Despite claims to the contrary, open access as such is not very complicated. Either publish your scholarly output with a publisher who will immediately make it available as open access, or put a copy of the (peer-reviewed) manuscript in a repository. What makes open access complicated is the myriad of policies that regulate it.
The Registry of Open Access Repository Mandates and Policies (ROARMAP) alone lists way over 700 OA policies – just from research organisations and funders. If you add publisher policies it gets even more confusing. As a sector we often complain about the difficulties publishers create with journal embargoes. We are also criticising funders for not aligning their policies. These criticisms are valid, but we tend to gloss over that universities are not always aligning their policies either. Policies that vary across universities make it more difficult for third parties to provide solutions as they need to map onto a wide range of workflows resulting partly from different policies. Different institutional policies also make it harder to communicate open access to academics.
I have on a few occasions suggested that we should aim to align institutional policies more, and that we should also simplify them. Thankfully, I am not the only one thinking about this. Jisc, SHERPA Services and ROARMAP have jointly developed a Schema for Open Access policies. The schema should help policymakers “to express their policies in a systematic manner”, as “an initial step to ensure greater clarity and uniformity in the way information about OA policies is recorded and made available”. Imperial College was one of 30 institutions that provided information to the new initiative. You can read more about the schema, initial findings and how to engage on the Jisc blog.
My ideal would be that over time we move to a single open access policy, or at least to a core policy to which institutions can add a selection of clearly defined elements to reflect their specific needs – where this is really necessary, of course. In the UK we do already have what could be considered the core of an OA policy, the Policy for open access in the post-2014 Research Excellence Framework. Leaving the details aside, the policy requires deposit on acceptance (for publication). Currently it only applies to scholarly articles and conference proceedings, but I would argue that that makes it ideal as a starting point as these more formalised outputs (compared to e.g. performances) are easier to deal with across institutions.
Therefore, my suggestion for a minimal universal OA policy would be:
- Publish in the journal of your choice, including full open access journals (subject to availability of funding).
- Deposit a copy of the peer reviewed manuscript of your journal article or conference proceeding into a repository on acceptance for publication.
Incidentally, that is effectively the OA policy at Imperial College. As the vast majority of College publications are articles or conference proceedings we can effectively limit the policy to these, at least for the moment. An institution with a more diverse range of outputs may decide to add monographs, videos, websites etc., and those who cover costs for hybrid open access (Imperial’s own fund does not support it) may want this included as well.
I fully understand that just two bullet points will not be enough. However, I would like to put out a challenge: look at your institution’s open access policy and think about which elements you really need, and how you could simplify it in a way that would help us moving towards a universal policy. And make sure to check out the schema!
The library has released a new workflow on how to make your publications REF compliant. Authors can now deposit their journal articles and conference proceedings on acceptance in Spiral via Symplectic. At the same time an application can be made for APC funding to pay open access fees.
Just in time before the College closes for the Christmas break I have found the time to write my overdue summary of recent developments in the world of open access and scholarly communication. Below are a few of the headlines and developments that caught my eye during the last couple of months.
Cost of Open Access
Commissioned by London Higher and SPARC Europe, Research Consulting have published Counting the Costs of Open Access. Using data provided by universities, including Imperial College, it concludes that there was a £9.2m cost to UK research organisations for achieving compliance with RCUK’s open access policy in 2013/14. Main conclusions are quoted below – the estimated costs for meeting REF open access requirements are particularly interesting seeing as HEFCE do not provide any funding for their in some ways even more ambitious open access policy:
- The time devoted to OA compliance is equivalent to 110 fulltime staff members across the UK.
- The cost of meeting the deposit requirements for a post-2014 REF is estimated at £4-5m per annum.
- Gold OA takes 2 hours per article, at a cost of £81.
- Green OA takes just over 45 minutes, at a cost of £33.
Pinfield, Salter and Bath published: The ‘total cost of publication’ in a hybrid open-access environment. The study analyses data from 23 UK institutions, including Imperial College, covering the period 2007 to 2014. It finds that while the mean value of APCs has been relatively stable, ‘hybrid’ subscription/OA journals were consistently more expensive than fully-OA journals. Modelling shows that APCs are now constituting 10% of the total cost of ownership for publishing (excluding administrative costs).
