Thursday, February 21, 2013

Open Access: A Tale of Two Tables

In an article published recently by the American Institute of Biological Sciences, Colin Macilwain concludes “If the open-access story started as a battle between open-access advocates and publishers, it seems to have morphed into a feud between gold and green open access, which cuts out the publishers.”

There is something in what Macilwain says, although in truth there has always been disagreement within the OA movement, and sometimes bitter wrangling between those who espouse Green OA and those who espouse Gold OA.

But there is no doubt that the publication last year of the Finch Report has brought a new intensity to this inter-movement discord, particularly after the UK Minister of State for Universities and Science David Willetts accepted all bar one of the Finch recommendations — making the Finch view official UK government policy on OA. This saw Research Councils UK (RCUK) immediately introduce a new OA policy in order to comply with Finch, a policy that will come into effect on 1st April.

RCUK’s new policy has had the effect of rekindling and intensifying a number of disagreements within the OA movement, including a disagreement over the relative merits of so-called libre and gratis OA (the RCUK Policy requires that where authors pay for Gold OA their paper must be published under a CC-BY licence to allow reuse), and a disagreement over what constitutes an appropriate embargo length when researchers opt for Green OA (the RCUK Policy specifies either 6 or 12 months, depending on the research field). 

But above all, as Macilwain noted, the new policy has reignited a long-standing disagreement between advocates of Green and Gold OA, especially over whether one form of OA should ever be prioritised over another (The RCUK policy controversially prefers Gold over Green, either pure Gold or Hybrid OA).

Historically, there was a consensus within the OA movement that it was not realistic to try and strong-arm researchers into publishing in OA journals, although you could mandate them to self-archive (Green OA).

Unfortunately, ambiguity over the precise requirements of the new policy has also led to considerable confusion and concern, not just over what the policy requires, but whether in its preference for Gold OA the UK is moving in step with other countries, or taking a risky new direction that could cost the country dear.

This confusion is graphically represented by two different tables that were drawn up to show how the UK policy compares with other OA policies around the world — one published by the RCUK (here) and one published by OA advocacy group SPARC Europe (here).

Strikingly, although both tables were intended to show the same thing, they offer a very different picture of the RCUK policy vis-à-vis the rest of the world.

Green or Gold?

However, to understand why these two tables are so different we need to explore the background to their creation.

First, let’s remind ourselves how the two “roads” to OA differ. With Green OA researches continue to publish in subscription journals, and then make their papers freely available on the Web (usually by posting them in an institutional repository, or a central repository like PubMed Central). This is usually done after an embargo period of anything between 6 months and three or more years — a delay intended to allow the publisher to recoup the costs it incurred in publishing the paper.

Gold OA, by contrast, is where researchers publish their papers in an OA journal. Here the publisher makes the papers immediately and freely available on the Internet for the researcher, but usually requires the payment of an article-processing charge (APC) for doing so. By making a one-off, up-front payment, researchers are able to ensure that their paper is made freely available to all, not locked behind a subscription paywall.

It is important to note that there is a variant of Gold OA: Hybrid OA. This is a form of Gold OA that allows authors to publish in a subscription journal but, on payment of an APC, request that their paper is placed outside the journal’s paywall

Critics of RCUK’s decision to prefer Gold over Green OA complain that it will be hugely wasteful of public money, since the costs of publishing research are already being paid by journal subscriptions — although Green does usually mean a delay before OA is provided.

Moreover, say critics, by telling subscription publishers that their UK authors must henceforth either pay to publish their papers, or make them freely available on the Internet 6 or 12 months after publication — while stating that Gold OA is the preferred route — RCUK will simply encourage publishers to introduce a Hybrid OA option for all their journals (which we are already seeing happen in preparation for the new policy), and then do all they can to prevent authors from opting for Green — because publishers dislike Green OA with a vengeance, especially when it allows for no more than a six month embargo.

Publishers are more likely to provide Hybrid OA (rather than pure Gold), critics predict, because if they converted all their subscription journals into OA journals they would have to forego their current subscription revenues.

Moreover, add critics, publishers are able to charge more for Hybrid OA than for pure Gold OA (Hybrid journals charge between $3,000 and $5,000 per paper). Consequently, the new policy will impose a significant new cost burden on UK universities at a time when their budgets are already straining at the seams. The Finch report estimated that the additional costs arising from its recommendations would be between £50 and £60 million a year. Critics predict that they will be much higher.

Above all, say critics, as the UK only accounts for around 6% of global research output, UK universities will still need to buy access to the 94% of research produced in the rest of the world (even though some of this will be published in OA journals). As such, they will not be able to cancel their journal subscriptions, and so will end up paying twice, first through APCs, second through subscriptions. For publishers, this will mean being able to “double dip”, increasing their earnings without providing additional value.

It is unsurprising, therefore, that when RCUK’s new policy was announced universities complained bitterly. And these complaints have persisted despite Willetts subsequently agreeing to provide an additional £10 million (not new money, we should note, but money from the existing science budget) to help research-intensive universities pay the additional costs.

