Jaron Lanier, in a prescient essay from 2006, lays out many of the problems with “online collectivism”, using Wikipedia as his central example. In the process, he attempts to work out when collective efforts are better than individual efforts, and when individuals are likely to surpass the collective.
He successfully predicts that one of the problems with collective efforts such as Wikipedia is that they’re “noisy”, prone to lots of random fluctuation that lead to feedback loops. He says that one of the advantages of representative democracy is that it acts as a “low-pass filter” on the noise of the collective, and predicts that Wikipedia would implement mechanisms to do the same thing for its most controversial articles. Indeed, we have seen just that, with the increasing levels of protection and filtering enjoyed by Wikipedia articles on controversial topics.