Search This Blog

Thursday, 24 November 2011

Wikipedia's Hive Mind

Since I started writing this blog, I've been doing a lot more internet research than I used to. And, almost inevitably, I've been using Wikipedia a lot more than I used to -- if only because it features so prominently in Google search results. For most mainstream subjects Wikipedia provides a reliable and objective source of information, but lurking under the surface there is something sinister. Wikipedia, by far the largest collaborative undertaking the world has ever seen, has developed a distinctive personality of its own... and it's not a pleasant personality. It comes across not so much in the visible content of the encyclopedia, but in what is omitted (or more accurately, deleted) from it, and the stridently authoritarian justifications to be found on the "Discussion" pages.

I've talked before about Wikipedia versus Forteana and the Shakespeare Authorship Question, but these are not isolated occurrences. The deletionists have targeted other heterodox ideas ranging from conspiracy theories to complementary medicine and climate change scepticism. If the material being deleted was simply the raving of an irrational crackpot, the situation would be understandable. But as often as not the deleted material was soberly written, well-balanced and carefully documented. The Wikipedia "hive mind" is systematically wiping out views it doesn't agree with... even if those views come from a Nobel Prize winner. The plasma physicist Hannes Alfvén is a case in point. Despite winning the Nobel Prize in 1970, his views on the subject of magnetic reconnection differed drastically from those of the scientific mainstream. In March this year, Michael Suede tried to insert a short, carefully researched section on Alfvén's views towards the end of the Wikipedia article on the subject. It was promptly tagged with a warning that "An editor has expressed a concern that this section lends undue weight to certain ideas, incidents, controversies or matters relative to the article subject as a whole". Shortly after, the whole section was deleted and Suede was banned from making further changes to Wikipedia.

So where does the Wikipedian Hive Mind come from? How can a collaboration between a potentially vast number of individuals result in such a clear-cut personality, with clear-cut interests, beliefs, opinions and prejudices? To answer these questions, we need to look at just what Wikipedia is, and how it functions.

Wikipedia was launched in 2001 as "the free encyclopedia that anyone can edit". In practice, of course, "anyone" means "anyone with a computer and access to the internet". And there are a few other selection effects which may be less obvious. Wikipedia is written using an HTML-like mark-up language, which while it's relatively straightforward may deter the less technically inclined. Wikipedia doesn't pay, so people who are motivated by financial gain are unlikely to contribute... while those who have full-time jobs, and families to look after, may simply not have time to become involved. And no-one looking for personal fame is going to waste their time on Wikipedia: the use of anonymizing handles guarantees zero recognition in the world at large.

A recent study by the United Nations University in Maastricht analyzed the demographics of Wikipedia contributors. Most contributors are male: 87 per cent compared to just 13 per cent female. The average age of contributors is 26 years. Amongst Wikipedians as a whole, roughly a quarter are under 18 years of age, half are under 22, and three-quarters are under 30. Two-thirds of those surveyed were unmarried, and only 14 per cent had any children.

These statistics paint a clear picture of Wikipedians as young males with no family responsibilities. This explains some of the well-known characteristics of Wikipedia, such as its bias towards left-wing politics and its bias towards contemporary popular culture... but does it explain the bias against non-mainstream ideas and opinions?

An article a few year ago on the Times Higher Education website suggested that "what matters on Wikipedia is not your sources but the support of the community... the Wikipedia community that is, within which there is much talk about consensus, civility and reliable sources. Yet on closer examination, Wikipedians seem an unappealing bunch -- computer fanatics, generally male, usually teenagers. They see the world only from a youthful cab driver's perspective. If anyone disagrees with the Wikipedian consensus, their edits are reverted and they can be banned indefinitely."

What has happened is a classic example of positive and negative feedback loops. Individuals whose interests and world-views match those of the Wikipedian Hive Mind find their edits retained, and their opinions applauded on Discussion pages. So they contribute more and more -- that's positive feedback. Dissenters are subjected to negative feedback: their contributions are constantly altered or deleted, and they find themselves on the receiving end of sarcastic, often viciously ad-hominem, attacks on the Discussion pages. So they go away, and switch to doing something more rewarding somewhere else. As a result, Wikipedia conforms to the "one per cent rule": half of all edits are done by less than one per cent of users, and over two-thirds of articles have been written by less than two per cent of users.

Over the years, Wikipedia's collective personality has become more and more one-dimensional, reflecting the personality of the majority of its successful contributors. You might be forgiven for imagining that, with its emphasis on factuality and neutrality, this personality of Wikipedia's would be calm and rational and supremely objective. But not a bit of it. As Nicholson Baker wrote in the Guardian a few years ago: "There are some people on Wikipedia now who are just bullies, who take pleasure in wrecking and mocking people's work... They poke articles full of warnings and citation-needed notes and deletion prods till the topics go away". He quotes one of the victims of deletionism as saying "I got the impression that they enjoyed this kind of thing as a kid enjoys kicking down others' sandcastles", while another victim "likened the organized deleters to book burners".

Wikipedia's editors make a big thing about a "neutral point of view", but really no-one over a certain age demands or expects everything they read to be written in a perfectly neutral way. Grown-ups just want the basic information, from whatever point of view, so they can form their own opinions. Instead, with Wikipedia's practice of deleting all material that is deemed to be "not notable", we've ended up with something much more insidious than a biased point-of-view... an encyclopedia that is biased by what is included and what is excluded.

Wikipedia dimly recognizes that the inclusion/exclusion problem exists... or at least, it did at one time. There used to be a Help page providing specific guidance on how to avoid what it called "Information Suppression". However, the page has not been updated since 9 January 2007, and it's now prefixed with a statement that "This page is currently inactive and is retained for historical reference. Either the page is no longer relevant or consensus on its purpose has become unclear."  In other words, the information has been suppressed!


Anonymous said...

Excellent article Andrew. It's disheartening the number of well done articles that have been deleted. And of those that remain they're usually watered down to the point where they're worthless. The only exception to this that I think I've noticed is with ufo related articles. The ufo proponents seem to be every bit as dedicated (fanatical?) as the pseudo-skeptics

Andrew May said...

Thanks for the comments. I can't say I'd noticed that Wikipedia is especially tolerant of ufology, but perhaps it's quite a recent breakthrough. It will be interesting to see how long it's allowed to last!

Anonymous said...

This subject is something that has bothered me about Wikipedia. 90% of the time if my search brings Wiki up I look at as much of the other information as I can BEFORE reading the Wiki article. It just seems to me that the whole premise of Wikipedia is flawed in that there is no concrete way to be sure that the information entered by users is accurate or that the editors know what they are talking about. The only way for something like that to be reliable is if the editor already knows everything there is to know thereby eliminating the need for contributors. Sadly, it appears that the editors of Wikipedia believe that they know it all and I'm pretty sure a human would have to be alive for thousands of years to accumulate that much knowledge. I'm quite certain that no human that old exists today and most likely never has.