A few years ago, I wrote that open access (OA) publications were gaining momentum. Based on a study of 2006 to 2010 in the biomedical literature database, PubMed, our access to publications was apparently growing somewhat steadily.
It seems to have stalled after that. With a push towards 100% open access for publicly funded research in major European countries and some academic institutions, though, this could turn out to be a long lull, not a plateau. It’s starting to look as though accessibility of publications increases in waves, with those waves arriving in different intervals, geographically and by academic discipline: 2020 looks like the year to watch now.
To get a rough idea of how much reader access there is for biomedicine, I ran some searches in PubMed. It’s not precise, for a few reasons. Because NIH-granted research is deposited in full in PubMed’s repository, PubMed Central, it means there are articles from some journals without the whole journal contents. On the other hand, the search for articles available in full text in PubMed underestimates how many articles are out there for free somewhere online. The contents of PubMed aren’t only research: access requirements tend to apply only to research reports. And the estimations of country of authors are very incomplete.
Only 27% of PubMed records added in the last 11 months had full texts. That’s strongly affected by public access requirements at the NIH and elsewhere that only require availability of full texts 12 months after publication. I looked at publications to the end of 2015 to see what happens after that embargo expires.
The level of accessibility for publications from 2005 is (23%). In 2010 it was up to 33%, but in 2013-2015 it had leveled out to about 40%. That varies by country, with less public access in some areas of becoming a bigger proportion of the literature. It’s 53% for papers listing at least one author in the USA and 48% with at least one in the UK. For Germany, it’s around 40%, and it’s 37% in China and France. (Details of my searches and the results are here.)
Making a big dent in that second half of literature – and the embargo – is going to take action by funders of research and libraries, at (supra)national level and at individual institutions. More of that is coming. Increasingly, the push isn’t just for access to read, but for accessible data and right to data and text mining, too. The embargo isn’t being challenged as widely, though.
The slow progress in accessibility of publications isn’t the only problem. There’s the high cost when author processing charges (APCs), which are also largely added to subscription costs instead of replacing them – and the exploitative “spam” journal industry financed by APCs.
There may be a silver lining in the cloud: the increased cost pressure introduced by APCs seems to be a force propelling funders and institutions towards deep reform. That gathered force in 2016.
Here are some standout developments, focused on Europe and the US, month by month. (You can catch up on previous years’ posts here.)
Open science was a focus of the Dutch year of EU presidency, kicking off with Bill Gates addressing a meeting of ministers. That seemed to increase momentum for change in Europe through the year. (Here’s a video from the Dutch education and science minister about why they consider open access so important.)
The International Committee of Medical Journal Editors (ICMJE) came down on the side of open data for clinical trials published in the group’s journals, starting in 2017:
share with others the deidentified individual patient data (IPD) underlying the results presented in the article (including tables, figures, and appendices or supplementary material) no later than six months after publication.
The League of European Research Universities (LERU) called upon the EU’s policymakers to do more to advance the open access agenda. They pointed to Harvard University’s judgment that the publishing system is “fiscally unsustainable and academically restrictive”:
In the era of Open Science, Open Access to publications is one of the cornerstones of the new research paradigm and business models must support this transition.
A concerted effort to get life science researchers to pick up physicists’ preprint habit took off with a high-profile, invitation-only, ASAPbio meeting at the Howard Hughes Medical Institute (HHMI) on 16/17 February.
Preprint repositories are managed by communities of researchers. They enable authors to upload manuscripts, alongside or instead of publishing in a journal. (I wrote more about preprints here.)
Another roadmap: this time, the European University Association (EUA). They want to steer a course to “a balanced of realistic costs and benefits shared between all stakeholders”, in sustainable models, including their own university presses.
Dutch negotiators have been working to a goal of 60% of all funded research being OA by 2019 and 100% by 2024. They netted another nationwide agreement, this time with Wiley: unlimited access to all research by Dutch academics, no author charge for a Dutch authors, and access to 1,400 Wiley publications. The cost wasn’t made public. Meanwhile, the cost of France’s deal with Elsevier for its universities was reported to be €34 million per year for 5 years.
