UK House of Lords short enquiry into open access: my response

Update February 28: the House of Lords has released its report. Excerpt:

The House of Lords Science and Technology Committee has today criticised Research Councils UK’s (RCUK) for failures in its communication of its open access policy. The report says the previous lack of clarity about RCUK’s policy and guidance was ‘unacceptable’.

Update January 28 - the House of Lords just released the evidence received to this consultation. My final comments begin on page 165. 

Update January 17 - note that I am working on a revised submission to conform to the Select Committee's guidelines and to correct some potential misinterpretations.

Important addition January 17 - it has come to my attention that my section addressing maximum embargoes could be misinterpreted as suggesting a recommended embargo. To be clear, my words are meant to address the absolute maximum for disciplines in the humanities and social sciences where UK based traditional journals do not yet have experience with the common practice of providing free access to back issues. My revised recommendation reads:

On maximum embargoes: an industry norm of free back issues to scholarly society journals about a year after publication appears to be emerging. For this reason, I recommend that a year's embargo be considered as the absolute maximum across the disciplines. The current 6-month embargo in STM should be retained, and all advice to publishers should clearly indicate that the practice of allowing embargoes is to facilitate a transition to full open access, and that the eventual goal is to gradually reduce and then eliminate embargoes. Embargoes are a concession to existing publishers; the public has a right to access the results of publicly funded research with no delay. (added Jan. 17, 2013).

There will be other changes in my final revision, however this is a particularly important one to note.

The UK House of Lords Select Committee on Science and Technology has launched a http://www.parliament.uk/business/committees/committees-a-z/lords-select/science-and-technology-committee/inquiries/parliament-2010/open-access/short enquiry into open access. Following is my response.

Heather Morrison, M.L.I.S., PhD

Coordinator, BC Electronic Library Network

Adjunct Faculty, University of British Columbia (scholarly communication & open access)

hgmorris at sfu dot ca

House of Lords

Select Committee on Science and Technology

hlscience at parliament dot uk

January 16, 2013

House of Lords Select Committee on Open Access: Call for Evidence

This is an individual submission, from an open access advocate and scholar focused on the area of scholarly communication and open access. In November 2012, I defended my dissertation, Freedom for scholarship in the internet age <https://theses.lib.sfu.ca/thesis/etd7530> which addresses some critical issues of relevance to open access policy development.

Highlights

The long-term leadership of the UK and the House of Lords in open access is acknowledged and applauded. It is recommended that researchers always be required to deposit work in UK based repositories, even when publishing work in open access venues, to ensure that UK funded research never becomes unavailable or unaffordable to people in the UK.

My research delves into mapping open access with the Creative Commons licenses, finding that, despite superficial similarities, the CC licenses are useful tools but no CC license is synonymous with open access and each license element has both useful and negative implications for scholarship. For example, allowing derivatives and commercial uses to anyone downstream will not always be compatible with research ethics requirements. A participant in a weight loss study giving permission to use a photo for a scholarly journal cannot be assumed to have granted permission for anyone to use this photo in a commercial advertisement. I recommend replacing the requirement that funded articles use the CC-BY license with a statement that when RCUK funds for open access publishing are used, there should be no restrictions placed on educational or research uses of the works.

As an open access advocate, I recommend against block funding for open access article processing fees, as this will interfere with the market, raising prices that will result in loss of support for this approach outside the UK, disadvantaging the very publishers who think that this approach will benefit them. Instead, I recommend that the UK follow the policies of the U.S. National Institutes of Health and Canada’s Canadian Institutes of Health Research in allowing researchers to use their research grants to pay open access article processing fees.

I suggest providing some funding to provide infrastructure and support and/or subsidies to assist scholarly society publishers, a common practice at university libraries throughout North America, and I further recommend that the UK set aside some seed funding to fund the future, that is, the next generation of scholarly communication, overlay journals built on institutional repositories, an area where the UK is well positioned to play a leadership role.

Finally, I present some data of relevance to the question of maximum permissible embargoes before works can be made open access. It can be argued that a new norm of scholarly journals providing free back issues on a voluntary basis, typically within a year of publication, has emerged in the past ten years. This is such a widespread and growing practice that the lack of evidence of harm to these journals is in itself evidence that a one-year’s embargo causes no harm to journals relying on subscriptions, even when all articles in the journal are made freely available. Therefore I suggest that it would be quite appropriate to set a maximum embargo of no more than one year regardless of discipline. Thank you very much for the opportunity to participate in this consultation.

Details

1.         Open access to scholarly knowledge, as noted in the 2002 Budapest Open Access

Initiative, is an unprecedented public good, an opportunity to make scholarly works as useful as they can be, and an opportunity for all, including those from the developing world and poorer regions in the developed world, to benefit from and participate in a global scholarly conversation. Congratulations and thanks to the UK for long-standing leadership in the area of open access. The intent of the RCUK Open Access Policy is a welcome push towards even greater support for open access. My submission is intended to address a few areas where minor changes could substantially improve this strong leadership in the push for the next phase of open access.

