Open Access: some facts and some thoughts

This blog post has been prompted by Open Access 2013, which began on Monday (21st October) and will run until 27th October. What I aim to do in the paragraphs that follow is describe what Open Access publication is, discuss some salient advantages and concerns, and provide readers with resources which I have found useful. It is by no means an exhaustive treatise on Open Access, but I hope that it can function as a primer for those among us for whom this is a new concept.

1. Simply put, Open Access publications are scholarly publications that are provided to readers free of charge. In the Open Access publication model,  publication costs are borne by the author(s), or by their sponsors, rather than the readers, as is the case in traditional publication. Open Access publications are peer-reviewed, ideally to the same standards of rigour as traditional publications (but more on that later). More information about Open Access publications can be found in this briefing document.

2. From an author’s perspective, choosing an Open Access journal as an outlet for dissemination of their work will likely lead to increased visibility and impact. The Open Access publication model is also advantageous to users who do not have institutional access to paywall-protected research. Even in the case of large, affluent universities, there are mounting concerns that the traditional model of publication, which relies on institutions purchasing subscriptions to journals, is no longer sustainable. A detailed listing of benefits of the Open Access model, along with supporting references, can be found here.

3. Several government agencies and other research funding bodies are gradually adopting Open Access policies. This means that researchers who receive public funding from these agencies undertake to make the findings available to the public through Open Access. Such policies have been implemented, or are being phased in, by the by the US federal government (1, 2), the Higher Education Funding Council for England,  and the European Commission, among others. By way of example, in the EU, research output produced with Horizon 2020 funding must be made accessible online no later than six months after initial publication (or 12 months for the social sciences and the humanities). By 2016, it is expected that 60% of publicly -funded in Europe will be available under Open Access.

4. There are two major variants of Open Access, Green and Gold:

• In the Green Open Access model, the author is allowed to archive a post-print version of a published article in an institutional repository (e.g. eScholar), a disciplinary repository or personal website, all of which should be publicly accessible. Post-prints are papers that have undergone peer-review, and have satisfied publication requirements, but have not been typeset or formatted by the journal yet. One should note that copyright remains with the journal, and not all journals allow this kind of archiving as a default option. Green Open Access might involve a fee, and/or there may be an ’embargo’ period, during which the paper is only available to journal subscribers. The SHERPA/RoMEO database lists the copyright policies for a large number of journals, and should be consulted by researchers who consider publishing under Green Open Access.
• Under Gold Open Access, the journal makes published papers immediately available for viewing to anyone interested, free of charge. To cover costs, Gold Open Access often requires substantial publication fees to be paid upfront. Regrettably, this publication model has attracted many publishers of dubious repute, who do not engage in rigorous peer review, and whose output cannot be considered a scholarly publication. Beall’s List is currently the most comprehensive listing of predatory, or potentially predatory publishers, and it should be consulted by researchers who consider publishing under Gold Open Access.

5. The Open Access publication model has come under sustained scrutiny, on the grounds of inconsistent quality of Open Access journals. Most recently, Science reported on a ‘sting’ operation, in which different machine-generated versions of spoof paper were accepted by no fewer than 157 Open Access journals. This was despite the fact that, in the author’s view, “[a]ny reviewer with more than a high-school knowledge of chemistry and the ability to understand a basic data plot should have spotted the paper’s short-comings immediately” and the experiments reported were “so hopelessly flawed” as to render the paper devoid of meaning. The conclusions of the ‘sting’ operation were criticised by several Open Access proponents (examples: 1, 2, 3), who claimed that the affair demonstrated a failure of the peer-review process, rather than the Open Access model; it was also pointed out that traditional publication was not immune to such disasters (e.g.,  the Sokal hoax, and other less famous cases: 1, 2). Nevertheless, it seems that the financial considerations involved in Open Access, particularly in the Gold model, are such that invite corruption in ways that traditional publishing doesn’t.

6. Discussions of the Open Access publication model are often framed in a discourse of academic ethics, and while Open Access appears to democratise the research process, an argument could be made that, in some respects at least, this approach to publishing might widen the gap between privileged and less privileged academic settings. There can be little doubt that researchers without access to funding will benefit from being able to peruse scholarly publications under Open Access; what is less obvious is that, since publication costs are transferred to the author, such researchers are likely to find it harder to publish. This seems particularly true in the case of researchers in developing countries, and in chronically underfunded disciplines, such as the Arts and Humanities. Even if funding agencies sponsor publication, it is difficult to see how the scope and quality of research will remain unaffected by higher publication costs. On the whole, it seems fair to say that eventually, in the majority of cases, Open Access will likely make it easier to consume research outputs, but harder to produce them. To the best of my knowledge, a discussion about how such risks might be avoided has yet to develop in earnest.

[NB A slightly different version was originally published in my own blog, but I hope it may be of use here as well.]