London, UK (Scicasts) — The house was full on the day when UCL held its Open Access Conference this year. The event, organized annually as part of an international Open Access Week, for the past three years has been attracting increasing numbers of librarians, researchers and policy makers – as has been the Open Access movement itself.
“Research is better performed when all buildings are open and all data is accessible,” Dr. Paul Ayris, Director of UCL Library Services and CEO of UCL Press, opened the conference stating the main objective of the movement – to make results of publicly funded research freely available to the community.
Open access has the potential to improve scientific research (through improved reproducibility), involve citizens and society as well as accelerate innovation.
Meanwhile, opponents of the open access movement, such as a University of Colorado librarian, Jeffrey Beall, characterize it as “an anti-corporatist movement that wants to deny the freedom of the press to companies it disagrees with” and argue that the rise of resistance against the traditional system leads to creation of low-quality new journals and fosters research misconduct.
However, as the debate about the future of publishing continues, both sides seem to agree that opening up the access to all research results today would require fundamental changes to the business models, the majority of which are currently based on subscription fees. This article is an attempt to outline the latest shift in publishing concepts and governing policies aimed at promoting the anticipated change.
The Need for Change
The Open Access (OA) movement dates back to the dawn of the internet in the 1990s, when the centuries-old tradition of print publishing was disrupted by the arrival of digital technology and publishing ventures were forced to adapt to the new challenges of making their content available online.
The digital challenge in publishing was aggravated by the so-called ‘serials crisis’ – steadily rising prices of subscriptions for academic periodicals sold by for-profit publishers, which over the years between 1975 and 1995 exceeded the inflation levels by over 300%, according to the report of the European Commission in 2006.
Among different reasons behind the crisis, such as the rising numbers of academic publications and new journals covering specialized sub-disciplines, was also the ascendance of commercial publishers back in the late 1950s.
Prior to World War II, academic journals were published primarily by the learned societies and governed by the scientific community, notes Dr. Robert Parker, CEO of the Royal Society of Chemistry, in his speech at the Conference of the Royal Society in April 2015.
“But we were using methods that were slow and outdated,” Dr. Parker admits. “There was insufficient engagement with the author community and we were winding the authors out by insisting on our editorial conventions. That left a perfect gap for entrepreneurial publishers like Robert Maxwell to be extraordinarily successful.”
We need to put the power to make purchasing decisions back into the hands of researchers as they directly benefit from publishing services.
Among the first to recognize the need for change back in the 1990s was Prof. Stevan Harnad, who in 1989 established Psycoloquy – the first online open access peer-reviewed journal, covering research in psychology. In his “Subversive Proposal” in 1994, he urged researchers to deposit all unrefereed preprints of their articles in publicly accessible online archives, arguing that the transition from paper publication to electronic publication would then happen almost immediately.
"Paper publishers,” wrote Prof. Harnad, “will then either restructure themselves […] so as to arrange for the minimal true costs and a fair return on electronic-only page costs […] to be paid out of advance subsidies […] or they will have to watch as the peer community spawns a brand new generation of electronic-only publishers who will.”
However, at the turn of the millennia, print publishers, who have successfully adapted their facilities to hosting digital content, kept generating revenue through both print and online subscriptions, the prices for which continued to rise.
Melting the Ice
The first major change occurred in 1998 when the London-based publishing entrepreneur Vitek Tracz founded BioMed Central – the first open access journal in biomedical sciences. Almost simultaneously, NIH established its online repository PubMed Central, which remains the largest open archive for published peer-review articles in medicine and life sciences to date.
The non-for-profit Public Library of Science (PLoS) followed soon after that, in 2000, with a similar business model – flipping the reader subscription charges into the system where authors would pay article processing costs for publications.
The EU Commission is setting up an Open Access Policy Platform to further investigate the transition to alternative business models.
