Can the Open-Source Model Work for Content?
October 2004

Now that Linux has proven the effectiveness of open, collaborative authoring of software, such open-source models could be adapted for building content. One example of such a model of open collaboration is Wikipedia, an online encyclopedia written by thousands of volunteer contributors. ("Wiki," the Hawaiian word for "quick," is a class of software that allows people to create and edit web pages collaboratively.) Started in January 2001, Wikipedia has over 300,000 entries written by 9,000 unpaid contributors who also function as editors. By comparison, Encyclopedia Britannica has 75,000 articles created by 20 in-house editors working with 2,500 outside authors. While the comparison may be unfair, it’s still intriguing. Whereas conventional encyclopedias have a structured authoring and review process, Wikipedia has no formal editorial process; anyone can write anything and can edit any existing entry, although all previous versions are maintained in an archive so changes can be undone if necessary. But being open means Wikipedia is vulnerable to vandals who can add biased information or make comments about socially or politically charged subjects such as abortion. Wikipedia’s volunteers police the site and make repairs as necessary.

Using a community of content contributors is not new to the internet. Communities have long shared postings on innumerable subjects via bulletin boards, listservs, and news groups. But while most such community postings are ephemeral in nature, Wikipedia’s entries, like those in a conventional encyclopedia, are intended to be durable. Moreover, since entries can be edited repeatedly by individual contributors, Wikipedia offers flexibility and currency that conventional encyclopedias can’t hope to match.

The biggest concern with open-source content is its authority and reliability. Unlike traditional published works, Wikipedia has no editorial review process. One way to make open-source content more credible is to institute levels of editorial refinement and certification. Once an entry has undergone a certain amount of review and modification, it could be frozen by an editorial board and designated as "stable." Contributors could continue to modify the original entry, but it would be designated as work in progress. Freezing entries once they reach a level of editorial approval could also reduce vandalism.

It remains to be seen whether traditional publishers will incorporate any elements of the open-source approach into their products. In a curious way, peer-reviewed books and journals share some characteristics of open-source, except that the community of traditional reviewers is carefully selected rather than self-appointed, and the reviewers have the right to comment on, but not rewrite, an author’s work.

Another question is whether open-source databases like Wikipedia are needed, given the vast amount of information already available from contributors on the internet at large. The weakness of much internet content--its unreliability–could be addressed by open-source publications like Wikipedia. If Wikipedia manages to find a way to meld the power of open collaboration with the discipline of traditional review, it might be on to something big.