Principle of proof

An open letter to academic journal publishers and editors: Okay...we get it…biotin and streptavidin bind each other. Luciferase and green fluorescent protein glow in the dark. I believe you. You don't have to prove it to me anymore. Enough with the proof-of-concept papers. How about proving something we don't already know.

 

Perhaps I aggravate the situation by writing synopses of these papers in the Bench Press section of DDN Online, but it seemed to me there was once a time when scientists were generally only published when they had something novel to contribute to our understanding of the universe. If you subcloned a fragment of DNA, no one asked you to prove that the enzyme EcoRI cut DNA where you said it did.

 

Over the last decade or so, however, the number of peer-reviewed and academic journals has risen dramatically and they have become increasingly specialized, with titles like The Journal of Left-Finned Zebrafish and Annals of GenoProteomic Yeast Metabolism. And the need to fill these journals has resulted in a lot of manuscripts with little or no significant scientific value. I firmly believe that in the last decade, I have learned 432 different ways to prove that biotin and streptavidin interact in vitro, in vivo and in silico (and a few other in's I'm not recalling at present).

 

Now, I have to admit that I am a method guy. I stood at the bench for years, looking for new ways to solve old problems. So, I like method papers. But I also recognize that there is life outside of method papers. If anything, my fascination with method papers was focused on finding something beyond the technique. I fully recognized that the method was a footnote to the discovery…that it had its own place in the manuscript and was not a result unto itself.

 

Remember, the expression is "a method to my madness", not "a madness to my method."

 

In trying to address an ever-widening spectrum of scientific techniques, publishers and editors have moved the focus away from novelty and toward trying to get as much money out of readerships that have dwindled in number until they are no longer economically feasible or fiscally responsible.

 

If your market is only 750 subscriptions, is the business model really feasible? Does enough truly significant work occur every month in combinatorial chemistry or crystallization to justify 12 journal issues per year? And who in blazes is going to read all this stuff?

 

When I was a graduate student, I used to joke with my fellow students that we should have been allowed to publish failed experiments, because that was often the only result we accomplished months at a time. Now, I am afraid that we are on the verge of losing the sarcastic humor of The Journal of Irreproducible Results and Annals of Improbable Research as the market tries to justify the validity of this research.

 

I'm not advocating the belief that only Nature and Science are worthy of consideration—as much as their respective publishers might appreciate it. There is room for specialization within the various biomedical and biochemical fields, but only as far as the science will carry it. So please, stop padding your journals with proof-of-principle papers.

 

For any of this work to be meaningful, science must lead publishing, not the other way around.