There is a myth of science that has developed over the centuries: science is an open process. Of course, scientists strive to publish their results, and educate their students. But these alone do not make for an open process. Perhaps centuries ago, these practices were more open than the previous practices, but in the twenty-first century, academic science is far from a model of openness. I agree that the scientific method and academic should be open, but it is far from open in reality. There are many problems with it, here are some off the top of my head:

- limited access publishing (e.g. Elsevier): It is so expensive to get access to most academic scientific journals that many universities and colleges in the US, let alone less rich nations, do not subscribe to a lot of them. The Public Library of Science is doing en excellent job of publishing true, open access journals: http://plos.org

- closed access data: Although there is a lot of pressure for academic scientists to publish their results, there no such pressure to publish the source, i.e. the raw data that the results were based on. If the gatherer of the data, doesn’t publish it, it remains totally locked away behind a wall of copyright. So this definitely isn’t open source, since no one can get to the source to try their own analysis. Science Commons is talking about this issue some, http://sciencecommons.org/

- completely closed process: many if not most scientific labs are super secretive about what they are working on because they want to be the first to publish, and they believe that restricting access to this process will help guarantee them future grants.

So compared to free software, current academic science completely fails in the these three regards, and perhaps others. The New York Times recently ran an article about what they call “open-source science”. It seems to be mostly about funding research via cash prizes and is very vague on the details. It almost reads like PR for the companies featured.

As for this funding model, I think that this model of using a competition for prize model can be useful for some developments, but it needs to be one of many approaches. If this was the only funding model, we would be worse off. For example, in the drug industry, it is well demonstrated that the for-profit R&D model is very good at successive improvement of existing techniques, but bad at all other types of research. Basically all groundbreaking drug research is done by government funded R&D that has very few direct returns to the funding agency. So these two different funding models complement each other.