English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

This is a commentary as part of the debate on Open Access. Editor’s Note: The paper was handled by the Department Editor for Debates. 379 A Comment on Open Access: the Whipping Boy for Problems in Scholarly Publishing Volume 34 Paper 19 In their paper, Kingsley and Kennan (2015) provide a spirited, but not entirely convincing, defense of the open access process/movement/system. In this rejoinder, I address what I see as some of the weaknesses of their argument and move on to open access’s weakness itself. My analysis of these issues is based on 15 years’ experience as a librarian with responsibilities such as dealing with academic publishers, developing and supporting repository software, and encouraging and supporting the research community to make their research outputs more open. To begin, I should state that I have no issue with the central rationale behind the push for open access. It is undeniable that academic journals are expensive to purchase at a level that seems difficult to justify—a 2014 Library Journal study found that the average cost of a chemistry journal is in excess of $US4,215 per year (Bosch & Henderson, 2014). And this represents an 11 percent increase on the 2010 figure of $3792 (Monbiot, 2011). Equally, it is clear that the current process, in which researchers provide the content and peer review and edit papers largely free of charge, has resulted in an ecosystem where a handful of large publishers are generating substantial profits at very high margins (Holcombe, 2013). In a period where technology allows for far cheaper distribution of information than was previously possible, it seems galling to many that this is not reflected in cheaper prices. In theory, the open access concept offers an elegant solution to the problem. Authors could make a copy of their work available online or with new publishers that would embrace alternative publishing models. In practice, this hasn’t actually worked. As Kingsley and Kennan (2015) note, a decade after the Budapest Open Access Initiative was announced, at best 20 percent of the world’s scholarly publications are “open”. Additionally, the definition of “open” has become less and less clear (Harnad, 2014) as copyright and publisher-resistance issues have enforced compromises and workarounds. The proliferation of open access flavors—green, gold, hybrid, libre, gratis—is confusing and contentious enough for those intimately connected to the process. For the average researcher, it is largely incomprehensible (Harnad, 2013). This has opened up the field for some unsavory behavior, the so-called “predatory publishers”, which brings me to Kingsley and Kennan’s (2015) first argument: that open access journals and publishers are predatory. There is no doubt that this perception about open access exists in academia. For many, the beauty of the current system is its perceived egalitarianism. A paper is judged on its merits by peers and is published regardless of one’s ability to pay. In the logic of many researchers I have spoken to, if you have to pay to be published, it is because your work wasn’t good enough to be published as part of the normal system. Paying to publish is, therefore, either vanity publishing or some sort of scam designed to trap the unwary. The widely reported “Bohannon sting” that Kingsley and Kennan (2015) reference helped to reinforce this view. As Bjorn Brembs (2013) has noted, the decision to target only OA journals that charge an APC and to not attempt a similar process with other traditional titles or proper controls limits the overall value of the work. Nevertheless, that half of the journals contacted were willing to publish a substantially flawed paper as long as money was paid does prove that some OA journals and publishers are predatory. This is not a binary question: it does not mean that all of them are, but it is undeniable that some are. Kingsley and Kennan (2015) attempt to dodge this question by pointing out the flaws in the current peer review process and in the training of researchers around sound scholarly publishing choices. That the current peer review system is flawed and that poor, fraudulent and useless research is published is undeniable. However, there is a difference between flawed peer review and no peer review at all. The “Bohannon sting” proved without doubt that some OA publishers will promise peer review, will claim to have done it, and will take money in exchange for no more than putting an unedited, unreviewed copy of an paper on a website. In some cases, this will be a website designed to look like another publisher’s website (Beall, 2014). The question of payment to publish leads me to Kingsley and Kennan’s (2015) second argument: that open access is too expensive. Kingsley and Kennan’s arguments rest on the fact that it is possible to publish in an OA journal for no charge and that the average cost of an APC across OA journals is quite low, (according to a 2011 study they cite, it was USD$906). However, more recent studies by Jisc in the UK that look at compliance with OA mandates in that country have estimated an average cost of around GBP£2000 per paper (Jacobs, 2014), or roughly USD$3100 at time of writing. Kinsley and Kennan (2015) rightly point out that much of this money is going to established publishers and journals, and their own solution is that researchers become more educated about the value of publication. I believe that researchers are already making this decision but that they value the reputation of the title Communications of the Association for Information Systems 380 Volume 34 Paper 19 they are publishing with more than the cost of the APC. This is especially true if there are options to charge the payment back to an institution or granting agency, in which case the relative difference between $1000 and $3000 is often marginal compared to the value certain publications provide in reputation. While the cost may be low for an individual or work group, especially compared to the relative value, this is not the case for the institutions that will end up supporting APCs in some form. My own institution, Monash University, published 3350 journal papers in 2013 (Monash University, 2014). Assuming a world where the institution paid all the APCs for its authors at the GBP£2000 rate, the cost per annum in Australian dollars would be roughly $12 million. This figure is roughly equivalent to the amount the library currently spends on all its journal subscriptions. Realistically, the total figure would be lower (some publications would be collaborations, which would split APCs between institutions for instance, and some publications would be made OA with no APC), but, even assuming the university only paid half, it is still a substantial number. And unless the library is able to cancel journals to recoup some of these funds, the OA would indeed be too expensive as the Monash University example above indicates. So, to pay for the OA model as it is currently operating, the current publishing system needs to change dramatically, or as Kingsley and Kennan (2015) put in their third argument: making work available open access in a repository means the publishing system will collapse. Theirs is a slightly odd argument because it ignores OA publishing and posits a non-existent world where there is only green OA. This doesn’t explain much about the argument because researchers have voted with their feet, and voluntary participation in open access has been very low, which Kingsley and Kennan admit. Even with mandates, it has been poor enough that large funders are now having to enforce compliance (Van Noorden, 2014). Oddly, in a paper that is generally positive about OA, the only defense that the authors can muster is that, because OA has not been successful, no one has been able to cancel journals yet. And that, in any case, there are many other reasons why journals are cancelled. This is true, but does not mean that this will hold true forever, which the authors admit. If a sufficient number of papers were available in a timely and discoverable fashion, libraries would cancel their host journals despite Kingsley and Kennan’s (2015) trying to imply that librarians didn’t understand the question when asked if they would do this. Even if librarians didn’t want to, administrators would. It is a pretty obvious cost-saving measure. The main impediment to this issue is probably the uneven pace of OA take-up across disciplines. arXiv.org is often cited as a longstanding discipline-based repository in which researchers in physics, mathematics, computer science, quantitative biology, quantitative finance, and statistics make pre-print versions of their papers available to share with little demonstrable impact on the subscription rate of their host journals (Tidor, n.d.). I would contend that there are three reasons for this. The first is that there is not a sufficient percentage of content made easily available in a timely fashion. My own library has looked at this and come to the conclusion that to cancel any journal would inconvenience our users. Secondly, the way that journals are sold now would make it difficult just to cancel specific journals. Most titles are now bought as part of “big deals”, packages of journals sold at high cost but less than the price of subscribing to all of them individually. These packages, which have been around for around 15 years, mean that, unless the institution is willing to lose all the titles, it can’t afford to lose individual ones. And, if those titles range across many disciplines, there will need to be a far wider range of disciplines where the OA tipping point has been reached before the really large packages can be cancelled. To cancel just the physics, mathematics, computer science, quantitative biology, quantitative finance and statistics in Elsevier’s ScienceDirect “Freedom Package” would probably end up costing the institution money. Thirdly, most researchers