There has been a story going around comparing Wikipedia, an extraordinary online encyclopedia that follows the open-source methodology in the sense that anyone can write an entry or correct/improve upon previous entries, to the encyclopedia britannica. I must confess I’m a huge fan of Wikipedia, I remember when it started I thought it had tremendous potential, but it has grown by leaps and bounds in a very short time. I can’t imagine what Wikipedia will look like in a decade. It’s probably the closest thing we have to a sum total of the basics of all human knowledge in one place, or it will be at some point in the future.
Anyway, the
first story that came out about this comparison from Nature magazine claimed that Wikipedia’s error rate was impressive, because it was only slightly greater than Encyclopedia Brittanica’s: in the 40-something entries they compared, Wikipedia averaged 4 errors per entry to Brittanica’s 3 per entry.
However, others have looked a bit more closely at the entries from each source, and found that on average, the Wikipedia articles are nearly 3 times longer (6.8KB per article, vs 2.6KB per article for Brittanica), with more information in them, so it would make sense that the error-per-article rate is larger in Wikipedia. But if you break it down to error-per-2KB of text, Wikipedia comes out much better, 1.3 errors per 2KB of text, vs Brittanica’s 3.6 errors per 2KB of text.
Anyway, it’s an interesting study that in my view validates Wikipedia as a useful tool for learning the basics on just about any subject. Though, any time vast quantities of information are summarized in encyclopedia format, there are bound to be errors. It’s just that the open-source methodology used by Wikipedia provides a means to correct these errors, which to me speaks of the inherent superiority of open, free information.