Gold or green: which is the best shade of open access?

Gold standard just right in the 'Goldilocks zone'

Michael Mabe hails the Finch report's consensus-building and its pragmatic support for the 'pay-to-publish' open-access model

Dame Janet Finch has my admiration and my sympathies. Having recently chaired a European Union-funded research project involving groups with very different frames of reference and viewpoints, I understand the challenge she faced in building and retaining an open-access consensus.

Make no mistake: the Finch report is a pragmatic compromise and an example in consensus-building. As Adam Tickell, pro vice-chancellor of the University of Birmingham and a member of the Working Group on Expanding Access to Published Research Findings, has said: "We ended up with a very much more consensual set of outcomes than any of us anticipated."

Not unexpectedly, voices have been raised in criticism from outside the Finch group. But the fact that some feel the report is too publisher-friendly, while some members of the association of publishers that I represent feel it goes too far, suggests that the balance may be closer to the "Goldilocks zone" than either the pros or the antis might suppose.

After extensive deliberations, the Finch group concluded (with some reservations) that "gold" open access - where the model is pay-to-publish rather than pay-to-read - would be the best way to maximise public and researcher access. Gold is not without its challenges. During the transition period, while waiting for other nations to follow suit, costs will rise: British funders will have to pay to publish the 6 per cent of world articles produced by the UK each year, while still subscribing to the other 94 per cent from the rest of the world. About 40 per cent of researchers are not supported by grants and for them, publication funds will have to be created. Ultimately, these funds will be balanced by the lowering (finally to zero) of library budgets for journals, but bridging funds will be needed in the interim.

The reward of gold is free universal access to the final published versions of articles immediately upon publication for academics and the public alike. For publishers, gold open access has the best chance of becoming a stable and sustainable business model.

The report does not ignore the alternative approach to open access - the "green" or self-archiving route - but is clear about its limitations and risks. Scholars want access to the peer-reviewed version of articles: the green route tries to do this by mandating authors to deposit their final peer-reviewed manuscripts in institutional or subject repositories. There is no business model for doing this, other than the assumption that whatever model is used will not be harmed by free access after an embargo period. The length of embargo crucially influences whether this "nobody pays" approach causes harm.

There is very little research on the effects of embargo periods, but it is clear that different fields will have different sensitivities. While preferring gold, the Finch group decided to include the green option, recognising the role of repositories in managing data and "grey literature", but was cautious on embargo periods, recommending that they be not less than 12 months. I think this is a reasonable compromise, but it may turn out to be too short a period for arts, humanities and social science journals, and even chemistry (where 50 per cent of lifetime downloads only occur after 18 months). However, it is certainly preferable, for most fields and most journals, to the widely chosen figure of six months.

I believe that in focusing on gold and how it can be turned into reality, the report made the right choice. The Peer project (Publishing and the Ecology of European Research), which I chaired and which ended in May, was set up to investigate the effects of green open access. It showed how difficult it is to source, manipulate and tag peer-reviewed author manuscripts. Green requires the development of an infrastructure and collaboration between publishers and repositories (unlikely outside experimental projects), including provision of high-level metadata and a mechanism to manage embargo periods, which most repositories lack.

Peer also showed that authors have little appetite for self-archiving. Invitations were extended to the authors of 11,800 research articles: only 170 were deposited. The other 20,000 manuscripts included in Peer were provided by participating publishers.

Discussing these difficulties, one European Commission official said: "If green is cumbersome, messy, involves assumptions about cooperation and investment in infrastructure, and still only delivers an imperfect version of the article, and then several months after publication, surely it's better to pay for the final version to be accessible upon publication?"

I couldn't agree more.

That's why the scientific, technical and medical publishing community is solidly behind gold open access. It is also why I warmly welcome the Finch report.

Michael Mabe is chief executive officer of the International Association of Scientific, Technical and Medical Publishers and was chair of the Peer project.

Don't deal in a debased currency - go green

A unilateral adoption of gold open access would come at the cost of UK competitiveness, David Price argues

It is a central irony of the Finch report that in seeking to maximise the accessibility of scholarly knowledge and evidence - and thereby encourage openness and transparency - its authors have failed either to consider fully the facts before them or to substantiate some of their assertions.

The Finch report is certainly a useful and important contribution to the open-access debate, and welcome in that it has driven this important issue further up the policy agenda. However, it represents a starting point for debate, not the conclusion.

A sustainable model of open access - sustainable, that is, for the producers, distributors and consumers of knowledge - is a goal I endorse. Access to leading-edge research will stimulate business innovation as well as academic research productivity. It will also greatly benefit the NHS and the health of the nation by facilitating the rapid translation of research findings into clinical practice and outcomes.

However, the transition to such a model must also be endured (and endurable). Someone will have to pick up the bill and the Finch report estimates that transition could cost more than £70 million a year, a tab to be met by the higher education sector.

Is there a viable alternative? The evidence presented by the Finch report does not provide an answer. It dismisses national licensing (under which, for agreed amounts, publishers allow access to their content by all sectors of society) as prohibitively expensive, without citing evidence (and despite its use in other European countries). Neither is green open access, under which authors self-archive their published papers in open-access repositories, comprehensively modelled.

There are other significant matters at stake. Would unilateral implementation of gold open access - a model that requires authors to pay fees to make their articles open access - really enhance the UK's competitiveness? Not only would the costs have to be met from existing research council and university budgets, which are already under significant pressure, but such unilateral implementation would also make all UK research globally available, with no requirement for the rest of the world to reciprocate. Certainly the UK should provide leadership in open access, but not at the cost of its own competitiveness. This element of the Finch report seems to me to be not merely an oversight but actually a fundamental flaw.

I think that the more preferable approach is to move to green open access, coupled with a national licensing arrangement. This would achieve the objectives set out in the Finch report of opening up access to the research literature while minimising the costs and associated risks.

Contrary to the report's claim that national licensing is too expensive, indicative modelling suggests that while unilateral gold open access would cost the UK £67.1 million a year, national licensing would cost only an additional £14 million or so. The crucial point here is that while gold open access would increase access outside the UK to the 6 per cent of articles written by British-based academics, national licensing would increase access inside the UK to 100 per cent of the research articles from around the world.

National licensing is clearly the more beneficial option for the UK: it offers a more effective use of the public purse, greater potential influence in negotiations with publishers, increased access to and dissemination of UK outputs, and greater access to a wealth of additional knowledge for research users in all sectors.

My suggested next step in the development of open access is, therefore, to see green open access fully implemented in the UK. The Finch report has missed a trick by failing to model green versus gold. Extensive economic modelling in a forthcoming Jisc-funded report by the Open Access Implementation Group shows that green is the cheapest option for universities. The study will also show that the best way to move from a subscription model to open access is via green, not gold. Green open access is also the route propounded by the European Union in its current funding programmes as a standard route for open-access publication.

At University College London, we have seen first-hand the value of green open access (our institutional repository, UCL Discovery, is the largest in the UK). With as yet only a small fraction of our output available, we are still recording more than 500,000 downloads a year, and the figure is growing.

The Finch report is right to conclude that the UK has an opportunity to demonstrate global leadership in the dissemination of research and in championing a knowledge-led economy. But it has surveyed the open-access landscape incompletely. Gold is not the answer: it is a so-called cure that is likely to be worse than the disease.

David Price is vice-provost (research) of University College London