It became clear, however, that achieving fundamental changes in the publishing “status quo” would not be possible without the additional support from funding and regulatory bodies. In an attempt to drive these changes, the Association of Research Libraries in Canada and the United States in 1998 founded SPARC (Scholarly Publishing and Academic Resources Coalition).
“SPARC was created by the library community in the US and Canada out of frustration over the rising prices to get access to scientific research journals,” says the Executive Director of SPARC, Heather Joseph.
The organization’s original goal was to look into alternative publishing models, which could provide the community with open access to all published research. Together with a number of international institutions as well as representatives from PLoS and BioMed Central, SPARC signed the Budapest Open Access Initiative in 2002, defining what later became known as the Green and the Gold strategies of achieving open access.
Gold Open Access, which relies on the principle of charging authors for the article processing, was soon noticed and taken on board by large commercial publishers, such as Springer (2004), and it seemed for a while that the model could efficiently replace the traditional subscription fees.
In an attempt to help researchers choose from the available open access journals, the University of Lund in 2003 established a Directory of Open Access Journals (DOAJ). “DOAJ is a white list of open access journals and aims to be the starting point for all information searches for quality, peer reviewed open access material,” states their website.
Research is not a commodity - you can't go and buy a competitor's product.
However, the idea that one has to pay to get their research published was viewed by many in the research community as “self-vanity publishing” and therefore regarded with suspicion. It also appeared that a growing number of new “predatory publishers” were able to exploit the system, collecting money from researchers without providing high quality peer-review or editorial services. The term “predatory publishing” was coined by Jeffrey Beall, who in 2008 began assembling a “black list” of dubious open access journals and publishing ventures.
The “black list” itself came under question over time, although DOAJ recently took a decision to scrutinize members of their “white list”, requiring every publisher added to the list before 2014, to re-apply for admission.
Green Open Access, or self-archiving as first proposed by Prof. Harnad, has also been initiated multiple times over the past two decades, although with varied success.
Unlike a preprint repository arXiv.org for physics and mathematical sciences, the idea of depositing unpublished articles in medicine and life sciences online initially was not met with equal enthusiasm.
A 5-10% profit margin would be a reasonable operating value that still allows re-investment into innovation.
One large attempt was made by Elsevier in 2002, when in light of the growing open access movement, the company established their own preprint repositories in Chemistry, Maths and Computer Science. However, it soon became clear, that many other prominent publishers refused to accept articles, which have already been made available online. Depositing already published articles, on the other hand, was inhibited by long embargo periods for most publishers.
The initiative was terminated two years later when Elsevier announced that their efforts to make the venture self-sufficient were unsuccessful.
In response to the situation, the open access advocates decided to change their course. “In 2005, we started looking at the possibility for policy advocacy to flip the rules of the game from the point of research funding,” says Ms. Joseph. “If an organization provides funding for researchers to do an investigation, the communication of the results is an important part of that funding.”
“Our first and biggest success came in 2008, when NIH requested the US Congress to pass a law advocating for open access to the research data, with a maximum embargo period of 12 months.”
The law, which was successfully passed, now requires researchers funded by the US public bodies such as the NIH to deposit their manuscripts in open access repositories online, allowing the government to retain a copy of the article.
“Many other agencies are already following the model by requesting authors to deposit articles in an openly accessible repository,” notes Ms. Joseph. “There are a handful of exceptions at this stage, where institutions don’t have working repositories yet – that is a future priority for SPARC as a library organization.”
Meanwhile, in Europe, funding bodies had begun adopting open access policies as early as 2005, when Wellcome Trust issued the first mandate requiring researchers to make their work available through PubMed Central (and later, its European counterpart, Europe PMC), or institutional repositories. The move was followed in 2006 by the UK Medical Research Council (MRC).
The Commission will adapt its policy if it finds that publishers are charging excessive article processing charges for opening access to articles.
Unlike their peers in the US, the European funding bodies, including the European Commission in 2007, also offered additional grant funding to cover publication fees in Gold Open Access journals and requested shorter embargo periods of 6 months for the articles in STEM disciplines in alternative cases.
