The last decade has seen enormous movement toward broadening access to scholarly literature. The number of articles published in an open-access (OA) manner continues to climb rapidly, as does the number of OA journals. Despite the growth and a near constant level of discussion and debate in the publishing industry, OA still has low priority for most researchers. Governments and funding agencies are working to counter the lack of uptake in the community by changing tactics, philosophically moving from the carrot to the stick.
Study after study shows that researchers place low priority on OA when choosing an outlet for their work. Participants in Ithaka’s most recent faculty survey1 put journal OA policies as the least important determining factor in choosing a venue for submission. Only 10% of respondents in the Research Information Network’s study on communicating knowledge2 felt that OA repositories were important dissemination channels for their work. Even the Study of Open Access Publishing (SOAP) survey3 , most of whose participants were actively publishing OA articles, ranked OA as only the 10th-most important factor in selecting an outlet for publication. When depositing research papers funded by the National Institutes of Health (NIH) in PubMed Central (PMC) was voluntary, rather than required, compliance peaked at 3.8%.
Those data points should not be taken as indicative of a lack of support for broader access to the research literature. They are instead probably a result of the career structure in academe and the priorities that it engenders. Academic research offers an increasingly demanding career path. More and more, researchers are forced to focus relentlessly on the few things that bring in funding and provide concrete career advancement. Everything else, even things that are seen as beneficial, falls by the wayside.
Public and private funding agencies see OA as an important means of maximizing society’s return on investments in research. Funders are slowly recognizing the institutionalized inertia and lack of motivation toward progress and are beginning to implement new policies and enforcement regimes to drive academe to better serve their purposes.
The Finch Report
In June 2012, a UK government-commissioned report, “Accessibility, sustainability, excellence: how to expand access to research publications”4 , known as the Finch report, was released.
The group that wrote the report was made up of representatives of funding agencies, libraries, academe, and publishing. The report is an incisive and fair document but in some ways a frustrating one. It does a nice job of making clear the concerns and issues for each set of stakeholders involved and does not sugarcoat a complex situation. It is somewhat lacking, though, in that it mostly discusses goals and roadblocks but offers little in the way of solutions. For example, the report goes into great depth about how important research societies are and how OA models threaten their continued existence, but it does not offer solutions other than to caution that efforts should be made to help them continue to exist.
The Finch report’s conclusions can be summarized essentially as stating that broader access is a good thing, but that it is important not to lose other important aspects in the quest for improved access, including the high quality of services and usability that the research community now experiences. The report estimates that the transition period to OA will be long and cost the UK alone an additional £50–60 million per year, and that to make the transition all interested parties need to work together in a managed process.
The Finch report posits “gold OA” as the way research should be published and made publicly accessible. There seems to be a high level of confusion over the terms gold and green OA. To pare them down to simple definitions, gold OA involves publishing through formal journals, and green OA involves using repositories and self-archiving to achieve OA. There are multiple variations in each route; the two routes are not incompatible and can occur together or separately.
The report suggests that papers be published through journals and made publicly available through the payment of article-processing charges (APCs). Funding must be provided for APCs, and the fewer restrictions on use and reuse of published articles, the better.
The authors offer strong cautions to funders in setting licensing terms and in setting policies, stating that great care should be taken in determining embargo periods for green OA: “We believe that it would be unreasonable to require that embargo periods are shorter than twelve months. . . . Moreover, in subject areas where the half-life of the articles in each issue of a journal is several years, there may be a case for a longer period”.
Many OA initiatives use Creative Commons licenses. The two most often used are CC-BY-NC, which means that anyone is free to reuse material for noncommercial purposes but must gain permission or pay a license fee for commercial uses, and CC-BY, which means that anyone can do anything with the work without compensation or permission, so long as attribution is provided.
The Finch Report warns against the use of the CC-BY license and the associated threat of lost reprint revenue, particularly for medical publishers, and also because it allows others to “harvest published content from repositories and present them on new platforms that would compete with the original publisher”.
RCUK Policies
Just after the Finch report was released, the Research Councils UK (RCUK) announced its new OA policy5. It hews closely to the Finch report in some ways but diverges from it in others, particularly in setting embargo length and licensing terms, which has resulted in a great deal of controversy. As of this writing, the UK House of Lords has held a hearing on the RCUK’s policy, and a second hearing, by the Business, Innovation and Skills Committee, is scheduled.
