Back to the archive

The encyclopedia · Product & Design · Product decision · 2000–2003

Nupedia's expert review produced 25 articles — the wiki side project became Wikipedia

Jimmy Wales built Nupedia, a peer-reviewed encyclopedia. It produced 25 articles in 3 years. The wiki feeder, Wikipedia, produced 18,000 in its first year

Bomis · Wikimedia Foundation · 2000-03-09

What happened

In March 2000, Jimmy Wales launched Nupedia, a free online encyclopedia intended to produce articles of professional quality. The project was funded by Bomis, a web-advertising company Wales owned with Tim Shell and Michael Davis. Larry Sanger was hired as editor-in-chief. Nupedia's articles were written by volunteer experts and subjected to a rigorous multi-step peer review process. The process was designed to ensure encyclopedic quality — but it was catastrophically slow. In its first year, Nupedia published only 12 articles. After three years, the total was 25.

On January 2, 2001, computer programmer Ben Kovitz explained the concept of a wiki to Sanger over dinner. Sanger proposed that a wiki be set up as a 'feeder' project — a place to draft articles before they entered Nupedia's formal review. Wales set up the first wiki on January 10, 2001 under the nupedia.com domain. On January 15, 2001, it moved to a new domain: wikipedia.com. The wiki was intended as a side project to support Nupedia. It never fed Nupedia a single article.

Wikipedia grew at a rate that Nupedia could not match. In its first month, Wikipedia had 200 articles. In its first year, it had 18,000. By the time Nupedia shut down in 2003, Wikipedia had hundreds of thousands of articles in dozens of languages. Nupedia was not just outpaced — it was made irrelevant. The feeder had eaten the main product. In March 2002, Bomis withdrew funding during the dot-com bust. Sanger left both projects. Nupedia's server crashed in 2003 and was never restored. The encyclopedia that produced 25 articles had spawned the largest in history.

Why it happened

  • Nupedia was designed by experts for experts. The multi-step peer review was correct on every article — but so slow the project was stillborn. 12 articles in a year proved the model could not scale.
  • The wiki was a feeder project — a side experiment to support the main effort. Never intended to be the product. The open, fast wiki model was what the world needed. The side project overtook the main.
  • Nupedia was not abandoned — it was outpaced by something that worked better. Wikipedia was faster, more inclusive, and more productive. The experts designed a perfect process that produced nothing.
What it costBomis' three-year funding, a salary, and 25 articleslucky mistake

The lesson

The best process wins by producing more output, not higher quality per unit. Nupedia's expert review was correct on every article — it produced 25. Wikipedia was messier and produced 18,000.

Sources

spotted an error? The club wants to know.

Comments · 0

    Sign in to join the comments.

    More like this

    Somewhere, someone solved the problem this company failed at. 2nd Opinion →