Pedias, Familiar and Otherwise
By Joseph Janes
American Libraries Columnist
Assistant Professor, Information School, University of Washington.
intlib@ischool.washington.edu
Column for October 2005
Matilda, empress and claimant to the English throne, died on September 10, 1167, near Rouen. I know this peculiar little factoid because the Encyclopaedia Britannica told me so. And who are we to question them?
For the record, Wikipedia gives the same death date for Matilda. In fact, its entry on Matilda is longer and more vividly and engagingly written. It gives more information, including her epitaph; links to related information, including historical fiction; sources discussing her life; and her familial coat of arms; and, unlike Britannica’s article, is signed.
Well, sort of signed. In typical wiki fashion, five people have contributed to the entry so far, one of whom wrote the original version. Others have since made minor edits, links, and tweaks. I know this because each page has a history telling who wrote what and when. Matilda’s article was initiated on July 5; the most recent edit was almost two months later.
Obviously, this is a very different way to build an encyclopedia. We are used to printed encyclopedias that are centrally controlled. The editors believe they are in a position to adequately decide what topics should be included, represent appropriate points of view, and choose people who can write authoritatively based on their experience and ability to use research that embodies the current state of knowledge and opinion. There are always myriad sources of bias and preferences at work, visible and invisible, obvious and subtle, which editing presumably deals with.
In the world of wiki, it’s entirely different. As a contributory form, the current state of a wiki is at best a compromise, a shared, collective, fluid work that reflects the biases and perspectives of the contributors in a more overt fashion. In theory, this yields a work that is more democratic and assumes that “exposing an article to many users will result in accuracy,” as the Wikipedia entry on Wikipedia currently states. (During the time I was writing this column, that entry was vandalized, and “Wikipedia is a piece of crap” was inserted by someone who’d made several edits to other entries in the past.)
Jimmy Wales, the futures-trading millionaire who founded Wikipedia, describes it as “an effort to create and distribute a free encyclopedia of the highest possible quality to every single person on the planet in their own language,” intending it to achieve a “Britannica or better” quality and eventually be published in print. He retains sole and final judgment over entries.
So while a traditional encyclopedia can—must, in fact—decide on what topics to cover and to what depth and by whom, the Wikipedia is what everybody involved wants to include—and more importantly can agree on—with some (minimal) centralized editorial intervention. Obvious questions emerge. How well and frequently will multiple perspectives be dealt with? Controversial topics? Will coverage be spotty, based more on popularity than perceived importance?
There is great potential here, of course, to have a reference work that is broader and more representative of wider points of view. There is also potential for cacophony, an encyclopedia with no serious claim to authority or accuracy other than revision after revision, always subject to new errors benign and otherwise, accompanied by people locked in eternal ideological death grips. These are all empirical questions, and the proof will be in the pudding.
Accuracy is a tricky business
Back to Matilda. How do we know that date is right? Britannica falls back on their institutional authority, presumably, since the entry is unsigned; the Wikipedia contributor could be asked for a source, or corrected it if was found to be in error. But who says? If you dig deep enough, accuracy is a very tricky business; shy of some original record, who knows when she really died? We often use authority—much easier to assess on the fly—as an avatar for accuracy when assessing a fact or a source, and the wiki world puts a whole new spin on that process.
Perhaps the most interesting tidbit I picked up at the OCLC luncheon at the ALA Annual Conference is a plan to add a wiki component to WorldCat, allowing normal people to contribute reviews à la Amazon, aimed toward creating a social recommendation system. Could this be true? The Fortress of Metadata, allowing the infidels to wipe their muddy feet on the Holy Grail of Libraryland? Gasp. This seemed truly seismic to me—or perhaps simply evolutionary, transitional, and maybe even inevitable. Some or all of the above, no doubt . . . but that’s another story.
|