EBSCO’s 2015 Serials Price Projection Report assumes price increase of 5-7%, not including a recommended additional 2-4% to allow for currency fluctuations.
John Ulmschneider, Librarian at the Virginia Commonwealth University, estimates that with current price increases the cost for subscription payments would “eat up the entire budget for this entire university in 20 years”. Partly in response to that, VCU has launched its own open access publishing platform.
UK Funder News
Arthritis Research UK, Breast Cancer Campaign, the British Heart Foundation (BHF), Cancer Research UK, Leukaemia & Lymphoma Research, and the Wellcome Trust have joined together to create the Charity Open Access Fund (COAF). COAF operates in essentially the same way as the WT fund it replaces.
An article summarising responses to the RCUK review of open access cites the Wellcome Trust saying that sanctions could accelerate the implementation of open-access.
The Wellcome Trust published a list of journals that do not provide a compliant publishing option.
International Funder News
A new Danish open access strategy sets the goal to reach Open Access to 80% of all publically funded peer-reviewed articles in 2017, concluding with 100% in 2022.
The Open Access policy of the Austrian FWF requires CC BY (if Gold OA) and deposit in a sustainable repository on publication. The FWF covers APCs up to a limit of €2500.
Research Information published a summary of international developments around open access: The Research Council of Norway is making funding available to cover up to 50% of OA publishing charges. The Chinese Academy of Sciences and the National Natural Science Foundation of China require deposit of papers in an OA repository within 12 months of publication. The Mexican president has signed an act to provide “Mexicans with free access to scientific and academic production, which has been partially or fully financed by public funds”.
Publishers and Open Access
In November, negotiations between Elsevier and the Dutch universities broke down following an Elsevier proposal that “totally fails to address this inevitable change [to open access]”. The universities have since reached an agreement with Springer; negotiations with Elsevier have resumed.
The launch of Science Advances, a journal of the American Association for the Advancement of Science (AAAS), prompted strong criticism of the AAAS approach to open access. Over a hundred scientists signed an open letter criticising AAAS for charging $1000 for the CC BY license as well as $1500 for papers longer than ten pages – on top of a $3000 base APC. This has been picked up by media including the New Statesman.
The Nature Publishing Group has had two major OA-related headlines. Generally well received was the announcement that NPG would switch the prestigious Nature Communications to full open access. On the other hand, the move to give, limited, read access to articles has been widely criticised as beggar access and a step back for open access: NPG allow those with a subscription to give others viewing (not printing) access to papers, through a proprietary software.
An open letter signed by nearly 60 open access advocates, publishers, library organisations and civil society bodies warns against model licenses governing copyright on open access articles proposed by the International Association of Scientific, Technical & Medical Publishers (STM). The letter says the STM licences “would limit the use, reuse and exploitation of research” and would “make it difficult, confusing or impossible to combine these research outputs with other public resources”. The STM licenses are seen as incompatible with Creative Commons licences.
Jisc and Wiley have negotiated a deal that provides credits for article processing charges (APCs) to universities that license Wiley journal content and have a Wiley OA account.
It is time for another round-up of news relating to open access and scholarly communication – here is a summary of interesting things that caught my eye during the past few weeks. I would like to highlight one of the miscellaneous items: analysing its publications, Chalmers University found that the open access articles deposited in the institutional repository have a 22% higher field normalized citation rate than the non-OA articles. So if you would like your citation rates to have a similar increase, why not deposit in Spiral, Imperial’s repository?
David Willetts has been replaced by Greg Clarke as minister for universities and science. Whether this will have an impact on the government’s OA policies remains to be seen, but Willetts has been an active supporter of open access.
Harvard was one of the earliest universities adopting an open access policy where academics grant the university a non-exclusive licence to distribute publications – this allows deposit in the institutional repository regardless of publishers’ OA policies. Other American universities have implemented similar policies – CalTech recently made such an announcement – and we now also see universities outside of the US following Harvard as KAUST has now adopted a similar OA policy.
