A new study analyzing how social norms are perpetuated within the Wikipedia community says they look a lot like a corporate bureaucracy, according to research journal Future Internet. The results bring into question the supposedly democratic and egalitarian functionality of an online encyclopedia that relies on crowdsourcing to operate.

Researchers analyzed 15 years' worth of editing and community management articles since the creation of the website, including 587 million editing events, 5 million articles and millions of "talk" pages, a sort of running conversation about improving individual articles. They looked for how social norms and ideas have shifted as Wikipedia has grown. The articles singled out as setting social norms for the website were instructional articles on how to moderate the website, including administrative information such as how to title or cite articles, along with big, guiding content principles, such as neutrality, verifiability and no propaganda.

Researchers found that these articles, which were mostly created very early in the website's life, have remained highly influential and more or less the same as the website has grown. These "governing" rules were mostly created in 2004, back when the website's population was less than 3 percent of its highest population peak in 2007 (the population has steadily declined since then). A small group of users developed most of these guiding principles, even as Wikipedia has gradually decentralized into smaller spheres of governance, where article editing is grouped by topic, such as "WikiProject Medicine," and run by its own editor.

The persistence of these overreaching norms centralize influence. Just as in big, bureaucratic companies, power structures are hierarchical and tend to collect at the top.  A study in Physical Review E from earlier this year showed that a small number of editors, "super editors," were largely influential on the website and that the influence of larger groups of participants tended to be very small.

Moreover, breaking in to the circle of "super editors" proved very rare for a newcomer. Lead author of this study, Jinhyuk Yun, told Physics Focus that the monopolization of content by a few editors was a problem, and that the site needed to work toward fostering a more democratic process.

"Why are there no normative revolutions on Wikipedia?" asks the Future Internet study, pointing to a kind of Matthew effect, a sociological term meaning an accumulation of economic capital that leads to social stratification, occurring within the editing community at Wikipedia. For example, the study points to the article "Civil POV Pushing," which argues that some articles that adhere closely to a "civil" tone, which is one defined norm, can get away with extremely biased and non-neutral articles. In this case, an article questioning the value of a specific, long-lasting norm, "civility," gained little momentum and failed to rise to the level of a much-read or referenced article.

This phenomenon has to do with the reinforced core social norms created by a small group of users who were there at the beginning, the study states, and all that come after them are at a disadvantage if they don't adhere. "Users who disagree with these norms may find that reinterpretation, rather than replacement, is a more effective response given the disproportionate allocation of attention to early pages," the study states.

What does that mean for this great experiment in online, community-driven information? "The evolution of this network is a remarkably conservative process," the study concludes. "Wikipedia is a paradigmatic example of a 21st Century knowledge commons. Yet, its core norms play a structural role analogous to the institutional myths of rationalized 20th Century bureaucracies."