House of Lords Inquiry

Such has been the ferocity of the criticism, in fact, that in January the House of Lords Science & Technology Committee launched an inquiry into RCUK’s implementation of the government’s OA policy.

Amongst those called to give evidence to the inquiry were Dame Janet Finch (Chair of the Finch Committee) and David Willetts (transcript here), as well as representatives from the RCUK  — RCUK Chair Rick Rylance and RCUK’s Information Champion, Douglas Kell — plus David Sweeney, Director (Research, Innovation and Skills) of the Higher Education Funding Council for England (HEFCE) — transcript here.

As the Committee began to explore the complaints it had received, a key issue emerged: Was RCUK’s policy in line with what the rest of the world was doing with OA, or was the country taking a risky gamble in the hope of acquiring a leadership role in the development of OA?

David Willetts was forthright on this, saying that he wanted the UK government to set a “clear position” on OA, and that he believed doing so would “influence the debate elsewhere and would have an influence across the EU and could have an influence in the US, and I think that there is a pretty clear trend in both the EU and the US to move in the same kind of direction.”

Rick Rylance also appeared keen to talk up the UK as an OA leader, which he suggested would offer benefits for the country (although I am not sure he spelled out what these benefits would be).  As he put it to the Committee, “There is some degree of advantage to be gained from being a first mover.”

Nevertheless, Sweeney conceded that there was a potential downside. “There are also risks, of course, involved in that, and that is what we have to try to calculate on a regular basis.”

It was these risks that the Committee was particularly keen to explore. And its concern was that they had not been adequately assessed prior to the new policy being introduced.

For this reason, the Committee repeatedly asked what analysis had been done by the government, and by RCUK, to assess the potential risks and benefits to the country of becoming an OA trailblazer. In replying, Willetts conceded that his department had “focused on a much narrow set of questions”, primarily looking at “public expenditure costs of different routes to green and gold.”

When the Committee took evidence from RCUK and HEFCE Lord Willis of Knaresborough asked the same question in this way: in developing its new policy, he suggested, RCUK must have done some work to estimate the risks, “and there must be some figures somewhere that indicate there is going to be some real advantage. Why on earth would you want to lead this on a global scale if you could not justify that?”

Responding to the question, Sweeney replied, “I do not understand the, ‘We are leading it on a global scale’. All over the world — and indeed the Research Councils will provide you with our table — people are adopting open access policies.”

In short, unlike Willetts and Rylance, Sweeney seemed inclined to dismiss the suggestion that the UK was aiming to be a trailblazer. Rather, he implied, it was marching down the OA road arm-in-arm with the rest of the world.

In defence of the RCUK policy Sweeney cited some research undertaken by the economist John Houghton into the differing costs of Green and Gold OA. When asked by the Earl of Selborne why he thought Gold was more sustainable, Sweeny replied, “The work from John Houghton in Australia, who I think is the premier economist in this area, suggests that that is so. He does raise concerns about the transition period and the additional costs that there may be from pursuing green and gold, but you started off by saying green is cheaper. That is not what the literature says.”

Critics later suggested that Sweeney had misrepresented Houghton’s findings. In fact, they said, Houghton had concluded that while Gold OA might prove cheaper in the long term, a more cost-effective transition strategy would be to prioritise Green OA.

But what was the table that Sweeney referred to? It was a table that Kell had mentioned earlier in the hearing. Explaining to the Committee that the UK was not acting alone, Kell had said, “In our evidence we set down some of the international context. A great number of the European countries are already implementing this. The Australian Research Council is already doing this. The ERC is going to be debating it in its meeting coming up, and the Global Research Council has it as its main agenda item in May in the USA, so there is a very great deal of interest and implementation going on, and much of it following what is coming in the UK.”

To support his argument Kell had added, “We have a large table, which we intend to submit as post-evidence, because we have looked at all of the implementations in Europe, for instance, and almost every country has a very clear implementation. While we are driving this thing, we are very much not alone.”

Nevertheless, the evidence session with RCUK and HEFCE appeared to leave the Committee in some doubt as to whether and how the UK policy differed from other OA policies being introduced around the world. And they were concerned that if the UK was taking a lone road then it was important to know if anyone had established what level of risk the country might be exposed to.

Table One

For this reason, presumably, in closing the hearing the Chairman stressed that the Committee would like to see the table that Sweeney and Kell had referred to. As he put it, the Committee wanted to see, “the international summary of where different countries are on open access and their plans and policies for green and gold open access in the future.”

In addition, the Chairman asked to see “whatever material you have on the cost-benefit modelling, both in terms of the costs of green in relation to gold, where you said, I think, there was a study that demonstrated that gold was a cheaper option than green, and on the cost-benefit modelling of the cost to the higher education sector in terms of transfer to open access and the benefits to UK industry, if you have any modelling on that.”

The RCUK table was subsequently submitted as supplementary written evidence, and can be viewed in this large pdf file. 

