Open access will remain a half-revolution

The world will surely get open access but the research community will probably fail to resolve the affordability problem that led many to join the OA movement in the first place, claims Richard Poynder in conversation with Michał Starczewski. Publishers appear to be the only “stakeholders” who are relatively organised and coherent about open access and it is to them that the paymasters are turning. OA advocates have failed to take responsibility for the movement. Furthermore, there does not appear to be much appetite in the research community for giving up publishing in prestigious journals, and abandoning the notorious Impact Factor. Therefore essentially, the focus needs to shift from, “How can we force researchers to embrace OA?” to, “How can we use the network to create a more effective and efficient scholarly communication system, one fit for the 21st Century?”









Richard Poynder is an independent journalist and blogger specialising in information technology, scholarly communication, professional online database services, open science, e-Science, and intellectual property. Richard takes a particular interest in the Open Access movement, whose development he has been following for more than a decade.


Michał Starczewski: Do you think that openness is already a new standard in the world of scholarly communication, or is it still an ongoing experiment?

Richard Poynder: Well, openness is certainly fast becoming a new standard in scholarly communication. What we don’t yet know, however, is exactly what openness means (or should mean) in this context, and exactly what processes and outputs it should apply to (and to what degree). We also don’t know who should best fund it, provide it, and manage it.

The OA movement is more than 20 years old. What surprised you most during this period?

What has surprised me most is the OA movement’s lack of organisation, or clear strategy on how to make OA a reality. As a consequence, we are now some 15 years away from the Budapest Open Access Initiative (where the term open access was adopted), and much still has to be achieved, not least clarifying the issues listed in my last answer. Apart from anything else, we still have no conclusive definition of open access. Given this, it is no surprise that there is a great deal of confusion about open access.

I think there are two main reasons for the failure of the OA movement to take a more structured approach. First, the research community is not actually very good at organising itself, particularly on a global scale. And it doesn’t help that researchers are increasingly incentivised to compete more than co-operate with one another.

Second, OA advocates have tended to approach open access more as if it were a religion than a pragmatic response to the possibilities the network provides to improve both the research process and scholarly communication (which should surely be the ultimate goal of open access).

These two factors have generated unhealthy schisms and disputes within the movement, with advocates spending too much time arguing over doctrine. We have also seen OA advocates become addicted to cheerleading and the shouting of slogans, which has deflected them from devoting sufficient time to developing practical strategies and tools to achieve open access. The assumption was that all that was required was to “convert” colleagues. When the movement failed to do that it began lobbying funders and institutions demanding that researchers be compelled to embrace OA, essentially they sought to offload the responsibility onto others.

It also has to be said that the strategies proposed and/or supported by OA advocates have often been cockeyed — not least the concept of the article-processing charge (APC). That anyone ever thought pay-to-publish was a sensible way of disseminating research is most odd. Not only is it impractical, but it has played into the hands of profit-hungry legacy publishers, and indeed any fly-by-night cowboy able to create a web site

I have also been surprised at how disconnected OA advocates are from the views of the wider research community — a tendency exacerbated by their habit of gathering together in their echo chamber of choice (conference hall, social media etc.) where their beliefs, prejudices and misconceptions are reinforced rather than subjected to a reality check. 

The recent Berlin 12 meeting suggests that this ghettoisation is increasing. As the meeting was entirely focused on “flipping” subscription journals to OA models it was “by invitation only” and the organisers chose not to invite any prominent green OA advocates, presumably to avoid any dissenting voices questioning the premise of the plan (although we cannot state this as a fact since the delegate list was secret).

All of which is to say that I have been surprised at how open access has been treated as a “cause” rather than a solution. And despite what OA advocates like to claim, the movement is not by nature democratic, but evangelical.

The French Christian philosopher Blaise Pascal said, “The heart has its reasons of which reason knows nothing”. OA advocates have sought to persuade colleagues by appealing to their hearts rather than their reason. While this approach may make sense in the context of deciding whether to believe in God (aka Pascal’s Wager), it is not very helpful when trying to persuade people of the need to change the way that research is disseminated.  

And it is my belief that this approach has not only slowed progress but is allowing legacy publishers to co-opt the movement for their own ends.

What do you think about the current role of publishers? Is there a point where researchers, librarians and publishers can meet? Is it possible to reach a compromise that would satisfy all stakeholders?

What is interesting here is that OA advocates have at least managed to persuade funders and institutions to force their colleagues down the OA road (often for reasons not directly related to open access), so it is the paymasters that are now driving events. Since this offers publishers little choice but to commit to open access, some form of OA looks sure to become the norm.

The problem is that since publishers appear to be the only “stakeholders” who are relatively organised and coherent about open access it is to them that the paymasters are turning. At the same time, publishers have frightened funders into believing that unless OA is implemented in a way that poses no threat to their profits the entire research process could be jeopardised. It is this that is allowing publishers to appropriate open access for their own ends.

Timing and costs aside, this looks set to limit the extent to which the network is leveraged to improve scholarly communication. So I would say that publishers’ current role is probably a negative one.

What makes compromise especially difficult is that OA advocates view the world in a completely binary way. They see a world populated by good guys and bad guys, with no one in between. And they have cast publishers in the role of bad guy. Much of the discussion about OA is therefore focused on demonising publishers, who have (unfairly) become scapegoats (in the literal sense) for all the ills of academia.

It is therefore hard to see any common meeting ground. Rather we are likely to see governments and funders increasingly step in and impose open access requirements on the research community. And these requirements will likely suit publishers more than they will suit the research community, not least because it will see the triumph of gold OA, and the legacy journal will be embedded in the new environment — which is not the best outcome.

Ironically, while librarians have been the most vocal supporters of OA, and the most vociferous critics of publishers, the triumph of gold OA may suit them just fine. After all, with new money being thrown at them to manage and pay for gold OA, and subscription costs set to fall as the world moves to open access, librarians will surely feel that their lives have improved. Moreover, with funders also now providing money to pay for OA, library budgets will no longer have to bear the full burden of scholarly communication.

You have published on your blog many interviews, i.e. the series “The state of open access” ( From your point of view, what are the most important conclusions?

In the hope of not repeating myself here I would say that the main conclusion I have reached is that OA advocates tend to delude themselves about OA — for the reasons stated above but also because they have consistently underestimated the difficulties of turning the great ship of scholarly communication around.

They have also tended to ignore or deny the new challenges and problems that open access introduces. I am talking here not just about the financial costs of trying to impose OA mandates (e.g. monitoring and managing them) but the negative effects that compulsory policies will inevitably have on the morale and job satisfaction of their colleagues. Consider, for instance, that the OA policy that the Higher Education Funding Council for England (HEFCE) is introducing effectively tells researchers that unless they make their research freely available they could lose their jobs. It does not help that it is proving horrendously difficult for researchers and librarians to navigate the plethora of new rules and processes that inevitably accompany these policies.

And of course we have seen the rise of predatory publishing, which OA advocates either deny is a problem, or assert that only the naïve fallen victim (and thus have only themselves to blame).

So OA advocates talk ad nauseam about the (at times questionable) benefits of OA, but say little or nothing about the costs (monetary, managerial etc.) of forcing it on their peers. They have also given little or no thought to the need to create the necessary infrastructure to implement OA policies. What infrastructure has been put in place has generally been inadequate and amateurish, and usually suffers a lingering death when funding runs out.

Indeed, most of the practical aspects of achieving OA have been an afterthought at best, with the underlying assumption that this is something others should worry about and provide. I think this is implicitly acknowledged in the recent Knowledge Exchange report “Putting Down Roots: Securing the Future of Open Access Policies, which notes that the success of open access policies relies on many disparate non-commercial services whose funding can disappear overnight. Moreover, the report adds, these services are now having to compete with profit-rich commercial organisations determined to maintain control of the scholarly communication process.

Even more damning, OA advocates frequently fail to comply with their own standards — not least when creating institutional repositories. As the above report notes, the potential value of what non-commercial tools have been created “is undermined by the limited adoption of the underlying standards and metadata on which they rely, and is also challenged by the presence of established commercial providers which offer more robust, proprietary datasets (e.g. Elsevier’s Scopus database and Thomson Reuters Web of Science).”

A good example of the latter point is the response we have seen to the OSTP Memorandum. In order to ensure that public access to research papers subject to the Memorandum will be mediated by them, and provided on their sites (rather than by repositories outside their control) publishers quickly created CHORUS. Librarians responded with SHARE, which appears to be modest by comparison, will have to rely on publishers’ co-operation, and will doubtless struggle for funding. 

What we see here is part and parcel of what Geoffrey Bilder likes to call “the enclosure of scholarly infrastructure”. This growing enclosure must limit what the OA movement is able to achieve, and by not having anticipated it OA advocates are all the more powerless to prevent it.

Consider also that the poster child of the green OA movement — arXiv — recently reported that it is facing substantial financial pressures. As a result, it said, it will need to “embark on a significant fund raising effort”. Strikingly, it added that it will first have to “create a compelling and coherent vision to be able to persuasively articulate our fund raising goals beyond the current sustainability plan that aims to support the baseline operation.” This comes as the service approaches its 25th Anniversary. Is it not therefore a little late coming?

And when reality intervenes OA advocates are not very good at adapting to circumstances. Rather, they tend to respond by doubling down and repeating the same old mantras, or reinventing history. They have always claimed, for instance, that (unlike the subscription model) gold OA will impose market forces and price discipline on scholarly communication, and so force down costs. We have not seen this happen, and it now seems highly unlikely that we ever will. OA advocates nevertheless continue to insist: “It will happen; it just hasn’t happened yet”. Or they change tack, insisting: “Well, we always said the main goal is access, not affordability”.

It is noteworthy that only recently (14 years after BOAI) OA advocate (and former employee of OA publisher PLOS) Cameron Neylon conceded on Twitter: “I may be completely wrong on APCs. No functional market is emerging and it might be the wrong economic model.”

What is distinctive about Neylon is that he is one of a small group of thoughtful OA advocates, and it is to his credit that he has finally acknowledged the problem. However, even if he managed to persuade his colleagues in the movement that there is a fundamental problem with the primary OA business model, and this encouraged them to take action, it would surely be a case of seeking to shut the stable door after the horse has bolted.

So my main conclusion is that OA advocates have failed to take responsibility for the movement. And as a result, publishers have repeatedly been able to outsmart them.

What differences are there in the attitudes towards OA and its implementation in poorer or richer countries? What impact does wealth have on discussions about open science?

One would certainly assume that wealth would influence attitudes and discussions about open access. One would therefore expect that the Global South would major on green OA and the North on gold OA, since there is more money sloshing around to pay for gold OA in the North, and green OA is (erroneously) viewed as costless. But I wonder if it is as simple as that. 

For instance, there is a greater focus in wealthy North America on green OA, and a greater focus in wealthy Europe on gold OA. Of course, this may be a superficial difference because green mandates can be fulfilled with gold OA. And as funding for gold OA increases we can expect even major green mandates like the NIH Public Access Policy to increasingly be fulfilled by researchers paying for gold OA — if only because it is so much easier to do so.

Of note here, Cambridge University’s Danny Kingsley has recently suggested that the HEFCE policy (widely celebrated as the quintessential green OA mandate) may turn out to be a Trojan Horse for gold OA. Kingsley reports that nearly half of the HEFCE eligible articles submitted in 2015 were published as gold OA. And of these, 74% were hybrid OA — that is, captured by legacy publishers, and at great cost to Cambridge University.

What about the Global South, where researchers don’t generally have access to funds to pay for gold OA? In theory, of course, gold OA is still possible as open access publishers operate various fee waiver schemes. But legacy publishers are less likely to offer waivers, and as Raghavendra Gadagkar explained in a 2008 letter to Nature these schemes are very problematic. So one possibility is that we could see a global divide, with the North embracing gold OA and the South green OA.

For me there is here an interesting puzzle: the OA movement was kicked into life by George Soros’ Open Society Foundations (formerly known as the Open Society Institute), which organised and funded BOAI, and donated $3 million to the cause.

OSI’s interest in open access grew from a concern about the information divide that former Soviet Bloc countries and the Global South faced, and the Budapest meeting came in the wake of the earlier (1999) OSI initiative Electronic Information for Libraries, or EIFL (formerly

EIFL was a recognition of the role that libraries play in the exchange of ideas, knowledge and information, and the development of open societies. To that end, OSI invested in library development and modernisation in the post-socialist countries of Central and Eastern Europe and the former Soviet Union. It also helped them secure subscription access to large portfolios of international journals by means of national Big Deals.

So we might want to wonder how an initiative that started out with concern about restricted information flows in less prosperous countries has given rise to a pay-to-publish regime that now threatens to limit their ability to share their research with the world. The information divide OSI set out to resolve appears to have simply been pulled inside out.

Again, however, the picture is more complex that it first appears, as the developing world is fast falling victim to the developed world’s obsession with publishing in prestigious journals, and embracing the same “publish or perish” culture that benefits bean counters and managers, but degrades the quality of published research. These prestigious journals, of course, invariably belong to international publishers based in the Global North. As a result, home-grown low-cost publishing solutions face a growing existential threat, as the publishing of research papers is effectively outsourced to the so-called publishing oligarchy, a development that also poses a threat to local OA distribution platforms like SciELO and Redalyc.

We saw an example of this in 2014, when the Brazilian funder CAPES announced its intention of “internationalising” 100 Brazilian journals by outsourcing them to a foreign publisher. To that end it organised a meeting to which a bunch of Brazilian editors and five large global for-profit publishers (Elsevier, Emerald, Springer, Taylor & Francis, and Wiley) were invited. Understandably, this sparked some heated local protest, not least from SciELO.

We should also note that while researchers in the Global South can currently make papers they publish in international journals freely available by means of green OA, the publishers of these journals are introducing increasingly restrictive self-archiving embargoes in the hope of emasculating green OA. In any case, as these publishers begin to flip their journals to OA, researchers will increasingly find that if they want to be published in international journals they have little choice but to embrace gold. 

Will this disenfranchise researchers in the Global South? Perhaps not. If you think about it, there are few governments that could not/would not pay legacy publishers their asking price — if they felt it was important (be it for subscription access or publishing fees). It is worth noting that when I spoke to librarians in Poland and Serbia in 2013 I was told that, due to national Big Deals, access to international research (via subscription journals) was not really a problem.

As publishers move to new-style Big Deals that combine subscriptions with APC costs, and pay-to-publish becomes the norm for international journals, it seems reasonable to assume that national governments in the Global South will find the money to pay for these new Big Deals, as they did for subscription access. And they will be able to justify doing so on the grounds that by paying to publish in international journals they can ensure that their research gets international visibility.

So while we can expect to see green OA policies continuing to be introduced in countries like Argentina, Mexico, Peru and India the accompanying repositories may increasingly be filled with papers for which a publication fee has been paid (i.e. gold OA) rather than self-archived papers. And given that many victims of predatory publishing are based in the developing world we should not doubt that at least some researchers there are able/willing to pay to publish, even if they have to meet the costs themselves. So I expect to see these countries gradually converted to national deals that buy publication rights as well as access rights, as we are seeing happening in Europe already.

Can openness improve research quality and how?

Openness certainly should improve quality — if it is done effectively. If for instance, it includes openness of both papers and data, and if open peer review is deployed.

Open data is important if we are to hope to address the reproducibility problem, and we can expect open peer review to improve the quality of papers, since reviewers would surely be more conscientious and thorough if their names were attached to reviews.

And if OA is combined with both open data and open peer review it should make it more difficult for researchers to publish erroneous, shoddy, fraudulent and/or fake research, or indeed to operate fake review scams.

However, open peer review needs to consist of more than just publishing reviewer names alongside papers. Their reports need to be freely available too. We also need to be cautious about calling for openness without considering any potential downsides — a point made recently by Stephan Lewandowsky and Dorothy Bishop.

Do you see libre OA getting more attention of policy-makers nowadays, or do most of them rest with OA gratis instead? What in your opinion will be the prevailing form of OA in the future? Will there be more libre OA, or will there be a consensus that gratis OA is enough?

Libre OA does certainly seem to be getting a lot of traction, and policy-makers do currently seem tempted to try and make it the norm. However, we can also see a lot of pushback from researchers, especially in the humanities, and especially where policy makers try to insist on CC BY. If funders persist in this there is a danger that all but the most dyed-in-the-wool OA advocates will become thoroughly put off OA. This would likely further hold back progress rather than advance it.

Many researchers view demands that they embrace libre OA as an attempt by funders and institutions to appropriate their intellectual property. Of course that is not true, but in the context of the increasing proletarianisation they are experiencing it is unsurprising that researchers should be suspicious. And while using CC BY does not mean giving up one’s IP, it does mean that the world at large can profit from your work. It also means forfeiting earnings that would otherwise be available.

I realise OA advocates insist that researchers only write for impact, not for payment, but many researchers do earn royalties from their work, particularly humanists. I do not know how common this is, but I know of one researcher who has just been offered a $130,000 advance on a book. How would researchers able to command royalties like this feel if they were told that all their future work has to be available under a CC BY licence?

And this is not an issue only for humanists. Scientists tend to express outrage when they discover that OA papers they have authored are appearing in book collections on Amazon, pulled together by some canny entrepreneur with his eye to the main chance, and sold at a high price.

That said, I can see the arguments for libre OA. I realise, for instance, that research published under an all rights reserved licence today will not enter the public domain for many, many years. I also understand that the tidal wave of papers now being produced each year means that few if any researchers can hope to process all the information relevant to their work, so machines will have to do much of the initial sifting (and indeed may eventually begin to make scientific discoveries themselves). So the right to mine research papers is becoming urgent.

But while I understand the “information overload” problem, and while I can see the potential offered by TDM, I feel bound to say that many of the papers published today appear to offer very little value, and so should probably never have been published in the first place. That they are is because most papers are now published to fill CVs, not to advance science. 

Leaving aside the fact that this obsession with filling CVs encourages cheating, fraud, and mediocrity, at some point CVs themselves start to suffer from information overload. Soon when a researcher applies for a job, or funding, it will be necessary to text mine their CV in order to extract the relevant data! What we need are fewer, better papers, not more low quality papers.

Regarding TDM I think a better approach might be for governments to amend copyright laws to ensure that researchers have an automatic right to mine papers to which they have lawful access. As I understand it, this is the approach that the UK government has taken, and it is an approach being proposing more widely (not least in France).

So which form of OA do I expect to become the prevailing form? In the short term I expect to see a mixed picture but as gold OA grows so libre OA will grow, and gratis will decline. However, I suspect we will see much less CC BY than some OA advocates would like.

Do you think that the struggle between the green and gold routes of OA will become stronger and stronger, or will a compromise be found? 

The struggle between gold and green may intensify in the short term, but unless something changes my expectation is that green OA will be increasingly side-lined. Green mandates will doubtless continue to grow, but as I noted they can be complied with by means of gold OA, and so as gold OA funding grows I would expect gold OA to become the default.

This is certainly what we are seeing in Europe right now: Research Councils UK is providing funds for UK institutions to pay for gold OA, universities in The Netherlands are signing new-style Big Deals that combine subscription access and gold OA publishing fees, and Germany’s Max Planck is seeking to engineer a mass “flipping” of journals from subscription to OA models.

Meanwhile publishers are persuading librarians that there is no need to host full-text journal articles in repositories, but simply to link to the OA version on the publisher’s site. This is the purpose of Elsevier’s Institutional Repository Pilot Project (which it expects other publishers to join at some point). In Elsevier’s terms, repositories will be reinvented as “vehicles for discovery” rather than sources of full text.

Repositories will of course also continue to grow and flourish. But while they will host things like theses, working papers, newsletters, conference papers and other grey literature in full-text, their role will increasingly be an archival one and a “shop window” for the institution’s research efforts. Research papers will be hosted on publisher sites and linked to.

Elsewhere, we can expect to see growing pressure on social networking platforms like and ResearchGate to redefine themselves as vehicles for discovery too (encouraged by the use of take-down notices). This appears already to be the situation with Mendeley, which was acquired by Elsevier in 2013. This tweet seems to say it all.

And as also noted, while green OA may appear to be the best choice for the Global South, the CAPES incident demonstrates the extent to which governments and funders in the developing world are likely to feel that in order to present themselves as serious centres of “research excellence” they need to have their scientists publish in international journals, which will increasingly mean paying to publish.

If there is any sort of compromise in sight I think it will be some form of what is variously called diamond or platinum OA, where there are no paywalls and authors pay no publishing charges. However, that still leaves us with the question of who will pay, and how much. Equally importantly, how transparent will the costs be?

And while platinum/diamond OA is currently discussed in terms of the research community’s “reclaiming of ownership of the mission of scholarly communication” and by referencing initiatives like the Episciences Project, we should note that commercial publisher De Gruyter has been offering what it calls a “publisher pays” model for some time. This allows research institutes, societies, universities and other organisations to pay all the costs of publishing, so that the content is made freely available and the author has to pay no publication fee. De Gruyter says this option is proving very popular, and currently it publishes over 500 open access society journals in this way.

For me the interesting question is: at what point does platinum open access become no different to the membership schemes that OA publishers have long operated, or the new-style Big Deals that Dutch universities are currently inking with legacy publishers.

Once again, however, this is a compromise that can be expected to suit publishers more than it suits the research community.

What areas have the most potential of embracing such experimental forms of conducting research as open notebooks or open peer review? Do you see any particular barriers that could prevent their wider adoption? Do you think they will become new standards some day?

I think it is too early to say whether experimental forms of openness like open notebooks and open peer review will become new standards. But I would certainly hope that open peer review becomes the norm.

Concerning open notebook science (a method pioneered by an organic chemist), I would think areas like chemistry and the medical sciences offer the most potential.  Beyond that, I suspect experimental practices and methods will remain niche — a point I think Toma Susi conceded when I spoke to him recently regarding his decision to publish a grant proposal in the new Research Ideas & Outcomes (RIO) Journal.

However, this is a future that RIO certainly anticipates, describing itself as it does as a journal that plans to publish “all outputs of the research cycle, including: project proposals, data, methods, workflows, software, project reports and research articles together on a single collaborative platform, with the most transparent, open and public peer-review process.”

What we have learned is that receptivity to openness tends to be discipline specific. The success of arXiv, for instance, owes a great deal to a pre-internet culture that existed among the physics and mathematics communities in which they routinely shared preprints with one another in paper form.

In terms of barriers, I would say that if legacy publishers do manage to appropriate open access in the way I anticipate then they will act as a drag on innovation and greater openness. They have come to accept OA because at some point they realised it was possible to provide it in a way that allowed them to continue publishing their legacy journals, and in a way that protected their profits. To push openness further would I think threaten those profits, and indeed probably moot the continued existence of the journal.

There are several mature business models for open journals at the moment. However, we are not sure what is the best solution for books. Do you think that the Open Library of Humanities model (a consortium of institutions paying for publishing without APCs) is a good answer for this issue?

Open Library of Humanities publishes journals rather than books I think, but do I feel its model is appropriate for books? Personally, I think it is too early to be definitive about OA books, and that seems to be the conclusion reached by Jisc.

For a start, books are mainly published by humanists, and humanists are most resistant to OA. Moreover, it is far from cheap to publish a book. It is sobering to note, for instance, that legacy publishers charge authors around £10,000 to publish an OA book (i.e. Routledge), or even £11,000 (i.e. Nature).

This latter point would certainly seem to suggest that some kind of consortial model is appropriate. And this seems to be the model envisaged by Lever Press (although it has also talked of using an “unlocking” model similar to that pioneered by Knowledge Unlatched, which could be a little different perhaps).

Once again, however, I would be concerned about the degree of financial transparency provided by a consortial model, particular when dealing with a for-profit publisher. As I noted earlier, OA advocates long argued that having authors pay APCs would introduce market discipline, on the grounds that authors would shop around and make a buying choice based on price. This has not happened, partly because we have seen a move to the bulk buying of publication rights, which strikes me as being somewhat similar to a consortial model. So the question is how would a consortial arrangement provide the price discipline we see in a true market? If nothing else, I suspect this aim will be confounded by what one might call “prestige hunger”.

This came home to me recently when I spoke to a researcher determined to make her new book open access. I pointed her towards Ubiquity Press (which, by the way, still appears to charge £9,340 if you want your book typeset, copyedited and with language checking and book index services provided), and a couple of other new-style OA book publishers (including Punctum Books). Her response: “Well, my co-author is a junior research and he does not have tenure, so he really needs to build up his CV. No one is going to be impressed with a book published by an upstart OA publisher”.

As a result, she is currently planning to go with a legacy publisher that offers an OA pay-to-publish option.

This all suggests to me that there is need to give a lot more thought to how OA books could be published effectively.

All over the world there are many OA strategies and policies, not always coherent. Do you see any effective mechanisms for the future international coordination of OA strategies and policies?

We are seeing some attempts here, most notably perhaps with the PASTEUR4OA project in Europe. However as you indicate, what is needed is a global approach. UNESCO might seem to be an organisation that could take responsibility here. I know it sees open access as something it ought to take a leadership role in, and it has published a “policy guidelines” document. But I fear UNESCO is too bureaucratic, and easily mistakes meetings and reports for practical action. It is also hostage to geopolitical forces that inevitably limit its ability to act decisively.

In the end it is the purse holders who are best placed to take this on. One funder organisation that has taken a global stance on open access is the Global Research Council (GRC). Again, however, if you look at its 2013 “Action Plan towards Open Access to Publications” you will see that its main focus is on persuading publishers to adopt OA. To that end it talked of the need to develop “an integrated funding stream for hybrid open access”. Hybrid OA is now widely viewed as antithetical to an effective transition to open access, not least because it plays into the hands of legacy publishers. GRC’s approach is a further reminder that the disorganised state of the OA movement is encouraging governments and funders to turn to publishers for solutions, and not coming up with the right ones as a result. And as we have seen, publishers are very adept at infiltrating policy committees and working groups — as they did so successfully with the UK Finch Committee).

As far as effective mechanisms for international coordination go, therefore, one might want to be a little sceptical about possible outcomes.

In any case, as the reality of what it will cost to monitor and manage OA policies comes more sharply into focus funders and institutions are likely to conclude that it is much simply to continue outsourcing everything to publishers. Publishers have always argued that there is a danger that open access will lead to a great deal of wasteful duplicated effort. They are right of course, particularly when seeking global solutions. So the choice for funders would seem to be: subcontract everything to publishers, or try to get a disorganised research community to “reclaim ownership” of scholarly communication. Which would you opt for?

The good news is that the world will surely get open access. The bad news is that the research community will probably fail to resolve the affordability problem that led many to join the OA movement in the first place. More worryingly, open access could end up as a half-finished revolution.

As Vitek Tracz has pointed out, scholarly communication will not be fit for purpose in the networked world until the kind of developments he outlined when he spoke to me last year have been implemented, including the abandonment of the traditional journal.

Essentially, the focus needs to shift from, “How can we force researchers to embrace OA?” to, “How can we use the network to create a more effective and efficient scholarly communication system, one fit for the 21st Century?”

And in order to do that it would appear that the research community would have disintermediate legacy publishers. This could be by creating “overlay journals”, or developing a range of other new publishing initiatives in which the whole process is managed and controlled by the research community itself. Examples of the latter include the use of institutional repositories as publishing platforms, and the founding of new OA university presses like Collabra and Lever Press.

Governments could also do more to fund and support low-cost national and regional publishing platforms like SciELO, Redalyc, AJOL and CyberLeninka.

One would also need to see the editors of legacy journals following the example of the editorial board of Lingua, by declaring independence from their publisher and setting up a rival journal, something the board of Cognition is currently also considering. However, as Peter Suber has pointed out, while there is a history of such actions, they are rare. The fact is that while researchers are happy to shout the odds, and sign petitions like the Cost of Knowledge, it is far from clear that many are willing to walk the talk.

In the end, the key question is whether the research community has the commitment, the stamina, the organisational chops and/or the resources to reclaim scholarly communication. While I would love to end on a positive note, I am personally doubtful that it has. The fact is that, OA advocates aside, there does not appear to be much appetite in the research community for giving up publishing in prestigious journals, and abandoning the notorious Impact Factor. More importantly, university managers and funders do not want to see anything that radical occur. We live in an age of bureaucratic scrutiny, and scrutineers crave simple and standard ways of practising their dark arts. That is exactly what the IF and legacy journals provide. If I am right, OA will surely remain a half-revolution, for now at least.

Additional information