Thursday, October 31, 2019

The OA Interviews: K. VijayRaghavan, Principal Scientific Adviser, Government of India

India has announced that it will not, after all, be joining cOAlition S. Instead it will focus on developing an open-access solution better suited to its needs. What has changed?

Last year a group of European research funders calling themselves cOAlition S launched a radical new open-access initiative dubbed Plan S. The aim: to ensure that all research papers resulting from work they fund are made immediately freely available on the internet.

Plan S has been controversial, not least because its principles outlaw** the use of hybrid OA (although a transition period is permitted if time-limited “transformative arrangements” are put in place). By banning** hybrid OA, some researchers have complained, cOAlition S is effectively telling them where they can and cannot publish. This they believe to be an infringement of their academic freedom.

For their part, publishers worry that the plan is being introduced too quickly and has not been thought through properly. As a result, some say, it could have “serious unintended consequences for the integrity of the scientific literature”. And while the initial start date for Plan S has been extended from 2020 to 2021, there continue to be concerns that the timescale is too aggressive.

More importantly, while the Plan S implementation guidelines permit open access to be provided by publishing in open access journals, on open access platforms, or by depositing papers in a repository without embargo, most believe that the practical outcome of the initiative will be a near-universal pay-to-publish environment in which the main beneficiaries will be the publishing oligopoly, not the research community. Essentially, say critics, courtesy of new-style Big Deals (aka “transformative agreements”) commercial publishers will be able to continue profiting excessively by migrating their current exorbitant profits to the new OA environment.

How successful has Plan S been? On launch cOAlition S consisted of 11 funders. Today that number is still just 23. Significantly, the coalition has failed to persuade public funders in the US or China to join (these two countries are the largest publishers of research papers in the world). At the same time, it has lost members: the Swedish funder Riksbankens Jubileumsfond pulled out in June, and the Italian funder Compagnia di San Paolo left cOAlition S in August. On the other hand, some believe that Plan S has so alarmed publishers that the coalition has already achieved its objective (depending on what one believes the objective to be).  

It is, however, clearly problematic that cOAlition S has remained an essentially European initiative. For this reason when, in February, the Indian Government’s Principal Scientific Adviser, Professor VijayRaghavan posted a series of tweets saying that India was joining cOAlition S the news was greeted with great excitement by cOAlition S membersas well as by Plan S supporters like the European Commissioner for Research, Science and Innovation Carlos Moedas.

The news was greeted with less enthusiasm back home in India, with concerns raised about the cost implications, the likely impact on small journals and publishers, and the way in which it would allow commercial publishers to continue to profit excessively from the research community – see, for instance, here, here and here.

Following Prof. VijayRaghavan’s tweets, however, radio silence set in, with no confirmation that India had formally joined, or any updates on the status of its plans. For this reason many ears pricked up last Friday when, during a lecture he gave at IISc Bangalore to mark Open Access Week, Prof. VijayRaghavan commented, “We are not committed to whatever Plan S does or does not do.” This sufficiently piqued the interest of Vasudevan Mukunth that he sought out Prof. VijayRaghavan and asked for clarification, which led to an interview in The Wire where it was confirmed that India no longer plans to join cOAlition S.

As I had been trying to interview Prof. VijayRaghavan for some months, I too was piqued by his comments and so took to Twitter to again invite him to answer the questions I had sent him in June. He agreed and below are his answers to an updated list of questions I emailed over to him.

The Indian context

So why did India change its mind? It turns out that while little was being said in public there has been a lot of internal discussion taking place in India about open access and Plan S. This has led to a consensus that India needs to find an OA solution more suited to its specific needs. As Prof. VijayRaghavan commented to The Wire, Plan S “is by definition a Eurocentric approach … We will work with them to learn about what they’re doing ... [but] ... What we do will be what we think is best for our context.”

What this will mean in detail is not entirely clear at the moment. However, current thinking is reflected in a document published recently by three Indian national academies called Suggestions for a National Framework for Publication of and Access to Literature in Science and Technology in India.

This document (as one might expect) argues that all scientific literature arising from public-funded research should be available in the public domain. It also recommends that there should be an explicit provision in research grants to pay publication charges (APCs) so that papers can be made open access. In addition, it says, preprints should be encouraged, and India’s various national repositories should be “synergised, strengthened, and harmonised into a single national institutional system.”

