The changing role of research funders in open science?

Over the Easter weekend, the European Commission released a call for tenders for “Open Research Europe”. This is to be a new publishing platform that enables recipients of H2020 research grants to publish their research open access without paying an APC. It follows similar initiatives by the Wellcome Trust, UCL, Health Research Board Ireland, the Gates Foundation, Montreal Neurological Institute and others. It appears that the problems that these funders and research institutions hope to solve by setting up these platforms include several or more of the following:

  1. the transition to open access is proceeding too slowly
  2. open access using the APC model is too expensive
  3. the APC model does not work well for research funded by time-limited grants
  4. the APC model can be administratively complicated for many of those involved
  5. wider shifts toward open science, including more sharing of research data and more transparent research indicators, are not gaining enough traction

Would a research funder publishing platform stand a good chance of solving these problems? Might it create new ones?

But first a bit of history. Research funders have established systems and platforms as part of their operations, and to help their grantees, for a long time. For example, PubMedCentral was started in 2000, and UK PubMedCentral (now Europe PubMedCentral) in 2007. Taking a UK parochial view for a moment, the UK Research Councils (now part of UK Research and Innovation) have used a variety of systems to help make their research more accessible, most recently via Gateway to Research and Konfer. However, the difference now is that funders are showing interest in moving up the supply chain, to intervene earlier in the dissemination of research outputs. Whereas the Gates, with Chronos, took one approach in line with the APC model, others are perhaps looking to build on the current momentum of preprint services. There has been a long-held view that “preprints plus a journal overlay” might be a model worth exploring (for example, see the RIOJA project from 10 years ago), and this is effectively what F1000 Research has developed and which lies behind many of the recently developed funder platforms.

Certainly having a publishing route that eliminates the need for financial transactions involving the author does reduce the administration for her and often for her institution, but sometimes the “big deal” models that include an OA component (rather than individual APCs) can cause problems for funders who need transparency in the use of their grants. Funders can also struggle to design instruments that enable payments to be made after the research grant has ended, and associating an output unambiguously with a grant can be difficult for researchers anyway, because research is often not done in discrete projects but builds on previous work in different ways. So, while a researcher lives in this messy reality, she is asked to help funders tell stories about the success of their grants and institutions tell stories about the success of their departments. While these stories rely on a close attribution of a research output to a particular grant or department, often authorship and attribution are tactical affairs. Funder publishing platforms like the EC’s that, unlike journals and book publishers, are only open for outputs from particular grant programmes, can only ever support one dimension of that messy reality, unless there is some concerted effort to link them up with others, and with institutional systems. Of course, such efforts are possible and have been proposed.

The funder publishing platforms will only accelerate moves toward open science either if they are adopted spontaneously by researchers, or if there is some degree of incentive applied by funders. There is some evidence from the Wellcome Trust of spontaneous adoption – if Wellcome Open Research had been a journal, then in 2017 it would have been the fifth most popular journal for Wellcome grantees, which is impressive so early in its life. The question of whether funders, as agencies that can add terms and conditions to their grants, might incentivise use of the platform remains open for now. It would need to be done very carefully, as there appears to be potential for conflicts of interest and other issues to arise.

Will funder publishing platforms make open access more sustainable? The EC call for tenders gives an indicative estimation of the costs at €1m over four years [*see comment below], to publish at least 5,600 research outputs, which equates to around €180 each – much cheaper than virtually any APC on the market. Journals publishers may, with justification, say that the comparison is not a fair one, and that the services they offer go well beyond those offered by funder publishing platforms such as Open Research Europe. Whether that €180 figure turns out to be accurate is not yet clear, but it could be a challenge to many journals if it did turn out to be an example of “low end disruption”, including to journals such as PLOS ONE that were once themselves the disruptors.

But I wonder whether there is a bigger opportunity here? Anecdotally, I have heard concerns from publishers that their manuscript submission systems can struggle to keep up with rapidly evolving requirements. We are certainly aware that authors deeply resent having to start from scratch every time they re-submit a paper to a new journal, if it is rejected from their first choice. We are aware of a burgeoning range of preprints services and, now, funder publishing platforms. We know about notification services and routers (eg, Jisc and OpenAIRE). We have ORCID. Surely it cannot be beyond us, collectively, to design some technical and data standards to allow preprint-like services to act as the submission point for research papers, allowing both overlay open peer review and/or easy routing of those papers to established journals by agreement with the authors and journals? Has this been done already? Jisc is a member of NISO; I may see whether they know of such an initiative or could be interested.

4 thoughts on “The changing role of research funders in open science?