“Our hope is of course that publishers will align their policies with that of Horizon 2020,” comments the European Commission on their open access policies for Scicasts.
“If publisher and Horizon 2020 rules are in conflict, we ask the researcher to contact the publisher in order to signal what his/her funder’s policy is and to find a way to comply with it. In case of further questions, the researcher can contact the European Commission via the Project Officer of the concerned project.”
The Future and Its Costs
As Mikael Laakso and colleagues concluded in their article in PLoS One in 2009, the Open Access movement had reached its consolidation phase, with the number of OA articles published per year growing consistently and novel infrastructure being developed to support OA.
In the wake of new policies, established commercial publishers also began to adapt their models to comply with funding regulations. In 2006, Elsevier, shortly followed by other major publishers, introduced a hybrid publishing model, which allowed researchers to choose to pay article processing charges and make their manuscript open access immediately upon its release.
Springer Nature, a joint venture of Springer Science + Business Media who acquired Biomed Central in 2008, together with Macmillan Science and Education now hold 553 fully open access journals in their portfolio and report that 63% of all original research content on Nature.com is fully open access.
“Our policies enable authors to comply with all major funder mandates for open access,” says Grace Baynes, Business Operations and Policy Director in the Open Research Group for Springer Nature. “We offer immediate open access funded by article processing charges or self-archiving a version of the manuscript after an embargo period.
“Our embargo periods are currently 6 months on Nature Publishing Group titles and 12 months on the majority of the rest of the titles."
The activities of SPARC have been quite influential on the Horizon 2020 policies.
Some authors have also been successful in negotiating with their publisher and were allowed to upload their post-reviewed manuscript copy in a public repository immediately upon publication. “Thanks to the immense efforts of the green open access movement, the problem of restricted access can easily be solved using existing infrastructures and with a small additional effort on behalf of the authors or their librarians,” writes the senior author of the above mentioned manuscript, Prof. Pandelis Perakakis, in his blog post.
A number of more innovative ways to move forward also emerged and some of them were presented at the recent workshop on the alternative open access publishing models, organized by the European Commission. The Commission also notes that: “As a general rule, [new] publishing models should be fairer, more transparent and pursue the primary goal of supporting European research.”
One increasingly popular model has been to put the responsibility for academic publishing back into the hands of universities and research institutions. Earlier this year, the University College London announced the launch of UCL Press – a fully open access university publishing venture for arts, humanities and social sciences. The initiative will be supported by the UCL funds and through the publication processing fees, which some of the non-UCL researchers will have to pay.
Price is a customizable product feature: By influencing the price they pay, buyers can signal information (e.g., their values) to themselves and to others, thereby expressing themselves and feeling empowered.
Another option available exclusively to research institutions, campuses and libraries for social sciences and humanities is the so-called Freemium, proposed by the open acces platform for academic communications in social sciences and humanities, OpenEdition. The business model offers open access to the journal articles and books from their catalogue in HTML format and charges for the additional services, such as downloading content in Pdf or Epub formats.
Perhaps, the most disruptive business model at the EC workshop was proposed by Prof. Martin Spann, Director of the Institute of Electronic Commerce and Digital Markets at Ludwig-Maximillians-University in Munich. Supporting his proposal by the conclusions of recent studies of social behaviour, he laid out a “pay what you want” option – “a participative pricing mechanism in which a seller delegates pricing power entirely to buyers. The buyer can offer any price (including zero), and the seller cannot reject it.”
Meanwhile, even the traditional subscription models are now beginning to melt. Last month, Springer announced two new agreements – one with the members of the Max-Planck Digital Library in Germany and another with Jisc, a UK charity providing digital solutions for education and research. The new scheme is named Springer Compact and combines reading and publishing in one scheme. It will allow researchers to publish open access articles in over 1600 Springer hybrid journals.
