FROM EUROPE TO NORTH AMERICA
The momentum has shifted to North America, where in February, the University of California (UC) terminated its $11 million-a-year subscription agreement with Elsevier. UC negotiators wanted a transformational read and publish deal like those sought by European consortia—that is, combining the cost to access or read with the cost for UC authors to make their papers OA (osc.universityofcalifornia.edu/open-access-at-uc/publisher-negotiations/ucand-elsevier). UC also wanted to pay less money than before, and it wanted this to be a transitional step, one that would “accelerate a shift to full open access” (senate.universityofcalifornia.edu/_files/reports/academic-council-statement-elsevier-feb28.pdf).
According to Jeffrey MacKie-Mason, co-chair of UC’s negotiation team, “Elsevier eventually offered to do something like what we wanted, for OA, but they wanted to charge us a lot more. Our current calculations are that they would have increased the amount of our payments by 80 percent—an additional $30 million over a three-year contract” (news.berkeley.edu/2019/02/28/why-uc-split-with-publishing-giant-elsevier).
The UC system is enormous—10 campuses, 238,000 students, and more than 198,000 faculty and staff—and produces an estimated 10% of the U.S. scholarly publishing output. That makes it extremely influential. Since February, heads of libraries at several institutions have made supportive statements, including those at the University of Virginia (tinyurl.com/yy92kx36) and University of North Carolina–Chapel Hill (library.unc.edu/2019/03/statement-uc-elsevier). No doubt, many others are now thinking carefully about how to approach their contract renewals with Elsevier and other publishers.
UC and the European consortia have said their goal in negotiating transformative deals with publishers is to help push the system toward full OA. This is also the goal of the Plan S initiative, launched by a dozen European funding bodies last September. Plan S will require researchers supported by these funders to publish their articles in fully OA (non-hybrid) journals or OA platforms starting in 2020. Though funders are still working out implementation details, so far, their preference is for publisher-made OA rather than self-archiving.
There has been a lot of concern and excitement around Plan S, and that’s because it is next-level OA. By tying research money to strict OA requirements—no embargoes, no hybrid journals, and caps on APCs—publishers and authors are forced to think and behave differently. In the case of publishers, they need to determine, and soon, publishing models that will fit with the plan. Holly Else notes that Plan S “would bar researchers from publishing in 85% of journals, including influential titles such as Nature and Science” (“Radical Open-Access Plan Could Spell End to Journal Subscriptions”; nature.com/articles/d41586-018-06178-7). Marcus Banks thinks that authors should follow the example of libraries and funding agencies when it comes to OA (“It’s 2019. Academic Papers Should Be Free”; Undark.org/2019/04/18/open-access-publishing).
Plan S is relevant in the United States, and not only because the Gates Foundation has joined the funder group. A report released earlier this year finds that 6.4% of papers published globally in 2017 were supported by Plan S funders. But several American institutions, including MIT and Caltech, have more than 15% of papers that list a Plan S funder, according to the ISI Report “The Plan S Footprint: Implications for the Scholarly Publishing Landscape” (clarivate.com/wp-content/uploads/dlm_uploads/2019/02/WS190021_ISI-Report-2019_013.pdf).
Transformative agreements and funder mandates asking publishers to change their economic models are potentially great for OA. However, they still leave control in the hands of publishers.
“Universities have done themselves terrible damage by outsourcing a lot of their infrastructure to corporate entities that do not have their interests at heart,” says Kathleen Fitzpatrick, director of digital humanities and English professor at Michigan State, in a December 2018 interview (“ARL-SSRC Open Scholarship Meeting Prework Interview Report,” Philip N. Cohen; opensociology.pubpub.org/pub/yjgp3jdk). Fitzpatrick, a former director of scholarly communication at the Modern Language Association, uses learning management systems as an example of infrastructure that individual universities buy from companies because it’s “shiny and pretty,” but then locks in data and doesn’t end up working the way they want it to. Why do they do it? Because no one institution can, on its own, effectively create and maintain such a system.
“But institutions working together, collaboratively, around this problem could build something open source, academy owned, not for profit that would be far better than anything the corporate world can sell us right now,” Fitzpatrick maintains. “And this is true of publishing as well. We need,” she says, “to be thinking about ways the academy can support and sustain the infrastructure that it relies upon in ways that sustain its value.”
It’s not necessarily a problem that many publishers are “commercial”; it’s that “too often they are commercial companies whose values are radically misaligned with the values of our community”—that is, those of research institutions, scholarly societies, nonprofit publishers, and libraries, writes Heather Joseph, executive director of the Scholarly Publishing and Academic Resources Coalition (SPARC). “The goal of these companies is profit maximization” (crln.acrl.org/index.php/crlnews/article/view/17246/18986).
