Introduction

I have to thank Lorraine Estelle, Co-Editor of Insights, for prompting me to set down some thoughts on the current open access (OA) landscape in the UK. She did so following a visit to my office at Imperial College early in 2015. Filling my whiteboard was a matrix of information relating to OA. It was full of acronyms and a really varied mix of initiatives including: systems – some in use at Imperial, some Jisc services, and some publisher and submissions systems; standards – including ORCID,1 RIOXX,2 Funding Data3 and CASRAI,4 publishing models – traditional and emerging; and a section on licensing and copyright, including service development possibilities. On another axis we had ‘dependencies, cost and coverage’ – the last referring to the number of academics who might benefit from the various initiatives captured on the matrix, who might be supported to ensure compliance with their funder, or whose work would be safely eligible for the next Research Excellent Framework (REF) submission. We had two columns headed ‘RCUK, etc.’ and ‘HEFCE,’ in which anything from no ticks to three ticks were entered, indicating the effectiveness of the initiative and the likelihood of the initiative delivering. Finally, there was a column labelled ‘action,’ for which entries varied between ‘wait,’ ‘doing,’ ‘investigate’ and ‘needs costing.’

Whilst I will describe some of our approaches at Imperial, I should like to stress that opinions expressed are very much my own and not necessarily those of my institution.

Policy

The recent OA policy development environment in the UK has made for a time which for libraries can probably be best described as sitting along a spectrum between ‘wildly exciting’ and ‘terrifying,’ and at both ends of the spectrum there are two common factors: significant institutional risk and significant potential effort and cost. Many have focused on the ‘terrifying’ end of the spectrum and on the short-term costs, rather than stepping back and seeing the potential for longer-term benefits and opportunities.

‘Policy’ sits on another spectrum, at one end of which is ‘law’ and at the other ‘effecting organizational change.’ I will confess that, having for a great part of my career found myself ‘embracing’ policy through gritted teeth, at this rather late stage in my career I have finally become fascinated by policy development. This, in part, is because I have also become deeply committed to working with teams to effect changes, small and large, and I can see that good policy has the potential to effect deep and lasting changes as well. I have realized that through policy, the writer can describe the end state of a desired change. Sometimes policy describes the ‘how;’ sometimes it mandates the ‘how;’ sometimes it has teeth and is effective (particularly when enshrined in law); sometimes it motivates and empowers; sometimes there is a local cost offset by an individual and societal gain. Where successful, it becomes second nature and we sometimes look back on the pre-policy era with shock and incredulity because we are thinking and acting differently and without realizing it. The various OA policies in the UK need to be considered with these transitions and end-points in mind, not least because there are individuals involved, individuals who, where they are carrying out research, are attempting to bring us insights, to answer research questions, and to deliver improved well-being, knowledge and potential economic benefit.

Fundamentally, the various UK OA policies are about ensuring that public funding for research results in public knowledge through early communication of the findings and outcomes of that research. Funders, researchers, publishers, HE institutions and their libraries, IT departments and research offices are all entangled in the transition to that state. On the way, some are incurring costs and others will reap financial reward for anything between maintaining the status quo to developing innovative solutions.

Open access: two words used to describe two different issues

Despite the fact that OA publishing has been with us for over 20 years now – both philosophically and practically – those who promote it are still largely regarded either as altruistic or as anti-capitalist. Today, for many hearing about OA for the first time, words like ‘compliance’ and ‘sanctions’ abound. Until relatively recently there has been precious little reward for the academic who chooses to publish OA – they are ‘compliant’ and that’s about as good as it gets. There are notable exceptions, of course. In the physics community arXiv5 is perhaps at the top of the list in the self-deposit green category, with PLOS6 in the pure gold corner. Meanwhile, hybrid gold – where an APC (article processing charge) is paid in order to make an individual article in an otherwise subscription-based journal open access – has grown as funders have provided financial support for APCs in order to facilitate immediate OA on publication. Generally speaking, though, academics are understandably cautious about choosing a pure gold OA publishing route as the whole academic reward mechanism still revolves around journal impact factors, citation counts and h-indices. Pure gold journals are, for the most part, relatively new and many are without significant status – it takes time to build up that journal impact factor so breaking into the market is considered a risky business. The trouble with policy-driven OA is that there is still not widespread acceptance that public funding for research resulting in public accessibility of that research will generate/result in public good. Meanwhile the academy instils terror in its newest recruits by telling them that they simply will not progress even to the lowest rung on the ladder of academe if they don’t have a ‘good journal’ (for which read ‘brand’) on their list of publications. Early career researchers can find themselves caught between the philosophical attraction of pure gold OA, or green repository deposit on the one hand, and on the other the messages they receive from their supervising professors pushing them towards publishing in a traditional subscription journal, with the option of publishing hybrid gold in some of those journals.

