March 3, 2015


An Illustration of Wikipedia's Vast Human Resources

The Ashley Todd incident has given us a nice illustration of the points I made on Friday about “free-riding” and Wikipedia. As Clay Shirky notes, there’s a quasi-ideological divide within Wikipedia between “deletionists” who want to tightly control the types of topics that are covered on Wikipedia and “inclusionists” who favor a more liberal policy. On Friday, the Wikipedia page on Ashley Todd became the latest front in the battle between them. You can see the argument play out here. For the record, both Shirky and I came down on the inclusionists’ side. The outcome of the debate was that the article was renamed from “Ashley Todd” to “Ashley Todd mugging hoax,” an outcome I was reasonably happy with.

Notice how the Wikipedia process reverses the normal editorial process. If Brittanica were considering an article on Ashley Todd, some Brittanica editor would first perform a cost-benefit analysis to decide whether the article would be interesting enough to readers to justify the the cost of creating the article. If she thought it was, then she would commission someone to write it, and pay the writer for his work. Once the article was written, she would almost always include the article in the encyclopedia, because she had paid good money for it.

In contrast, the Wikipedia process is that some people go ahead and create an article and then there is frequently an argument about whether the article should be kept. The cost of creating the article is so trivial, relative to Wikipedia’s ample resources of human time and attention, that it’s not even mentioned in the debate over whether to keep the article.

To get a sense for the magnitude of this, consider that in less than 24 hours, dozens of Wikipedians generated a combined total of about 5000 words of arguments for and against deleting an article that is itself only about 319 words. The effort people (including me) spent arguing about whether to have the article dwarfed the effort required to create the article in the first place.

Not only does Wikipedia have no difficulties overcoming a “free rider” problem, but the site actually has so many contributors that it can afford to squander vast amounts of human time and attention debating whether to toss out work that has already been done but may not meet the community’s standards.


  1. This is basically gushing over the discovery that Wikipedia is a cult.

    Consider this: “Religion solves the free rider problem. You just tell people a story about some old guy in the sky, and a scary creature who is going to eat them, and – OH MY GOD THEY GIVE YOU MONEY – THAT’S AMAZING!”
    And (channeling Richard Dawkins), in a way, it is pretty amazing, looked at through the lens of simple economic theory.

    But all that means is that simple economic theory doesn’t cover a lot about human behavior.

    This is much less of a revelation than it might appear.

    “One of Wikipedia’s major public relations successes has been in misdirecting observers into a narrative of technological miracles, diverting attention from analysing its old-fashioned cult appeal. While I don’t mean to imply that everyone involved in Wikipedia is wrapped up in delusion, that process is a key factor. A charismatic leader, who peddles a type of spiritual transcendence through selfless service to an ideal, finding a cadre of acolytes willing to devote their lives (without payment) to the organisation’s projects – that’s a story worth telling. But not abetting.”

  2. avatar Anonymous says:

    Most new articles do not produce that level of dispute. Indeed most AFDs are far shorter.

  3. I always struggle with whether to respond in comments or post on my own blog. With regard to your last post on Wikipedia: totally thought-provoking, but I disagree on most of it. I posted a response here:

    With regard to this post, I I think that ‘squander vast amounts of human time and attention’ is not a fair way to think about something like the Ashley Todd incident. The benefits that people get out of debates like that are not strictly (or even mostly) related to the product you now find on the Wikipedia page. The benefits come from exerting and building power and authority, status and reputation, building and reinforcing community. It’s not just a semantic point: squander implies that the effort is wasted, like who cares about a splash of water when you’ve got a whole pool of it. But if you take a slightly wider view on what makes up Wikipedia, it doesn’t seem like that effort is squandered at all.

    • Oh, I completely agree. I was being deliberately provocative with the word “squander” in order to highlight the point that if we’re modeling things as an economic production process, with human labor as the input, then we’d have to say that this is a really inefficient way of doing things. Of course, there are good reasons for the way Wikipedia is organized, and the seemingly wasteful decision-making process serves a variety of serve a variety of useful purposes. My point is simply that “overcoming the free-rider problem” doesn’t rank very high on the list of problems the Wikipedia process needs to solve. Because the amount of work required to actually produce an article is a trivial share of the raw manpower the Wikipedia community can muster.

  4. The effort people (including me) spent arguing about whether to have the article dwarfed the effort required to create the article in the first place.

    Not really–the argument itself was part of the process of creating the article–you can’t draw a distinction between the two.

    I think, on a parallel note, that it is significant that wikipedia subsumes the work of its critics:

  5. Example of a Wikipedia article lifespan:

    - Someone writes an article about a fictional foodstuff in Babylon 5 that amused the B5 USENET group a long time ago
    - It reaches “featured” status without sources
    - After four separate debates it is finally demoted
    - Now I find this article and notice it is not an encyclopedic summary of academic literature but a synthesis of several USENET posts, so I try to delete it
    - The meta-article battle continues…

  6. Another way of saying what Seth says is that your perception of a potential free-rider problem is the result of a classification problem. You are thinking of Wikipedia as an online publication rather than as an online discussion group some of whose products the general public wants to read.

    (As an aside, one of the things keeping the community going may be the decision not to aggressively monetize the product — the history of the net is littered with projects that took publicly-generated content and turned them around for private profit, with varying degrees of financial success and fairly uniform degrees of disapprobation and withering of unpaid contributions.)