Skip to content

Wikipedia’s Reality Check

The responses from some corners of the Wikipedia community were predictable when Larry Sanger, part of the team that developed Wikipedia, published a long article on kuro5shin arguing that “anti-elitism” and “trolls” were undermining this user-contributed “encyclopedia”–or as Jason Scott nailed it in an earlier article that inspired Sanger’s cry in the desert, “a low barrier to entry leads to crap.”

One commenter fumed, “Many of the critics of Wikipedia are either technophobes or the type that refuses to contribute to it to help solve the reliability problems.” Another took to great heights, arguing that “Because Wikipedia does not pretend to the utopian Enlightenment ideal of an authorative [sic] telling of the truth it often better portrays a richer tapesty [sic] of truths.” Wow, that’s a lot to put on the shoulders of a user-contributed “encyclopedia” largely ignored by serious scholars.

Some commenters agreed with Sanger on his point about trolls and Wikis, noting that “Wikipedia has an unrecognized problem with cranks on obscure topics.” Well, perhaps not that unrecognized, now that Sanger and some supporters have outed themselves on this issue, which is an admission of sorts that there is a logical basis for the scholarly community’s lack of enthusiasm for Wikipedia. Of course, librarians know that the ability of specialty encyclopedias to authoritatively address “obscure” topics is part of their value.

What I find most intriguing about Sanger’s long confessional is that he helped develop Wikipedia in the first place, given his respect for expertise and scholarship, and particularly in light of his familiarity with the chronic social problems faced by computer-mediated communications. To paraphrase Samuel Johnson, for Sanger to participate in the development of Wikipedia was the triumph of hope over experience. Sanger touches on the pernicious problems with social software in his article, but the concepts are elucidated best by Scott and also by Clay Shirky in his 2003 piece, “A Group Is Its Own Worst Enemy,” where in talking about designing high-quality, troll-resistant social software, Shirky insists, “There needs to be some kind of segmentation of capabilities.” Scott, in turn, offers a devastating analogy with the user reviews of the Internet Movie Database, which, as he points out, are “a garbage pile of one-line off-the-cuff nothings.”

As Scott notes, far more significant to IMDB are what are politely called “external reviews,” a euphemism for reviews by recognized film critics, and that points up another problem with Wikipedia, alluded to but not directly discussed by Sanger (or Scott or Shirky): Wikipedia, for such a “democratic” tool, facilitates a one-answer mindset to scholarship. For me (a point made also by Scott), not just the fun but the value of IMDB is not seeing what Ebert or Manohla Dargis said about “Sideways,” but seeing what they both said–a concept completely at odds with Wikipedia’s infamous “Neutral Point of View” (NPOV), which would be a concept strange to our profession. Imagine a library whose collection was developed on the premise that there was one true book on hamster care, divorce, or chicken recipes! Librarians (and IMDB) know that the only way to present a truly neutral point of view is to present all reasonably sane, reasonably reliable points of view.

(Ironically, Shirky deemed Wikipedia healthy at the time of his article, July 2003, due to their “volunteer fire department,” but 18 months is a long time on the Web, and I wonder what he’d say about Wikipedia now; as Scott notes, “the issue is not that damage will occur immediately, but will occur over time.”)

In Wikipedia’s current manifestation, all contributors are on equal footing in terms of editing power and authority (and that’s not even getting into the issue of scope; Wikipedia is built haphazardly, like building a library by buying the first fifty books you find walking into Borders). With no editorial workflow, no significant mechanisms for (and no emphasis on) acknowledging authority, and no way to give expertise its due, Wikipedia functions primarily as high-octane troll food, fueling lengthy “Lord of the Flies”-style shout-downs between, on the one hand, cranks and malcontents, and on the other, the vastly outnumbered experts who care to be bothered to contribute in the first place.

One of the commenters, arguing on Wiki’s unvetted behalf, said that people use Google searches all the time. Indeed we do, and some of us–such as the team at My Place Of Work–often use Google searches to find potential resources which we then evaluate for inclusion. But MPOW is primarily valuable for what we don’t include. Wikipedia entries rank high in Google searches, and we do get asked to include Wikipedia entries in Google. But it merely underscores what Sanger is saying for me to tell you that we don’t add entries from Wikipedia to MPOW because we don’t have the time or resources to constantly monitor the Wikipedia entries or dig through long, heavily trolled discussions to verify the authority of a resource.

Shirky points out that since 1978 the electronic communities have gone through endless cycles with social software. A brave new world is posited; software is written that supports it; people behave as they always do in similar circumstances. Majorities shout down minorities; experts are drowned out by loudmouths; the attempt to develop rules is labeled censorship; the truly interested and dedicated lose steam and interest, and turn elsewhere; what remains is an echo chamber for the bored and the borderlines.

What is the answer to Wikipedia? I wouldn’t bet on adding mechanisms for authority and expertise. “Good luck getting the ball rolling on that,” observed one commenter. Once the trolls run the bridge, game over. My guess is Wikipedia will either implode like most of Usenet or become abandoned for the next starry-eyed project that gives the inmates the run of the asylum.

How about a high-quality encyclopedia built by teams of experts and updated as needed? But it’s not “free,” you might argue. TANSTAAFL (There Ain’t No Such Thing As A Free Lunch)! Wikipedia isn’t free, given the labor required to create and maintain high-quality scholarly content (something I am intimately familiar with). And it has the hidden overhead of massive futz factors.

Scott (who unlike Sanger has the distance from Wikipedia to diagnose the problem) compares Wikipedia to a auto repair shop where “every one else, and I mean everyone else in the garage can work on your car with you.” It would be as if we let anyone edit any content in MPOW, rather than steering content through a highly codified workflow that maximizes the quality of the sites we enter (a workflow that includes extensive post-publication review). While we do have a number of treasured long-time volunteers, most of the work for MPOW comes after the volunteers finish their entries, in a process where our editors ultimately decide what stays and what goes, and bring the records up to consistent stylistic and bibliographic standards. (Addressing the issue of scope, the staff also decide what priority-topic items need to be added to produce the weekly newsletter as we wish to see it.)

Elitist? Arrogant? Anti-whatever? Yeah, probably. When your tagline is “Information You Can Trust,” you try to deliver the goods.

Yet on Wikipedia, every entry is subject to revision, no matter how expert the person who submitted the resource; as Scott says, “any huckleberry can wander along and scrawl crayon on it.” In which case, why bother?

Good encyclopedias already exist. Wikipedia is fixing a problem that isn’t there, and in doing so, with its endemic, unsolvable, inherent problems, it is revealing the naivete of its creators and the predictable characteristics of unmanaged electronic territory.

Posted on this day, other years: