Category Archives: roy_rosenweig

rosenzweig on wikipedia

Roy Rosenzweig, a history professor at George Mason University and colleague of the institute, recently published a very good article on Wikipedia from the perspective of a historian. “Can History be Open Source? Wikipedia and the Future of the Past” as a historian’s analysis complements the discussion from the important but different lens of journalists and scientists. Therefore, Rosenzweig focuses on, not just factual accuracy, but also the quality of prose and the historical context of entry subjects. He begins with in depth overview of how Wikipedia was created by Jimmy Wales and Larry Sanger and describes their previous attempts to create a free online encyclopedia. Wales and Sanger’s first attempt at a vetted resource, called Nupedia, sheds light on how from the very beginning of the project, vetting and reliability of authorship were at the forefront of the creators.
Rosenzweig adds to a growing body of research trying to determine the accuracy of Wikipedia, in his comparative analysis of it with other online history references, along similar lines of the Nature study. He compares entries in Wikipedia with Microsoft’s online resource Encarta and American National Biography Online out of the Oxford University Press and the American Council of Learned Societies. Where Encarta is for a mass audience, American National Biography Online is a more specialized history resource. Rosenzweig takes a sample of 52 entries from the 18,000 found in ANBO and compares them with entries in Encarta and Wikipeida. In coverage, Wikipedia contain more of from the sample than Encarta. Although the length of the articles didn’t reach the level of ANBO, Wikipedia articles were more lengthy than the entries than Encarta. Further, in terms of accuracy, Wikipedia and Encarta seem basically on par with each other, which confirms a similar conclusion (although debated) that the Nature study reached in its comparison of Wikipedia and the Encyclopedia Britannica.
The discussion gets more interesting when Rosenzweig discusses the effect of collaborative writing in more qualitative ways. He rightfully notes that collaborative writing often leads to less compelling prose. Multiple stlyes of writing, competing interests and motivations, varying levels of writing ability are all factors in the quality of a written text. Wikipedia entries may be for the most part factually correct, but are often not that well written or historically relevant in terms of what receives emphasis. Due to piecemeal authorship, the articles often miss out on adding coherency to the larger historical conversation. ANBO has well crafted entries, however, they are often authored by well known historians, including the likes of Alan Brinkley covering Franklin Roosevelt and T. H. Watkins penning an entry on Harold Ickes.
However, the quality of writing needs to be balanced with accessibility. ANBO is subscription based, where as Wikipedia is free, which reveals how access to a resource plays a role in its purpose. As a product of the amateur historian, Rosenzweig comments upon the tension created when professional historians engage with Wikipedia. For example, he notes that it tends to be full of interesting trivia, but the seasoned historian will question its historic significance. As well, the professional historian has great concern for citation and sourcing references, which is not as rigorously enforced in Wikipedia.
Because of Wikipedia’s widespread and growing use, it challenges the authority of the professional historian, and therefore cannot be ignored. The tension is interesting because it raises questions about the professional historians obligation to Wikipedia. I am curious to know if Rosenzweig or any of the other authors of similar studies went back and corrected errors that were discovered. Even if they do not, once errors are published, an article quickly gets corrected. However, in the process of research, when should the researcher step in and make correction they discover? Rosenzweig documents the “burn out” that any experts feels when authors attempt to moderate of entries, including early expert authors. In general, what is the professional ethical obligation for any expert to engage maintaining Wikipedia? To this point, Rosenzweig notes there is an obligation and need to provide the public with quality information in Wikipedia or some other venue.
Rosenzweig has written a comprehensive description of Wikipedia and how it relates to the scholarship of the professional historian. He concludes by looking forward and describes what the professional historian can learn from open collaborative production models. Further, he notes interesting possibilities such as the collaborative open source textbook as well as challenges such as how to properly cite (a currency of the academy) collaborative efforts. My hope is that this article will begin to bring more historians and others in the humanities into productive discussion on how open collaboration is changing traditional roles and methods of scholarship.