Research Fortnight published a summary of FOI requests on RCUK open access compliance. “The average rate across the 27 universities that responded was 49 per cent, just above RCUK’s target. However, at least 11 universities have not hit the target—and the real number may well be higher, given that 57 universities did not respond.” Regarding publishers actually delivering OA, Research Fortnight conclude that “on average, 8 per cent of articles that should have been made open access had not been, and that 12 per cent carried no clear indication of their open-access or publishing-licence status.” As RCUK have now made the details of their review public, we are in the process of bringing together the relevant data at Imperial College; a challenge is to identify which of the roughly 10K articles annually published by our academics fall under the RCUK policy, and also the articles where the authors have paid for open access from their own budgets or gone down the green route in an external repository without alerting the College.
Wiley Exchanges published an interview with Mark Thorley, who coordinates OA and RDM for RCUK. He is overall positive about progress and defends gold OA, but he is also concerned about fluctuating embargos and some universities “acting in an ‘anti-Gold’ OA manner”. He commends UCL on launching a new OA university press. It is interesting to compare his comments with those of the Wellcome Trust’s Robert Kiley who at a recent Jisc-CNI event was very critical of hybrid journals and publishers progress in switching to OA publishing.
Subscriptions, Gold OA and cost of scholarly publishing
A study on Evaluating big deal journal bundles finds that journal pricing is not necessarily related to size or research outputs of the subscribing universities. It also concludes that academia is receiving significantly less value from commercial publishers than from non-profits: “Among the commercial publishers in our study, Elsevier’s prices per citation are nearly 3 times those charged by the nonprofits, whereas Emerald, Sage, and Taylor & Francis have prices per citation that are roughly 10 times those of the nonprofits.” The study has been picked up by media, including the Guardian.
Notes of the RCUK International Meeting on Open Access (20th March 2014) have been made public. They contain some interesting information. For example, the Wellcome Trust has imposed sanctions in respect to its OA policy 62 times in the last twelve months (none at Imperial College), and the overall compliance rate is now 66%. Cameron Neylon estimates that gold OA is now 20-25% of the global market. The paper has a useful summary of global OA activities, by country.
Stuart Lawson suggests that the Finch Report may have missed evidence when it estimated the average price of APCs. According to his article, a comprehensive study that had not been acknowledged in the report set the average APC at about 1/3 of the price eventually published by Finch. As the current APCs from hybrid journals fall into the much higher bracket given by Finch, Lawson speculates that the Finch estimate may have become a “self-fulfilling prophecy”.
A lecturer in New Zealand has taken inspiration from Tim Gowers and has sent FOI requests about subscription payments to NZ universities.
UC Davis have set out a plan for a Mellon-funded project to investigate the institutional costs of gold open access, partnering with North American universities including Harvard. We are doing something similar, but on a smaller scale, as part of the OA project.
Other open access related news
A Swedish study shows that OA articles deposited in the repository of the Chalmers University of Technology have a 22% higher field normalized citation rate than the non-OA articles.
Taylor & Francis apologised for interfering with and delaying the publication of an article that criticised the profits of major academic publishers. The apology followed from the editorial board considering resignation over what could be perceived as censorship.
An analysis of data from the Directory of Open Access Journals (DOAJ) shows that the US has the highest number of OA journals, followed by Brazil and the UK. The UK is the country with the second highest percentage of OA journals that charge a fee – about 64%, compared to Germany’s 30% and Egypt’s 87%. Egypt ranking highly is probably due to the OA publisher Hindawi being based in Cairo, whereas the high percentage of OA journals that charge authors a fee in the UK may be due to funders’ focus on paid-for OA.
The Research Council of Norway is making available money to cover 50% of open access publishing costs in a new, five year funding scheme.
“A Subversive Proposal”, a message sent by Stevan Harnad encouraging academics to make their publications freely available online, has just had its 20th anniversary. It is generally seen as one of the founding documents of the OA movement. Being asked what he would change if he were to write it today, Harnad responded: “Knowing now, in 2014, that researchers won’t do it of their own accord, I would have addressed the proposal instead to their institutions and funders.”
It is time for another round-up of news relating to open access and scholarly communication – here is a summary of interesting things that caught my eye during the past few weeks.