So what is the RCUK table? Essentially it consists of a list of 41 OA policies introduced by research funders in different countries. Anyone scanning the table would doubtless conclude that it confirmed Sweeney’s claim that the RCUK policy is much of a muchness with all the other OA policies being introduced around the world — although we should note that the table only covers Europe.  

Thus, for instance, anyone examining the table (which can be viewed in full here) would probably see little or no difference between the RCUK policy and the policies introduced by the European Commission and the European Research Council.

Critics, however, point out that this apparent conformity is a function of the way in which the table has been designed. Specifically, it does not indicate what the different policies require. As such, they suggest, the table occludes the fact that the RCUK preference for Gold is not echoed in any of the other policies listed. Rather than indicating what the policies require, the table merely records the different flavours of OA encompassed by them.

After reviewing the table de facto leader of the OA movement Peter Suber commented:

“I find the table very misleading. It suggests that nearly all the European funder policies are ‘gold’ policies. But none of these policies, with the possible exception of the RCUK policy, actually requires gold. Hence, the table must mean that these policies encourage gold, permit gold to satisfy a 'colourless' OA requirement, or authorize paying fees at fee-based OA journals.But it's not clear which. Moreover, the uninformed reader will think that the policies simply require gold, which would give false momentum to a bad idea. An undifferentiated ‘gold yes’ label will only muddy the waters already muddied by the Finch report.”

OA advocate Stevan Harnad agreed. Moreover, he says, there are some significant errors in the table:

“This document unfortunately contains several important errors, and, in addition, there is an ambiguity about ‘Green’ and ‘Gold’ Open Access that needs to be clarified, because the UK government is using this document as part of its evidence base for a controversial new policy, and the ambiguity risks engendering a major misunderstanding”

He adds:

 It is not sufficient to indicate whether an OA Policy includes Green, Gold, or both. The point at issue is about what the policy requires, i.e., mandates.

“According to ROARMAP and other sources, all OA policies worldwide include Green, without exception ([although the RCUK table] erroneously lists two exceptions). 

“Many OA policies (but not all) also include Gold. (And fewer policies offer to pay Gold APCs.)

Harnad concludes:

“The most important datum, however, is that many of the policies require Green, but none (with the possible, exception of RCUK) prefer or require Gold.

“Most of the policies simply require OA, and allow it to be provided either via Green or via Gold.”

When I put these points to an RCUK spokeswoman she responded, “The table that has been pulled together to show the positions of European funders has been based on information provided by the funders themselves. They have also all checked it for accuracy.”

She added, “[T]his table is a work in progress, having been originally only pulled together for briefing purposes, and we will continue to work with research funders across Europe to improve, refine and add to it in what is a fast moving agenda.”

In fact, the table was not produced by RCUK itself, but by the Open Access Working Group of Science Europe, and it was compiled from the results of a survey of its member organisations that Science Europe conducted.

Since it covers only Europe, the table cannot fairly be said to show how RCUK’s policy fits with what is happening globally, even if it had included the details of exactly what all these policies require. We should also note that only 41 of Science Europe’s 51 members (80%) appear to have responded to the survey. It is for this reason, presumably, that RCUK describes it as a work in progress.

But are there several important errors in the table, as Harnad believes? RCUK suggests not. For further clarification I emailed over three questions to Science Europe. However, I was not able to get a reply to these prior to publishing this.

One is inclined to conclude that even after the evidence sessions it held the Committee was probably unclear as to whether the RCUK policy is unique in requiring that its funded authors prefer Gold OA and, if it is, what the implications of that are likely to be. And it is not apparent that the RCUK table will have helped very much here. But perhaps the rest of the supplementary written evidence requested by the Committee will have clarified matters.

If the Committee is still unclear, it will not be alone in that. There remains widespread confusion within the research community itself as to exactly what the policy requires. Can authors, for instance, only opt for Green OA if their publisher does not offer a Gold option (Hybrid OA or pure Gold OA) — as some critics maintain?

To do it justice, RCUK has on a number of occasions sought to clarify matters.

Nine words

On 28th September last year, for instance, Mark Thorley (chair of the RCUK Research Outputs Network) published a blog post explaining when authors can opt for Green and when they should opt for Gold. In doing so, he argued that there was greater flexibility than critics claimed.

As he put it:

If the journal they want to publish in only offers policy compliance through a Gold route, they must use that journal’s Gold option. If the journal only offers compliance through the Green route, the author must ensure that a copy of the post-print is deposited in an appropriate repository – for example, UKPMC for papers arising from MRC funded research.  If the journal offers both a Gold and a Green route to compliance (and some journals already do this), it is up to the author and their institution to decide on the most appropriate route to use.

While this seems clear enough, critics argue that it is not what the policy actually says. Moreover, they add, RCUK fails to appreciate that publishers (most of whom believe the policy’s embargo period is far too short) will probably respond by offering a Gold option, and then set their embargo period outside the 6 or 12 months stipulated by RCUK, thereby making their own Green policy non-compliant with RCUK’s. In such a scenario, critics point out, authors would have little option but to take the Gold route, and doubtless the more expensive Hybrid version at that.

