Organisation pour le Développement de l’Éducation, des Ressources et des Savoirs AccessiblesNonprofit association · Paris

← All articles

Published

How does an encyclopedia written by anyone stay standing?

Anyone can edit it, yet it holds together. Here is the actual machinery behind a volunteer-written encyclopedia, and how to read one of its pages like an insider.

A teenager cites an online encyclopedia for a homework assignment and gets told, again, that it ’doesn’t count’ because ’anyone can just write anything on it’. The line has been repeated for two decades, and it captures something true: anyone really can open an article and change a sentence. What it misses is everything that happens next, the part that turns a page anyone can touch into one that mostly holds up.

Anyone can edit, but not unnoticed

Every single change to every single page is kept, forever, on a public ’history’ tab: who edited what, when, and exactly which words moved. That record is not decoration. Volunteers watch live streams of recent edits across the site, and according to the Wikimedia Foundation, obvious vandalism on Wikipedia is typically reverted within minutes, sometimes seconds, spotted by watchful editors and automated filters working side by side. A page with a lot of eyes on it heals fast. A page nobody watches can sit wrong for longer, which is exactly why the size of the volunteer community, not the openness of the edit button, is what actually keeps the whole thing standing.

Three rules under the hood

Behind the editing screen sits a set of core policies that every article is supposed to follow, and the central one is called verifiability. As Wikipedia’s own policy page puts it, people must be able to check that facts and claims in an article ’are not just made up’: readers need to be able to trace a statement back to a published source, not take an editor’s word for it. Alongside verifiability sit two companion rules: no original research, meaning an article reports what has already been published rather than an editor’s own theory, and a neutral point of view, meaning a contested topic is described from more than one side rather than argued for. None of the three rules guarantees a perfect article. Together, they mean a disagreement gets settled by pointing at sources rather than by whoever shouts loudest.

That settling happens in the open, too. Every article carries a ’talk’ page where editors argue about which source is reliable, whether a sentence oversteps what its source actually says, and how to phrase something contested fairly. Reading that page is like watching the editorial meeting that produced the article you just read.

What the actual studies found

The honest question is not whether mistakes exist (they do, on any reference work made by humans) but how the error rate compares to what came before. In 2005, the journal Nature sent 42 matched science articles from Wikipedia and from Encyclopaedia Britannica to outside experts for blind review. The tally came back close: about 3.9 errors per article on average for Wikipedia against about 2.9 for Britannica, a real gap but a narrow one for a work built entirely by volunteers competing against a professionally edited encyclopedia with a century of institutional backing. Britannica publicly disputed the study’s methodology afterward, and that dispute is itself worth knowing about: even a favorable-looking result got, and deserved, scrutiny rather than a free pass.

The fix for one wrong sentence is not a single expert. It is the next volunteer who happens to read that sentence and knows better.

Reading a page like someone who knows how it works

That short checklist turns five minutes of scrolling into a real judgment about how much to trust what you’re reading, on this kind of page or any other. It is the same instinct that makes a good research habit in general: not ’is this the kind of site people trust’, but ’can I actually see where this specific sentence came from’.

Contributing is smaller than it sounds, too. Fixing a typo, adding a missing citation to a fact you happen to know the source for, flagging a sentence on a talk page instead of arguing in the article itself: every one of those small acts is what keeps the machinery running for the next reader who shows up needing an answer fast.

Sources

Keyboard shortcuts

TabMove from link to link
EnterOpen the link, or expand and collapse a question
?Open this help
EscClose the panel or this help