  1. Pablo de Castro

    Dear Neil,
    Thanks for sharing your thoughts on the potential implications of the ORE call.
    This is to point out that the actual budget for the tender is €6.4m as mentioned in the Information Note at (and also in the contract to be signed by any bidder). The figure of €1m that you mention is the budget the EC are making available for the initial development/tailoring of the platform that will support the ORE services. The subsequent average ‘hidden APC fee’ is significantly higher than this €180 figure you estimate.
    My main concern from an institutional viewpoint is that early attempts to test the water with our researchers show that they tend to have issues with a deliberately JIF-less publishing venue like the Wellcome Open Research because it clashes with the research assessment framework they’re used to live with. DORA would need to be more effectively promoted (and especially adopted) in combination with these platforms.
    Another aspect I’d like to mention is that although these funder-driven platforms are funded from the public purse, they’re not particularly friendly to institutions (i.e. they’re the typical publisher platform in that it’s not possible — as far as I can see — to filter articles by institutional affiliation in order to find out if any of ‘our’ Wellcome-funded researchers has already ventured to submit her manuscript there). Again, given their publicly-funded character and also the very specific aims these platforms are supporting, it might make sense for them to join the data provider queue for brokering services like the Jisc Publications Router asap (perhaps even to be allowed to skip the queue altogether!)
    Finally, regarding the suggestion for further system integration (I believe this is what the idea for preprint-like services to act as default submission points is at least partially about), there’s two steps, the first submission itself and the later ‘cascading’ as publishers call it. Given the amount of metadata that manuscript processing systems are typically requiring (funding info comes to mind), the first bit is bound to be a hard ask. There could however be clear options here for streamlining the process for the second step since any required metadata are likely to be already in the system.

    1. Neil Jacobs Post author

      Thanks Pablo, yes i misread the documents, so the implied per article fee then is nearer €1150, which is much less of a challenge to pure gold OA megajournals like PLOS ONE (though still cheaper). Apologies to all for getting that wrong.
      It is disheartening that the JIF still weighs so heavily among researchers. The prestige of the publishing venue is, of course, important to authors, but clearly there is a way to go to get DORA and the proposals in the Leiden Manifesto widely accepted and adopted.
      I would hope that proper use of trusted identifier registries such as ORCID (and soon others, we hope), plus notification technologies, ought to help us with the streamlining of data flows that I think everyone is looking for. What I really hope is that we keep looking at the big picture across stakeholders, and that the author is not forgotten in this streamlining.

  2. Ashley Farley

    Hi Neil,

    Thank you for your post!
    I wanted to offer some clarity from a funder’s perspective.
    Chronos is a system that we built to help manage the payment & compliancy process for our open access policy. It is not a standalone publisher or repository. It is a way for the foundation to easily track grant research outputs in a way we have not been able to do so previously. The system also allows for easy resubmission of articles to different journals no matter the publisher.

    Overall, you outline great reasons why funders are going this route. I would like highlight that these platforms still charge an APC that is (in the Gates foundation case) covered by the funder. We realize that there is a cost associated with publishing – however, typically APC pricing is inflated and does not reflect the actual cost. I like to think of these endeavors as publishers putting their brand and clout behind a different model of publication. One that supports transparency and speed, to ensure maximum impact of research grant funding outputs. These platforms are restricted to grantees and sub-grantees, so publications are tied to a grant the foundation has done due diligence to give.

    One way to avoid grant end dates is for the foundation/institution to create a centralized budget to cover APC’s. These alleviates the financial burden from authors.

    The funder publishing platforms that are built on the technology of F1000 benefit from their expertise and partnership support. The F1000 team invests resources heavily in working with grantees, educating them on the publishing experience, as well as, assisting in the editorial process. So, I think “spontaneous adoption” is a very small piece of the puzzle – both the funders and F1000 are very serious about ensuring success of the platform and the individual articles and have completed outreach to reach these goals.

    Hope this helps in the discussion! I recommend watching the work of the Open Research Funders Group for more progress in this space.

    1. Neil Jacobs Post author

      Thanks for the additional information Ashley, and I will continue to watch the Open Research Funders Group with interest. You mention that papers that go through Chronos can be resubmitted to other publishers easily; this is interesting, as that is a really hard use case. However, since writing the post, I have spoken with Todd and Nettie at NISO, and they pointed me to Manuscript Exchange Common Approach (MECA)
      This looks exactly what I was describing in my last paragraph, so apologies to those who already knew about it. Potentially a significant development, and I hope Jisc can help in some way.


Leave a Reply

Your email address will not be published. Required fields are marked *