This, critics add, is an inevitable consequence of RCUK’s stated preference for Gold OA. And RCUK’s preference for Gold is the one certain part of this confusing story. We know that because on 24th October Thorley published another blog post, which was headed, “RCUK Open Access Policy — Our Preference for Gold”. And he went on to explain why it was preferred.

He wrote:

The first of our four key principles is that the ideas and knowledge derived from publicly-funded research must be made available and accessible for public use, interrogation and scrutiny, as widely, rapidly and effectively as practicable. It is this principle which is at the heart of our preference for Gold.

Thorley added:

“For us ‘use’ means much more than just being able to read research papers – it means having the ability to re-use and exploit research papers in the widest possible sense – be that text and data mining to advance new areas of research, to re-presenting collections of research papers in particular areas, to mashing together elements of research papers with other information to create new information products.

This latter point explains why RCUK favours CC-BY, which it assumes is only possible with Gold OA. We could note, however, that Suber believes this to be an incorrect assumption.

What is also evident from Thorley’s second post is that RCUK dislikes the delay that generally accompanies Green OA. However, RCUK seems unconcerned that its policy could see Gold OA become the only option available to authors, and at a high a price too — presumably because it prefers Gold.

Moreover, even Thorley’s posts did not fully clarify the situation. And this is partly because of the wording of the policy which many appear to misinterpret, and in a way RCUK did not perhaps anticipate.

The policy wording includes the following: 

The Research Councils will recognise a journal as being compliant with their policy on Open Access if:

1. The journal provides via its own website immediate and unrestricted access to the publisher’s final version of the paper (the Version of Record), and allows immediate deposit of the Version of Record in other repositories without restriction on re-use.  This may involve payment of an ‘Article Processing Charge’ (APC) to the publisher. The CC-BY license should be used in this case.


2. Where a publisher does not offer option 1 above,* the journal must allow deposit of Accepted Manuscripts that include all changes resulting from peer review (but not necessarily incorporating the publisher’s formatting) in other repositories, without restrictions on non-commercial re-use [presumably CC-BY-NC] and within a defined period. In this option no ‘Article Processing Charge’ will be payable to the publisher. Research Councils will accept a delay of no more than six months between on-line publication and a research paper becoming Open Access, except in the case of research papers arising from research funded by the AHRC and the ESRC where the maximum embargo period is 12 months.

* My bold

Critics point out that this wording inevitably leads researchers to conclude that they can only opt for Green if their publisher offers no Gold option.

But this is surely a misunderstanding, and based on the assumption that the wording is a direct instruction to authors (i.e. “Only where a publisher does not offer option 1 can you opt for option 2”), whereas in reality it is a description of the two ways in which a journal can be compliant. It is no doubt because of this misunderstanding that RCUK has often appeared genuinely perplexed by some of the criticism its OA policy has attracted.

Harnad has suggested that this ambiguity and confusion could be avoided if RCUK were simply to strike out the nine words emboldened above. After all, he points out, they are supernumerary.

But is seems unlikely that RCUK will want to do this. After all, Thorley has made clear that RCUK prefers Gold, and so it presumably wants authors to choose that route. Anything that might drive them down that road, presumably, is to be welcomed — regardless apparently of the cost implications.

In September Peter Suber said this of the policy:

The RCUK policy is not a gold OA mandate, or not a simple one, because in some circumstances it can be satisfied with green. But it deliberately steers authors toward OA journals and in that sense approaches a gold OA mandate. When the author's journal offers no suitable green option, the policy becomes a definite gold OA mandate. (Moreover, it's a gold policy with incentives for journals not to offer suitable green options.)

And he made this prediction:

Non-hybrid TA [subscription, or Toll Access] journals will add APC-based gold options, or convert from TA to hybrid OA, in order to collect fees they are not now collecting. Nothing I've seen in the RCUK policy or Finch recommendations even prohibits double-dipping (charging APCs for OA articles and subscriptions for all articles, including the OA articles). Adding a double-dipping hybrid option is an easy move for a journal to make, and it's easy money.

No-fee OA journals will add APC-based gold options, or convert from no-fee OA to fee-based OA, for the same reason, to collect fees they are not now collecting.

If I'm right, we'll see a decline in full-TA journals and a corresponding rises in hybrid OA journals. And we'll see a decline in no-fee OA journals and a corresponding rise in fee-based OA journals.

If Suber is correct, we can presumably expect to see a community that has for several decades now been groaning under the yoke of the so-called “serials crisis” face an even greater financial burden.

Table Two

So let’s move on to Table 2, which was published on February 11th by OA advocacy group SPARC Europe. SPARC’s aim was to provide what it views as a more accurate picture of how the RCUK policy compares to those in the rest of the world.

The SPARC table lists 48 funder polices from the ROARMAP registry. And it covers not just Europe but the world (including North America). It also takes into account what the policies actually require. The result: 33 of the policies (68%) require Green OA, 14 (29%) state that either Green or Gold will suffice, and just one (RCUK) is listed as preferring Gold.


