Wiki's wild world - Researchers should read Wikipedia cautiously and amend it enthusiastically.
A recent article from Nature shows that the number of errors found in Wikipedia science articles is not significantly higher than the number of errors found in articles in Encyclopedia Brittanica. An editorial that went along with it encouraged scientists to get more involved in the creation and review of Wikipedia entries. What does the comparative accuracy of Wikipedia vs. EB tell us? Perhaps it doesn't tell us so much about Wikipedia, but about the encyclopedias that we hold up as the "gold standard". Not to be flip, but the postmodernist in me says that - just because something is written in a book does not make it the truth. I know it sounds heretical coming from a librarian, but researchers should always question - regardless of how "authoritative" a resource may seem. Go to iCONN databases to double-check something you've found on the web. Verification and confirmation are among the researcher's best friends (along with a willingness to question the "facts").
Collaborative work can be a key to high reliability (hence, the importance of peer-review to scholarly publications), strong arguments, and powerful ideas. But we should not downplay the reality that an open forum without a mechanism for accountability built into it does not lend itself to the provision of authoritative information. Certainly, if the scientists and experts in academic fields would heed the message that Nature is sending and take the time to get involved with Wikipedia's efforts, it would be a positive for everyone.