Harvard Library and MIT Libraries provide recommendations for Plan S implementation

Plan S is an initiative for open access that was launched in September 2018. It requires that, from 2020, scientific publications that result from research funded by cOAlition S members must be published in compliant open-access journals or platforms, or deposited in compliant open-access repositories. cOAlition S, a group of national research funding organizations and charitable foundations who have agreed to implement the 10 principles of Plan S in a coordinated way, has solicited public feedback on its guidance for implementation. Harvard Library and the MIT Libraries have jointly issued the following recommendations in response.


Harvard Library and the MIT Libraries are committed to fostering a scholarly communications environment which advances the values of openness, equitable access, transparency, responsible stewardship, and flexible reuse. Aligned with these commitments, we are in broad support of Plan S and its goals to ensure that publicly funded research is made openly available to a global audience. We applaud the clear, unmistakable intention behind Plan S: to provide strong, meaningful incentives to make new research open access.

To support the Plan S coalition in realizing its intention, we recommend certain adjustments to the implementation details. We’d like to see Plan S make better use of the global network of open-access repositories. We’d like to see Plan S reinforce and expand – rather than neglect or unintentionally hinder – the power of open-access repositories to democratize access to science and scholarship.  We offer these suggestions out of our long experience implementing open-access policies and managing open-access repositories on our campuses.

Suggestions for the Green Open Access (Repository) Compliance Option

We commend the Plan S recommendation that “all publications and also other research outputs [be] deposited in open repositories.” However, the current Plan provides an incomplete picture of the benefits of OA repositories, indicating that “Deposit of research outputs in open repositories is recommended to ensure long-term archiving, research management, and to support maximum re-use.” We recommend that Plan S add “open access” to this list of benefits. OA repositories can provide bona fide open access. Suggesting otherwise is simply inaccurate, and more importantly, limits the coalition’s ability to grow the total corpus of OA research.

Along the same lines, we recommend that Plan S broaden the green OA option (OA through repositories), to make it less onerous and more viable for researchers. In its current form, the Plan S green option is needlessly and even harmfully narrow and difficult.

There are two good reasons to broaden the green road. First, green OA is a workable and inexpensive path to OA in all academic fields and regions of the world. Second, barriers to green OA put researchers, particularly early-career researchers, in an untenable situation. A reasonable green OA option will let researchers publish where they must in order to advance their careers, and still satisfy their funders by making their work OA. Without a reasonable green OA option, early-career researchers will be torn between the demands of their funders and the demands of their promotion and tenure committees.

A good green OA option enables authors to submit new work to the journals of their choice, and thereby answers an objection based on academic freedom. If an author’s journal of choice is not OA (or does not satisfy the Plan S criteria for eligible OA journals), then a green option would let the author comply with Plan S by making the work OA in a repository. Plan S has already expanded its original green OA option by allowing deposit of the Author’s Accepted Manuscript or the Version of Record (AAM or VOR), and by making the green OA option permanent rather than limiting it to a transition period.  These are important ways to support a viable green OA option. By adjusting a few other conditions on green OA, Plan S could fully realize its vision of openness to science and scholarship while avoiding needless and damaging barriers to  those who create that science and scholarship.

With this aim in mind, specifically, Harvard Library and the MIT Libraries recommend that Plan S:

  • Change four required features of OA repositories to recommended features:
    • Automated manuscript ingest facility
    • Open API to allow others (including machines) to access the content
    • QA process to integrate full text with core abstract and indexing services (for example PubMed)
    • XML format for repository contents (more on this below)
  • Be more explicit and clear on the distinction between repositories and platforms.The existing language is likely to leave readers with the view that the rules about “platform OA” apply to repository (or green) OA.
  • Reduce the barrier to meeting the green OA requirements by allowing Creative Commons licenses other than CC-BY and CC-BY-SA.
  • Clarify what we understand to be Plan S’s intent: to require green OA “without publisher embargoes.” This language should supplant the existing requirement that the work “be fully available OA at the time of publication.” While immediate OA is a laudable goal, the existing language leaves no room for inevitable logistical delays in obtaining and depositing articles.
  • Reduce the requirement that green OA texts be in XML to a recommendation. As a requirement, this would generate high and frequently insurmountable barriers to meeting Plan S’s green OA requirements. Making it a recommendation, rather than a requirement, would also mirror the Plan S’s rules for goldOA texts.
  • Add a rights-retention requirement to facilitate the green option. There are several ways to do this.
    • One model for this approach exists in the US, where there is a standing “federal-purpose license” that makes it unnecessary for federal agencies to ask authors to retain rights. (For complicated reasons, the NIH does not currently use that standing license, but instead requires that grantees retain sufficient rights to post the works in accordance with the NIH’s Public Access Policy.) The Wellcome Trust uses a method similar but not identical to the NIH.
    • We recommend that coalition members check to see whether there are comparable standing licenses in their own countries. Even if there are not, we recommend that they build the retention of nonexclusive rights into the language of their funding contracts. That would give them all the rights they need to meet Plan S requirements. It would streamline the rights-retention or license-acquiring part of the policy and avoid trapping authors in situations where publishers balk at allowing authors to retain sufficient rights.
  • Clarify that authors are not required to make double deposits. The current language says the Accepted Author Manuscript or the Version of Record (AAM or VOR) must be available on a certain timetable, “including the early view version.”  This language could be interpreted to mean that authors must always deposit a preprint, even when they are also depositing the AAM or VOR.  This is unnecessary and adds a needless burden for researchers.

Suggestions for the Gold Open Access (Journal) Compliance Option

  • Require both DOIs and ORCIDs. The implementation guidelines require eligible journals to use DOIs, but merely recommend that they use ORCIDs. ORCIDs and DOIs are both critical infrastructure that can help us realize the potential of the digital age to improve access to science and scholarship, and Plan S should capitalize on the opportunity to widen use of both. Requiring ORCIDs would align with the view of several members of the Plan S coalition who signed an open letter (December 6, 2018) calling on funders to user ORCIDs and committing to using ORCIDs themselves.
  • Provide financial support for no-fee OA journals, as coalition members are already willing to do for fee-based (APC-based) OA journals. The implementation guidelines acknowledge the existence and value of no-fee OA journals (“cOAlition S explicitly acknowledges the importance of a diversity of models and non-APC based outlets”). Providing support for no-fee OA journals will avoid the perverse effect of giving no-fee journals an incentive to start charging fees. While the coalition thinks about the best specific ways to support no-fee OA journals, it could offer financial support in general terms, as it does now for OA infrastructure.

We appreciate the opportunity to comment on Plan S, and look forward with great optimism regarding its potential to advance equitable access to science and scholarship for all.


Chris Bourg, Director, MIT Libraries
Amy Brand, Director, MIT Press
Greg Eow, Associate Director for Collections, MIT Libraries
Ellen Finnie, Head, Scholarly Communications and Collections Strategy, MIT Libraries
Peter Suber, Director, Office for Scholarly Communication, Harvard University