Study on an open access scholarly publishing cooperative

John Willinsky, founder and Director, Public Knowledge Project (PKP); Professor, Graduate School of Education, Stanford University.

Willinsky discussed the future of open access from the point of view of financial capital. The future of open access scholarly journals is a question of memory and heritage, but also of money, in terms of capital, of investment in scholarly publishing, which will neither necessarily make a profit in the near future nor generate a return comparable to the 30-40% profits made by major commercial publishers, though it has already begun producing benefits in terms of knowledge circulating in the public domain, not only on the local level, but also internationally.


1. Failures related to using a model of scholarly publishing inherited from biomedical sciences for the human and social sciences.

Scholarly publishing developed on the foundation of a very specific field, namely, that of biomedical science, which has very little in common with the human and social sciences, whether we are talking about the impact of a 12-month embargo, revision of articles or, first and foremost, the funding available.  The economic future of scholarly publishing that we are in the process of building is dominated by the biomedical science model alone, and this produces inherent, integrated, innate, a priori inequality that ensure that we will not have universal access to knowledge. In other words, speaking in economic terms, we will not have equitable distribution of the means of production. Instead, we need to begin by asserting the vital, indispensable nature of the human and social sciences, and take them as a point of departure in our reflection on the economic aspects of scholarly publishing.

The first problem with the green route for the human and social sciences is the loss of the value added by journals to articles. The green route, in other words, self-archiving of articles, designed for health sciences, leaves aside all the revision and re-writing that is the added value of scholarly publishing in the human sciences. For example, PloS One does not revise articles that are submitted. It simply selects manuscripts, and that alone is the guarantee of their quality. Since no bibliographical standards are applied, that section of articles is very irregular.  


2. The subscription system: problems and beliefs

Another problem with the green route is the 12-month embargo. Let us take an example from biomedical science itself. At Stanford University, to see whether research could be useful to an audience broader than that of researchers alone, we did an experiment on use of open access by doctors. We gave 350 doctors free access to the Stanford University library for 12 months. Two thirds of the participants did not use their access, but the third who did read on average one article per week. The articles those doctors consulted led them to change standards in their hospitals and create new policies, to verify and confirm that their practices were well-founded or, on the contrary, to question and then amend them. However, what we found most interesting was the age of the articles consulted. Half of the articles consulted had been published less than 12 months previously. This means that half of the articles could have be under the 12-month open access embargo. Thus, if we think that this two-track system is beneficial for the economics of knowledge because of the subscription income it brings in, we nonetheless have to realize that it is based on the (false) pretext that articles need to be exclusive and that is works by excluding certain readers.

Let us look at the issue of exclusiveness in another way. Is it important to the author, or for libraries and librarians? The University of British Columbia library considers itself to be a major provincial library: terminals are reserved for the general public so that everyone can use the collections. When they are arguing for funding for their institutions, it is not in librarians’ interest for some scholarly articles to not be available to all of the audiences targeted by the library, in other words, researchers, doctors, lawyers, teachers and the general public. This is why it is not in our interest to continue with a subscription system. Even major publishers agree on this, which is what had led them to create a form of open access based on publication fees paid by authors.  

This brings us to two forms of open access that would provide a way around the system of subscriptions to scholarly journals. Certainly, such subscriptions do provide a transfer of revenue, but both models of open access would provide perfectly equivalent transfers of revenue. The majority, if not the totality with only a few exceptions, of subscriptions are paid for by the libraries of educational institutions and not by individuals. The amount of money is enormous - $10 billion. It is spent mainly on subscriptions to science, technology and medical journals, while journals in the human sciences and letters have almost no impact on the total. In order to escape this system, this enormous amount of money could be invested in two forms of open access: (1) the author-pays model, based on an article processing charge (APC), or (2) the cooperation-based model.  