Note, the table was altered after publication. (Previously it had said Gold OA was "required". This was later changed to "preferred"), as will become clear below. (The current table can be viewed here: In addition, the number of Green Required policies was changed to 36 from 33, and the total changed to 48) Nevertheless, whichever version of the table one views, the RCUK policy stands out as unique, an outlier different to all the other policies SPARC examined, and perhaps different to all other OA policies in the world.

When I asked the RCUK spokeswoman why she thought the two tables convey such a different picture, she replied, “Inevitably providing any information like this in table format is going to lose some of the nuances. For example, the SPARC table could be taken to imply that the Wellcome Trust only supports Green OA.”

It is clear that RCUK and SPARC Europe see things differently.  However, the more important point for the OA movement is surely the one highlighted by Colin Macilwain: The continuing row over Finch/RCUK is sowing discord and bad feeling. And some fear that this discord could prove more damaging for the OA movement than previous disagreements.

This danger became all too evident when SPARC Europe published its table.

Posting to the Open Knowledge Foundation (OKF) OA mailing list two days after its publication, OA advocate Ross Mounce commented:

“In what I infer to be a rather political move, SPARC Europe has just published a classification of open access funder mandates. RCUK is listed alone under the classification title: ‘Gold (journal-based) Open Access required:’

Whilst there are 14 funders listed under: ‘Either Green or Gold routes satisfy policy requirements’ my understanding of the UK policy was that it allowed either gold or green OA, with at most gold preferred.

Mounce added:

Why then is SPARC Europe making such a point of classifying RCUK as ‘Gold OA required’? This is flat out wrong IMO. Preferred does not equal required! In no way is gold absolutely *required* by the new RCUK policy. This is clear misrepresentation.

Why did they do this? Are SPARC Europe trying to help or hinder the Open Access movement?

Velichka Dimitrova agreed, “I think to single out the RCUK in this way is not justifiable.”

And Peter Murray-Rust suggested that a formal statement be published explaining the concerns that had been raised over the SPARC table: “This is important in that any inconsistency will be seized on by opponents of funder mandates in general.”

Subsequently, Mounce posted a response he had received from Director of European Advocacy at SPARC, Alma Swan. Mounce had emailed Swan to say that he was perplexed as to why the RCUK policy had been placed in a class of its own, and to ask why SPARC had described the policy as “requiring” Gold OA. “I fear RCUK's lone listing may adversely affect opinion of it," he wrote. "Furthermore it undermines the credibility of SPARC if they publish untrue statements such as this.”

Swan replied:

I am sorry you are perplexed. Our classification, unlike that of Science Europe, is trying to show the differences between policies and the directions in which policymakers appear to wish to travel.

This classification acknowledges the ground-breaking move that RCUK has made. The policy requires publication in an RCUK-compliant journal, which it defines as one that provides immediate OA (on payment of an article-processing fee if it requires to be paid). If the journal does not provide OA, then the Green route can be used. No other policy in the world is the same as this and the classification highlights this individual stance.

Fortunately, however, this particular disagreement had a happy ending. When I later emailed Mounce he replied, “Please do note that the wording of the SPARC Europe classification has changed from ‘required’ to ‘preferred’ so it now reflects reality.”

He added, “It would be good if you could mention that everything is harmonious once again with this change. I have no objection to the wording as it is now. OKF & SPARC Europe are collaborators on projects together so it’s good that disputes like this can be so quickly resolved.”

Nevertheless, it is hard not to conclude that the ambiguous wording of the RCUK policy, RCUK’s preference for Gold OA, and the fact that it appears to be the only funder in the world to state this preference in its OA policy,  will have an uncertain, and likely negative, impact on the UK research community. And if the UK policy is seen to fail it could hold back the development of OA more widely.

Moreover, the discord that Finch/RCUK have sown amongst members of the OA movement (of which RCUK is now presumably also a member) will surely make it harder to end the confusion surrounding the policy.

In short, the big problem facing the UK is that its national OA policy is sufficiently confusing, and confused, that there is likely to be continuing uncertainly and disagreement about exactly what it requires, and what implications it might have for the country. And this was clearly what worried the House of Lords Science & Technology Committee.

Who’s leading now?

But whatever the meaning, whatever the outcome, and whatever the implications of the RCUK policy, recent developments in the US look set to undermine any hope the UK might have had of taking a leadership role in the OA movement. These developments could also further increase the risk the UK faces as a result of its new policy.

Some would argue that the US has long been the natural leader of the OA movement, a leadership role it could be said to have acquired in 2005, when the US National Institutes of Health (NIH) — the largest medical funder in the world — introduced its controversial Public Access Policy. This, we should note, is a Green OA policy, requiring scientists to submit final peer-reviewed journal manuscripts arising from NIH funds to the digital archive PubMed Central upon acceptance for publication.  The Policy then requires that these papers are made accessible to the public on PubMed Central no later than 12 months after publication.

Thus as the UK continues to wrangle over its OA policy, the US may be about to cement its leadership of the movement. On February 13th Representatives Mike Doyle (D-PA), Zoe Lofgren (D-CA), and Kevin Yoder (R-KS) introduced the Fair Access to Science and Technology Research (FASTR) Act into Congress (More here).

