Tonight's Sky
Sun
Sun
Moon
Moon
Mercury
Mercury
Venus
Venus
Mars
Mars
Jupiter
Jupiter
Saturn
Saturn

Tonight's Sky — Change location

OR

Searching...

Tonight's Sky — Select location

Tonight's Sky — Enter coordinates

° '
° '

Giants fall hard

Professional scientific journals deserve an ignominious death.
Hester_Jeff
In Research Land, publish or perish is the law of the jungle. That’s not a bad thing. Research is pointless if nobody hears about it. And when it comes time to get jobs, promotions, tenure, raises, and professional accolades, peer-reviewed publications are the coin of the realm.

So how come a slew of prominent researchers working on artificial intelligence is staging a revolt, boycotting Nature Machine Intelligence, a new journal from an ultra-prestigious publisher? How can they get away with such impudence, and why would they try?

All that work
For starters, journal publication is a boatload of work that can take forever. When your paper lands on a journal editor’s desk, they have to find a bona fide expert in your field to review it. For a referee, that means days going over your paper with a fine-tooth comb, then writing up a thorough assessment of its strengths and weaknesses. What credit or compensation does the anonymous referee get for doing yeoman’s work? Nada.

Fortunately for publishers, researchers have trouble saying no to anything. So after a few tries, the editor finds a sucker, er, a good referee. When your paper arrives, the grossly overcommitted referee puts it on the “Important” pile, where it stays until the editor sends a gentle reminder that the report was due a month ago. Whoops! Time to get started! (I hate to admit it, but . . . um . . . guilty.) You can bet the family jewels that the referee will want modifications, which means another pile of work for you and your colleagues.

But eventually, hooray — your paper is accepted for publication! Now begins the potentially lengthy process of getting it into print.

Open a scholarly journal and look at the submission, acceptance, and publication dates of a few papers. Don’t be surprised that the process can literally take years.

Ouch! Research moves fast these days, and it’s getting faster. By the time your groundbreaking work arrives in mailboxes, you hope it hasn’t become passé, or even irrelevant.

ASYJH1018_01
Watcher Panyajun/Dreamstime
Worse yet, a lot of work never gets published at all. In particular, who wants to jump through all those hoops just to publish a paper that says, “It didn’t work”? That’s catastrophic. Falsification of expectations is the bedrock of scientific knowledge.

If the current process of publication is so bad for science, why does it persist? Easy. There’s an 800-pound gorilla in the room with dollar signs for eyes. Scientific publishing is a $25 billion-a-year business. That’s a lot of zeros. With researchers doing all the work for free, then buying back access to their own research, scientific journals are what you might call a racket. In 2010, Elsevier, one of the industry’s largest publishers, had a profit margin of 36 percent, netting them $960 million! That’s a breathtaking return in anybody’s book.

Things have changed
Maybe it’s no wonder that researchers are starting to balk, especially since alternatives are popping up organically. In 1991, Paul Ginsparg, a physicist at Cornell University, started a repository where researchers can directly post their work. By 2016, his site, arXiv.org, was adding papers at a rate of over 10,000 a month in astronomy, physics, mathematics, and a few other fields. Add your paper to arXiv.org, and it is available instantly. Is it any surprise that these days people read arXiv.org while The Astrophysical Journal collects dust on the shelf?

What about peer review, you ask? Publishers beat that drum loudly. Post your preprints, they say. We’ll keep the job of getting papers refereed and giving them our seal of approval. Huh? Pay $25 billion a year for somebody to inefficiently moderate conversations between authors and reviewers. Come again?

There are better ways. ArXiv.org already has moderators who look after the legitimacy of posted papers. Why not go a step further? Let experts who really care about a paper offer their own comments, some of which moderators flag as the equivalent of refereed reports. As comments, related research reports, cross-links and the like accumulate (including failed attempts to replicate results), the paper and its associated thread become a living document, timely from the start and perpetually relevant as the field advances. Add blockchain (the technology behind cryptocurrencies) to the mix, and you have ironclad security for the whole ball of wax.

The culture of science has to change for new models of publishing to gain acceptance, but $25 billion a year seems ample incentive for even the oldest dogs to learn some new tricks! Let there be no doubt that traditional publishers will fight tooth and nail to keep their gravy train steaming along, but it’s past time for the mighty to fall. Journal publishers might heed the immortal words of Nobel Laureate Bob Dylan as they ponder the coming revolution. “It’ll soon shake your windows and rattle your walls / For the times they are a-changin’.”

0

JOIN THE DISCUSSION

Read and share your comments on this article
Comment on this article
Want to leave a comment?
Only registered members of Astronomy.com are allowed to comment on this article. Registration is FREE and only takes a couple minutes.

Login or Register now.
0 comments
ADVERTISEMENT

FREE EMAIL NEWSLETTER

Receive news, sky-event information, observing tips, and more from Astronomy's weekly email newsletter. View our Privacy Policy.

ADVERTISEMENT
ADVERTISEMENT
Apollo_RightRail

Click here to receive a FREE e-Guide exclusively from Astronomy magazine.

Find us on Facebook