Researchers Have the Power to Change Publication Incentives. Nobody Else.

That’s right, Smokey

In an excellent new feature for the Atlantic, James Somers convincingly argues that the scientific paper is obsolete. Reinforcing arguments made by many others, including myself, Somers points out that a PDF is an amazingly antiquated means of communicating scientific knowledge in the 21st century. Mimicking the output of a printing press, in an age of software code and big data, is anachronistic to say the very least.

In addition to arguing that the process of representing science should be just as dynamic as the work of doing science, Somers presents two alternative visions of what the future state of scientific publishing could look like. One option for publishing would be the walled garden, proprietary software approach — as represented by the program Wolfram Mathematica. The other is a completely open source approach, with no centralized direction— as represented by Project Jupyter.

Somers shows that, so far, Project Jupyter has achieved more penetration into the everyday working lives of researchers. I hope this continues. All of science — indeed, all of scholarship — rests on sharing knowledge and building upon what has come before. Doing so within the bounds of a proprietary format is antithetical to that mission.

Thinking about these alternative futures is getting very far ahead of ourselves, though. At the moment the traditional paper (or PDF) remains the coin of the realm for sharing new scholarly knowledge. This may seem ridiculous, because it is, but Somers points out that it often takes decades for new communication technologies to take hold: “After Gutenberg, the printing press was mostly used to mimic the calligraphy in bibles. It took nearly 100 years of technical and conceptual improvements to invent the modern book.” (Bold mine).

Given that the internet has been in common use for a mere quarter-century, we may have a long way to go before scholarly publishing fully embraces the capabilities of the web.

In any discussion of the lethargy of scholarly publishing practices the topic of “incentives” always comes up. As long as scholars are rewarded for publishing a standard paper, and discouraged from experimenting or trying anything new, they will keep on doing what they have always done.

Diane Harley, of Berkeley’s Center for Studies in Higher Education, has done fantastic work that validates this thesis. And the thesis makes intuitive sense. Everyone does more of what is rewarded than what is frowned upon — scholars are simply responding to their prevailing incentives. But in his piece, this is the one area where Somers misses the mark: “Until journals require scientists to submit notebooks, and until sharing your work and your data becomes the way to earn prestige, or funding, people will likely just keep doing what they’re doing.”

Here is the usual sleight of hand at work. If only the journals changed. Or the promotion and tenure committees. Or the provost. Or all of academia. In each of these scenarios, the researcher has absolutely no individual agency and is powerless in the face of larger forces.

Not true. If enough researchers conceived their data to have equal standing to a paper, or treated their data visualizations as a publication, or shared their data as it was being developed, scholarly publishing would evolve. The current ways of operating are deeply entrenched and very hard to change, but they are not natural and ironclad laws.

Some researchers can make this transition much more easily than others. For example, authors in the global south are in a very tenuous position for challenging these prevailing norms. But researchers at wealthy western institutions have much power than they realize (or care to admit). Here’s hoping they start to use it.