Esteem

Academic esteem accrues over time, and via multiple mechanisms. Publishing remains one of those. Academics like publishing in elite journals – indeed, they are sometimes required to by their departments. Equally, they like working in elite institutions and they like the elite status of ‘Professor’ or ‘Fellow of ….’ Journal impact factors, as much as global discipline and institutional rankings, are proxies for quality and are the established route to academic stardom. But why have academics in other disciplines not followed the physicists? Why are the contents of many institutional repositories so under-representative of the scholarly outputs of their institutions? Why is this also true of disciplinary repositories? Why has it taken funders so long to act? And why are they acting so variably?

What began as a means of scholarly communication has turned into big business which, on one hand, funds learned society activity and, on the other, fuels commercial publisher profits. Furthermore, the bigger publishers, who over time are supporting the publication of many learned society journals, are now mining the mass of data associated with all that publishing activity and with usage of the published outputs, and are using that data to drive new products to sell back to institutions. From the academic perspective, important benefits from this business-driven activity are those academic esteem and reward mechanisms that arise from big brand-fuelled publishing. Yet a growing body of evidence, including that assembled by the Open Citation project7 (and being evidenced by data from my own institution), is demonstrating that we can now safely consider separating ‘scholarly communication’ from ‘esteem.’ Moreover, as Pascal Rocha da Silva8 has recently demonstrated, there is now compelling evidence to suggest that there is no meaningful correlation between those journals with a high rejection rate and the impact factor.

With success came growth and brand and impact factor and ‘must have on CV.’ Somewhere in all of this was the funder, and sometimes the funder was also the publisher, and sometimes the institution in receipt of funding was also itself the publisher. What is certain is that businesses continue to grow as a result of the publication of research, and sometimes that growth fuels further research and sometimes it feeds shareholder pockets. However you look at it, it is complex. It has 350 years of history.

OA emerged as a result of the growth of the internet and the communication possibilities that it brought, combined with the age of straitened library budgets and inexorably increasing journal costs. Its proponents argued, very reasonably, that now that it was possible to disseminate the fruits of research relatively cheaply and very quickly, the expensive business of journal printing, distribution, acquisition, documentation, storage and delivery might no longer be the most effective way of distributing research findings.

From the librarian’s perspective, OA may have started as a means of using the disruptive innovation that came about with the development of the internet in order to seek to drive down journal subscription costs, but it is increasingly seen as a mechanism by which publicly funded research can be communicated to the public at the earliest possible opportunity and without having to pay at the point of use. OA means moving the costs of publishing upstream, from the point of consumption to the point of production. At the moment, though, we are in a transition period where costs of publishing as well as the costs of subscribing are incurred by institutions. This is widely referred to as ‘double dipping’ and it is not sustainable.

The policy stack

There is something bitter-sweet about the fact that so many institutions and funders have seen the opportunities that OA might bring. ‘Bitter-sweet’ because, having acted in what they considered to be the best interests of all in exploiting the opportunity, they are lambasted for challenging academic freedom, for not going far enough with their policy, or for being different and therefore introducing complexity downstream, or for introducing overly severe sanctions. There are times when I want to substitute ‘policy’ for ‘standards’ in the old adage that goes something like ‘standards are like toothbrushes – everyone thinks they are a great idea but no-one wants to use anyone else’s.’ The policies start stacking up, so we have funder policies, then institutional ones. Then, in parallel, publishers introduce their own policies and some create different policies for the same journal which are dependent on who has funded the research in the first place. Academics seeking publication ask for a simple set of instructions and become suspicious that librarians have made everything really complex just for the fun of it. Believe me, we have not, but we are becoming more and more aware of the factors that have contributed to this complexity and are working within and across our institutions to seek means by which we can reduce that friction.

Academic support

In the UK, the funder policy environment is complex, and the institutional stakes are exceptionally high. Following the publication of the Finch report,9 RCUK (Research Councils UK) funders now have OA policies. In order to pump-prime a transition to OA, funding has been made available at the institutional level, and only to some institutions, to support gold OA publishing. Where it exists, that funding is almost certain to be finite, yet we are simply not seeing that transition to OA amongst the traditional publishers that Finch envisaged. Instead, we’ve seen some ‘feeding frenzy’ behaviours exhibited by some. I heard one funder representative estimate that, at the current rate of change, one particularly large publisher would take just shy of 1,500 years to transition to OA. I am quite sure that this was not the transition timeline envisaged by the Finch group or RCUK themselves. The addition of the HEFCE policy which, uniquely, implies sanctions from the date of implementation, has both added to the complexity of the policy environment and to the institutional risk.