Importantly, the document also says that researchers should have “the choice of final destination for the publication of their work”. This assertion alone would seem to put India’s thinking at variance with Plan S, which by banning** hybrid OA is expected to significantly reduce the number of journals that European researchers will be able to publish in.

Finally, the framework document proposes as an optimum choice what it calls a “One Nation-One Subscription” model. This would seem to imply that the government will negotiate the licensing of paywalled content directly with publishers – which would be a departure from current practice where individual Indian research organisations negotiate their own licensing contracts, the details of which are kept secret.

In short, it seems that India is still moving towards a national open access policy, but wants to develop its own approach rather than join cOAlition S.

It has taken India a while to get to this point. OA advocates have been calling for a national OA policy for some years, and in 2017 they even drafted a proposed policy and invited organisations like the University Grants Commission (UGC), the Department of Biotechnology (DBT) and the Department of Science and Technology (DST) to sign up to it.

While nothing came of this effort, a number of institutional policies have been in place in India for some years, including the policy introduced in 2014 by the DBT and the DST (here). These policies are essentially green open access policies in which researchers are asked to post all their final accepted manuscripts (post-prints) either in a relevant institutional repository or in one of the Ministry’s central repositories (here and here). These policies, however, are recommendations rather than requirements, and most researchers appear to have ignored them.

Judging by the framework document, it would seem that all the necessary elements for a national open access policy are in place and have been considered. The secret will lie in putting these elements together in a way that works, and which does not produce unintended consequences. Creating a national open access policy is not unlike playing a difficult game of chess.

It is worth noting here that the open access movement has always had two primary objectives. First to solve the accessibility problem (i.e. provide open access). Second, to solve the affordability problem (i.e. reduce the costs of scholarly communication). It has been learned over the years, however, that getting researchers to engage with open access is very difficult and OA policies often have the effect of increasing rather than reducing costs – not least because the research community needs to continue paying licensing fees to get access to paywalled content plus pay publishing fees to enable researchers to make their papers OA in the journals they want to be seen in.

It is for this reason that Plan S has banned** hybrid open access (which leads to double dipping) and it is why transformative agreements are becoming so popular. The latter combine licensing fees with publishing fees in a single agreement that is (theoretically) cost-neutral. It does not, however, solve the affordability problem.

India seems to be taking a different approach, insisting that publishers agree to “capped subscriptions” and “capped APCs”, with the aim both of making its research open access while also lowering costs. How this will work in practice, and whether it is possible to persuade publishers to buy into the idea, has yet to be established. Discussions are apparently only just beginning.

What does not help is that researchers are addicted to the Impact Factor and other proxy measures of quality. It is this that has allowed commercial publishers to acquire the power over the research community that they have, and which allows them to effectively name their own price for the services they provide. As Prof. VijayRaghavan puts it below, researchers have been chasing the “derivatives of science” rather than science itself that. “The publishing industry feeds this monster of narcissism and the monster is addicted to the food.”

Sridhar Gutam, Founder & Convenor of Open Access India agrees with Prof. VijayRaghavan. “Authors in India are obsessed with commercial publishers and impact factors, and they are willing to pool money in order to pay the APCs needed to get published in OA journals like Nature Reports, Frontiers, and PLOS,” he says.

And as noted, simply exchanging subscription fees for APCs is unlikely to reduce the costs of scholarly communication (unless India can persuade publishers otherwise). “What India should do therefore is introduce a national mandatory green open access policy requiring that when researchers publish in subscription journals they self-archive copies in online repositories. This provides open access without the need to pay an APC, says Gutam.

In addition, he adds, Indian scholarly societies should be publishing their own journals using open source software like Open Journal Systems (OJS). “Peer reviewing is voluntary, publication software like OJS is free of charge and open-source, and AmeliCA XML – which allows journals to convert Word documents to the various online formats – is now also free and open source. Why should we be incurring huge costs in APCs?”

Of course, India would still need to pay licensing fees if it wants continuing access to the huge backlog of paywalled content publishers have.