2.         I have explored the question of the mapping of Creative Commons licenses with open access and have come to the conclusion that despite superficial similarities there are important differences. Therefore, I recommend against equating open access with Creative Commons licenses, and particularly with the CC-BY license. My work on this topic will not fit within the 6-page limit, but the following points feature a few highlights.

3.         The RCUK policy states:  The Research Councils will recognise a journal as being compliant with their policy on Open Access if 1. The journal provides via its own website immediate and unrestricted access to the publisher’s final version of the paper (the Version of Record), and allows immediate deposit of the Version of Record in other repositories without restriction”.

Comment: It is not sufficient to insist that a journal allows deposit in other repositories. Researchers should be required to deposit in other (UK-based) repositories for open access. The reason is that the RCUK policy applies to researchers, not to publishers or journals. The Creative Commons licenses are means by which license holders can relinquish certain rights that they have under copyright, which do not place any obligation on the licensor. If a researcher publishers in an open access journal but fails to deposit in an open access repository, then if the journal or publisher ceases to make the work open access, access to the work could be lost to the UK research community.  For example, an open access publishing company could be sold to another company which could issue the same works under toll access only, with all rights reserved, and no obligation to sell products at prices UK universities can afford.

4.         As an open access advocate, I argue that the CC-BY is not optimal as a default for open access, but on the contrary a weak and problematic license that raises the possibility of wholesale loss of open access downstream. See point 3 above regarding the example of an open access publishing company selling to another company that decides to go with a toll access model.

In addition, there are valid scholarly reasons why CC-BY cannot be used with every research article. CC-BY is not always compatible with research ethics. An example is a person whose picture is taken and published as part of a study on weight loss. The researcher’s right to publish this picture (with permission) does not give the researcher a right to grant anyone, anywhere blanket permission to re-publish the picture for commercial purposes (such as using it in a weight loss ad), as use of the CC-BY license does.

It is not uncommon for scholars to use works created by third parties in their works, in which case copyright belongs to the third party and scholars cannot grant permission to others to use these works.

Scholars may have valid reasons for preferring that NoDerivatives be specified. The quality of work reflects on a scholar in a way that can make or break a career, and a poor quality derivative could reflect poorly on the scholar. Third parties whose work is included in scholarly articles may have similar concerns about quality with derivatives.

The CC-BY provision in the RCUK policy places both scholars and journals in an awkward position, with a choice of violating the rights of others, or refraining from using works that truly are useful and could be included, just not under the CC-BY license.

For all these reasons, I recommend replacing the requirement that funded articles use the CC-BY license with a statement that with a statement that when RCUK funds for open access publishing are used, there should be no restrictions placed on educational or research uses of the works (RCUK policy point one). This should be sufficient to prevent such potential abuses by publishers as retaining rights to charge for coursepacks or for use by commercial companies, while avoiding the potential abuses of the works per se opened up by CC-BY.

5.          As an open access advocate, I argue against block grants for open access article processing charges to universities. This is an interference with the market that I consider to be against the interests of open access, and even the open access publishers that one might think would benefit from this policy, although I doubt that they would agree.

One reason is because the way these grants have been set up encourages high prices for article processing fees. It is highly unlikely that other countries will follow the UK’s lead on this. The UK is aiming to protect a positive balance in trade, while for virtually every other country, the incentive is in the opposite direction, i.e. for most countries scholarly publishing involves a negative balance in trade, and propping up the existing system is counter-productive.

If prices for open access article processing fees are inflated due to the RCUK’s generosity, this is a disincentive for voluntary initiatives to support article processing fees. Ross Mounce recently released some research illustrating a 5% increase in BioMedCentral’s open access article processing fees over the past year. This is above inflation, and for library subscriptions, an increase of this amount would be sufficient to trigger a “review before renewal” decision, and possibly cancellation. In other words, RCUK’s generosity in its support of open access via article processing fee publishers may actually cause a drop in support for this publishing model from outside the UK. What I recommend is the approach of the NIH and the Canadian Institutes of Health Research, that is, allowing researchers to use grant funds to pay open access article processing fees, but not providing targeted funds. This is an approach that provides support in a manner that allows the market to do its work. 

6.         To support scholarly societies in making a move to open access, I recommend subsidizing scholarly and university press publishing either directly through block publishing grants or indirectly through providing infrastructure support for their publishing. Direct subsidy is much more cost-effective than indirect subsidy through APFs. One example of such a program in Canada’s Social Sciences and Humanities Research Council’s Aid to Scholarly Journals, and the Synergies project that has provided support for Canada’s social sciences and humanities journals to move online and facilitated the move to open access for a number of these journals. In North America, it is common for university libraries to provide hosting and support services for journals local faculty are involved with, an option that facilitates high quality open access publishing at a modest cost. 

