Open source software proponents sometimes suggest scientific research and peer-reviewed publication as a parallel to open source software development. This topic is worth exploring a little.
Some Distinct Similarities
In their 2006 report Open Source Software: A Primer for Health Care Leaders, Goulde and Brown say:
Open source has evolved a model for software development that emphasizes collaboration among a community of people who share common interests. Frequent releases are also typical of this model. The collaborative approach is similar to peer review, a prevalent practice in medical research.
Especially in some research fields, research efforts are indeed highly collaborative. For example, scientific papers in these fields may be published as electronic preprints to speed communication among researchers.
Some Important Differences
However, if open source software is indeed a new paradigm, its development process must differ in significant ways from scientific collaboration. One of these ways has to do with ownership of the “raw material” – source code in the case of open source software, datasets and publications in the case of scientific research.
Another is the role of competition. Open source software is a non-competitive culture, while scientific research, especially in fields like biochemistry, is a highly competitive one.
And open source software culture is sometimes described as a meritocracy, whereas prestige and reputation are important in scientific research.
Ownership in Scientific Research
What do individual scientific researchers own? They can own intellectual property in the form of patents and of copyrights on their publications. Plagiarism of another scientist’s work is not only a violation of copyright, but a cardinal sin in the eyes of the profession.
However, individual scientists do not own conclusions or findings once they are published or reported at a scientific meeting, Individual scientists and groups of scientists may compile research datasets which are usually technically the property of their organization or employer. In some scientific disciplines such as meteorology, datasets like satellite photographs are routinely made publicly available soon after their collection. In other fields, scientists are sometimes accused of “hording” their data to keep it from being used by their colleagues. This may indeed be the case in many instances, since individual scientists have a natural inclination to fear someone else taking advantage of their work and receiving credit for it. On the other hand, scientists also have a natural reluctance to release datasets that have not been checked for inaccuracies for fear of gaining a reputation for sloppiness, so what appears to be hording may actually be prudence.
Competition in Scientific Research
As discussed above, competition is inherent to scientific research, and this is quite different from open source software development. Scientists expect credit for their findings, even if they do not actually “own” them, and deliberate failure to attribute findings to their original authors is also a professional sin. Fierce debates sometimes rage over who was in fact the first to report specific findings and thus deserves credit for them.
Prestige and Reputation in Scientific Research
As previously mentioned, open source software development is sometimes described as a meritocracy in which individual contributions are considered without regard for the reputation of their author. This is far from the case in scientific research, much as many scientists would like to believe the opposite. Particularly in the case of grant proposals to organizations like the National Science Foundation, the reputation of the principal investigator is important because it helps reviewers weigh whether the proposed work will actually be carried out successfully.
The Peer-Reviewed Publication Process
The first step in publishing a peer-reviewed scientific paper is, of course, to write it, which is often done as a collaborative effort among several researchers. Scientists value their professional reputations and will request to have their names taken off scientific papers if they strongly disagree with conclusions insisted on by coauthors.
Although open source software has some features in common with peer-reviewed scientific research, the parallel is far from exact. In particular, the role of competition in scientific research is much different than in open source software development. However, scientific researchers, especially those at non-profit institutions like Seattle’s Institute for Systems Biology, founded by prominent molecular biologist Leroy Hood, use open source software by preference when it is suitable for their purposes.