Critiquing Crowdsourcing
Earlier today a college student named Tabish Bhimani asked whether I had "restricted [my] opinion on crowdsourcing to its benefits." Daren C. Brabham, a University of Utah graduate student and occasional Crowdsourcing.com contributor, wrote an interesting comment on the subject and while I started to reply in the comment thread in which Tabish asked the question, I realized the thread might hold interest for a wider audience. So, without abbreviation (or editing!), I upgraded the thread to a post. (Ah, the pleasures of administrator access!) Here's Daren's comment, followed by my response:
Daren:
@Tabish: Plenty of criticisms out there on crowdsourcing, particularly in its potential to exploit creative labor for little reward. Also some criticism along digital divide lines, in that we can't assume the crowd has a diversity of opinions unless we've measured them and find the crowd is more diverse than internet access data indicate. Otherwise we have to change our theories. Shamless plug: I have an article coming out in the journal Convergence in February which addresses some of these potential criticisms. Email me for a pre-publication copy: daren.brabham utah.edu
@everyone: Another shameless plug--MIT media scholar Henry Jenkins mentioned crowdsourcing in his recent keynote lecture here at the University of Utah on participatory culture. Crowdsourcing mention is at 36:50 in this podcast: http://www2.utah.edu/podcast/category.php?id=6 Jenkins also did an encore of this lecture a few weeks later at the Association of Internet Researchers conference in Vancouver, with the same mention of crowdsourcing.
Jeff:
@Tabish: Funny, I was just discussing this with a friend yesterday. I do think this blog has been a little thin on robust critiques of of crowdsourcing. I wouldn't use the word "restricted," however, as that implies a conscious attempt to ignore any emerging critique. The comment I made to my friend, in essence, was that so much of my labor these past 18 months has revolved around simply gathering and analyzing examples of crowdsourcing in action that it's been difficult to find the time to develop critiques on my own. And I'd disagree with Daren that there's a lot of literature out there, which is exactly what makes his own efforts so valuable. I imagine in the next year or two a lot of critical material will emerge, but so far what we've seen is a lot of Gee Whiz coverage in the mainstream press that essentially just recaps the cases I examined in my original article.
It's important to remember that the migration of peer-production, or open source, models of economic production has just begun. In fact, as I enter the final stages of writing my book, the most prominent critique that comes to mind is that while we can say without any doubt that crowdsourcing thrives in the wild, it's also evident that it's hard to recreate artificially. By that I mean that our purest examples, so to speak--threadless.com, istockphoto.com, topcoder.com--emerged organically from existing communities.
There are exceptions to this: Dell did a wonderful job of soliciting product development ideas and (crucially) acting on the suggestions (It's going to release a Linux laptop in response to consumer demand), but there are also abundant examples of what I would call "limited successes," such as Current TV, which has been disappointed by the number of viewers submitting content to the network, and our own Assignment Zero. While we produced some wonderful work, and we certainly were able to recruit a number of contributors, we underestimated the massive administrative labor that would be required to properly leverage the initial enthusiasm we received. (And those weren't the only mistakes we made! Here's a longer list in a piece I wrote for Wired.com, which counts as a critique of crowdsourcing I guess.)
@Daren, et. al.: I remain, well, let's call it "cautiously skeptical" about the exploitation critique. I dealt in a bit greater depth on this issue in a June blog post, but since writing that I've done a great deal of reporting and an even greater deal of thinking and, if anything, my initial position has hardened: Crowdsourcing is enabled by communities, and communities are held together through shared passion. I just can't square that with any concept of exploitation, per se. iStockers are stoked to get paid anything for their work; ditto Current contributors; ditto Threadless designers.
That said, I still emphatically agree that it's a healthy addition to the discourse around these rapidly developing participatory models. One can envision a day when the cost of various forms of content drops so low that it becomes a buyers market. In photography, that's already the case. But so far it just doesn't resemble anything I think we can call exploitation. Are people like Mark Zuckerberg getting rich off user-generated content? You bet. I personally don't have a problem with that. I really enjoy the five or ten minutes a week I spend futzing with my Facebook page. Zuckerberg created a neat service, and in response people used it. Now he reaps the benefits of his invention. That's the way the market works. But then, I haven't read Daren's paper yet, and he may well have uncovered some cases of which I'm still ignorant.
Regarding diversity, I really couldn't agree more. It's important to draw a distinction here, though. What I would call "group intelligence applications" are really a distinct genre of crowdsourcing. Should we be concerned that Threadless draws a fairly homogenous demographic to its community (I'm guessing here, but I bet it's young, fairly affluent and skews toward whites and Asians). I would say not. But does that same homogeneity reduce the usefulness of Digg? I'd say yes, or at least to someone like me, whose interests include stories on the visual arts and American history. And when it comes to strict group intelligence applications like prediction markets, the lack of diversity is a downright monkeywrench in the works. (A few of you have asked me for a reading list. Here's one recommendation at least: Scott Page does a superb job of examining how diversity fuels models like prediction markets in his book, "The Difference: How the Power of Diversity Creates Better Groups, Firms, Schools, and Societies.") If everyone participating in a prediction market has had similar experiences and training, it won't provide any advantage over just asking one of the participants to predict an outcome.
At any rate, there's much more to say on the subject. Here's hoping all of you will say some of it here.
Recent Comments