We attended the workshop on Alternative Open Access Publishing Models organised by the European Commision (Directorate General Communications Networks, Content and Technology and Directorate General for Research and Innovation). This blogpost summarises our impressions.
The workshop
When we signed up, we were expected a small workshop with a lot of discussion. We were surprised to see about 100 people in the audience and a packed program with lots of information transfer from the speakers to the audience, but little room for interaction or discussion. The speakers were very well selected and gave a good overview of various approaches to OA publishing in Europe. It was at this meeting that I realised how diverse and vibrant the OA scene is in Europe.
New models
One of the basic question in OA is funding. Of the various solutions proposed, I will highlight some here.
Open Library of Humanities
The OLH has a very well set up model where libraries become “partners” of the OLH for a sum of 850 USD a year on average. From that money, OLH finances the shift of subscription journals to OA journals. The result: everybody has access and libraries actually save money since subscription fees for the shifted journals disappear. This is a very nice example of a successful implementation of a collective action via a convincing model and professional communication. Up to now, over 100 libraries have signed up. Doing the math, we arrive at a yearly revenue of about 100,000 USD. This is certainly enough to run a couple of journals, but I would guess that with overhead and IT (setup, support, R&D), a larger sum will be required in the long run.
Freemium at Open Edition
Open Edition from France presented their Freemium model, where HTML content is available for free, but pdf and epub are available for a fee. With that model, they generate 400,000 EUR a year. This looks like a very healthy model. For Language Science Press, however, it does not look promising. We make books, and reading books in HTML is tedious. We could say that the poor folks can still access the HTML content, but for our many readers in Indonesia for instance, we would feel that pushing them to the second rank because of the low economic power of their country is actually not an option.
Pay what you want
The Thieme Verlag tried a very interesting model: pay what you want article processing charges. Martin Spann, an economics professor from Munich explained why neoclassical theory would predict that PWYW cannot work, and why it works nevertheless. He offered some preliminary empirical data (n=9). The take home message is that there was only one freerider (0€), while the other 8 authors paid some money. Authors came from different countries and different continents. The psychological and social discussion of the workings of PWYW were also very interesting, but would require a blogpost of their own.
SCOAP3
High Energy Physics has been a leader in green OA. Since the early days of arxiv.org, preprints have been available for nearly all papers in that subfield. Actually, articles are very often cited before they appear in a “regular” journal. The aim of the SCOAP3 project was to switch the publication of the final articles to OA as well.
By using a very nice game theoretic approach, they got publishers to actually bid to be included into the project. Publishers had to signal the APCs they would charge. Only the most competitive publishers were eligible. This led to the APCs being much closer to the real costs (~1000 EUR) than normal
News
Max Planck Digital Library study
There were also very interesting talks not tied to any particular project. The Max Planck Society presented their study of the cost of closed access: The sum spent by all libraries world wide on journal subscription is about 8bn EUR. There are about 1.5 M articles per year according to the Web of Science. Easy math gives you the price the world tax payers are charged for a closed access article: ~5000€. A parallel survey of APC yielded a figure between 1000 and 2000€ per article. Take home message: the current closed access model overcharges the tax payer by at least 100%, possibly more
LingOA
There are not many linguistics journals at Elsevier. Lingua is one of them. The editors got very unhappy with the closed access model Elsevier uses and formulates the principles of Fair Open Access under the umbrella LingOA:
- The editorial board owns the title of the journals.
- The author owns the copyright of his articles, and a CC-BY license applies.
- All articles are published in Full Open Access (no subscriptions, no ‘double dipping’).
- Article processing charges (APCs) are low (around 400 euros),
transparent, and in proportion to the work carried out by the publisher.
They have approached Elsevier to either publish the journal Lingua compliant to these principles or otherwise they would shift to OpenLibHums (see above). A very nice proof that the OpenLibHums concept has already an impact on the publishing world!
The Netherlands
Gerard Meijer reported on his dealings with the major Dutch publishers and was visibly fed up. Next to overcharging libraries, the non-disclosure agreements in the contracts particularly met his wrath. As civil servants or politicians accountable to the public, it is simply not admissible that money is spent on contracts whose precise deliverables and costs are unknown. In my personal view, Meijer and the Dutch deputy minister Dekker have had it with the big publishers and will push hard for OA with acceptable terms. Also see the joint statement of Dekker and Commissioner Moedas.
Other issues
Book publishing in the Humanities
One point which was repeatedly made was that the Humanities had less money in the system than STEM disciplines. As a result, author charges are less of an option. If the budget of a research project is only 20,000 EUR, an APC of 1000 EUR or a book processing charge of 5,000 EUR hurts. Therefore, Humanities should follow a different route than the STEM disciplines, it is argued. While it is true that Humanities have less money, I am not convinced that APCs/BPCs are not affordable. As the MPDL study has shown, the money is actually available in the system, but it is tied up in subscription/library costs. It is in principle conceivable to free that money and divert it to APCs/BPCs.
But this raises another question: if the state gives money to the universities, which give it to the libraries, and if many many libraries give their money to Open Lib Hums: would the state not be better off directly funding publishing initiatives like Open Lib Hums rather than incurring the administrative overhead of splitting up the money among 100 institutions, channeling it and then pooling it again? The answer to this question is left as an exercise for the reader.
Summary
The meeting was very informative and very exciting. My apologies to all presenters who are not included in this summary. The meeting was a thriving fair of ideas, and most certainly inspired more than one person to try whether they could take over some of the concepts presented.
One thing I found curious: the meeting was organised by a political body, the European Commission. But actually, there were very few calls for political action like laws or investment programs. Other OA meetings I have attended were rife with whining or rants against the publishers. This one was different. No whining, no rants. If the terms of the publishers don’t fit, we will simply walk away, as LingOA shows on a small scale and the Netherlands show on a large scale.
Still, I do think that there is room for action by the EU/EC. I would suggest to fund a study to evaluate the different funding models as to their soundness, robustness and scalability. Furthermore, I would suggest that there should be an independent source of funding for APCs/BPCs for disciplines outside the money-heavy STEM fields. Max Planck, again, has shown that setting up an extra publication fond outside of the institution budgets will make the transition much easier. And as their study has shown, that money is easily earned back by lower subscription/purchase costs.