BY ANNA BROOK
It is not uncommon for Wikipedia to pop up as one of the top matches of a Google search. Although Wikipedia is “the online encyclopedia,” the trick is figuring out if the information is valid or a figment of a bored teenager’s imagination. On Thursday, March 16, David Weinberger, a fellow at the Berkman Center, provided tips on how to assess the authority of Wikipedia articles to a group that included Berkman fellows, visitors from MIT, a librarian, a Crimson reporter, and of course one from The Record.
The talk started with a discussion of traditional sources of authority: encyclopedias and recognized experts. Unlike Wikipedia, the Encyclopedia Britannica has teams of editors check every article and then publishes authoritative information on the topics it selects. The limits of this, Weinberger says, are that only certain content is published and once it is printed it cannot be modified without issuing a new multi-volume edition. With any printed source of authority, the problem is that “the universe is very, very big, and the journal is very, very small,” said Weinberger.
Wikipedia solves the space limit problem; it already has over 1 million entries, compared to Britannica’s 65,000. All of the entries, however, need to be read with a grain of salt. There is no multi-step process as there is at Britannica, but there are ways to gauge an article’s credibility, according to Weinberger. Articles that have a long history of edits and active “talk” pages are more reliable than those written by a single author with no comments. Citations and the tone of an article give further clues to its accuracy. S.J. Klein, a Wikipedia administrator and steward, mentioned that articles with one author that have been viewed many times without significant changes are also authoritative since no one thought the information needed editing.
Weinberger pointed out the difference between traditional authorities who take credit for their work, and “Wikipedians,” who have no self-identification. This leads to the question of why anyone would spend time to research and edit an entry. People do, however, take pride in their Wikipedia postings. They can either create a pseudonym or edit as a guest. Guest edits are actually less anonymous than using a pseudonym because it records the person’s IP address. Some pseudonyms become experts in a particular field in the online community and are recognized by others who visit posts to which they contribute.
Weinberger summarized the editing process by stating that individuals who may be the foremost authority on a topic in the off-line world but refuse to negotiate tend to fare worse than those who can accept other viewpoints. Wikipedia is a collaborative effort that requires authors to be able to justify their position and consider changing it to accommodate others. A librarian attending the talk commented that such compromises often lead to bland middle-ground entries on controversial issues that fail to reflect the full flavor of the controversy.
Anyone can create or edit a Wikipedia entry, but there are approximately 850 administrators to prevent chaos. Administrators monitor changes and sometimes have to intervene. Two of these administrators were at the talk and provided insight into their duties. They check for vandalism, un-encyclopedic entries and “edit-wars”. To demonstrate how quickly invalid entries can be caught, a new article about a Berkman fellow entitled “Rebecca MacKinnon is a kitten-eating cyborg” was detected and deleted within 80 seconds. The creator received a polite message informing him that his test was successful. Other, less obvious “inaccuracies,” such as putting the same line into MacKinnon’s biography, can stay up longer, but tend to be caught eventually.
Wikipedia is still not accepted as a scholarly source. Weinberger pointed out that even Jimmy Wales, Wikipedia’s founder, insists that there are inaccuracies, that Wikipedia is a work in progress, and that it should not be cited in scholarly work. Instead, it should be used as a starting point to find background information and other sources.
Last semester, a Wikipedia posting by HLS students demonstrated the site’s editing checks. A page was created for “Nick Rose Day,” an HLS holiday initiated by Section 5 members on December 8, 2004 to celebrate Rose, a 2L, for being a nice guy. After approximately 5 days of negotiation, administrators deleted the page. Although The Record ran a notice wishing everyone a Happy Nick Rose Day, there was no other indication that the holiday exists. What may have doomed the entry, however, was its tone. The history of Nick Rose Day started off, “A number of students of the Harvard Law School class of 2007 dreamt up Nick Rose Day last year when I noticed that Nick Rose was the most genuinely nice person I knew at HLS.” This apparently was too biased and personal for Wikipedia, and the article was axed.
Suggested encyclopedic reading is the article on Japanese toilets, found at http://en.wikipedia.org/wiki/Japanese_toilet, which includes excellent citations and a thorough history.