In the last few years, Elsevier, for example, has bought up companies whose services run the length of the scholarly publishing pipeline (crunchbase.com/organization/elsevier/acquisitions/acquisitions_list). It now owns SSRN, a preprint server; bepress, a repository and publishing platform; Aries Systems, a submission and peer-review system for journals; Mendeley, a reference manager; and Plum Analytics, an altmetrics provider. Wiley, too, has bought several companies, including the publishing platform Atypon.
How can academy-owned publishing help with this? There are many ideas and projects in the works; below are just a few examples.
Because of the growing concern about a concentration of control in the hands of a small number of companies, the Confederation of Open Access Repositories (COAR) and SPARC recently developed a set of “good practice principles” for librarians and other negotiators to keep in mind when making purchasing decisions to ensure that services are open and support the aims of scholarship. Principles include making sure that terms and conditions of contracts with vendors are not covered by non-disclosure agreements, and that companies’ data-gathering policies are transparent (sparcopen.org/our-work/good-practice-principles-for-scholarly-communication-services).
Another example of academy-owned publishing is keeping the technology and infrastructure community controlled as open platforms managed by universities. In 2016, MIT launched a homegrown publishing platform called PubPub; it was developed in the Media Lab and is now supported by the Knowledge Futures Group at MIT Press. PubPub, which is OA and open source, is experimenting with ways to do publishing, peer review, hosting, and collaboration. The site hosts reports, journals, and books. Last year, on the 200th anniversary of Frankenstein’s publishing, MIT Press and Arizona State University collaborated to create Frankenbook (frankenbook.org), an OA version of Mary Shelley’s book, annotated by experts and lay readers around themes of equity and inclusion, medicine, philosophy, politics, science, and technology. The MIT Libraries is also collaborating with the MIT Press on a hybrid print and OA book series called <strong> Ideas. The OA versions will appear on PubPub.
MIT Libraries has given financial and other support to the Public Access Submission System (PASS), an open-source platform developed at Johns Hopkins whose aim is to help researchers comply with OA policies and mandates from federal funders, grant institutions, and their own universities. PASS allows researchers to simultaneously submit articles to different repositories.
Academy-owned publishing could also mean keeping journals community-controlled, which can include journal flipping. In January, the editorial board of the Elsevier-owned Journal of Informetrics resigned (insidehighered.com/news/2019/01/14/elsevier-journal-editors-resign-start-rival-open-access-journal). As they wrote to the publisher, it was “[b]ecause our position on ownership, open access, and open citations is fundamentally irreconcilable with the position of Elsevier. …” (documentcloud.org/documents/5683932-Resignation-Letter.html). The board has since been working with the MIT Press to launch a fully OA journal called Quantitative Science Studies.
Keeping publishing in the hands of non-corporate entities is not a new idea—Paul Ginsparg, founder of the preprint repository arXiv, wrote about scholar-run journals called “overlays” (because they overlay on top of open repositories) back in 1997 (“Winners and Losers in the Global Research Village,” The Serials Librarian, v. 30, issue 3–4; doi.org/10.1300/J123v30n03_13). But it is now hitting the mainstream. There are now guides to starting OA journals or flipping closed ones to OA (blog.scholasticahq.com/post/how-to-start-flip-open-access-academic-journal) and to broaching these conversations with faculty editors (osc.universityofcalifornia.edu/open-access-at-uc/transitioning-journals-to-oa/hosting-a-roundtable). Earlier this year, many people celebrated the first Academic-Led Publishing Day (academicledpublishingday.com).
MORE THAN ONE ROAD TO OA
There is no one way to make OA happen. In the MIT Libraries, we’re regularly engaged in conversations with publishers, asking them to experiment with us on new models that move the system toward openness (intheopen.net/2019/03/subscribe-to-open-as-a-model-for-voting-with-our-dollars). We also support academy-owned publishing in the ways laid out above. Both approaches to OA will soon be strengthened at MIT because the institute-wide Task Force on Open Access to MIT’s Research recently released a set of draft recommendations that include, for example, broadening the faculty OA policy to cover all MIT authors; adopting an OA policy for monographs; and asking department heads to develop discipline-specific plans to encourage and support open sharing from their faculty, students, and staff (mitoataskforce.pubpub.org/pub/draft-recommendations).
In a key recommendation in the report, faculty members are asked to ratify a set of principles for open science, which include the stipulation that authors should retain copyright for their own work, scholarly work should be open to computational analysis, and the full cycle of research should be part of the scholarly record. If ratified, these principles would be implicit in any MIT approach to OA, as they offer guidance for individual researchers in making decisions about communicating their work and are useful when the MIT Libraries negotiations team works on contracts with publishers.
There is more than one way to get to open in scholarly publishing. The point of recommendations like those at MIT is to double down on responsible ways to make it easier for researchers to produce open scholarship and for others to read, share, and use it.