Meanwhile, academics are taking the opportunity to take up gold OA publishing by choosing to pay APCs to enable immediate publication of the final published version. With the growth in pure gold publishing, funds available for hybrid will certainly decline over time. Libraries, too, need to be part of the transition thinking. We need to acknowledge that our library budget profiles will look different as this transition progresses. We will move towards supporting a different ecosystem, supporting shared initiatives including the SHERPA services,10 the Directory of open access books (DOAB),11 the Directory of open access journals (DOAJ),12Open access publishing in European networks (OAPEN),13 the Open Library of Humanities (OLH),14 and Knowledge Unlatched,15 alongside pure gold publishing. We need to accept that some of what was once paying for content will now need to be put towards supporting shared OA services and infrastructures.

In the UK this complex policy environment has opened up fruitful and engaging discussions within institutions. No longer are journal prices just a library problem, the whole business of publishing has become a discussion point amongst all the different parts of the scholarly communications ecosystem within the institution: academics, research administrators, finance offices, ICT, strategy and planning, and libraries. Library services have moved from being the custodian of expensive book warehouses and open plan hot-desking areas to having the opportunity to work at a strategic level within their institution on supporting the scholarly communications enterprise through supporting compliance with those funders. The positioning in that discussion is delicate – librarians are not around to tell academics what to do. What we can do, though, is work with the institution and with the academic departments in order to understand what support might look like and how we can minimize the risk of financial sanctions further down the line. In doing so, we can also think where our taxpayer money is going.

In libraries we are acutely conscious that we are trying to develop frictionless services to support academic endeavour and yet this layering of policy complexity combined with the required data capture points mean that we find ourselves having to insert new services into the academic research and publishing process where no natural touchpoint with the library previously existed. Inevitably, we find ourselves opening any discussion on the subject with the words ‘Please don’t shoot the messenger.’

Our institution-wide discussions now mean that we think about publisher services. That includes publishing services to our academics, licensing published resources back to the institution, and enabling timely and frictionless data exchange to support local and funder reporting.

Seeking out the root cause

When compiling the grid that Lorraine saw, I did what I do in any situation where I feel there is excessive complexity – I sought out the root cause. With scholarly communication, the root cause is that academics are asked to assign copyright at the point of acceptance, and most take the line of least resistance and do so right there and then. We have all got used to the ‘Click here to accept our terms and conditions’ culture – we do it all the time on websites – and academics are no exception. A few will read and modify a publisher agreement, but the majority will sign without really thinking. Some will secretly harbour a desire to challenge, but fear doing so in case the journal retracts the acceptance. Many don’t know that there might be options, including a ‘right to first publish’ licence of the type used by UKSG and the Royal Society, to name but two. It remains, though, that the elephant in the room, the root cause of the challenge we face, is at the point of signing with a publisher. So at the bottom of my grid was ‘new licence?’

Scholarly communication and the development of a scholarly communications licence

Going back, then, to scholarly communication, I found myself asking ‘What if we really do separate out “scholarly communication” and “esteem”?’ It is early and open scholarly communication that the funders are seeking. Physicists have achieved this separation very successfully and their model has achieved early scholarly communication without compromising the esteem factor which arises from publishing with big brand journals owned by publicly listed publishers. The journals have not suffered business collapse in the wake of early scholarly communication via arXiv.16 Physicists are still being promoted and are still moving up the institutional ranks. Other disciplines are following suit. Better still, the work done to look at the effect of early scholarly communication on the citation counts for the published articles is demonstrating a strong citation benefit where there is early scholarly communication. In other words, the citation advantage brought about by the early scholarly communication of research findings may actually be underpinning journal impact factors (See Figure 1).

Figure 1 

Cumulative citation count as a function of the age of the paper relative to its publication date, 4839 articles from five major HEP journals published in 200517

At Imperial our early communication about the addition of the HEFCE REF policy to the UK’s growing OA policy environment, coupled with the growing complexity of achieving compliance amongst academics, has opened up the dialogue about publishing choices. We are asked which publishers are moving to support academics in this process, e.g. by notification back to the institution of the acceptance of a manuscript for publication. Some publishers (Springer Nature, for example) are being particularly supportive in this respect. Others are simply refusing to engage, some saying, essentially, ‘Tough, that’s an institutional problem.’ With that knowledge, though, we can help our academic colleagues understand any risks associated with choosing a particular publisher, and we can help those acting as journal editors and reviewers to understand the consequences of certain decisions that they might be taking (or might be being encouraged to take by their parent publisher). These discussions have resulted in a very strong steer by our academic colleagues to explore the development of a local policy: a policy which might replace our current OA policy, which might transcend the myriad funder and publisher policies, and which might include a licence that would enable single-step compliance with multiple funder and publisher policies whilst preserving academics’ freedom to publish where most appropriate.

