Neil Waters had never seen anything quite like it.
“I was looking at a stack of final examinations,” said Waters, a professor of Japanese studies at Middlebury College in Vermont, “and I found several instances of misinformation that [were] identical from one student to another.”
All of those students in Waters’ Japanese history class late last year had been steered wrong by the same source — Wikipedia, the sprawling online encyclopedia that has revolutionized how ordinary people find information.
Wikipedia is a marvel of Web innovation and utility, but the incident in Waters’ class, added to several celebrated controversies in which entries for famous people were found to be false, raises a troubling question: Just how accurate is Wikipedia, and can you trust what it tells you?
For Middlebury College’s history department, the answer is plain: Not totally, and not always. The department banned students from using it as a source in their papers, although they are allowed to consult it for background material, a move that was quickly mimicked by professors at other schools, including UCLA and the University of Pennsylvania.
Harnessing the wisdom of the masses
Wikipedia is different from traditional encyclopedias in one crucial respect. Instead of seeking out recognized authorities in hundreds or thousands of fields to write its articles, it lets anybody — everybody — write them. And it also lets anybody edit nearly all of them at will.
The idea is that the large Wikipedia usership will yield experts on a particular topic. The back and forth as they debate and tweak entries should, in turn, yield a deeply reviewed and credible consensus article.
But the sheer size of that usership means tens of thousands of changes are made each day to Wikipedia’s nearly 1.7 million entries (that’s in the English version — there are Wikipedias for nearly every significant language on Earth, including Esperanto and even Tok Pisin, a Creole spoken in northern Papua New Guinea). And while Wikipedia has a large staff of moderators and trusted editors, it can take a while for entries to be reviewed.
If you happen to consult an entry that hasn’t been fully vetted or edited — or one that’s fallen victim to a flurry of disputed edits by folks with axes to grind — you can get into trouble.
Just this year, a Wikipedia entry falsely proclaimed that the comedian Sinbad was dead. (“Saturday, I rose from the dead,” he said.) Golfer Fuzzy Zoeller sued last month to find out who anonymously posted, falsely, that he abused drugs. And a prolific and highly trusted contributor believed to be a professor was unmasked as a 24-year-old college dropout.
Wikipedia comes clean
Wikipedia co-founder Jimmy Wales presides over a burgeoning empire. Wikimedia, the site’s host, has expanded into textbooks, republishable content, news, shared media and online project coordination. It all rests on Wikipedia’s reputation as an always available, convenient and reliable repository of the world’s knowledge.
But as controversies have grown, Wikipedia has had to fight to uphold its reputation. One way it now does so is by acknowledging its shortcomings.
“Reaching neutrality is occasionally made harder by extreme-viewpoint contributors,” it says, and it warns that “Wikipedia makes no guarantee of validity.”
“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,” it says in a general disclaimer.
‘Hacked to bits by hoi polloi’
That unreliability draws critics who say Wikipedia allows a forum for information vandals and propagandists. One of them is Larry Sanger, a co-founder of Wikipedia with Wales and its first editor.
While making it clear that he appreciates the merits of a project like Wikipedia, Sanger said in an article on the technology site Kuro5hin in 2005 that users are forced to take authors’ claims of expertise on faith and can be sandbagged by vandals at any time.
“If the project was lucky enough to have a writer or two well-informed about some specialized subject, and if their work was not degraded in quality by the majority of people, whose knowledge of the subject is based on paragraphs in books and mere mentions in college classes, then there might be a good, credible article on that specialized subject,” Sanger wrote.
“Otherwise, there will be no article at all, a very amateurish-sounding article, or an article that looks like it might once have been pretty good, but which has been hacked to bits by hoi polloi,” he added.
The conclusion is that users who rely on Wikipedia are running a risk. And for students whose research will be graded by real, honest-to-goodness experts in the classroom, that is probably too big a risk, said Sree Srinivasan, a journalism professor at Columbia University and visiting professor of new media at the Poynter Institute, a journalism education organization.
“We need to teach our students that, basically, information on Wikipedia can be updated very easily,” Srinivasan stressed.
On that point, even Wikipedia agrees.
The ban at Middlebury College “is a great idea,” said Jim Redmond, a Wikipedia administrator and editor. “Students shouldn’t even be tempted to use Wikipedia as an original source.”