Open Access: “Come the Revolution”

First published on 

I spent most of last week in the company of revolutionary librarians.

The event was the first conference of Open Scholarship Initiative, a working group in the vanguard in the movement for unrestricted, cost-free access to academic research and the demise or transformation of traditional academic publishing. The implications for scholarship and education are enormous and already well-demonstrated by Public Library of Science (PLoS), a unique success for open access.

The conference brought together an enthusiastic group of advocates, principally information scholars and library managers, provosts (chief academic officers of universities) and their representatives, representatives of funding agencies and foundations, and public or subsidized publishers. In this mix there was also a much smaller and less vocal group of skeptics, including commercial publishers, intellectual property advocates, and a few concerned authors. The absence of a large contingent of authors with a stake in the outcome was obvious and troubling for the initiative.

Essentially, the problem is that the current system involves cost-recovery (by publicly-supported and well-financed private publishers) and profit generation (by commercial publishers who bundle highly used journals with less frequently-accessed journals in bundles, not unlike cable television options). Authors submit their manuscripts, which are managed by (usually unpaid) editors who seek (unpaid) reviewers from among peers in the field (and who are increasingly overburdened and so are balking at doing this work); the reviewers’ responses guide and in some cases determine the editor’s decision for acceptance. This mechanism, called “pre-publication review” serves as a gatekeeper to keep badly written papers and bad science out of the system. It requires a lot of handling and for this the publisher needs compensation, which comes from revenues paid to the publisher. Higher prestige and impact generates greater demand and so commands a premium price, but also constricts the free flow of scientific information.

Once accepted, the manuscript is published in a journal that someone interested in the topic has to pay for if they are not working at an institution that subscribes to the journal (usually as part of a package). For example, it costs me $30 to $50 to download from the publisher a single paper that I may need in my work (which often involves many papers), and I cannot tell which papers have the essential information I need until I have already paid for it, because until I pay I am only allowed to read the abstract.

“Open access”, or “open” as its advocates frequently say, would post scholarly articles on line for anyone to download at no cost. The technology is no obstacle and the cost would be marginal. The social benefits might (would?) be huge, in the form of accelerated and more thorough research and application, for example in biomedical advances.

However, in a pure open access model the publishers would derive no revenue stream from the publication and so without a subsidy there would be no means to support an infrastructure of editors, manage a peer-review process (and the reviewer would still not be paid), or support a sophisticated publishing platform that would accommodate different media (for example, video is now useful for showing the kinetics of chemical reactions) and evaluate the significance of the paper. Another cost is for saving data sets in repositories, so that a study can be re-analyzed, repurposed for other research questions, and easily audited. These costs would have to be carried by funding agencies, authors, or employers (institutions). On the one hand, doing these things would make research more efficient, and possibly reduce cost for conducting a study over time, but it would add to the cost of a grant, possibly (in the short term) at the expense of budget for the research work itself.

Closely related to open access is the problem of peer review. Many open access advocates would like to break up the gate keeping function of “pre-publication” peer review and replace it with “post-publication” review (basically, posted detailed commentary after the paper is made public), and see the open access movement as one way to do this. It is hard to see how this would work in practice, without flooding open access journals with junk, and that is, in fact, what has happened to the many (almost certainly most) commercial on-line publishers that have sprung up over the last ten years. Opponents point to the deplorable proliferation of “predatory journals”, which charge authors (sometimes exorbitantly) for publishing their papers and provide weak (or more often nonexistent) editorial review. On the other hand, mainstream publishers are increasingly setting up open access platforms (for example, Taylor & Francis supports Cogent open-access journals) and often provide open-access options for authors for a fee (usually running into hundreds of dollars). For the moment, it is hard to see how post-publication review would improve things and open access does not necessarily require it, but the economics of publishing suggest that it would be hard to maintain a high standard of pre-publication peer review if major journals went to open access.

The publishing industry is still searching for a viable business model. The proponents of open access often feel that no business model is needed – that just as “information wants to be free”, scholarship should be a public good and should flow wherever it is needed. Similarly, opinion is growing that public funding of research, and funding by organizations that have tax-exempt status, requires investigators to share their work, including the data. Data-sharing is particularly sensitive, because although sharing data is a good thing in principle, in practice it is difficult, because it requires many hours of extensive documentation and formatting for another investigator to use, and must be filed in a repository, which normally charges a fee. Journals do not want to handle large data sets and so repositories have sprung up at institutions.

I participated in a breakout group on the “morality” of open access publishing, which really examined ethics. It seems to me that there are two incompatible and therefore mutually exclusive ethical models. One point of view (the open access model) is that there is an ethical imperative (categorical or otherwise) to make information/scholarship available as widely as possible because of the good it can do. The other (the existing model) is that authors have rights to benefit from their work (including data sets they have collected) and that publishers too have ownership rights, which right now drive compensation in the form of rent-seeking behavior (journal subscriptions and fees). This strikes some as immoral and selfish because knowledge is or should be a public good, but the premise of a capitalist, classically liberal society is that social benefit is maximized when individuals have an incentive to do well for themselves.

Which is more ethical, the model that makes scholarship open to all but denies the author control over the use of his or her own work or the model that works within the ethical model we already have and promotes dissemination of knowledge through business incentives and subsidy? Members of our breakout group were all prepared to go unequivocally with the former, but when the issue came to the plenary for discussion, the loudest voices were those concerned about a breakdown in the latter, which they saw as catastrophic for scholarly publishing. This appears to be why there were so few authors at the meeting and those who were there were not very vocal. Authors, in general, would love for their work to be easily accessed but not so keen on surrendering their hard-won data and having their contributions marginalized or even misused by other users.

Although each model can be tweaked and improved insofar as operations are concerned, they are fundamentally incompatible. There is no obvious way to protect and balance the rights and interests of authors, publishers, funding agencies, institutions, users, and knowledge managers (including librarians and managers of data repositories), who are increasingly emerging as major players in this debate.

This conference was a little like eavesdropping on someone else’s conspiracy. When I was young, the catch-phrase “come the revolution!” was already an ironic trope, used to suggest the way one wishes society to be, not a serious declaration of post-revolutionary agenda. The revolution in scholarly publishing is still working on its manifesto and is still trying to figure out what it means by “open access”. But come the revolution, scholarship will be freely available to all.