By Edward Buckner
Despite the acknowledged shortcomings of Wikipedia’s governance, the strongest argument against reforming it is its apparent extraordinary success. Editing Wikipedia is like making sausages, they say: it’s a nasty process that you really don’t want to see . As long as the end product is nice tasty sausages, does it really matter how it ended up on our plate?
Even when you do find mistakes, Wikipedia tells us that it doesn’t matter in the long run. Anyone can edit Wikipedia, and so there is a quasi-Darwinian process which ensures that only the fittest edits – i.e. the good ones – will survive, and the final result will be perfect. Wikipedia doesn’t need experts, or an editorial board, any more than natural selection needs a design committee. As Kevin Kelly of Wired once said, the crowdsourced wiki is like an invisible hand which emerges from its ‘very dumb members’. Why then would it need a visible hand? 
More sausagemakers, better sausages?
The idea that more is better is a fundamental principle of Wikipedia, and is the reason why such a large chunk of the Wikimedia Foundation’s annual budget – probably about $10m of it  – is dedicated to combating the decline in the editorial base. The idea underpins the unofficial policy of purging the ‘old guard’ of established editors, on the grounds that their impatience or uncivil behaviour towards well-meaning but incompetent editors is putting off newcomers. A respected member of the Wikipedia military history project, and another long-standing editor of many of Wikipedia’s Featured Articles have recently been driven off as a result of the policy. The survival of Wikipedia, they say, depends not on keeping a small number of skilled individuals, but on attracting large numbers of unskilled ones.
Is it true? It’s hard to test. Yet there is much evidence that this theory is wrong, and that the quality of Wikipedia pages, such as it is, is mostly down to a handful of editors who look after them. I know this from experience. I am a specialist on medieval philosophy and I follow all the articles about it, although I no longer bother to improve them or save them from the gradual deterioration caused by vandals or incompetent contributors. All the evidence suggests that when competent editors or specialists do not look after them, they go off. There is no ‘invisible hand’, and the policy of encouraging more contributors to Wikipedia, without any consideration of the quality of what they contribute, is likely to accelerate the process of deterioration.
For example, I keep a close eye on the article about the medieval philosopher Duns Scotus. I know a little about Scotus – I am publishing a book about him later this year. The Wikipedia article about him is always in a terrible state, simply because there are no specialists on Scotus who are willing to maintain it. They know it will soon deteriorate, and they will have to watch it like a hawk to stop the accumulation of errors.
Despite myself, I tried to clean it up in August 2012. A vandal had changed his birth date to c. 1250 – an apparently simple error to spot, but Wikipedia’s vandal patrollers focus on obscenities and other forms of crude vandalism, and it was missed. The claim that “Duns Scotus is usually considered the beginning of the formal Scottish tradition of philosophy”, was nonsense. The old myth about Scotus being buried alive had been uncritically recycled. A whole paragraph from the old Catholic Encyclopedia had been incorporated wholesale, even though it was obsolete and incorrect. There was much that was irrelevant, or incorrect, and there was an awful lot that was missing. We don’t know much about the lives of medieval philosophers, yet we have their work, in compendious volumes. The article did not, and still does not, explain Scotus’s work in a way that is accessible and interesting to the average reader, and which shows his importance to later generations of thinkers.
Almost immediately after the cleanup the article began to degrade. Someone replaced the obsolete passage from the Catholic Encyclopedia, without removing my own corrections, so that the article now contradicts itself. At the beginning of March 2013, an obvious hoax was added, still not removed by the end of March.
In 2011, John Duns Scotus was honored by the University of Oxford, his alma mater, as one of its 100 most distinguished members from 10 centuries. Other honorees included Duns Scotus, William of Ockham, Erasmus of Rotterdam, Saint Thomas More, John Locke, Adam Smith, Christopher Wren, Lawrence of Arabia, Oscar Wilde, J.R.R. Tolkien and living University members Rupert Murdoch, Bill Clinton, Stephen Hawking and Florian Henckel von Donnersmarck. For the cover of the 2011 Prospectus, Oxford University named 100 streets in Oxford historical centre after these graduates. Westminster Way was renamed for John Duns Scotus.
The pages linked by the article are also in a poor state. I added in August that “According to the fifteenth century writer William Vorilong, his departure was sudden and unexpected”. This prompted another editor to create an article about Vorilong. It is precisely two sentences long, and says that “William Vorilong (died 1643) was a philosopher”, which is certainly true, but hardly worth a whole article, and that he was ‘a contemporary of Duns Scotus’, which is obviously false, given that Scotus died in 1308, and Vorilong – a biographer of Scotus, rather than a contemporary – died in 1464 (not 1643). The second sentence says that he ‘contributed to influence his views about the square of opposition’. This is blatantly false, given that they lived over a hundred years apart, and in any case Scotus, as far as I know, never wrote anything about the ‘square of opposition’ although, as I discuss in detail in my book, he did have a lot to say about contradictory statements and propositions about non-existent objects, but that’s a different matter.
Professionally made sausages
To appreciate the difference between the traditional approach to writing encyclopedias, and the Wikipedia ‘crowdsourcing’ approach, you only have to compare any article in the Stanford Encyclopedia of Philosophy, or in the Internet Encyclopedia of Philosophy, with the corresponding version in Wikipedia, assuming there even is one. On the one hand, you are likely to find a well-written well-researched article by a writer who has a command of their subject. These are produced using traditional method of using credentialed scholars and experts, the kind of method that Wikipedians blame Nupedia for, and it shows. On the other, you are likely to find a jumble of incoherent nonsense. There is no evidence at all that crowdsourcing, i.e. collecting a bunch of non-specialist editors together to write about specialist subjects, works in the way that Wikipedia says it should. Perhaps it does not matter. As one Wikipedian noted last month “When it comes to improving the articles, I would prefer to have 58 clerical errors fixed in mega-article Jennifer Lopez rather than revise article John Duns Scotus to better describe his major works.”
 However, see this article about how sausages are really made.
 Kevin Kelly, Out of Control, chapter 2.
 Based on analysis of the 2012-13 budget, including $4.2m on software to improve ‘editor experience’, and $5m on projects to increase international participation.