The British science journal Nature recently compared the quality and accuracy of content in Wikipedia and Encyclopedia Britannica, as reported in BBC. In short, the researchers found few differences between these two reference sources. Just as a refresher, what makes Wikipedia so unique is that it is "based on wikis, open-source software which lets anyone fiddle with a webpage, anyone reading a subject entry can disagree, edit, add, delete, or replace the entry" (BBC).
I find this latest comparison to be quite fascinating as an academic because it leads to a broader question of "open source science." The latest debate surrounding the South Korean stem cell researcher Dr Hwang Woo-suk highlights this very same issue (see BBC news article). That is, for science to progress, there needs to be an openness to the process. There needs to be room to challenge traditional conventions as well as recent innovations.
I can imagine a day when psychology research, for example, becomes more open-source. Imagine allowing the public to become a part of the research project from the beginning. Let's take the study of international adoption. Rather than relying on advisory boards of adoptive parents, adoptees and adoption experts, we could create an open-source that allows input from everyone with interests in the topic. It might be unwieldy at first to manage all this input but there could be an oversight or meta-advisory group that monitors and edits.
Research then could be conducted online (which is already occurring in a more limited way), but it can be designed to self-correct (or self-evolve) based on sample responses and queries. Over time, large amounts of data can be amassed from around the world. Additionally, preliminary results can be published in real time, thereby reducing the lag between data collection and publication. Knowledge generated about a problem and answer will then evolve over time. Interpretations of the findings also will evolve, as more experts and lay scholars contribute to the process.
A current half-example of such research (or as akin as possible without wikis) is the Implicit Attitude Test which can be taken online. The IAT assesses unconscious bias toward various social groups. Unfortunately, it does not allow readers to modify the demand characteristics. This is a key methodological aspect of open-source science that needs to be worked out.
Obviously, there would be lots of methodological and ethical problems to overcome with this type of open-source science, but the possibilities are intriguing.
Posted by richlee at December 15, 2005 12:15 PMOpen scientific publishing is an area of active (and sometimes acrimonious) debate in the sci-tech communications community. New non-commercial ways of sharing research, such as the PLOS (Public Library of Science) free online journals and the arXiv.org physics preprint server, are making waves with traditional for-profit academic publishers like Elsevier and Proquest.
Many objections to "open source-y" scientific publishing center on the need for thorough and reliable peer review of research. Traditional publishers can support well-paid editorial staffs and even in-house scientists, and the prestige associated with journal publishing makes outside academics willing to participate in anonymous peer review. One way of looking at this is that the credibility and ethos of traditional scientific publishing is created through economic structures -- the power of creating "official" knowledge is in the hands of a small number of experts whose authority rests on the exclusivity of the publishing process.
Then along comes the Internet, a technology with the potential to let anyone distribute information on equal terms with the biggest institutions. Given this factor, how do we (as a society) separate good science from fraud? And on what basis do we even make that distinction? It wasn't as obvious a problem before, when shoddy research or outright charlatanism was suppressed by the economic realities of "getting published."
This are not new questions though. Philosophers and cultural critics have long debated the ways individuals and societies produce knowledge/meaning. Levi-Strauss and the structuralist anthropologists were into it, and Mr. Foucault cracked his bald head on the question too.
Posted by: Peter Park Nelson at December 15, 2005 04:56 PM