source to cite. While "according to Wikipedia" attributions do crop up, they are relatively rare.

To be sure, many Wikipedia citations probably sneak into print simply because editors don't catch them. Other times, the reference is tongue-in-cheek: The Wall Street Journal, for example, cited Wikipedia as a source for an item on "turducken" (a bizarre concoction in which a chicken is stuffed into a duck that is stuffed into a turkey) in a subscriber e-mail update just before Thanksgiving. In the e-mail, the Journal reporter wrote that some of his information was "courtesy of Wikipedia's highly informative turducken entry. As my hero Dave Barry says, 'I'm not making this up. Although, I'll admit that somebody on Wikipedia might have.'"

And when Time Inc. Editor-in-Chief John Huey was asked how his staffers made sure their stories were correct, he jokingly responded, "Wikipedia."

It's unclear if many newsrooms have formal policies banning Wikipedia attribution in their stories, but many have informal ones. At the Philadelphia Inquirer, which cited Wikipedia in an article about the death of television personality Tom Snyder last July, Managing Editor Mike Leary recently sent an e-mail to staff members reminding them they are never to use Wikipedia "to verify facts or to augment information in a story." A news database search indicates that "according to Wikipedia" mentions are few and far between in U.S. papers, and are found most frequently in opinion columns, letters to the editor and feature stories. They also turn up occasionally in graphics and information boxes.

Such caution is understandable, as for all its enticements, Wikipedia is maddeningly uneven. It can be impressive in one entry (the one on the Naval Battle of Guadalcanal includes 138 endnotes, 18 references and seven external links) and sloppy in another (it misspells the name of AJR's editor). Its topics range from the weighty (the Darfur conflict) to the inconsequential (a list of all episodes of the TV series "Canada's Worst Handyman"). Its talk pages can include sophisticated discussions of whether fluorescent light bulbs will cause significant mercury pollution or silly minutiae like the real birth date of Paris Hilton's Chihuahua. Some of its commentary is remarkable but some contributors are comically dense, like the person who demanded proof that 18th-century satirist Jonathan Swift wasn't serious when he wrote that landlords should eat the children of their impoverished Irish tenants.

Hubble Smith, the Review-Journal business reporter who wrote the crane story, says he was simply looking for background on construction cranes for a feature on the Las Vegas building boom when the Wikipedia entry popped up during a search. It was among the most interesting information he found, so he used it. But after his story went to the desk, a copy editor flagged it.

"He said, 'Do you realize that Wikipedia is just made up of people who contribute all of this?'" Smith recalls. "I had never used it before." The reference was checked and allowed to remain in the story.

Indeed, the primary knock against Wikipedia is that its authors and editors are also its users — an unpaid, partially anonymous army, some of whom insert jokes, exaggeration and even outright lies in their material. About one-fifth of the editing is done by anonymous users, but a tight-knit community of 600 to 1,000 volunteers does the bulk of the work, according to Wikipedia cofounder Jimmy Wales. Members of this group can delete material or, in extreme cases, even lock particularly outrageous entries while they are massaged.

The extent of the potential for misinformation became clearer in August, when a new tool called WikiScanner (wikiscanner.virgil.gr/) began providing an ingenious database to identify propagandists and hoaxers. It gave Wikipedia critics plenty of new ammunition, as it revealed that among those surreptitiously rewriting entries were employees of major corporations, politicians and the CIA trying to make their bosses look better. And then there was the John Seigenthaler Sr. episode, in which someone edited the prominent retired journalist's Wikipedia biography to insinuate that he briefly had been a suspect in the assassinations of John and Robert F. Kennedy. In an op-ed piece for USA Today in 2005, Seigenthaler, who once worked for Bobby Kennedy and was one of his pallbearers, railed against Wikipedia, calling it "a flawed and irresponsible research tool." (A Nashville man later admitted inserting the material as a joke aimed at a coworker, and apologized.)

