It’s not even two months since the tectonic plates shifted underneath academic publishing in the UK. But in the few weeks since the government’s response to the Finch report and the announcement of the new open access (OA) policy of the UK Research Councils (RCUK), the ground has settled. The contours of the new landscape can be more clearly discerned but still lack definition in places. Not everyone is happy with what they see.
From April 2013 RCUK will require all its funded researchers to publish in OA journals. They will be able do this by paying an article processing charge (APC) from funds awarded by the research councils to the their institution so that the publisher makes the paper freely available on publication (gold OA). Alternatively, the researcher can go down the green OA route, in which the author’s unformatted, peer-reviewed manuscript can be posted online, typically in an institutional repository, several months after journal publication. This route is free but permitted under the new policy only if the publisher’s embargo is less than 6 months (12 for papers in the humanities and social sciences).
In principle, both gold and green routes to OA comply with the RCUK new policy, but in practice the emphasis is on gold.
Although I raised questions about the costs and implementation of the new policy, I was generally positive in my initial response to what I saw — and still see — as a bold move. But I have been reassessing in the light of analyses of Finch and the RCUK policy by people who have thought about open access for a lot longer than me: Alma Swan, Peter Suber and Stevan Harnad. Between them, these three have written a great deal about the ramifications of policy changes in the UK. Their analyses are well worth reading but they are pretty long, so I thought I would try to distill the key questions as we start to figure how the policy will work in practice, something that RCUK acknowledges still needs to be sorted out.
Is gold OA the best route to worldwide OA? In Going for Gold (PDF), published in June this year, Alma Swan and John Houghton looked at the likely costs of gold OA for a range of APCs and universities with different levels of research output. They concluded that all institutions — even the most research-intensive — would save money from worldwide Gold OA as long as APCs were kept under £2000. Savings would be considerable if the APC were to be held at the current average, which is £571.
But this refers to a post-transition world where there are no journal subscriptions and we are some way from that. The same report finds that use of green OA during the transition would be about 80% cheaper than the cost of gold OA.
RCUK have gone for gold OA because, if I have understood the argument on their blog, they want to insist on a CC-BY licence for papers to maximise their re-use since that “allows others to modify, build upon and/or distribute the licensed work, including for commercial purposes, as long as the original author is credited”. Certainly the licence facilitates exploitation of the research literature but should CC-BY be a priority for the transition?
In the comments on the RCUK blog Peter Murray-Rust has argued that ‘billions’ are wasted because the current literature is not machine accessible (which requires CC-BY), but Harnad sees licensing as a secondary issue and countered that the current restrictions due to expensive subscriptions costs ‘trillions’. For him the absolute priority is to establish free green OA so that the research community (the primary user of the academic literature) can get access. But who is right? Where do these numbers come from? The importance of text-mining has been analysed by JISC: they see clear benefits but don’t attach numbers to them. This is hardly surprising given the forecasting difficulties but clearer acknowledgement of this by opposing sides of the CC-BY argument might lead to more productive discussion.
Can we pave the transition with gold? Swan, Suber and Harnad all have concerns that the emphasis on gold OA for the transition period is too costly. Finch estimated that £50-60m per year would be needed to cover excess costs during the transition (when APCs and subscriptions would both have to be paid), but these figures are very uncertain. In any case, just as important is the need to consider how the RCUK policy is likely to affect the behaviour of publishers.
Harnad and Suber have both expressed the fear that the policy is a gift to publishers because they could simply extend their green OA embargo period to beyond 6 months in order to oblige authors to pay gold OA APCs to comply with the RCUK stipulations. The temptation to adopt this stratagem seems irresistible; it makes good business sense, especially for journals that trade on their impact factors. The policy could therefore simultaneously inhibit the spread of green OA options and lead to hikes in APCs. Suber also points out that journals that currently offer free gold OA publishing will be induced by the new RCUK policy to start charging. These are perverse outcomes for a policy designed to promote open access.
Swan sees a threat to costs from another direction, arguing that RCUK’s preference for gold over green OA favours the status quo by protecting the income streams of publishers and so inhibiting the entry to the market of publishing innovators who are likely to offer better value for money.