I am confident that by 2020, the UK will be publishing almost all of our scientific output through open access. The advantages of immediate ‘gold’ access are well-recognised, and I want the UK to continue its preference for gold routes where this is realistic and affordable.
And the Research Libraries of the UK (RLUK) and others published a thought-piece on OA and the costs of academic publishing in the UK. (“Gold” OA is in journals, “green” OA is a version in a repository.) The Publishers Association responded in April, including data on APCs and a defense of “hybrid” journals – non-OA journals that charge subscriptions as well as APCs for OA articles [PDF]. RLUK responded with a critique of the higher costs of “hybrid” journals, and pointing to compliance problems even when fees are paid.
The cost burden is leading some institutions to cancel long-held journal subscriptions. In February, Memorial University of Newfoundland (MUN) joined several Canadian universities going down this road. Also from Canada in February: another report from the OA Publishing Cooperative Project, exploring issues involved with returning scholarly publishing to scholar communities.
Another model is the charitable Open Library of Humanities (OLH), where libraries kick in to fully fund OA journals, with no charges to authors: in February, MIT joined. As of writing, there are 218 institutions on board.
And again on the government side, the White House published a summary of progress on public access to the results of federally-funded research from 16 agencies.
Germany’s Max Planck Institute launched OA2020. As of writing, over 70 institutions had signed on:
We are pursuing the large-scale implementation of free online access to, and largely unrestricted use and re-use of scholarly research articles.
“Diamond” OA: that’s what mathematician Tim Gowers calls it when it’s free to read and publish. He and his colleagues launched Discrete Analysis, an overlay journal – a journal structure piggybacking on preprints.
Most academics, though, continue to put their volunteer editing efforts into commercial publishing. The majority of the scholarly community remain wedded to traditional publishing, and as long as that’s the case, OA will remain a “half-revolution”, argued Richard Poynder in a March interview on the half-full, half-empty OA glass.
Germany’s OA strategy was released: their goal is OA with economic efficiency, via the national library network [German]. Without a coordinated approach to the issue of discounting subscriptions to account for APC expenditure, they fear, OA would increase cost. Just how serious they are became clear later in the year…
The Open Science Framework (OSF) launched a preprint aggregator. As of writing, it’s approaching 1.9 million preprints (February 2017) and OSF’s enables scholar communities to create their own specialized preprint repository. The aggregator now combines preprints from 8 repositories, plus OSF itself.
In the US, FASTR, a 2015 proposed bill to increase access to science and technology research, moved on to the Senate’s to do list.
The Wellcome Trust posted detailed recent data on what it spent, where, and how much on OA for research it funded:
The good news is that we have seen an improvement in correct and programmatically identifiable licences (from 61% of papers in ’13-‘14, to 70% in ’14-‘15) and a similar increase in overall compliance from 61% to 70%. The bad news, however, is that in 30% of cases we are not getting what we are paying for.
The Netherlands released some information on the incremental progress they are making in striking a better deal with Elsevier: their current deal lasts to the end of 2018. Dutch academics will be able to publish with no APC in 30% of Elsevier’s journals by 2018.
The open access requirement of the Higher Education Funding Council for England’s (HEFCE) came into effect. It affects universities’ core funding: only open access articles count as evidence of university excellence. By 1 April 2017, publications that aren’t “gold” access have to be deposited in a public repository when a journal accepts them.
And bereaved parent and US Vice President Joe Biden’s Cancer Moonshot turned him into the world’s most powerful advocate open science and no-embargo OA. Speaking to researchers:
…[M]easure progress by improving patient outcomes, not just publications…Right now, you work for years to come up with a significant breakthrough, and if you do, you get to publish a paper in one of the top journals. For anyone to get access to that publication, they have to pay hundreds or even thousands of dollars to subscribe to a single journal. And here’s the kicker – the journal owns the data for a year. Your outfit does this.
And by the way, the taxpayers fund $5 billion a year in cancer research every year, but once it’s published, nearly all of that taxpayer-funded research sits behind walls. Tell me how this is moving the process along more rapidly.
