The second place where this line of attack fails is more obvious: it affects to assume that there exist comparators that are 100 per cent reliable. A (much-debated) study in Nature compared the accuracy of several Wikipedia articles with their opposite numbers in Britannica, with favourable results. Books make and propagate mistakes, too. Errors in books remain in error. Errors in Wikipedia need not.

The key thing is form, not content. Wikipedia’s brilliance lies in giving its community the tools to identify and correct error. Neutrality, as I’ve mentioned above, is one among the handful of core content policies. Verifiability — that the information has already been published in a reliable external source — is another. If you simply copy facts from the front page of a Wikipedia entry, you deserve what you get. Check the footnotes; use the links.

If mistakes in books stay wrong and mistakes in Wikipedia don’t, the converse, of course, is also true. If you get something right in a book, it stays right. That’s not always the case with Wikipedia. Policing the encyclopedia, ensuring its stability, is an ongoing task: a Forth Bridge paint- job in the commons of knowledge. With its now vast size and influence, difficult questions arise about the practical structures of authority. Not long ago, for instance, unregistered users were prevented from creating new articles.

Wikipedia — high in the top ten websites in the world — is now entering a third phase of its life. Most of the big subjects are done. The piranha-rush of new recruits is slowing a little. Will enough people stick around to do the janitorial work? Or will the enemies of good faith — like the snivelling Tory creep who vandalised the entry on Titian earlier this year in the hopes of sparing David Cameron embarrassment — overrun it? Lih concludes with a warning: ‘The greatest enemy of a revolution is its success.’

Blackwell Bookshop

Purchase your copy here, 10% off RRP