We shall have to see how that plays out — much remains frustratingly uncertain — but the development of the market will depend very much on the behaviour of purchaser and under the new policy these will be researchers’ institutions. To fund APCs the research councils will allocate block grants (from existing budgets) to the institutions where their grant-holders work. But how will APC funds be determined by research councils and how will universities allocate them? Given that the over-stretched UK research budget is in decline, it is unlikely that universities will get all the funds they need and I see scope aplenty for administrative headaches. Universities and their researchers will be torn between maximising value for money and — as ever, alas — chasing after expensive high impact factor journals to feed their REF returns and grant applications.
These tensions will be exacerbated if publishers move to eliminate green OA options in the quest for gold OA fees from UK researchers.
Swan, Suber and Harnad all agree that incorporation of a strong mandate for green OA in the RCUK policy would help to remedy many of the problems that have been identified. The lack of emphasis on green OA is seen as the direct result of the influence of publishers on the Finch working group, who were understandably seeking to protect their interests. But the trio of commentators agree that UK policy should first serve the interests of researchers (and the taxpayer) rather than those of publishers. The government is oddly conflicted in its attitude to the policy: although keen on value for money and the winnowing power of market forces, it appears to want protect publishing companies based in the UK from the storms of uncertainty that the push for OA is creating.
The lack of resolution of the gold vs green emphasis is problematic, particularly because no-one, and I really mean no-one, has an idea how long it will take to complete the transition to worldwide OA (when subscription monies can finally be transferred to pay APCs). For how many years is the UK prepared to pay excess transition costs? That question immediately raises another one: why isn’t the RCUK policy aiming to minimise the length and therefore the cost of the transition?
Part of the answer is surely that we cannot calculate how long it will take other countries to make the same commitment to OA as the UK. It is clear that concerted international action will be important, as Science Minister David Willetts has already acknowledged. But if coordination between governments is so crucial, asks Swan, why is the UK pursuing a policy that is out of kilter with the US, the EU, Denmark and Australia where green OA mandates are gaining traction?
I wish I knew the answers to these questions. I remain upbeat about policy developments in the UK: we have made huge strides this year. But in my darker moments I fear that we will still be arguing about these issues in 10 or 20 years time and we cannot afford to do that. I hope that all sides will recognise that the current position cannot be final, that the landscape is still shifting and that it needs to be shaped by us to make sure that worldwide OA works.
Just FYI,
one estimate of the potential value of textmining (content mining) can be found in this Guardian article http://www.guardian.co.uk/science/2012/may/23/text-mining-research-tool-forbidden
“A report published by McKinsey Global Institute last year said that “big data” technologies such as text and data mining had the potential to create €250bn (£200bn) of annual value to Europe’s economy, if researchers were allowed to make full use of it.”
(and here’s the link to that McKinsey report [I haven’t read it myself]: http://www.mckinsey.com/Insights/MGI/Research/Technology_and_Innovation/Big_data_The_next_frontier_for_innovation )
Furthermore, I often think we forget that researchers are not ‘forced’ to submit their articles to one journal or another – they have *choices*. If the traditional big publishing companies offer extremely expensive poor-value Gold OA for publishing in their traditional journals, then researchers should simply look to newer, cheaper, more efficient journals to publish in.
I do of course realise that some researchers feel they need to play ‘the Impact Factor’ game but if we can get rid of this mentality then I think you’ll find Gold OA won’t be very expensive. It is a big ‘if’ though to change researcher behaviour in this way.
Thanks for those links. I guess the McKinsey report will have to be read carefully to gauge the validity of their numbers. But it does point (as does JISC) to significant value.
Building evidence-based policy is critical to the success of funding the advancement of science, and focussing on what the users (readers, authors, +) need is probably even more important (Governmant’s can’t pay for everything!).
Publishers who undestand this – who foster the best technologies, aim for the right efficiencies, target the real cost savings, and build sustainable business models with their user communities – are well placed in ensuring a role in the long term evolution of academic and professional publishing.
The fundamental here is working together with researchers to achieve what they want out of the market, and where possible to do better still. So while gold and green OA seem the only options to some in achieving this, the question remains of how the market will grow, mature and what the means and needs will be in the future, where costs will still be present.
(My opinion of course.)
I can’t pretend to fully understand the economics of academic publishing but, though some companies will want to fight for the status quo, as Willetts has said, that’s the wrong battle. I agree that nimble-minded OA publishers will make a decent living.
Nor can I pretend to know all the political angles! But, as a publisher, I would agree that good policy enables individual researchers to make a choice (even if to a subscription rather than OA publisher) – the degree to which this is qualified by funder mandates remains a contentious one given the disparities between disciplines and fields within them.