The original RCUK policy applies to any paper for which the RCUK has funded any portion of the research and goes into effect for any paper that is submitted to a journal editorial office after 1 April 2013, regardless of when the funding itself was actually issued.
Any paper published must meet one of two requirements:
Option 1 is to publish via gold OA and have the article become immediately freely available in the journal through payment of an APC. Articles that take this route must be deposited immediately on publication into the relevant repository and must use a CC-BY license. Note that the payment of an APC comes with an assumption that no page charges will also be paid.
Option 2 takes the green route. The final accepted manuscript version of the paper must be deposited in the appropriate repository and made freely available within 6 months (or 12 months for papers in the humanities or social sciences). There can be no restrictions on noncommercial reuse, and no APC is paid to the publisher.
A letter signed by the editors of 21 leading British history journals announced that they will not accept these terms as doing so will harm the sustainability of their journals and, more importantly, damage the integrity of the research and the authors’ reputations6 . Instead, they have proposed use of a CC-BY-NC-ND license, which offers unlimited noncommercial reuse of articles and blocks the generation of derivative works. The editors are willing to work with a 36-month embargo for the green route because, compared with papers published in biomedical research journals, the citation lives of history papers are extremely long.
At the House of Lords’ hearing on the subject, these and other concerns were raised. Representatives from the RCUK suggested that embargo periods of 12 and 24 months might be acceptable and noted that there are no plans to actively enforce the policy for at least 5 years.
The Wellcome Trust Policy
The Wellcome Trust has had an OA policy in place since 2005 and continues to refine it. Its current policy expects funded authors to make their publications freely available, and Wellcome supplies funds to pay APCs. Deposit in PMC and UKPubMed Central is required within 6 months of publication. As of April 2013, all articles must be published under a CC-BY license.
Compliance has been an issue for Wellcome, with only 55%–60% of researchers following through on its rules, so it is adding stricter enforcement, withholding the final 10% of grant funding if papers are not in compliance, and preventing the award of future grants to non-compliant researchers.
Compliance
Compliance with funder mandates offers a competitive advantage for journals, and taking care of required tasks, such as depositing articles in repositories, is an attractive benefit for authors. But given the broad spread of funding agencies, each with its own policy (or lack of policy), determining compliance requirements can be difficult. It is also important to note that funding agencies are still refining their policies. Funder OA policies will be iterative, requiring continual readjustment as new issues or unexpected consequences arise.
The first step in compliance by publishers is determining which agencies fund your authors. Tools such as that provided by Web of Science list the funding agencies behind the papers that a journal publishes. Knowing which agencies are most relevant to your journal will help publishers narrow down the policies to track.
More funders will be issuing mandates, though, so it is likely that publishers will need to have a plan, regardless of whether the funding agencies connected with papers in your journal have made any policy announcements. Putting into place an immediate hybrid OA option for your journals, offering the CC-BY license for articles when an APC is paid, and implementing the current NIH policy on deposit after 12 months will probably bring a journal into compliance with most currently announced requirements.
Open Questions
Scientific publishing is a complex world, and any far-reaching policy, no matter how well intentioned, creates challenges. New OA policies leave many unanswered questions and generate some unexpected consequences that should be considered. None of these concerns is unsolvable, but they will take further thought and nuance to sort out.
An immediate positive result of the recent policy announcements is that many new voices have entered the conversation. Most academic researchers have not been involved so far in the debate; access issues are not on their radar. However, funding and researcher freedom are key concerns and it seems that everyone in research and in publishing wants to broaden access to the literature. The question is how to do so effectively. The involvement of the mainstream research community in conversations about policy can help forge a better path forward.
Many of the questions raised revolve around money. The transition will be expensive, and it is unclear where the money to pay for it will be found. There is a notion that the money now spent on subscriptions will eventually be moved over to APCs, but that time may be a long way off, and to get there will require some redundant spending. If you are the Wellcome Trust, you can probably afford that. But if you are a smaller funding agency, a government funding agency in an age of belt tightening, or a poorly funded university or researcher, finding those extra dollars is not a trivial matter, and they will probably come from funds that could otherwise be spent on research.
Both Wellcome and RCUK have announced that they will pay APCs through block grants to institutions rather than through individual grants. That has created concern that funds will be unfairly concentrated at major research centers, disadvantaging smaller institutions. It also may result in the creation of an additional round of prepublication peer review as universities decide who gets the funds and where people are allowed to publish. That will reduce researcher freedom and may delay publication as more hurdles must be cleared.
