The Canadian Research Knowledge Network (CRKN) welcomes the opportunity to provide feedback on the recent revised Tri-Agency Draft Open Access Policy on Publications. The draft revised policy is a critical mechanism for establishing open access as the default for Canadian research publications and a step towards a sustainable open science ecosystem in Canada. In the short window allotted for gathering feedback, CRKN reached out to our broad membership across the academic library community to gather their input, which our Board of Directors has also reviewed. We are pleased to provide the following feedback that we hope the Tri-Agency will take under consideration.
Diamond Open Access
With the requirement to deposit all funded articles in an institutional repository (IR) regardless of open access status, articles published in fully open access diamond journals (no subscription costs, no author payments) no longer satisfy the requirements of the policy. This appears to be at odds with the recent changes announced for the SSHRC Aid to Scholarly Journals program which are deliberately supportive of the diamond model. Canada has a strong diamond open access ecosystem, particularly in the humanities and social sciences. The proposed revisions to the policy would make the Canadian journals in the robust diamond ecosystem non-compliant with the Tri-Agency policy, despite their fully open access status and the lack of article publication charges payable by researchers. CRKN recommends that funded articles published by Canadian researchers in diamond open access journals should be considered compliant without requirement of deposit into IRs.
Author Rights Retention and Pre-Prints
The Tri-Agency’s decision to support author rights retention is a significant improvement in the revised draft policy. In particular, the requirement placed on authors to declare their intention to retain rights over their work when submitting articles for publication provides CRKN with increased leverage during negotiations with for-profit commercial publishers.
However, the exception in the policy that permits the deposit of pre-print or “un-refereed” articles undermines this leverage, as publishers will likely deny unembargoed deposits of “version of record” (VOR) or “author-accepted manuscript” (AAM) copies, leading to the proliferation of articles in IRs that have not been vetted by the established peer review process. This possibility puts at risk the quality and reliability of Canadian research and may further erode the public trust in science. CRKN recommends that minimum compliance must require deposit of the AAM, and that pre-print copies should not be considered to be compliant. Furthermore, we suggest that the Tri-Agency consider adopting a standardized naming convention for article versions, such as NISO-RP-8-2008: Journal Article Versions.
With a stronger call for AAM or VOR as the version to be deposited, CRKN will have more leverage in negotiations with the largest journal publishers to assist their authors with compliance by developing workflows that can better support automatic deposit, rather than adding an extra burden on either researchers or the libraries at their institutions.
Discovery and Indirect Costs of Research
University libraries are a key component of the research process and in most cases act as hosts for institutional repositories, which are a critical element of the revised OA policy proposed by the Tri-Agency. Therefore, it is expected that increased workloads for libraries resulting from the new policy are inevitable, and that costs for managing the substantial growth of deposit in IRs will follow. CRKN notes that the discrepancy in capacity of smaller universities to support researchers in complying with deposit, as compared to the larger research institutions, will pose a significant challenge to compliance with the policy.
The scholarly communications community widely accepts that discovery of research deposited in IRs is a challenge. Although the revised policy is a significant step towards increasing the availability and openness of Canadian research, it does not address the question of how researchers and the public search for and obtain access to these materials. Ensuring interoperability and efficient discovery of material deposited in IRs will fall to the libraries managing those systems. In many cases there may not be resources within academic libraries to support those efforts, putting the goal to make publicly-funded research widely available at risk.
Naturally, we consider our member libraries’ investments in this area to be in keeping with the concept of indirect costs of research. Considering this, we recommend that the delivery of the revised policy be accompanied by messaging to encourage institutions to allocate a portion of indirect costs of research funds to support the enhancement of IR infrastructure currently funded only out of library budgets.
Compliance Verification
We hope that the Tri-Agency will consider adding compliance verification elements to the revised policy, insofar that requirements to deposit in IRs without a method for ensuring or monitoring compliance leaves a significant gap in our collective ability to measure the effectiveness of the policy and its effect on open access for Canadian research. For example, the Tri-Agency could consider requiring researchers to provide links to deposited articles that were funded under the new OA policy when submitting final reports and when applying for future grants. This approach would make verification of past compliance with the policy a criterion in awarding said grants. Ensuring that researchers flag OA publication in their application CV attachments, as they currently do for SSHRC-funded publications, would also support evaluation according to this criterion and raise awareness of the policy amongst researchers. Some CRKN members also indicated that compliance could be facilitated by the Tri-Agency encouraging universities to adopt Open Science policies at the institutional level.
Embargoes
Implicit in the draft revised OA policy is the elimination of any embargoes on articles deposited within institutional repositories. To strengthen the clarity of the policy and to assist authors and CRKN when negotiating with publishers, CRKN recommends explicitly stating in the policy that articles deposited into IRs must not be embargoed.
Persistent Identifiers
A broad-spectrum implementation of persistent identifiers (PIDs) including Open Researcher Contributor IDs (ORCIDs - author identifiers), Digital Object Identifiers (DOIs - research output identifiers, commonly applied to scholarly articles or data-sets), and funder IDs such as Crossref’s Open Funder Registry are essential to evaluate the impact of this OA policy, to track compliance over time, and to identify areas needing attention or improvement. There is already strong institutional support for PIDs in Canada thanks to the activities of two CRKN-supported consortia, ORCID-CA and DataCite Canada, and a national strategy toward implementing PIDs has been developed through the Canadian Persistent Identifier Advisory Committee (CPIDAC), which includes SSHRC and NSERC as members. The Tri-Agency can strengthen this OA policy by requiring that grant funded articles include an appropriate funding identifier, for example from Crossref’s Open Funder Registry, as part of the funder acknowledgement process. Additionally, the Tri-Agency can strongly encourage the inclusion of author ORCIDs in all stages of the research process (for example when applying for a grant; when submitting and publishing a manuscript; and when depositing content to repositories) so that this research activity is tracked, researcher labour is reduced, and policy compliance activities are strengthened. This would strengthen the ability to understand what Canadian scholarly articles are being published, cited, deposited and otherwise used and where that happens.