The official press release from Springer states: “A pilot in nature, [the scheme’s] aim is to gain experience to help prepare a framework for a sustainable model of open access publishing and access to Springer’s subscription journals. Springer has already made similar pilot agreements with the Association of Dutch Universities and the Austrian Academic Library Consortium.”
The sustainability and the true costs of open access have so far been assessed in several reports, including a report on expanding access to research results chaired by Dame Janet Finch and released in 2012 as well as the open Access Policies White Paper produced by the Max-Planck Digital Library (MPDL) in 2015. Both reports conclude that the Gold Open Access model can be sustainable in the long term and estimate that the combined expenses for researchers under the Gold OA publishing model may be lower than the costs currently paid to publishers through subscriptions.
There are not many companies that have a 40% operating profit margin. Even Apple couldn't beat that.
The white paper from the MPDL estimates that publishing an open access article currently costs the research community between 3,800 and 5,000 Euro (£2,600-3,500). The report on open access costs for 2013-14 by Wellcome Trust declares an average APC of £1,837 paid by researchers who applied for their Open Access grant. Meanwhile, Nature states that their internal costs of producing a research article can reach as high as £20,000–30,000 per paper.
To clarify the true price of open access publishing, Scicasts asked the Executive Director of eLife and former Director of Publishing at PLoS, Dr. Mark Patterson.
“There are several layers of costs associated with publishing,” explains Dr. Patterson. “Firstly, there are core costs, which include activities like administering the peer review process and the infrastructure associated with that.
“Once the article is accepted for publication, the next stage is the content processing, which might involve copyediting, precise tagging or structuring of the content for the XML or HTML formats. It would also involve layout work to produce a formatted pdf to be printed.
“The third layer includes costs associated with presenting that content on a website, maintaining the website and ensuring that the content is then deposited on other services, which can be used as a way to discover that content.”
What contributes to the difference in publishing costs, he says, would usually be the editorial process or additional content such as news, commentary or opinion articles. “There would be journals that have paid full-time editors, but in the majority of cases journals are run by academics who are volunteers. eLife, for instance, is run by academic editors but in our case they are paid for their effort of making decisions and processing manuscripts.”
There is currently already enough money in the system. A large-scale transformation from subscription to open access publishing is possible without added expense.
eLife currently does not charge article processing costs, which was made possible by the support of three major funding organizations – the Wellcome Trust, the Max-Planck Society and The Howard Hughes Medical Institute. However, Dr. Patterson says he would not make any predictions on the long term sustainability of the venture at this stage.
“We are a very new organization and have only been publishing for three years. We have had a six-fold increase in submissions over the last two years and a comparable level of increase in the amount of published content. As we grow and develop new products and services, the costs of our publishing, in terms of costs per article, will evolve as well.”
It is likely, suggests Dr. Patterson, that in the future eLife will also have to introduce article processing costs.
Yet another layer of expenses associated with publishing today is the assisting technology that provides the reader with the most convenient way to access, read or write an article. Large publishers like Springer Nature and Elsevier are investing huge sums in acquiring innovative startups, whereas eLife developed their own in-house tool eLife Lens. "eLife continues to invest into technology development, which is part of our broader mandate to support innovation across the landscape of scholarly communication," says Dr. Patterson.
For researchers who do not wish to depend upon a publisher for dissemination of their data, there is the next wave of Green Open Access publishing platforms and repositories. Among the most recent is a public repository Zenodo, created by OpenAIRE and CERN, and supported by the European Commission. It allows researchers and institutions “to share, preserve and showcase multidisciplinary research results (data and publications) that are not part of the existing institutional or subject-based repositories of the research communities.”
Another solution is a commercial platform FigShare, which currently allows scientists to upload any part of their project and make it openly accessible at no cost.
In general, the pure-OA publishers are costing less than the legacy subscription publishers.