The emerging policy takes the Harvard model18 as its starting point, but differs in certain important respects in order that it can work within UK law. The terms of the licence itself have been developed in consultation with representatives from the major UK funders, RCUK, Wellcome and HEFCE. During those discussions the concept of a ‘moving wall’ licence was developed whereby, on deposit in a repository, metadata is made immediately visible and, over a period of time, the output itself transitions until it is available under a CC-BY licence. Depositing under the terms of this licence would mean that outputs would be compliant with RCUK, HEFCE and Wellcome requirements, and with the major European funders with OA mandates. At the same time, the citation advantage of early scholarly communication of the research findings would begin to accrue.

We have sought to draw up a licence which can be seen as ‘reasonable’ by relevant parties: funders, academics and institutions. We want a licence which minimizes the widespread claiming of waivers by publishers and, where a waiver is claimed, that claim might be seen as unreasonable by an academic to the extent that they then might question their choice of publisher. Over 60 institutions worldwide have implemented a similar policy, including a number in the Ivy League. Harvard itself reports that waivers are claimed in less than 5% of cases. Widespread claiming of waivers under a similar policy in the UK would most certainly be seen as unreasonable.

There is growing interest in this approach at senior levels within UK HEIs, and more widely in the UK and Europe. A number of us in the UK are working together on adoption and technical practicalities.

To summarize the potential benefits of the implementation of the policy and licence:

  • implementation of such a policy would give academics a single means by which they can comply with RCUK, HEFCE and many other funder policies
  • academic freedom to publish in the journal of academic choice remains
  • the policy supports academics and their institutions to ensure maximum REF eligibility
  • early scholarly communication of research increases citation levels and therefore maximizes research impact. The policy supports this early communication
  • the licence minimizes the reliance on and expense of hybrid gold publishing in a period of longer-term funding uncertainty. Hybrid publishing is not contributing to the transition to OA as envisaged by the UK Finch group, but is just eating up money – 80% of our spend at Imperial – as well as resulting in the need for significant additional staff effort as a consequence of ineffective publisher processes
  • the licence maximizes the funding available to support pure gold publishing. As uptake of pure gold publishing increases, institutions will wish to protect available funding for APCs to enable academics that choose to do so to publish pure gold. Pure gold publishers have much more efficient processes in place than publishers of hybrid journals and continue to work with institutions to continue to drive further efficiencies in that process
  • adoption of the policy and licence results in a reduction of ‘double dipping’ whereby institutions pay twice through both subscriptions and APCs
  • adoption of the policy and licence minimizes both local and publisher efforts to manage hybrid APC payments – these are the messiest
  • adoption of the policy and licence minimizes the need for complex hybrid publisher offsetting negotiations and minimizes the complexity required to manage subsequent arrangements
  • wider adoption of the policy would enable institutions to repurpose available funds to support other pure gold initiatives, including those emerging beyond the journals market
  • institutional adoption of the policy would allow institutions to concentrate staff effort on something which supports the funder aims – a true transition towards greater openness and early scholarly communication of publicly funded research
  • adoption of the policy would align funder and institutional support for academics.

Conclusion

The key overall message is that the dialogue between HEIs and publishers is changing. HEIs now think about total cost of ownership and those costs no longer simply comprise the subscription costs to journal content. They include the support services that the publisher gives to the academic; the infrastructure that the publisher puts in place in order to minimize – or, ideally, all but eliminate – the transactional costs, particularly at the article level; the extent to which the publisher, alongside funders and institutions, adopts and implements the many standards which now exist to support easy and timely exchange of data; the extent to which they work with academics, funders and institutions.

We still need publishers: publishing is still a necessary and professional activity. What has been remarkable in the light of the complex policy environment has been the way in which that environment is now driving change and innovation in the sector. Those that continue to cling to a 17th-century model of fixed journal issue sizes and paywall-only access, and who give little back to the academic, will find themselves being challenged. Those.that innovate and that rethink their models in the light of this new ecosystem will thrive.

Abbreviations and Acronyms

A list of the abbreviations and acronyms used in this and other Insights articles can be accessed here – click on the URL below and then select the ‘Abbreviations and Acronyms’ link at the top of the page it directs you to: http://www.uksg.org/publications#aa

Competing interests

The author has declared no competing interests.