No one is more aware of such pitfalls than the leadership of Wikipedia, whose online disclaimer reminds users that "anyone with an Internet connection" can alter the content and cautions, "please be advised that nothing found here has necessarily been reviewed by people with the expertise required to provide you with complete, accurate or reliable information." An even more blunt assessment appears in the encyclopedia's "Ten things you may not know about Wikipedia" posting: "We do not expect you to trust us. It is in the nature of an ever-changing work like Wikipedia that, while some articles are of the highest quality of scholarship, others are admittedly complete rubbish." It also reminds users not to use Wikipedia as a primary source or for making "critical decisions."

Wales says it doesn't surprise him to hear that some journalists are cautiously trying it out. "I think that people are sort of slowly learning how to use Wikipedia, and learning its strengths and its weaknesses," he says. "Of course, any reasonable person has to be up front that there are weaknesses... On the other hand, there are lots of sources that have weaknesses." Wales thinks the encyclopedia's best journalistic use is for background research rather than as a source to be quoted.

Wales, a board member and chairman emeritus of the nonprofit Wikimedia Foundation Inc., which owns Wikipedia, says the company constantly strives to improve its product. "Right now we're tightly focused on making sure that, for example, the biographies are well sourced," he says. The foundation is also developing new tools "to block people who are misbehaving," including one for new German-language Wikipedia users that will vet their contributions. If it works, Wales says, it can be rolled out for Wikipedia encyclopedias in other languages.

He also defends the right of Wikipedia — and perhaps even reporters — to have a little fun. "I subscribe to Google alerts and I saw that turducken [item in the Wall Street Journal e-mail] and I thought, well, what other source would you use? Britannica doesn't cover this nonsense," he says.

There are still plenty of journalists who aren't convinced of Wikipedia's worth, among them the denizens of testycopyeditors.org, where contributors to the online conversation have names like "crabby editor" and "wordnerdy." Asked his opinion of Wikipedia, Phillip Blanchard, the Washington Post copy editor who started testycopyeditors, responds, "I'm not sure what I could add, beyond 'don't use it' and 'it's junk.'"

While the Post has no written policy against it, "I can't imagine a circumstance under which a fact would be attributed to Wikipedia," says Blanchard, who works on the financial desk. "'According to Wikipedia' has appeared only a couple of times in the Washington Post, once in a humor column and once in a movie review."

Gilbert Gaul, a Pulitzer Prize-winning reporter at the Post, describes himself as a "dinosaur in the changing world" when it comes to rules about sourcing stories. Wikipedia, he says, doesn't meet his personal test — for one thing, "there is no way for me to verify the information without fact-checking, in which case it isn't really saving me any time." He prefers to do his own research, so he can "see and touch everything," rather than rely on the mostly anonymous content of Wikipedia.

"I like much of the new technology... But to me rules, borders, guidelines and transparency matter a lot," Gaul said in an e-mail interview. "I need and want to be able to trust the people I am reading or chatting with. If I can't, what is the point?"

Other journalists, though, are at least somewhat won over by what can be an impressive feature: those sometimes lengthy Wikipedia citations that lead to other, more authoritative sources. David Cay Johnston, a Pulitzer-winning reporter for the New York Times, says he recently looked up "thermodynamics" to see where it led him, and found that Wikipedia's entry listed numerous references from reliable sources.

"I have a solid understanding of the concept, but once we get into fine points, I have nothing beyond my skepticism as a reporter to judge the accuracy, validity and reliability of what is there," he says. "However, this entry appears to be useful as a source guide. It has names of researchers whose books were published by eminent organizations, and you can take that as a quick way to find sources. So as a tip sheet, as a road map to reliable sources, Wikipedia seems valuable."

Jim Thomsen, a copy editor at the Kitsap Sun in Bremerton, Washington, has no problem with attributing information to the online encyclopedia in certain cases. "If I see something in Wikipedia I might want to cite for background and context for a story, I trace back the cites to their original sources," Thomsen said in an e-mail interview. "If I feel the origins are solid, I'll use the info.