“We work with academic publishers, to support data within their infrastructures and with universities, to provide them with their own repositories for digital content,” Dr. Mark Hahnel, founder of Figshare, explains the concept to Scicasts. “We have 10 different types of formats we support, including the program code, for which we have partnered with GitHub. FigShare also provides a content engine, for which our sister company ReadCube analyzes the papers you are currently reading and suggests other literature based on that.”
Both Figshare and Zenodo assign each file with a unique Digital Object Identifier (DOI), which allows the content citation. In addition, OpenAIRE is now piloting open peer review in two projects, one of which will introduce an open peer review module for (Green) open access repositories and enable the conversion of repositories into functional evaluation platforms.
"I think it's a very positive sign that open-access publishing is gaining more and more ground, that new business models are being explored and entirely new approaches to research communication are being investigated," says Dr. Patterson.
The Reasons Not to Publish Open Access
New policies and business ventures have certainly changed the face of science publishing over the last decade. However, the main obstacle in their way is perhaps not the resistance of large corporations or the lack of funding support. It is the very essence of communicating academic research results, often named as a “publish or perish” mentality – the notion that the number of high impact publications define the success of a researcher’s academic career.
“There is a common opinion that if you publish two articles in a high impact journal like Nature, you are set for life and you will certainly get to the next stage of the funding pyramid,” says Dr. Hahnel, who is a former stem cell biology researcher. “It is very hard to change the mentality, which is why it has gone all the way up to the level of the government and funders.”
“Concerns about perceptions of the quality of OA publications is still the leading factor in authors choosing not to publish OA,” explains Ms. Baynes.
Open Science does not only mean doing science in an open way, but also in transparent, ethical and accountable way, for science itself and also for society.
However, the most recent data suggests that even these opinions are starting to change. According to the Author Insights Survey conducted by Nature Publishing Group and Palgrave Macmillan in 2014, 40% of scientists who had not published open access in the last three years said "I am concerned about perceptions of the quality of OA publications." In contrast, this year only 27% said they were concerned. In the humanities, business and social sciences (HSS), the drop was similar; from 54% in 2014 to 41% in 2015.
The younger generation of group leaders, especially in the technology-related fields, are increasingly looking for other ways to assess the impact of their research and argue that “impact factors” should be assigned to individual papers rather than to whole journals.
“The impact factor is not about whether a paper is good or not,” comments Dr. Robert Davey, Group Leader at The Genome Analysis Centre (TGAC) in Norwich. “It’s about how prepared you are to say whether the journal is valuable. But look at it another way – if there were no papers in that journal, the journal wouldn’t have any impact.
“The value of an article, on the other hand, is demonstrated when someone picks up the work, reads it, finds it of value and cites it.”
For the Open Access to take off and become the solution we envisioned it to be, we have to find ways to credit our researchers for open behaviour.
An innovative way of counting direct citations has been offered by the Altmetric platform, which allows tracking the article’s mentions on social media platforms alongside traditional citations. The system has been gaining popularity over the recent years and is now integrated within many publishing platforms, such as Springer Nature or F1000Research.
“As far as I’m concerned,” concludes Dr. Davey, “I have absolutely no interest in publishing in a high impact journal. Of course, I wouldn’t like to dictate where my junior faculty wish to publish their results - as we all know, the [career evaluation] system is changing very slowly. But in terms of my own research, I would never consider publishing in a paywalled journal.
“I think that if you wish to show that your research is valuable, you should let as many people as possible read it without any hindrance. And if fundamental science today isn’t open access, you are doing the humanity a disservice.”
UPDATE: On November 25, this article was updated to correct the following information - Springer Nature is made up of Springer Science + Business Media and Macmillan Science and Education (not Springer and NPG, as originally stated). The 63% original research content on Nature.com is fully open access (not 63% of Springer Nature content). The Author Insights Survey was run by NPG and Palgrave Macmillan (not Springer Nature). UPDATE: On November 30, this article was updated to clarify the references provided by the EU Commission and the link to the joint statement of Commissioner Moedas and Secretary of State Dekker was added. Information about two OpenAIRE projects piloting open peer review was also clarified.