Wednesday, March 14, 2012

Research Councils UK draft new open access policy: my comments

Following are my comments to the Research Councils U.K. on the their proposed new open access policy

Dear RCUK Open Access Policy group,

First of all let me say congratulations and thank you to RCUK for your continuing inspiring leadership on open access policy. Following are my comments, based on many years of experience in open access policy advocacy, my work as a professional librarian and adjunct faculty at the University of British Columbia's School of Library, Archival and Information Studies, where I have developed and taught courses on scholarly communication, and my doctoral studies (communications, in progress) in the area of scholarly communication and open access.

Overall, from my perspective this draft policy introduces two important innovations: reducing the permitted embargo period, and pushing towards libre open access (e.g. allowing use for data and text-mining). In brief, I recommend strengthening the language on shortening embargo periods, and eliminating reference to CC-BY in favor of broader language against restrictions and requiring formats usable for text and data-mining purposes. Also, I recommend that the policy specify immediate deposit, with optional delayed release to accommodate the permitted embargoes.
With respect to the embargo period, I recommend strengthening the language indicating that any permitted embargo periods are designed as a temporary measure to give publishers time to adjust to an open access environment, with a view to eventually requiring open access immediately on publication. This language can be found on page 4, I recommend including this in the introductory language to underscore this point.

Kudos to RCUK for adopting a leadership position on libre open access.  However, I would recommend against specifying the Creative Commons CC-BY license. While many open access advocates understandably see CC-BY as the expression of the BOAI definition of open access, my considered opinion is that CC-BY is a weak license for libre OA which fails to protect OA downstream and will not accomplish the Budapest vision of open access,. My perspective is that the best license for libre open access is Creative Commons - Attribution - Noncommercial - Sharealike (CC-BY-NC-SA), as this protects OA downstream (recognizing that the current CC NC definition is problematic, and noting that commercial rights should be retained by authors, not publishers). As one example of where open access might need such protection, because CC-BY allows for resale of open access materials: if all of PubMedCentral were CC-BY, a commercial company could copy the whole thing, perhaps add some value, and sell their version of PMC. They could not legally stop PMC from providing free access. However, I very much doubt that CC-BY could prevent such a company from lobbying to remove funding for the public version. If this sounds ludicrous and unconscionable, may I present as evidence that just such a scenario is realistic: 1) the efforts a few years ago by the American Chemical Society to prevent the U.S. government from providing PubChem on the grounds that this was competition with a private entity; 2) the Research Works Act, and 3) the current anti-FRPAA lobbying in the U.S., which, similarly to the Research Works Act, claims that published research funded by the public is "private research works" which should belong solely to the publisher.

Another reason for avoiding CC-BY is that while the contributions of funders are very important, so are the contributions of scholar authors. Many scholars do not wish to see others who have contributed nothing to a scholarly work sell their work and pocket the money; I certainly don't. For example, Peter Suber recently posted this note to the SPARC Open Access Forum which expresses the distress of an author who published CC-BY in a BMC journal and then found a bogus publisher selling her article for $3. https://groups.google.com/a/arl.org/group/sparc-oaforum/browse_thread/thread/fc977cabd0d59bcc#.  The more work that is published CC-BY, the more I believe we can expect to see this kind of scam, and this risks turning researchers off OA. Also, when faculty members develop their own open access policies (e.g. Harvard, MIT), they insist that articles not be sold for a profit. Links to these and other institutional repositories are available through the Registry of Open Access Material Archiving Policies (ROARMAP) at http://roarmap.eprints.org/.

To illustrate how CC-BY does not necessarily result in the Budapest open access initative's vision of "sharing of the poor with the rich and the rich with the poor": those who give away their work for commercial purposes may not be able to afford the results. For example, if a scholar from a poorer area gives away their medical articles as CC-BY, images and other elements from these articles could be used to develop point-of-care tools that could be sold at prices that the health care professionals serving the scholar and their families could not afford. That is, despite the best of intentions, CC-BY could easily result in a one-way sharing of the poor with the rich. This is one of the reasons why I strongly recommend that the developing world avoid CC-BY.

I cover this topic in more depth in the third chapter of my draft thesis - from the link below, search for open access and creative commons:
http://pages.cmns.sfu.ca/heather-morrison/chapter-3-open-access-as-solution-to-the-enclosure-of-knowledge/

For practical reasons, to further text and data-mining I would suggest that the article format is more to the point than licensing. An author's final manuscript may be more likely to meet this requirement than the so-called "Article of Record". For example, an author's own version in an open format that allows for text and data-mining, with no licensing language, is much better for text and data-mining purposes than a publisher's "Article of Record" in a locked-down PDF format with a CC-BY license. My recommendation is to specify useable format rather than license. Also, I would recommend against encouraging deposit of the "Article of Record", as scholarly communication needs to evolve beyond the print-based journal article format, and this specification may tend to further entrench a system that needs some shaking up.

Regarding p. 5 - working with individual institutions to develop open access funds from indirect costs - good!!! I recommend looking at the Compact for Open Access Publishing Equity http://www.oacompact.org/compact/ for guidance, and for institutions to join. When such funds are developed, it is very important to build in efficiencies to prevent against double dipping, avoid paying excessive costs, and planning for education about the growing pool of open access scam companies is an area that is growing in importance. I differ from some of my colleagues in recommending against funding agencies paying OA article processing fees.

What RCUK might want to consider if, similar to North America, some of the publishers experiencing difficulty transitioning are the smaller society publishers, is a journal subsidy program. Canada's Social Sciences and Humanities Research Council has such a program, called Aid to Scholarly Journals. If RCUK does not yet have such a program, that would make it much easier to start up with stronger OA expectations than SSHRC has been able to do to date. Canada also has a program to help scholarly journals transition to the online environment called Synergies which is a good model. In North America, most academic libraries nowadays are providing journal hosting and support services. This sector is by far the most efficient in scholarly publishing, with costs on average less than 10% of the current system. See chapter 4 of my draft thesis for details http://pages.cmns.sfu.ca/heather-morrison/chapter-4-economics-of-scholarly-communication-in-transition/).

Finally, a minor point: the introductory paragraph, talking about benefits of open access, appears to prioritize business interest. I fully agree that scholarship and open access to scholarship is a huge potential benefit to business, but would submit that this is not, nor should it be, the main point of scholarship and research. May I suggest that the final sentence of the first paragraph refer to the public first and foremost, and then perhaps speak to business benefits?

Many thanks for the opportunity to comment, and best wishes to RCUK in the next stage of your leadership on OA policy.

Heather Morrison
Doctoral Candidate, Simon Fraser University School of Communication
http://pages.cmns.sfu.ca/heather-morrison/