"I know there's been a lot of hullabaloo about people with agendas seeding Wikipedia with slanted or even false information, but as I see it, that sort of stuff can be easily sniffed out — by looking at the cites, and tracking them back. No cites? Fuhgeddaboudit. The bottom line is that Wikipedia can be a great tool as a central clearinghouse for contextual information. But not a single syllable there should be taken at face value."

The Los Angeles Times is one of many newspapers that have allowed an occasional "according to Wikipedia" in their pages in the last several months. One was in a commentary piece about Barack Obama; another appeared in a staff-written story about a professional "man in the street" who managed to be interviewed repeatedly. The reference in the latter story drew rapid fire on testycopyeditors.org, with comments including "Shame on the Los Angeles Times" and "No, no, a thousand times no."

Melissa McCoy, the Times' deputy managing editor in charge of copy desks, says the paper occasionally allows Wikipedia attribution. "We're certainly not going to use Wikipedia as a stand-alone news source, but we're not going to exclude it if it takes us somewhere," she says. "If a reporter spots something in there and it makes them do an extra phone call, it's silly" not to use it.

There's no unanimity about Wikipedia among academic experts, who have engaged in vigorous debates about the online encyclopedia. While many professors refuse to allow students to cite it, it has attracted some prominent defenders, including historians and scientists who have analyzed its content.

"If a journalist were to find something surprising on Wikipedia and the journalistic instincts suggested it was correct, the journalist might add that as an unsubstantiated Wiki-fact and invite comment," says Cathy Davidson, a professor at Duke University and cofounder of HASTAC (Humanities, Arts, Science, and Technology Advanced Collaboratory, www.hastac.org), a network of researchers developing new ways to collect and share information via technology. "Perhaps an online version of the printed piece, for example, might include a blog inviting people to comment on the Wiki-fact. It may be that there would be Wiki-facts online that were not in the printed piece. In other words, why not use the new technologies available to expand knowledge in all kinds of ways?"

Journalists also should consider, Davidson says, whether some of the sources they deem reliable have their own inadequacies. For example, when she recently researched the origins of calculus, she found that standard Western histories generally credited England's Isaac Newton and Germany's Gottfried Wilhelm Leibniz. But Wikipedia went much further, tracing the discovery of basic calculus functions back to the Egyptians in 1800 BC, and then to China, India and Mesopotamia — all hundreds of years before the Europeans.

So while journalists should be cautious no matter what resources they use, "What Wikipedia does reveal to those in the Euro-American world is knowledge which most of our sources, even the most scholarly, have, in the past, neglected because it did not fit in our intellectual genealogies, in our history of ideas," Davidson says.

In December 2005, the science journal Nature published a survey of several experts about the content of comparable Wikipedia and online Encyclopedia Britannica entries. In a conclusion hotly disputed by Britannica, Nature said that Wikipedia "comes close to Britannica in terms of the accuracy of its science entries," in that the average Wikipedia article contained four errors to Britannica's three. Britannica's 20-page response said that "almost everything about the journal's investigation...was wrong and misleading...the study was so poorly carried out and its findings so error-laden that it was completely without merit." The company further asserted that Nature had misrepresented its own data — its numbers, after all, showed that Wikipedia had a third more inaccuracies than Britannica — and asked for "a full and public retraction of the article." Nature stood by its story.

"The Nature piece profoundly undermined the authority upon which Britannica depends," says Gregory Crane, editor in chief of the Perseus Digital Library at Tufts University. He is a recent convert to the pro-Wikipedia camp, calling it "the most important intellectual phenomenon of the early 21st century."

He recognizes its faults, especially when Wikipedians write about controversial topics. So "people have to do some critical thinking," Crane says, by evaluating their sources, "whether it's Wikipedia or the New York Times."

In an article he wrote in 2005, Crane acknowledged that Wikipedia "is an extreme case whose success so far has shocked skeptical scholars." But he noted as well that other, more mainstream reference works had similar foundations — for example, the Oxford English Dictionary was written over a period of 70 years by thousands of people, including "an inmate at an asylum for the criminally insane."

