Wednesday, October 16, 2013

Canada's tricouncil draft open access policy: my comments

Canada's tricouncil funding agencies (NSERC, CIHR and SSHRC) have posted a consultation on their draft open access policy. Comments are due by December 13, 2013. Following are my comments.

Kudos to the tricouncil for an overall sensible draft open access policy that in many respects can serve as a model for other funding agencies.

Overall

One of the strengths of the policy is that open access publishing is an option, not a requirement, with use of grant funds allowed for this purpose. The reason this is a strength is that this permits market forces to operate. This is important to address a long-term market dysfunction in which a few large commercial scholarly publishers enjoy hefty profit margins in the range of 30-40% in an inelastic market that does not respond to market signals such as deep cuts to universities and research budgets, while on the other hand a significant portion of scholarly journal and monograph publishing requires subsidy. Today's technology makes it possible to conduct high quality scholarly publishing (peer review coordination and editing) at a fraction of the costs of some of the large commercial publishers, suggesting that it is very much worthwhile pursuing a competitive market so that funding can be re-directed from paying for publication to more money for research per se, as well as new needs such as research data retention, management and preservation. See the Appendix to this letter for more detail.

3.1 Option #1: Grant recipients submit their manuscript to a journal that offers immediate open access to published articles, or offers open access to published articles within 12 months.

Suggested change - add: and deposit a copy of the final peer-reviewed full-text manuscript in a Canadian open access archive on acceptance for publication, with open access delayed if necessary. 

Rationale: option #1 is not sufficient to ensure ongoing open access. The policy applies to grantees, not to journals. If a grantee meets the requirements by publishing in a journal that fits the criteria for this option, there is nothing to stop the journal from subsequently changing journal policy, for example changing its business model to toll access or extending the embargo period. This is true even for journals that use Creative Commons licenses; these extend and clarify permissions downstream but pose no obligations on the original copyright holder.

Option #2: Grant recipients archive the final peer-reviewed full-text manuscript in a digital archive where it will be freely accessible within 12 months (e.g., institutional repository or discipline-based repository). It is the responsibility of the grant recipient to determine which publishers allow authors to retain copyright and/or allow authors to archive journal publications in accordance with funding agency policies.

Suggested change to:

Option #2: Grant recipients archive the final peer-reviewed full-text manuscript in a Canadian-based digital archive on acceptance for publication where it will be freely accessible within 12 months (e.g., institutional repository or discipline-based repository). It is the responsibility of the grant recipient to ensure that they retain rights to archive journal publications in accordance with funding agency policies.

Rationale:

Ensuring ongoing open access to Canadian works is best served through archives based in Canada. Archives based elsewhere are subject to funding contingencies that are not under the control of Canada and access to archives outside of the country could be impacted by forces not under the control of Canada or Canadians. For example, if a grantee fulfills the requirement of this policy by depositing a copy of a work into an archive in a foreign country, it is within the realm of the possible that the foreign archive could be under the control of a country with which Canada is at war or in a trade dispute.

To facilitate compliance, authors should be required to deposit a copy of their post-peer-reviewed manuscript on acceptance for publication, with delay in open access if necessary. This greatly simplifies the process of accountability.

The original statement places more emphasis on publisher rights than is warranted. Research funding agencies have the right to set policies with which grantees must comply, and copyright begins with the author. Authors should insist on rights retention rather than passively complying with publisher policy.

Requiring deposit in a local open access archive is not an onerous burden, and is much less of a burden today than it was a few years ago. The majority of Canada's universities already have institutional repositories in place, and all Canadian research institutions will need to develop repository services in the near future, whether for this policy or for other purposes. For example, every institution with graduate students either has, or will soon need, a repository for electronic deposit of theses. Every institution with researchers will soon need one or more repositories for storing research data. A few years ago development of an institutional repository required local expertise and considerable investment, but today hosted solutions are available and affordable, and smaller institutions can share repositories and save quite a bit on the costs.

Embargo period (option # 1 and option # 2): suggest 6 months, not 12

The embargo period of 12 months is, in my opinion, far too generous. Scholarly publishers have had more than a decade to adjust to open access. The purpose of scholarly research is to advance our knowledge and serve the public interest; any financial benefits to scholarly publishers is incidental and should be given an accordingly lower priority. The embargo period should be shortened to 6 months, with an indication of review with a view to eventually eliminating the embargo period.

Another suggestion

One suggestion to facilitate both acceptance and compliance would be for the tricouncil to accept URLs to works in institutional repositories in CVs for grant application purposes. This would mean that the small steps needed to make works open access would also free up time for researchers.

Thank you for a good model for open access policy and for the opportunity to participate in this consultation.

Heather Morrison
Assistant Professor
University of Ottawa School of Information Studies


Appendix: how researchers can save money when open access publishing is a choice

Researchers can save money by selecting more cost effective options such as publishing in one of the majority of open access journals that do not charge article processing fees. Of the close to 10,000 fully open access, peer-reviewed scholarly journals listed in the Directory of Open Access Journals, 6,555 or approximately two-thirds have "no article processing charge" (DOAJ, 2013). Researchers considering open access journals that do charge article processing fees have an incentive to prioritize journals equivalent in scholarly value but lower in cost where possible, as seeking lower costs means more research funds can go to other priorities such as support for student training and purchase of necessary equipment. Today's technology has made it possible to complete the important scholarly work of coordinating peer review and editing at a very high level of quality at a fraction of the cost associated with some of the large commercial scholarly publishers. According to my analysis (Morrison, 2013), the average global expenditure by university libraries for each peer-reviewed scholarly article is approximately $4,500. To illustrate the potential for savings, note (merely as one of many examples) that the profitable Hindawi manages the process charging article processing fees as little as a tenth of this amount.

References

DOAJ (2013). Browse by publication charges. Retrieved October 16, 2013 from http://www.doaj.org/doaj?func=byPublicationFee&uiLanguage=en

MORRISON, Heather. Economics of scholarly communication in transition. First Monday, [S.l.], may. 2013. ISSN 13960466. Available at: <http://firstmonday.org/ojs/index.php/fm/article/view/4370/3685>. Date accessed: 16 Oct. 2013. doi:10.5210/fm.v18i6.4370.

No comments:

Post a Comment

Thank you for your comment. Comments on IJPE are moderated.