Timothy Gowers's first big assault on academic publishing started almost by accident.
In 2012, the Cambridge mathematician took to his blog to write a post bemoaning the exorbitant prices that journals charge for access to research. Gowers vowed to stop sending his papers to any journal from the world's largest academic publisher, Elsevier.
To his surprise, the post went viral — and spurred a worldwide boycott of Elsevier, which publishes 2,000 journals that can cost up to 10 or 20 thousand dollars a year for a subscription. Within days, hundreds of researchers left comments commiserating with Gowers, a winner of the prestigious Fields Medal. They were upset that prohibitive prices meant much of their research was out of reach, and that the current state of affairs slowed the spread of knowledge. One reader even set up a website, The Cost of Knowledge, to organize the boycott. More than 15,000 researchers have joined the protest since.
But in the four years that have passed, little has changed. Despite a decades-old "open access" movement — which aims to put research findings in the public domain instead of languishing behind expensive paywalls — the traditional approach to publishing remains firmly entrenched.
So Gowers is now launching his second attack, this time with a lot more intention.
This week, he debuted a new online mathematics journal called Discrete Analysis. The nonprofit venture is owned and published by a team of scholars. With no publisher middlemen, access will be completely free for all.
Can academic publishing actually cost nothing?
Right now, American universities and government groups spend about $10 billion each year to access published science.
To understand how we got to this absurd business model, you need to understand some of the history of academic publishing.
One of the best summaries I came across was from Michael Eisen, a biologist at UC Berkeley and co-founder of the Public Library of Science (or PLoS), one of the world's preeminent open-access publishing projects.
He has explained that scientific journals sprouted up alongside the printing industry in the 17th century to disseminate science and information about scientific meetings. Journals were enabled by the printing press and mail services, and like all pre-internet publishing models, sold subscriptions and reprints of articles to stay afloat and recover their costs.
But then along came the internet. Instead of adopting a new business and pricing model to match the new means of no-cost dissemination, many academic publishers hung on to their old one — costly subscription fees and all. Prices even grew in the post-internet environment, as a handful of big publishers (like Elsevier) bought up more and more journals, creating mini knowledge fiefdoms. (In 2014, Elsevier reported a profit margin of nearly 40 percent and revenues close to $3 billion.)
That fact that has increasingly annoyed researchers like Gowers. "The thing that concerns me," he says, "is the idea that my university and others around the world are throwing money on a product that should cost almost nothing."
Right now, taxpayers fund a lot of the science that gets done, academics (many of whom are also funded by public money) peer review it for free, and then journals charge users (many of whom paid for the science in the first place) ludicrous sums of money to view the finished product.
Gowers thinks there's a better way. He wants Discrete Analysis to be a sort of experiment to test whether academic-run, digital-only journals can actually cost nothing and be sustainable.
What medical researchers can learn from math and physics
Gowers chose to focus on this question for a simple reason: Many papers in his area of research are already open access. Researchers in math (and other fields, particularly physics) have had a long tradition of "pre-printing" articles. That is, they are posted on an open website called the ArXiv.org, often before being peer-reviewed and published in journals. There, the articles are sorted and commented on by a community of moderators.
Contrast that with most medical journals, where researchers are mostly forbidden from sharing the results of their research beforehand.
Gowers says this practice exposes the flaw in the journals' business model. "If the publishers are so convinced that they add value, why do they insist [on medical researchers] not posting pre-prints?" he asks. "Why do they insist on embargo of their work? If they added lots of value, they shouldn't mind about the less valuable pre-print."
Gowers is betting traditional journals don't add value, particularly when it comes to papers that are mostly math. "We don't need all the services journals provide — like formatting in a journal house style and copy-editing — because the pre-prints are good enough already, and there are lots of examples where copy-editing makes articles worse and introduces mistakes."
So the new journal will work like this: Researchers can submit a paper to the ArXiv, where the article is posted online for commenting. They can then go to the Discrete Analysis website and simply paste the URL for the ArXiv pre-print, add their contact details, and submit.
The journal will then coordinate peer review (using Scholastica software) and if a paper is accepted, ask the authors to produce a revision of their article to respond to any comments. Once it's in good shape, the article will get posted to the Discrete Analysis website and the authors can post the final version to the ArXiv, says Gowers.
"Our journal will be one of many potential low-cost models of academic publishing," he says. "I would like to see many more very low-cost journals being set up by academics in the control of academics rather than publishers."
This is just one of many alternative publishing models popping up
There are other disgruntled researchers like Gowers out there, trying out their own alternative academic publishing models.
A new journal called the Winnower abolished pre-publication peer review. Sites like PubPeer have popped up to facilitate commenting on already published articles (or "post-publication" peer review). Matters, meanwhile, is a new online journal that will publish fragments of academic papers — single observations. All of these alternative sources consolidate their own peer review and charge no subscription fee. They all take aim at the traditional model of academic publishing.
At the more extreme end, one Russia-based researcher pirated what she claims is every paywalled academic paper and put them on a website for free.
The question is which of these alternatives will actually stick. Gowers thinks it'll largely depend on how quickly journals like his can build up a positive reputation.
Right now, an academic's career advancement hinges on the quality of journals he or she publishes in, which is determined, in part, by impact factor.
"It’s important [that these alternative models] acquire a reputation and prestige that people can feel it’s okay to submit to them — rather than the more established traditional journals — without damaging their careers," Gowers says. "We need an alternative, cheap system sitting there — at which point the commercial publishers will become redundant."
So far, these alternative ventures have had little success dismantling the knowledge fiefdoms like Elsevier. The ArXiv (which launched in 1991) and open-access publishers like PLoS (established in 2000) still haven't displaced traditional journals. But maybe, as more and more mini ventures chip away at the incumbent publishers, the revolution will take shape.