The same day a Senate version of the bill was introduced by John Cornyn (R-TX) and Ron Wyden (D-OR) — providing the bill bipartisan support. And bipartisan support, points out Scientific American, is “rarer than hen’s teeth” so far as US bills are concerned.

FASTR would extend the NIH policy to all US funding agencies with annual budgets of more than $100 million, which Nature estimates would approximately double the number of research papers publicly available each year.

At the same time, FASTR would reduce the embargo period from 12 to six months.

In addition, says Suber, the bill includes three new provisions calling for libre OA or open licensing, suggesting that it may not be necessary (as the RCUK maintains) to require Gold OA in order to ensure re-use rights.

Reporting the news, Nature pointed out that the FASTR approach is very different from that being adopted by the UK, “where government-funded science agencies have plumped for another idea: authors paying publishers up-front to make their work free to read.”

Nature added, “Under that ‘gold’ policy, some cash is being taken out of the UK science budget to help authors pay for their publications, while UK libraries are continuing to pay subscriptions.”

The hope, explained Nature, is “that this extra-cost ‘transition’ period will be short as other countries leap to embrace the UK’s way of thinking, but it’s looking less likely that the US and Europe (which wants open-access in its 2014-2020 research programme, Horizon 2020) will follow the UK’s lead.”

At a stroke, the risk the RCUK policy could pose for the UK begins to look much greater. If the rest of the world follows the US lead, rather than the lead of RCUK, the UK will likely discover that the extra transition costs it anticipates (paying both APCs and subscription) could continue indefinitely.

The House of Lords Science & Technology Committee has announced that it plans to publish its report tomorrow (Friday, 22nd February).  OA advocates will doubtless be keen to see what the Committee has made of the confusion and unrest surrounding the RCUK policy.

But UK navel gazing will not end there: On 21st January the House of Commons Business, Innovation and Skills Committee said that it too plans to hold an inquiry into OA. 

*** UPDATE 22nd February 2013 ***

The House of Lords Science and Technology Committee has now published its report. In doing so it has criticised Research Councils UK for failures in its communication of its open access policy. The report says the previous lack of clarity about RCUK’s policy and guidance was ‘unacceptable’.

Commenting, Lord Krebs, Chairman of the House of Lords Science and Technology Committee, said:

“RCUK did not consult or communicate effectively with key stakeholders in the publishing and academic communities when implementing its open access policy. While we are delighted that our inquiry has shown that RCUK are proposing to phase in their open access policy during the initial five-year implementation phase, this should have been made clear much earlier. That is why we call upon the Department for Business, Innovation and Skills to review how RCUK communicated this important change.

“There are still many unknowns concerning the impact of the open access policy, which is why RCUK must commit to a wide rangeing review of its policy in 2014, 2016 and before it expects full compliance in 2018. We heard significant concern about the policy’s ‘one-size-fits-all’ approach, and are pleased that RCUK are both aware of these concerns and prepared to act on them.

“Open access is an inexorable trend. The Government must ensure that in further developing our capabilities to share research they do not inadvertently damage the ‘complex ecosystem’ of research communication in the UK.”

The Committee's report is available here

RCUK has welcomed the report.

"We acknowledge that communication and engagement around the policy, including its development, has been challenging despite trying to respond to the need to engage early on with a broad range of stakeholders. Lessons have been learned and we will continue to actively engage with the academic and publishing sectors as well as learned societies and other international stakeholders throughout the implementation period and beyond. This will allow us to address any immediate concerns as well as to keep a watching brief on the implementation process.

"We are keen to gather evidence of both intended and unintended consequences of the policy that can be reflected on as part of the 2014 review. We will consider the committee's recommendations for monitoring closely as we devise the terms of that review to ensure that it is both focussed and far-reaching. It is our intention to continue to monitor the implementation and consequences of the policy beyond the initial review period in 2014. Open Access is not limited to the RCUK policy so we will also continue to work with partners within the UK and internationally to share best practice and monitor developments in this fast-paced agenda"

RCUK's response is here.

*** FURTHER UPDATE 22nd February  ***

The White House has today announced its official policy on public access to federally funded research:

"The Obama Administration is committed to the proposition that citizens deserve easy access to the results of scientific research their tax dollars have paid for. That’s why, in a policy memorandum released today, OSTP Director John Holdren has directed Federal agencies with more than $100M in R&D expenditures to develop plans to make the published results of federally funded research freely available to the public within one year of publication and requiring researchers to better account for and manage the digital data resulting from federally funded scientific research. OSTP has been looking into this issue for some time, soliciting broad public input on multiple occasions and convening an interagency working group to develop a policy. The final policy reflects substantial inputs from scientists and scientific organizations, publishers, members of Congress, and other members of the public—over 65 thousand of whom recently signed a We the People petition asking for expanded public access to the results of taxpayer-funded research."

Details here

For an explanation of the key points of similarity and difference between the White House directive and FASTR see Peter Suber's post here.