7.         Fund the future. To prepare the UK for leadership in scholarly communication in the future I recommend providing some seed funding for the most transformative possibility for scholarly communication reported by Houghton and Oppenheimer in 2009 http://www.jisc.ac.uk/news/stories/2009/01/houghton.aspx, that is, building an overlay journal system on top of institutional repositories. Britain’s strength in institutional repositories makes the UK a natural leader in this area. UK-based mathematician Timothy Gowers recently posted about a new system that will make it easy to create overlay journals with arXiv, suggesting that this approach may be feasible much sooner than most of us had thought: http://gowers.wordpress.com/2013/01/16/why-ive-also-joined-the-good-guys/

8.         On maximum embargoes: an industry norm of free back issues to scholarly society journals about a year after publication appears to be emerging. For this reason, I recommend that a year's embargo be considered as the absolute maximum across the disciplines. The current 6-month embargo in STM should be retained, and all advice to publishers should clearly indicate that the practice of allowing embargoes is to facilitate a transition to full open access, and that the eventual goal is to gradually reduce and then eliminate embargoes. Embargoes are a concession to existing publishers; the public has a right to access the results of publicly funded research with no delay. (added Jan. 17, 2013).

            
           The following is intended to illustrate why the permitted embargo should be no more than 1 year at absolute maximum for journals in the social sciences and humanities that do not yet have experience with free access practices. (Added Jan. 17, 2013). This practice of making back issues free is widely practiced by traditional publishers. There is no evidence that providing this access has caused any harm to the publishers. The large and growing number of journals following this practice supports my assertion that this is becoming the standard.

On March 16, 2004, traditional not-for-profit publishers in Washington, D.C. made a commitment to the Washington D.C. Principles For Free Access to Science - A Statement from Not-for-Profit Publishers <http://www.dcprinciples.org/>, through which "representatives from the nation’s leading not-for-profit medical/scientific societies and publishers announced their commitment to providing free access and wide dissemination of published research findings".

The publishing principles and practices supported by this group include:

“3.  As not-for-profit publishers, we have introduced and will continue to support the following forms of free access:…

The full text of our journals is freely available to everyone worldwide either immediately or within months of publication, depending on each publisher’s business and publishing requirements”

In other words, this very traditional group of scholarly society publishers committed, back in 2004, to making their journals freely available either immediately or within months of publication. There are a number of indications that this is now a common practice. There are now more than 8,000 fully open access scholarly journals, and many more that provide free access to back issues on a purely voluntary basis, frequently with a 12-month delay.

The extent of this practice may be best viewed in the Electronic Journals Library (EZB). The EZB is a collaborative project of 589 libraries, based in Germany that collects both subscription and free "scientific and academic full text journals". EZB currently includes 38,066 journals - close to 30,000 more titles than are listed in the Directory of Open Access Journals, which is limited to fully OA journals. Among the 30,000 journals are a very large number of journals that voluntarily provide free access to back issues, with no policy requirement. Details about the EZB can be found here:

http://rzblx1.uniegensburg.de/ezeit/about.phtml?bibid=AAAAA&colors=7&lang=en

A quick scan of the journals participating in the US-based Highwire Free hosting service illustrates that a 12-month embargo is very common for the society journals participating in this service:

http://highwire.stanford.edu/lists/freeart.dtl

The number of journals voluntarily contributing to PubMedCentral has been growing steadily, from 410 in March 2008 to 1,464 at the end of 2012. Of these, over 1,000 voluntarily provide all content for immediate free access. Data is from The Dramatic Growth of Open Access:

http://poeticeconomics.blogspot.ca/2012/12/dramatic-growth-of-open-access-2012.html

A key point is that this data illustrates that a great many traditional scholarly journals have made the decision to provide free back to their journals with a minimal embargo period, with one year being common, and no embargo at all being not at all unusual. Many such journals made this step with some concern about the potential impact on their subscriptions and revenue. If there had been dire consequences for such journals, there would be plenty of data today to demonstrate that providing free access after a brief embargo harms subscriptions. No such data has ever been brought forward to my knowledge. Providing free access to back issues appears to be rapidly becoming the norm for scholarly journals, and so I recommend a maximum embargo for OA policy of 12 months.

Thanks once more for the opportunity to participate.

Heather Morrison, PhD

Freedom for scholarship in the internet age

https://theses.lib.sfu.ca/thesis/etd7530

Following are details about the short enquiry from the House of Lords' website

This short inquiry will focus on the implementation of the Government Open Access policy.

The Committee will consider a range of concerns including:

• embargo periods;
• arrangements for article processing charge (APC) funds;
• international issues; and
• risks for learned societies.

The Committee expect to produce an output in mid-February, to inform the development of Research Council UK’s policies.
The Committee has issued a targeted call for evidence to key stakeholders for this short inquiry, any party interested in submitting written evidence should contact the Clerk to the Committee on hlscience at parliament dot uk. The deadline for submissions is Friday 18 January 2013.