In copyright news, the US Supreme Court ruled that Google Books’ scanning is fair use.
The International Association for Scientific, Technical and Medical Publishers (STM) responded to the European Union’s call to action on OA with “disappointment”, concern about preserving the embargoes delaying accessibility, and this claim about costs without data [PDF]:
Any transition to open access publishing where costs are paid up front – often called gold open access – will cost research intensive countries more than the current system. Therefore, a transition to this form of Open Access requires joint ownership. All stakeholders – universities, funders, researchers, policy makers and publishers – need to take responsibility and share the economic burden.
There was no shortage of data, though, when Stuart Lawson and colleagues published an analysis of “the black box” public funding of subscriptions and APCs in the UK. It included 2010-2014 public funding on subscriptions gained by freedom of information requests to 10 major institutions: they paid 10 publishers £93,766,870 (nearly 40 million of which went to Elsevier – Wiley came in next, with nearly 17 million). Lawson’s article prompted a librarian blogger, Andrew, to add more black boxes of economic unknowns.
New principles for offset agreements in the UK were published by Jisc, laying down 5 positions designed to ensure discounted subscription levels for institutions footing the bill for journal APCs:
Systems should ensure that publishers do not charge the same institutions twice, through the payment of subscriptions and the payment of APCs.
But the disputes with Elsevier that got the most attention in April were its battles with the website providing “pirate” access to the scholarly literature, Sci-Hub, and Sci-Hub’s founder, Alexandra Elbakyan.
It had been a year with a lot of publicity for Sci-Hub – see for example this February story by Kaveh Waddell in The Atlantic. In March, David Rosenthal argued it was a Streisand effect – Elsevier’s efforts to have it shut down was giving the website enormous publicity.
In April, Sci-Hub made it to the front page of the print edition of the Washington Post. And Science’s editor, Marcia McNutt, wrote an editorial, “My love-hate of Sci-Hub”, alongside a report showing users got access to 28 million publications there in the previous 6 months. Even researchers with legal access to libraries were using it, because it was so easy: just cut and paste the URL or an identifier.
Elsevier legal action shut down Sci-Hub’s web domain – but there is no shortage of web domains, and it was business as usual elsewhere on the web: “Meanwhile, academic pirates continue to flood to Sci-Hub, domain seizure or not”.
A score for preprints: Crossref announced citations to preprints would now be “counted”. And a loss: Elsevier bought the SSRN’s major social sciences preprint server, raising concern about its longterm fate.
UNESCO released a joint statement on OA with the Council of Academic Repositories (COAR). It speaks of a “large-scale shift from subscription-based” publishing, urging concern for institutions and countries with smaller budgets and the need to “avoid further concentration in the international publishing industry”.
How do “indie OA journals” survive without APCs? Bo-Christer Björk and colleagues answered that with a longitudinal study in PeerJ.
The Swiss National Science Foundation (SNF) announced that 40% of all publicly funded research in Switzerland is OA.
The University of Montreal joined the subscription cancellation club, announcing it was ending its subscription to more than 2,100 Springer journals.
The Medical University of South Carolina had a poster at the annual Medical Library Association conference in Toronto about its experience of canceling its “big deal” with Wiley.
Denmark’s OA goals are 80% of the peer-reviewed scientific articles produced by the country’s universities in 2016 and 100% in 2021. They launched OA indicators – for across all of academia.
And the EU member states agreed to a goal of OA for all science articles by 2020. Dutch Minister Sander Dekker:
The time for talking about Open Access is now past. With these agreements, we are going to achieve it in practice.
Meanwhile the Netherlands continued its publisher-by-publisher subscription deal bargaining. In May, they announced they reached a deal with the American Chemical Society.
The Alliance of Science Organizations in Germany called for all academic institutions to disclose publication charges, to help “the transformation process from the traditional subscription system”. A viable new system required working above the individual institutional level, they argued [PDF].