As Suber puts, it "The White House directive and FASTR pull in the same general direction, but they are not identical." Rather, he says, the two approaches "complement one another.

*** FURTHER UPDATE 25th February  ***

Today the Higher Education Funding Council for England (HEFCE) published a "Call for advice on Open Access."

Of note in the call is the following statement:

"As the transition to full open access will occur over a period of time, we propose to accept material published via either gold or green routes as eligible, recognising that it is not appropriate to express any preference in the context of research assessment."

*** FURTHER UPDATE 8th March  ***

On 6th March RCUK published a revised guidance document for its OA policy. This has not been greeted enthusiastically. See here, for instance, and here.

One interesting sentence from the new guidelines:

Where an author’s preference is ‘pay-to-publish’ and their first choice of journal offers this option, but there are insufficient funds to pay for the APC, in order to meet the spirit of the RCUK policy, the Councils prefer the author to seek an alternative journal with an affordable ‘pay-to-publish’ option or with an option with embargo periods of six or twelve months.

And see here for a response to the new policy from an astrophysicist.

*** FURTHER UPDATE 22nd March  ***

On 21st March the House of Lords Science and Technology Committee recommended further changes to RCUK's revised open access policy and guidance.

Commenting Lord Krebs, Chairman of the House of Lords Science and Technology Committee, said: 

“We are pleased that RCUK has listened to the majority of concerns we raised in our recent report on open access and has taken steps to improve its policy and guidance. However, RCUK still has some way to go. It must simply state that longer embargo periods are acceptable where funding for gold open access is not available - with no ifs, buts or caveats.

“One of the key safeguards to the policy is a commitment to review the impact of open access in 2014. We welcome that commitment but call on RCUK to confirm that the impact of open access on peer review and collaboration between academics will form a part of that review. We also want RCUK to give some indication of what action or strategy for action will result from the review.”


Stevan Harnad said...


Another excellent, timely, comprehensive overview of OA goings-on by Richard Poynder.

Three comments:

1. RP: "Some would argue that the US has long been the natural leader of the OA movement, a leadership role it could be said to have acquired in 2005 [with] the US National Institutes of Health (NIH) [Green OA mandate]"

I for one would not say the the US has been the leader of the worldwide OA movement (though it is certainly naturally placed to do so): The historic leader to date has been the UK. The world's first Green OA repository software was created in the UK (2000); the world's first Green OA mandate was adopted in the UK (2003); the UK parliamentary Select Committee was the first in the world to recommend that all institutions and funders mandate Green OA (2004); all of the UK's research funding councils (RCUK) have mandated OA (2006-2011) and the UK today has more funder (16) and institutional (25) Green OA mandates than any other country in the world.

(The US is second with 4 funder mandates and 19 institutional mandates. Little Finland leads in institutional mandates with 28; it has no funder mandates, but with all Finnish universities mandated, it hardly needs them!)

It is only now, with its flawed BIS/Finch/RCUK Gold-Preferential policy that the UK has lost its worldwide lead: In fact, as shown by the SPARC Europe Table, all other countries are now following the path that the UK pioneered in 2003-2004: the only country not following the UK's historic lead now is the UK itself!

But the good news is that the UK's lead can easily be regained, if the UK simply drops its gratuitous preference for Gold and throws its full weight behind implementing an effectively verified Green OA mandate, leaving the option of publishing and paying for Gold as purely a matter of author choice.

2. RP: "Green does usually mean a delay before OA is provided… usually... an embargo period of anything between 6 months and three or more years — a delay intended to allow the publisher to recoup the costs it incurred in publishing the paper."

This too is one of the unanticipated negative consequences of the new RCUK OA policy. It is not true that Green OA means delayed/embargoed OA. At the moment, over 60% of subscription journals, including almost all the top journals in most fields, endorse immediate, un-embargoed Green OA self-archiving by their authors. (See the SHERPA/Romeo registry.)

Fewer than 40% of journals try to impose a Green OA embargo, and even for those, there is a compromise solution that is "Almost-OA":

All papers (100%) need to be deposited in the author's institutional repository immediately upon acceptance for publication, but access to the deposit can be set as "Closed Access" instead of Open Access during the embargo period. During the embargo, the repositories have an email-eprint-request Button that allows individual users to request, and authors to provide, with one click each, a single eprint for research use.

This means that an effective Green OA immediate-deposit mandate can immediately provide at least 60% immediate-OA plus 40% Almost-OA.

But RCUK's flawed policy, by providing an irresistible incentive for subscription publishers to offer UK authors hybrid Gold OA for an extra fee encourages publishers, by the same stroke, to adopt and to lengthen Green OA embargoes beyond RCUK's allowable limit in order to make sure that UK authors must pick the paid Gold option (the UK's "preferred" one) rather than the cost-free Green one.

This too is easily fixed if the UK simply drops its gratuitous preference for Gold and throws its full weight behind implementing an effectively verified Green OA mandate, leaving the option of publishing and paying for Gold as purely a matter of author choice.