Individual behaviours define the evidence base, and so empowering choice and encouraging moves to OA (rather than absolute compulsion), inclusive of options for re-use licensing (not all researchers are fans of CC-BY), can help to both build a better understanding of how different fields/researchers work as well as building evidence of where funding can be best deployed.
Having nimble/agile publications (and in aggregate, publishers) – or at least thinking in these terms more – will help drive the transformation of online publishing, but it’s interesting to note too that primary publications, and minimal publishing units, will need additional layers and integrated platforms to operate effectively – another realm of financing/funding, given the additional infrastructure and operation demands over and above the foundation of what the Internet offers.
I’m very interested in seeing how this fits best into the research and workflows of individuals, and how the wide range of offerings we’ll see can be effectively integrated rather than potentially siloed if without the right level of interoperability. Hopefully the money-go-round will see us to a better place for science and scholarship, and this can only be achieved by thinking this all through, and of course with effective policy frameworks and support.
I like to think we should all apply some clearer scientific method in our thinking to guide how we proceed. Having already seen some experiments working well, and if they endure we could well move to a ‘new normal’ (a great political phrase I heard again yesterday).
Thanks for such an insightful comment. I would really like to see more scientists getting involved in discussing this issue. Activity on the blogosphere translates only slowly into raising of consciousness within the wider academic community.
Interested to know why some scientists are not fans of CC-BY for papers (assuming any IP for the research can be protected before publication). I think funders should have a strong say in this. But people’s view will be coloured by how important they think text mining is.
Another great post, Stephen.
Don’t have *much* to add other than…
a) It’s good to openly discuss these issues especially in light of “tectonic plates” shifting that has occurred here. I certainly welcome events such as http://scicommforum.eventbrite.co.uk/ later this month which will stimulate just that
b) Horses for Courses. Different methods of making research OA will suit different academics in different fields, so no one “size of glove fit’s all” but as Jan Velterop pointed out when I recently interviewed him, the issue of reuse is an important one:- http://figshare.com/blog/More_Than_One_Road_Leads_to_Rome_-_An_interview_with_Jan_Velterop/46
Also of note is Velterop’s comment of
“What always surprises me in these discussions is their national focus, whereas science is one of the most global enterprises on earth. The most positive developments for OA have been the greater awareness of it, even in the general media. Little else is new. And even attention to open access by the Guardian isn’t, as this article from February 2005 shows”. http://www.guardian.co.uk/business/2005/feb/02/politics.science
c) Whilst I can’t find the *exact* quote, Bill Hooker made a great statement about the changing landscape of STM Publishing circa 2008 along the lines of “Who knows how all of this will end up, but we’ll like it when we get there in the end”.
And to that, I find the last para. of this post rather poignant !!
Thanks for persevering Graham. Looking forward to the depate!
Thanks for the useful summary, Steven.
I’m still left wondering about one important question – what happens to back catalogues? IIRC, Nature (and others) have used a shifting time-window for subscriptions, offering access to only the last (e.g.) 10 years of material electronically. If a subscriber wants to access articles that are >10 years old, they often have to pay extra. Once OA has passed through the teething stage and is aiming for it’s tweens and full independence from the subscription model, what will happen to the archives? Will we still need subscriptions for access to that?
Any ideas whether this has been brought up in the discussions yet?
That’s a good question. I don’t recall the issue of back catalogues being addressed in the Finch report. I wonder if some agreement could be reached between the major (and wealthiest) research nations to make a one-off payment or arrangement with publishers to make sure that these back catalogues are placed permanently in the public domain. The most important (valuable?) are likely to be papers from the last 20 yrs. Before that, the work is starting to be mainly of historical interest.
I should add that some publishers, e.g., the Royal Society, have already done a great job of making their archives freeely available. However, others have a bit of catching up, and justifying, to do, if they plan to continue charging for access to archives.
I can think of 2 very important caveats to this:
(1) The field of study
(2) The age of the researcher :p (new students won’t have had time to build up a pdf database)
Your caveats are important. I know from my work on human serum albumin that I have often found myself digging out papers from the ’60s and ’70s.
Plus the historical worth is also very important — I think there are likely to be many hidden gems as this recent translation of the classic Michaelis & Menten paper showed.
And – the Journal of Biological Chemistry has put all of its catalogue online – back as far as 1905. Make you wonder why others won’t simply do the same…
Probably something about a belief in money and old rope…