“Cutting without cursing”: Salisbury University Libraries reported on their experience of subscription cancellation – and Peter Suber noted some other published reports on cancellation experiences in 2011 and 2012.
The Wellcome Trust took another major step down the OA road: it announced Open Research, a publishing platform for research it funds, using f1000 Research services.
The Vienna Principles: 12 principles for scholarly communication in the future developed by the OA Network of Austria.
PubSpace arrived: NASA science publications added to PMC (PubMed Central). You can check the full list of US federal agencies now depositing in PMC in the “Research Funder” filters down on the left side of search results.
Traditional publisher, the American Chemical Society, got on the preprint repository bandwagon, announcing ChemRxiv.
Following on from the problems with OA policy compliance they had identified (see March above), the Wellcome Trust (WT) published its new requirements of publishers – researchers who publish in journals that don’t meet the requirements will still have to meet OA conditions, but won’t be able to use WT money for the APC. The requirements will come into effect on 1 April 2017.
Another perspective on Germany’s path to OA, this time from a major public funder, the Federal Ministry of Education and Research (BMBF) [PDF]. A key goal: cost-effective use of funds for communication of scholarly knowledge.
First news out of Germany’s implementation of its OA policy: some universities announced (e.g. in Berlin) that they could not guarantee there would be access to all journals in 2017 – including Elsevier journal, The Lancet. Moving to a national negotiating approach was affecting those universities with deals falling due for renewal first.
ASAPbio gathered data on the growth of preprints from 2003 to July 2016, when they reached about 900 uploads a month.
The UK struck a 5-year deal with Elsevier – but other than its length, not much is known, according to Stuart Lawson.
However in Finland, negotiators announced that bargaining for reasonable terms in 2017 might not succeed. Academics could sign a statement of support for the negotiators – as of writing, over 2,700 have signed. A few weeks later, they announced Finland is pausing at the brink: a deal for a 1-year extension for the nation’s academic institutions while bargaining continues.
PsyArXiv launches, a preprint repository for psychology – hands down winning the stakes for in the preprint name/logo race!
Nature named Sci-Hub founder, Alexandra Elbakyan, one of their “10 who mattered” in 2016:
It took Alexandra Elbakyan just a few years to go from information-technology student to famous fugitive.
In 2009, when she was a graduate student working on her final-year research project in Almaty, Kazakhstan, Elbakyan became frustrated at being unable to read many scholarly papers because she couldn’t afford them. So she learnt how to circumvent publishers’ paywalls.
But perhaps the most critical story of the year for the longterm was Germany’s resolve in exercising its bargaining power in line with the country’s science community decision to achieve OA. There is a good explanation about the German situation in English in Nature: Quirin Schiermeier and Emiliano Rodriguez Mega reported that scientists in Taiwan and Peru are in a bit of a similar boat.
As at year’s end, although German negotiations with Elsevier were still open, but no deal was reached. The 60 institutions going into 2017 without renewed Elsevier subscription could be joined, they report, by more in 2018. So…
Previous annual open access roundups:
The photo of the Pitt Building at Cambridge University was taken by Hannah Gregory (via Wikimedia Commons).
The image of Joe Biden is a screenshot from an Obama White House video, October 2016 (via Wikimedia Commons).
Data gathered from PubMed* (accessed 5 February 2017)
|Subset||Time||Free full text||All records||Percent|
|All||Last 11 months||284254||1068326||27%|
|All||2013 – 2015||1320229||3289247||40%|
|USA||2013 – 2015||273788||516827||53%|
|UK||2013 – 2015||73047||166976||48%|
|China||2013 – 2015||129251||345856||37%|
|France||2013 – 2015||41440||112107||37%|
* Searches in PubMed
Country: Name of country limited to [ad] – at least one author listed with that address.
Time limits: using [dp] field (e.g. AND 2013:2015 [dp])
Free full text link: (free full text[sb] NOT pmcbook)
All records: (all[sb] NOT pmcbook)
* The thoughts Hilda Bastian expresses here at Absolutely Maybe are personal, and do not necessarily reflect the views of the National Institutes of Health or the U.S. Department of Health and Human Services.