Stevan Harnad said...


3. RP: "the SPARC table could be taken to imply that the Wellcome Trust only supports Green OA"

The Wellcome policy allows either Green or Gold,

But, without announcing it explicitly, and without placing any pressure on authors, Wellcome too prefers Gold (and most of the OA that is generated by its policy is Gold OA). This is no coincidence, for the new UK policy was strongly influenced by, and to a great extent modelled upon, the Wellcome policy.

Wellcome gets the historic credit for having been the first funder in the world to mandate OA. (They did it before NIH.) But the Wellcome policy is deeply flawed and was for several years ineffective because compliance was in no way monitored and there were no consequences for noncompliance.

Now, both NIH and Wellcome monitor compliance: funding may not be provided or renewed if fundees fail to comply. But NIH still only mandates Green, whereas Wellcome, a private charity, has adopted the (simplistic) maxim that "Publication costs are part of research costs (1.5%) and a research funder should be prepared to pay them."

That is Wellcome's rationale for (implicitly) preferring Gold: "We fund the research: we're ready to pay its publication costs too."

The trouble is that most research publication is still subscription based. And institutions still have to pay those subscription costs, so their users can access the research. Wellcome is not offering to pay for that: just for the Gold OA costs of publishing the research Wellcome funds. Subscription journals are happy to take the extra Wellcome money, and duly offer a hybrid Gold choice for any author who wants to pay for it -- but they also continue to collect subscriptions, and institutions continue to have to pay for them. So Wellcome is merely subsidizing a 1.5% double-payment to publishers in exchange for Gold OA.

This absurd subsidy to publishers is fine when offered by a private funder that has nothing to spend its money on other than research (98.5%) and its publication (1.5%).

But this simplistic formula doesn't work for the UK (or any) government, or any public research funder. For unlike private charities, governments are using tax-payer money not only to pay for research (100%), but also to pay for journal subscriptions (100%). Hence if they foolishly elect to pay publishers even more -- 100% for subscriptions plus 1.5% more for Gold OA -- they are throwing taxpayer money away to double-pay publishing costs that they are already paying via subscriptions.

Hence, paradoxically, the very first funder to mandate OA, the Wellcome Trust, is definitely not the model to follow. Yet the UK has now done just that, adding to the Wellcome Trust's generosity to publishers an explicit preferential pressure on UK authors, with perverse consequences for the UK as well as the rest of the world.

(For a clear grasp of the contingencies, complementarity, and time-course of Green and Gold OA, the reader could do no better than to consult Houghton & Swan's "Planting the Green Seeds for a Golden Harvest": )

Steve Hitchcock said...

One simple way around the issue of the encouragement by policy towards expensive hybrid OA journals and extended green embargoes is for policy makers preferring gold and prepared to pay hybrid OA fees to require those hybrid journals are also green within the terms of the policy, i.e. within the policy embargo.

It is argued that green OA has not impacted on subscription fees because you can't know in advance which papers within journals will become OA, so you have to continue to subscribe. The same is true for hybrid OA journals, which being based on existing value-adding infrastructure makes them not only the most expensive route to OA but also makes them ripe for 'double-dipping' of OA fees and subscriptions. Requiring that hybrid journals are green would place a constraint on the possibility of exploitation of high OA fees and double dipping.

Hybrid OA tends to be associated with gold, because it provides open access to selected papers based on fees to provide OA. But these journals are not gold, and it is even potentially misleading to call them hybrid-gold. Instead, better for OA, and clearer terminology, would be to require them to be hybrid-green.

Stevan Harnad said...


@Steve Hitchcock

RCUK allowing hybrid Gold payment only if the publisher allows the Green option within the RCUK 6-12-24+ embargo limits is no solution for the perverse effects of the new RCUK policy.

The only solution is for RCUK to allow hybrid Gold payment only if the publisher allows an immediate un-embargoed Green option -- and RCUK must leave the choice between Green or Gold options completely up to the author (no "preference," no "decision tree").

A subscription publisher that pits paid hybrid Gold against embargoed Green is practicing extortion, with or without the help of RCUK's perverse policy.

Embargoes are a complicated story that will soon have to be told forthrightly.

Publishers embargo green under the pretext that it's the only way to protect themselves from sure ruin.

That is utter nonsense, of course.

What embargoes really do is to delay -- i.e. embargo -- the natural, inevitable evolution from subscription publishing to Fair-Gold OA publishing, at a fair, affordable, sustainable price by "protecting" double-payment at today's grotesquely inflated Fool's-Gold price.

Embargoes embargo both OA and Fair Gold, in order to lock in current subscription revenues and Fool's Gold.

Think about it….

But the compromise of an immediate-deposit/optional-access (ID/OA) mandate (in which deposit must be immediate but access to the deposit may be embargoed) not only provides a basic one-size-fits-all mandate for universal adoption, but it will also ensure that publishers will be unable to embargo the optimal and inevitable outcome for researchers much longer.

Whatever else it does, RCUK should immediately and unambiguously adopt (and ensure compliance with) an ID/OA mandate.