Interestingly, the approach being adopted by Plan S is theoretically compatible with what Gutam calls for. The problem is that the aggressive timescale set by cOAlition S does not allow sufficient time for the development of much in the way of alternative platforms and open-source journals, and the guidelines it has set for repository-based open access are too onerous and too costly. Meanwhile, the fad for transformative deals is being driven by the narcissism that Prof. VijayRaghavan complains of, plus the convenience they offer for librarians.

Whether India can find a better model than Plan S remains to be seen. The truth is that there are no easy options here. That is why 18 years after the Budapest Open Access Initiative people are still scrabbling around for solutions.

The interview begins …

Q: In February you posted a series of tweets saying that India was planning to join cOAlition S. I understand that India has changed its mind? Can you say why? What about Plan S no longer seems appropriate for India?

A: We have not had formal interactions with cOAlition S since February. Internally (within India), we have better clarity on the ways we would like to proceed. This is what we will focus on and should there be substantive overlap with what cOAlition S finalises, that will be good, and also likely is my guess. Fundamentally, the nature and complexities in India are substantially different from Europe – our focus is on what will be meaningful here.

Q: What is it about Plan S specifically that no longer seems appropriate for India?

A: That’s perhaps is inappropriately posed. We are focusing on what is appropriate for India, and I’ve outlined that. As you can see there will be aspects that overlap with cOAlition S. As we have delved deeper, we see complexities and requirements here that we need to address. We will hear from Plan S and others on their progress and tell them about ours.

Q: Three Indian national academies recently published a set of recommendations for a national policy on open access. I assume this was at the request of the Indian government. I believe the optimum choice that the academies have proposed is what it calls a “One Nation-One Subscription” model. That seems to imply licensing content from publishers. An article in The Wire indicated that India will likely be introducing a green OA model. How do you envisage these two approaches working together in practice?

A: We welcome the academies’ views. We will also continue to have wide engagements with researchers and students. We then hope to have formal discussions with major publishers. It’s too early to commit to a route. 

Q: What is a “One Nation-One Subscription” model?

A: That’s the academies’ phrase, is it not? So, best to ask them. From the government’s perspective, publishers now have portfolios that are akin to making cable TV channel choices with different rates for each portfolio. Some normalisation is needed. 

Q: The academies’ recommendations also say that government research grants in India should include an explicit provision for paying APCs. Might this not mean that costs will go up for India as a result?

A: Not if there is an APC cap and that is paid independent of your grant. The aim is to capture the current costs accurately. Separately calculate what we assess as reasonable costs for services such as quality reviewing etc. And see how we can go from the first, down to the second. 

Q: In order to keep costs down Plan S seems more focused on “transformative” deals in which agreements are signed with publishers where the publisher provides both access to its paywalled content plus publishing rights to allow researchers to publish their papers open access. What are your views on this approach and why would it not work for India?

A: In our view, all content should be accessible and not behind a paywall, including archival content. All publicly funded research should be freely accessible. These are areas where there is much overlap with Plan S.

That value has a price

Q: In the lecture you gave last week, you said India plans to negotiate “capped subscriptions” and “capped APCs” with publishers for the country. How hopeful are you that you will achieve that? I do not believe that India is eligible for initiatives like Research4Life or for reduced or waived APCs. How likely is it that publishers will agree to these capped fees? What leverage can India bring to bear on publishers here?

A: ‘Classical” publishers, learned societies, open-access journals all bring value. That value has a price. On the other side, publicly supported knowledge should not be behind a paywall, and the costs of publishing should not be exorbitant, and should be calibrated for economies such as India, where the rupee-dollar exchange rate means steep differentials in the ability to pay.

There are sensible solutions possible, and we will put these on the table. I am sure that the publishers will participate to deliver a viable, correct and principled solution. The present situation is none of the three. 

Q: In your lecture, you also stressed the need for access to research to be provided not just for the research community, but for every citizen. Uruguay’s Portal Timbo and Egypt’s Knowledge Bank are national portals where content has been licensed from a range of scholarly publishers and aggregated on one site, and anyone in the country can access that content. Is that what you have in mind?

A: The National Digital Library of India is doing this. Not yet through licensing. But, yes, we will explore other models here too and will discuss with their proponents. 