FOI request reveals cost of scholarly communication
In late April, the well-known mathematician Tim Gowers published the responses to an FOI request to the Russell Group universities. In the request, he asked how much each university “currently spends annually for access to Elsevier journals”. Due to the terms of the agreements between publishers and libraries, universities are not allowed to make this information public or share it with our staff. Because of those legal concerns, some universities initially declined the request (here is some context as to why), but with data from the LSE added on 31st May we do now have numbers from each university. Rumour has it that FOI requests regarding payments for other publishers are in preparation, so it may very well be that we will have more data by the end of the year. This would give us a more holistic view on the cost of scholarly communication and allow us to assess the value the academic sector gets for the money. Following the publication of the data, concerns have been raised about the overall amount that universities pay, but to me the more interesting question may be about the relative value that publishers add to the process. Tim Gower’s blog post includes some information on how his colleagues at Cambridge view this; it may be one of the longest posts I have read, but it is worth having a look at if you are interested in these topics.
The publication of the FOI data has led to a broader discussion on subscriptions internationally. Zen Faulkes has correlated subscription payments in the UK and US to student enrolment numbers (no significant link) and income from students.
The Open Knowledge Foundation blog contains reflections on the data, including calculations resulting in the claim that switching to an open access model where all articles are paid for by the author/university would result in a 76% reduction of the overall cost of publishing. It should be noted that the calculations are based on the APC for one open access journal and some assumptions that may or may not be accurate. Therefore one could easily arrive at a vastly different result – both higher and lower. As I have discussed in a previous blog post it is clear though that the publication charges for OA journals are lower than those of journals that are funded from both OA charges and subscriptions (the so called “hybrid” journals).
Establishing the overall cost of Gold OA publications is surprisingly difficult as the money comes from many sources including individual research grants; if you want to delve further into this, have a look at a blog post by the Australian Open Access Support Group. They estimate that in 2013 Australian researchers may have spent US $9m on gold OA, as opposed to $4m across the Netherlands.
Other countries are ahead of the UK in collating subscription data. For example, German libraries spent €170m on books and €130m on subscriptions in 2011, with an average of €660k per library on subscriptions, according to Bjoern Brems. From the Gowers data it would appear that UK universities on average pay in the region of 40% more than the German libraries. However, the UK data only includes Elsevier and the Russell Group universities, so we are talking about the largest subscription deals, and you have to factor in the exchange rate and different tax regime – we will have to wait for data from further FOI requests that will allow a more systematic comparison.
One of the concerns in the sector is the so called “double dipping”, where institutions that already have paid for subscriptions then also pay open access publication charges (APC) for individual articles in “hybrid” journals. There is not yet a working model on how to address this, but SAGE Publications and Jisc Collections have announced that they are working together to develop one. SAGE is offering discounted APCs to subscribers, and from 2015 will globally discount subscription charges for journals with 5%+ gold OA articles. Journals below that threshold will be reviewed individually. While discounts will probably be seen as positive, global discounts effectively result in the UK using its research budget to subsidise subscriptions abroad. IoP are also launching offsetting schemes.
HEFCE have released an invitation to tender for an Economic analysis of business models for open-access monographs.
If you are interested in an overview of what UK universities are doing about OA, have a look at a series of OA case studies published by Jisc Collections.
The University of Edinburgh has released new data on its open access activities. 23% of publications listed in the university’s research information system are available as open access. The percentage of journal articles is higher – in 2013 for instance 51% of all humanities articles have been made available as OA. Since the beginning of this year, they have on average published around 50 paid-for (“gold”) Open Access articles per month with funding from the Wellcome or RCUK. MIT has recently announced that since 2009 37% of papers published by their academics have been made available through their repository, a number they hope to increase significantly.
If you want some reasons why open access publishing is positive, have a look at a summary of a presentation given by Alma Swan in Bournemouth: “The case for Open Access within a university”.
Not everyone is convinced of open access though. Scholarly societies in particular are concerned about the impact OA might have on their business models. EDP Open released a report on Learned Society Attitudes towards Open Access (PDF) that summarises these attitudes. A majority of societies think OA might put some of them in financial jeopardy and two-thirds are looking for help, especially with regards to funders mandates. Interestingly, about two-thirds would also like to offer gold OA publishing.