A 2006 analysis by another scholar and Wikipedia fan, George Mason University historian Roy Rosenzweig, found some inaccuracies, omissions, uneven writing and even plagiarism in selected entries. But his comparison of several Wikipedia biographies against comparable entries in two other encyclopedias found that Wikipedia "roughly matches" Microsoft's Encarta in accuracy while still falling short of the Oxford University Press' American National Biography Online. "This general conclusion is supported by studies comparing Wikipedia to other major encyclopedias," wrote Rosenzweig, who was director of the university's Center for History and New Media until his death last year.

Still, many if not most in the academic community think that Wikipedia, if used at all, should be no more than a secondary source, and they frequently tell their students as much. For Cornell University professor Ross Brann, that position was reinforced in early 2007, after the outing of a salaried Wikipedia employee and editor who called himself "Essjay" and claimed to be a tenured professor with doctorates in theology and canon law. Turns out he had seriously padded his résumé: The New Yorker discovered after interviewing Essjay that he was actually a 24-year-old community college dropout. To Brann, a professor of Judeo-Islamic studies and director of graduate studies for the Department of Near Eastern Studies, the incident confirmed that Wikipedia could not be trusted as a primary source.

"I just tell students, 'Do not use Wikipedia, do not cite it, do not go there for my classes.' We're trying to teach them how to use sources, how to evaluate different sources, and I think that in general, although obviously a wonderful resource, for a student who just uses a search engine and they use the first thing that pops up..this undermines the kind of thing we're trying to teach them," Brann says.

Brann notes that Wikipedia's popularity probably has a lot to do with the fact that its entries so frequently pop up first, because that's the nature of search engines. "Many of them just work by the multiplicity of uses, others by virtue of ad arrangements — somebody is deciding for you what you're going to look at," he says.

And what about college journalists, a group that has never known life without computers? A news database search suggests that they are just as reluctant to cite Wikipedia as their professional colleagues. In August, for example, the University of Iowa newspaper, the Daily Iowan, used the WikiScanner database to determine that thousands of Wikipedia entries had been made or modified by people using the campus computer network. Some involved obvious but harmless enough vandalism: "Hawkeyes Rule" was inserted into text about the college's football stadium; less generously, a former university president was called an "eater of monkey brains," according to the paper's story.

Jason Brummond, editor in chief of the Daily Iowan, says he considers Wikipedia a good initial source, "but you go from there to find what most people would consider a more reputable source." Reporters in his newsroom generally understand that, he adds.

Brummond thinks the age of the journalist doesn't necessarily have that much to do with accepting Wikipedia: "It's more a personal awareness of how Wikipedia works."

In September, the University of Kansas student newspaper ran an editorial calling upon Wikipedia to do a better job of restoring "adulterated pages," noting that "despite a thousand recitations by our professors that Wikipedia is not a genuine source, students trust the site to give them accurate information." Nevertheless, Erick Schmidt, editor of the University Daily Kansan, says he doesn't rely much on Wikipedia, in part because his reporters write mostly about college and community issues. Plus, "we're taught to be cautious of things and skeptical," he says.

Schmidt rejects the notion that college students uncritically accept Wikipedia because they are infatuated with all things Internet. "We don't want to move things to technology because we think it's cool or paper is lame," he says. "But honestly, we are pressed for time, and if technology speeds things up.. that's why we're being drawn to it."

For his part, Wales maintains that the more people use Wikipedia, the more they'll come to understand and accept it. His conclusion, he says, "comes from people who have used the site for a long time and know, 'I have to be careful'.. which is what good reporting is supposed to be about anyway."

But whatever the verdict on Wikipedia, one thing should not change, says the New York Times' Johnston: "No matter who your sources are, when you sign your name, you are responsible for every word, every thought, every concept." Contributing writer Donna Shaw (shaw@tcnj.edu) has written about front-page ads, hyperlocal Web sites and Pulitzer Prizes for AJR.