Determining a sustainable embargo period is another important issue. Policies should be designed to be rational and evidence based. A good deal of information is available on the 12-month embargo period, at least for biomedical publications, courtesy of the NIH policy. The rationale behind the 6-month embargo is unclear, and, as the Finch report notes, such short embargoes are a worrisome unknown.
Policies can instead set rational criteria for embargo length, including a mechanism for lengthening or shortening embargoes as evidence is collected on their effect. There are huge differences among fields. A history journal sees a peak in citation around 5 years after publication, which is vastly different from a bioinformatics journal, where developments move much more rapidly. Embargo periods should reflect such differences.
As the Finch report also recommends, licensing terms need to be carefully approached. If we are to move away from the subscription business model, we need to find new ways to generate revenue to pay for the services that we render. The more we can shift the financial burden away from the researcher, the better. There seems to be a great deal of confusion over the differences between copyright of journal articles and patents of research results, and the proposed CC-BY license may not be the best route to maximize return on funder investment.
For many publications, particularly medical journals, secondary rights licensing brings in a significant amount of revenue. Current APC prices are subsidized by this revenue. If the CC-BY license is required, this revenue is lost, and must be made up elsewhere. The Nature Publishing Group already charges an additional fee for the use of a CC-BY license in some journals, and other publishers are likely to follow suit. Asking researchers to pay more to support for-profit companies and pharma marketing does not seem like a particularly good trade-off.
The CC-BY-NC license provides the same benefits and the same open access to the academic research community but asks for-profit companies to pay their fair share of the costs needed to support the raw materials that they are using to generate income and may provide a more sustainable alternative.
Even so, these CC licenses are incompatible with the inclusion of previously copyrighted material, making them impossible to implement for some journals. An article about a work of art could not include a reproduction of the actual artwork, or a literary journal could not include a substantive excerpt from a written work. Review articles could not reuse figures from previously published papers. There are also ethical concerns that CC-BY licenses would allow reuses of patient information and images that would violate research subject consent agreements.
Finally, enforcement of the new policies is another cost sink, requiring time and effort to track each grant recipient and to check on the status of each publication. Each dollar spent in enforcing OA policies is a dollar diverted from funding research.
Taking the Lead
A key conclusion of the Finch report is that all the players involved need to contribute to progress, and we all need to work together and coordinate our actions. So far, many individual groups are driving their own policies, at times in a vacuum without input from other parties. Scholarly publishers have an enormous level of expertise and experience in the publishing process. It can be frustrating to watch scientists and bureaucrats play amateur publisher and offer unrealistic plans—but instead of just complaining about it, we need to take a leadership role and offer better alternatives.
Funder mandates are happening. Complaining about them or wishing them away will not make any difference. They are now part of the landscape.
Scholarly publishing is a service industry. Our job is to provide the services that our customers require. If the community wants to broaden access to knowledge, to put vital knowledge into the hands of as many people as possible, then that is what we need to provide.
That means a cooperative approach rather than a combative one. It means working with funders, librarians, and researchers to prise out the details of a complex system, to remain vigilant against and correct for unexpected consequences, and ultimately to meet the needs of our customers. It will not be an easy task, but it is not impossible. More scholarly publishers must take the lead, to turn an antagonistic relationship into a cooperative one, and to offer realistic and sustainable solutions that meet the needs of the research community.
References
- Schonfeld RC, Housewright R., Ithaka S+R. Faculty Survey 2009: Key insights for libraries, publishers, and societies. http://www.sr.ithaka.org/research-publications/faculty-survey-2009.
- Research Information Network. 2012. Communicating knowledge: how and why researchers publish and disseminate their findings. http://www.rin.ac.uk/our-work/communicating-and-disseminating-research/communicatingknowledge-how-and-why-researchers-pu.
- Davis P, SOAP survey requires clean interpretation of data. 2011. The Scholarly Kitchen. http://scholarlykitchen.sspnet.org/2011/02/02/soapopen-access-survey/.
- Research Information Network. Finch report: report of the working group on expanding access to published research findings – the Finch group. 2012. http://www.researchinfonet.org/publish/finch/.
- Research Councils UK. RCUK announces new open access policy. http://www.rcuk.ac.uk/media/news/2012news/Pages/120716.aspx.
- Institute of Historical Research. 2012. Statement of position in relation to open access. http://www.history.ac.uk/news/2012-12-10/statementposition-relation-open-access.
DAVID CROTTY is senior editor at Oxford University Press, New York, New York.