Wednesday, September 02, 2009

The Guardian gets it half right

A recent article in the Guardian, reacting to the news that the English Wikipedia is about to hit 3 million articles, and the news that Wikipedia's growth appears to have fallen out of the logistic phase, made the mistake of, for all intents and purposes, listening to Aaron Swartz.  The first part of their piece, covering the statistical findings from PARC regarding Wikipedia's growth dropoff and increasingly hostile environment for casual editors, is all spot on, but the second part, in which they blather on about inclusionism and deletionism being a "significant battle" in Wikipedia, is significantly misinformed.  I suppose that's what they get for talking to Aaron, who has never been a significant player in Wikipedia's community.  I suppose they picked him as their "random Wikipedian" because he's something of a internet celebrity for reasons entirely unrelated to Wikipedia, and because he's got an ego larger than Montana.

The "struggle" between "inclusionists" and "deletionists" pales to irrelevance in comparison to the more serious struggles among all the pitched camps of ideologues in Wikipedia. Both "inclusionist" and "deletionist" are reasonable philosophical attitudes that one can take toward the activity of editing an encyclopedia. However, the ideologues do not fall reliably into either category, because they each individually favor including only that content that furthers their respective personal agendas, and favor deleting content that opposes those agendas.  The friction in Wikipedia's community comes largely from these ideological battles (which can be on nearly any topic, although perennial ones are the Middle East, animal rights, and Northern Ireland), not from any dispute over philosophical attitudes related to encyclopedic worth.  Wikipedia has never developed any meaningful way to resolve content disputes, so these matters usually end up being settled with one side goading the other into breaking enough of Wikipedia's polymorphic conduct rules badly enough to get themselves banned.

Unfortunately, there's no good statistics on how biased Wikipedia content is.  It's not easy to measure bias.  It's not even easy to define bias in any objective way.  At least the inclusionist/deletionist axis is simple to define and one can identify individual Wikipedians on that axis by examining their deletion votes (which, while tedious to do, presents no serious evaluatory challenge).  Given that, I suppose it's excusable that the Guardian fell for it when Aaron told them that this was actually an important division within Wikipedia and that the current factionalism of its community (which is very real) is somehow derived from that.  To be honest, Wikipedia would be a much better place if the inclusion question represented the most serious division within its community.