3. Open access: the author-pays model

The author-pays model is the dominant one among commercial publishers. It is a common system for science, technology and medical journals, and its roots can be traced back to the page-fee publication system, which dates from the eighteenth century and lasted into the 1950s and 1960s for physics and astronomy journals. From the point of view of the human and social sciences, we can wonder how to distinguish it from the vanity press model. How can corruption be avoided in such a system? In any case, it is the dominant financial model at this point. For example, PloS One, which functions according to this principle, published 30,000 articles last year, at $1350 per article, for total revenue of $40 million. Of the authors, 10% came from developing countries and were exempted from paying the fees. This model works and is attractive to the five major commercial publishers, who could convert to it. By dividing their revenue by the number of libraries that pay fees, they have found that they could publish in open access with an APC of $3000 per article. One of the advantages of the author-pays system is that it has established a degree of competition among scholarly journals by introducing a relationship between the price of an article and its quality, which is not the case under the subscription system. The APC model is a leading candidate for open access: it allows authors to retain ownership of copyright, it makes articles universally and immediately available, and it is simple. However, it has disadvantages: the fee exemption policy for authors from the developing world leads to some condescension; it results in financial penalties for the most productive authors, who will have to pay more APCs; and it will oblige the libraries of institutions to reserve funds for APCs so as to lighten professors’ financial burden, which could penalize institutions where professors publish a lot. According to the logic of this system, the purpose would be to gradually transfer the money reserved for subscriptions into funding for APCs. However, the central problem remains the dependency on the monopoly of commercial publishers: we should anticipate annual augmentations of APCs similar to the increases that these publishers have imposed on libraries for subscriptions. Thus, the open access system based on APCs has advantages, but is in no way a sustainable solution.


4. Open access: the cooperative model

According to Willinsky, the cooperative model is the best. Since the 2000s, the degree of cooperation between scholarly publishing and libraries has been constantly growing owing to various projects; the PKP is one of them. Founded in 1998 at the University of British Columbia, the PKP has been working in a cooperative manner, thanks to grants and donations, to develop open access software, including Open Journal Systems, to meet the needs of scholarly journals. Today, more than 8600 journals around the world use that software, and most publish in open access. We should note the major role of Canadian libraries, which are collaborating in the initiative by providing hosting and web services to 3000 of these open access journals. We could mention other open access initiatives based on cooperation; there are many.

Now we need to take this cooperative model even further. In Québec, a province with an impressive number of cooperatives, this project is especially meaningful. In the notion of “cooperative,” the idea of shared management of resources needs to be highlighted. We need to think about knowledge not as a good to be disseminated, but rather as common heritage. All of the players involved in scholarly publishing could be considered as members of a cooperative, from authors to the students entering revisions, and including university presses, and all of the money we spend on subscriptions could be paid to the cooperative. This means that a major scholarly publishing cooperative would have $10 billion to distribute equitably among disciplines – for while health sciences need more money than the human sciences to do research, when what is in question is writing an article and disseminating it, all disciplines should be equal. The scholarly publishing cooperative would be an autonomous, democratic organization dedicated to increasing the quality of published research. Since open access is now obvious, what is really at stake is the quality of the open access archive that we are creating, the quality of the investment we are making in knowledge.

Among other things, the cooperative would allow us to achieve a level of transparency and accountability that is presently impossible with either the subscription or the APC system. At present, we need the patient research and data analysis work done by Vincent Larivière and his research chair in order to discover the trends and schemas underlying scholarly publishing, whereas in the cooperative model, the data would be directly available.

In Canada, the CRKN is a pioneer as a cooperative of libraries that was formed in order to purchase research findings as a collective. The fact that today, thanks to the partnership with Érudit, Canadian scholarly journals can sit down with the CRKN and form an independent organization dedicated to funding, development and innovation in scholarly publishing, as well as support for the creation of new journals within the framework of a cooperative structure, is a perfect example of what has been described as the second economic model for open access – but which should be the first in terms of the choice to be made.  


5. Conclusion

In conclusion, the APC-based model, which is in fact the open access model used in biomedical science, is attractive, but it was forged thanks to the most well-endowed funding system in the world: the United States National Institute of Health, which has no equivalent anywhere else in the world, or in any other discipline. In contrast, other countries and other disciplines, in particular in Canada and Québec, are working on a collaborative, cooperative model in which we now need to involve all of the links in the scholarly publishing chain. At this time in Québec, Érudit functions like a cooperative of scholarly journals, but we now need to involve libraries in the discussion. The PKP has thus agreed to work in collaboration with Érudit and the CRKN to consider the membership of 300-400 Canadian scholarly journals in both languages in such a cooperative system. The partnership implies not only that the money spent on subscriptions would now be reinvested in open access, but also that the various stakeholders, journals and libraries would have to rethink in a cooperative manner the funding of scholarly journals so as to improve quality and distribution, and thus to establish real leadership with respect to sharing of knowledge. In this way, Canada will not only be able to disseminate the knowledge it produces, but ensure it circulates throughout the whole world, in an open, transparent way, while increasing quality. The question that now needs to be asked is: can we all make a commitment today to such a cooperative undertaking so that the collaborative spirit, which has united us, spreads to all players in Canadian scholarly publishing? Can we ensure that the subscription system will be relegated to the rank of an artifact of the culture of the printed page?