Wikipedia as efficient market
Dick Clark has an interesting view on Wikipedia’s philosophical underpinnings and its efficient allocation of resources over at the Mises Institute. It was prompted by a Reason magazine article on Jimmy Wales, its founder, in which he says: “One can’t understand my ideas about Wikipedia without understanding [libertarian economist F.A.] Hayek.”
Writes Clark:
But how does [Wikipedia’s] polycentric — even anarchic — system, composed of editors acting independently and for their own reasons, result in such an utterly useful resource? The answer goes back to the Hayekian inspiration for the project. Because editors receive both psychological satisfaction and material usefulness from their contributions, the project has grown to include safeguards that help guarantee that the development of the project will move in a positive direction — towards broad, accurate articles that depend on reliable, verifiable sources.
… Wikipedia’s reflection of market dynamics is most easily observed in what many people view as the project’s weakest areas: obscure articles that draw little traffic. In articles about … topics of limited interest, one will often find factual and typographical errors at a much higher rate than in high-traffic articles … . The much higher demand for information about the latter topics means that many more eyes will be combing those much-demanded articles for mistakes.
Since Wikipedia is open to correction by anyone, it stands to reason that the articles attracting more potential editors will be of a higher quality. Rather than a failure, this is a great demonstration of Wikipedia’s efficient allocation of resources. The project, like any other, has a finite amount of productivity to apply to its various activities. It is a positive thing that those articles in greatest demand — those about topics of popular curiosity — would be the ones that are the most complete and reliable.
This explains the usefulness of much of Wikipedia, but doesn’t address the common criticism that Wikipedia cannot be cited as a source of any authority, or even as a source equivalent in authority to, say, Encyclopaedia Britannica. That is to a large extent a red herring, however, since an encyclopaedia is anyway not a citable source in any academic or journalistic work. What is to me a much greater problem with Wikipedia is the fact that it battles to consolidate opposing views on matters of opinion. Its utility declines dramatically, even with popular articles, on subjects such as politics or environmentalism, which are controversial or on which views are strongly polarised. It has a tendency to converge on a populist view on such topics, which isn’t particularly useful.
Wikipedia’s fundamental content policy of Neutral Point of View is, in my opinion, its greatest innovation - far greater than merely letting anyone edit the website. The goal is not to lecture on the Truth (capital T), but to present relevant views on a subject. (Ideally with documentation.) This does lead to battles, and there are obviously going to be a lot of people who don’t understand or don’t care or consider their personal views *obviously* neutral or whatever. But as an ideal, I’ve found it a fabulously useful editorial compass to follow. And I think it mostly works well in practice - an article might not be well-written (our average prose quality is well below our content coverage), but should at least provide the relevant information and views on a given topic.
This is not addressing the endless examples of this failing, of course. But I hope it reassures you that those of us deeply into the project are *painfully* aware of our defects and do work on how to fix them systemically :-)
Fair comment, and thanks for pointing it out. My skepticism comes from having run into some of the examples of this failing. Instead of providing one set of competing arguments, with references, I got presented with three different points of view at three different times. That experience has made me wary when it comes to controversial subjects.
That said, I think the WikiTrust project at the UCSD WikiLabs, which I blogged about here, could go a long way to resolving these problems. In general, I’ve been very impressed with how the Wikipedia project is developing.