I believe the most important component of Wikipedia is not its commitment to collaborative and ongoing creation, commendable though that is, but its attitude towards its own history -- specifically, the history of each Wikipedia page. On the Wikipedia model, history is a necessary component of the presentation of information, and that history is subject to endless investigation and assessment.

bug_wikipedia.jpgConsider this one small example: go to the homepage of The W. H. Auden Society -- maintained by that exemplary scholar and literary executor Edward Mendelson -- and you'll find this interesting note:

A highly accurate, thoroughly revised version of the Wikipedia.org entry on Auden is now available. This site strongly recommends that online researchers make reference to the archived version of the page, in the link above, rather than to current versions, which may be less accurate or may be subject to vandalism.

Perhaps no one can step into the same river twice, but anyone who knows how to look behind the front page of a Wikipedia entry can step into its history -- can step into any moment in that history, and can do so as often as he or she wants. That's pretty astounding.

No one understands the true genius of Wikipedia if they look only at the current version of any given page. James Bridle makes this clear when he explains why he decided to record the Wikipedia debate about how to recount the story of the Iraq war: "Wikipedia is a useful subset of the entire Internet, and as such a subset of all human culture. It's not only a resource for collating all human knowledge, but a framework for understanding how that knowledge came to be and to be understood; what was allowed to stand and what was not; what we agree on, and what we cannot." (The books are cool, but they are frozen in time: who knows how Wikipedia's account of the war may change in the coming years or even decades?)

This extraordinary temporal inclusiveness does not just dramatically increase the usefulness of Wikipedia as a "resource for collating" and a "framework for understanding;" it also means that Wikipedia contains, within itself, the tools for its own evaluation. We don't just have to assert that the whole project is on an endless upward trajectory, or, conversely, that it's always regressing towards the mediocre mean: by studying the history, we can go a hell of a long way towards finding out.

But if, as I contend, its self-recording function is the coolest and most important element of Wikipedia, it is also one of the more vulnerable elements. Much of the project's budget goes towards keeping that history alive: vast as Wikipedia is, it would be only fractionally as expensive to maintain if its history was allowed to dissolve into the ether. Which is why, if Wikipedia ever does lurch towards collapse, we had better be ready to show up with the money necessary to keep it -- all of it -- alive.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.