Titus Brown’s The Cost of Open Science is very similar in spirit to my claims that open source research software is a collective action problem.
This kind of thing is one of the hidden “costs” of open science — the cost of pushing science forward as a whole, sometimes at the expense of one’s own career — and something I’m watching happen with some of our software approaches. I’m being asked with increasing frequency to review papers that extend some of our approaches (especially diginorm) and while it’s really exciting to see people building on our work, it’s also bittersweet, because we could have done some of this stuff quite easily ourselves, and (the dogma goes) more publications is better. If we’d sat on our code and eked as many publications as possible out of it, we’d probably be in a better position with respect to a monopoly on certain kinds of sequence analysis. But it’s also clear that (IMNSHO) some sequence analysis would be a harder — the most obvious example is Trinity’s inclusion of a diginorm-inspired approach. I think we did the right thing, but it’s hard to convey this to the people in charge of convincing MSU to retain me, and I’m not sure too many granting agencies care, either
Open science is often good for the world and bad for you. That’s the very definition of a collective action problem.
Meta Rabbit 