Q: Plan S is not the only open-access game in town today. Under the aegis of AmeliCA, for instance, Latin America is proposing a different approach based on national journals and open access platforms. It has also co-founded (with UNESCO) the Global Alliance of Open Access Scholarly Communication Platforms (GLOALL) in order to propagate this model. What do you think are the pros and cons of this approach as compared with Plan S? Might it turn out to be closer to Indian needs than Plan S? Do you see a possibility that the open access movement could splinter, and the scholarly communication system splinter in the process, with the South taking a different approach to the North?

A: We have not yet explored these models in detail, though I am aware of them. We will discuss with them and with cOAlition S too, of course. There are key differences between North and South, there is a South in the North and a North in the South. So, a binary is simplistic.

We are discussing access for all of India. I think this will actually integrate the scholarly communication system, rather than splinter it. This is about scientists and citizens everywhere. Publishers are the vehicles that carry the message, but the directions, the speed, the access and the cost of travel should not be determined by the marketplace, in the main. 

Nationally capped

Q: There has been some concern about Plan S’s decision to ban** hybrid OA, and to insist that the liberal CC BY licence must be attached to all papers. People have complained that these requirements will infringe researchers’ academic freedom and allow Northern-based publishers to inappropriately monetise research produced in the South. What are your views on these particular issues?

A: All research should be openly available upon publication. The use of ArXiv, bioRxiv etc. will be encouraged. Subscription fees should allow access to all articles in the journal, with the costs of subscription nationally capped – APCs should be borne by funding agencies and there should be negotiated caps on all these. 

Q: Is open access in your view solely about increasing access to research, or should it also be viewed as a way of reducing the costs of scholarly publishing?

A: Both. Access to quality publishing has costs, of course. But context-dependent reasonable costs will be there.

Q: In your lecture you said that India currently pays 1,500 crores ($211 million) a year for subscriptions and 150 crores ($21 million) for APCs. In your interview with The Wire the journalist suggested that the APC figures are likely much higher. (He thought the figure should properly be more in the region of 985 crore). I get the feeling that no country in the world knows how much its researchers are paying for open access and this might be a particular issue in the Global South. For instance, a recent study indicated that 60% of researchers that had paid an APC in the developing world had had to fund it themselves. How problematic is this when trying to develop a national policy on open access?

A: The publishers know. We can get better and better estimates, of course, but it will be simplest for the publishers to put that on the table. 

Q: In your lecture you seemed to imply that there is really no need for intermediaries in scholarly communication anymore. Do you think that publishers are no longer needed? If they are still needed, how does their role need to change?

A: Publishers will not perish, nor should they. But their current forms will change. The way research is assessed and used in academia needs to change too. Publishers need to change. Quality need not be excessively expensive and does not have to be the enemy of access. 

Q: One of the benefits of a national OA policy, perhaps, is that researchers can also be encouraged to post their papers in preprint servers like arXiv prior to the peer review process taking place. This allows research to be shared more quickly. What future role do you see for preprints in scholarly communication?

A: Preprint servers such as arXiv are very important. They are the foundation for scholarly communication and ‘journal publishing’ should be built on this foundation and follow this and not be separate or confrontational. 

The monster of narcissism


Q: In your lecture, you also said words to the effect that any achievement made by a researcher needs to be viewed as an achievement not for the individual but for the whole community. Can you expand on that and its implications for the Indian research community?

A: Globally, not just India. All, with very, very rare exceptions, scientific success is built principally by the efforts of many contemporaries and on the work of many more predecessors. Singling out individuals as icons through awards, fellowships and honours of certain kinds has resulted in many scientists chasing these derivatives of science rather than scientific questions. 

The publishing industry feeds this monster of narcissism and the monster is addicted to the food. When confronted, the successful (by these metrics) point to the many good features of this egregious system, and there is a point here. There is indeed science of different qualities and valuing those and recognising quality is very important. Therefore, those individuals who are so recognised must see themselves as one of the many of high-quality who will never have any recognition. 

They must also strive to enhance the spread of quality-science. Hiring, awards and promotion committees need to step back, think, and not conflate precision (IF etc) with accuracy (the value of the work). 

Q: You say, “All, with very, very rare exceptions, scientific success is built principally by the efforts of many contemporaries and on the work or many more predecessors. Singling out individuals as icons through awards, fellowships and honours of certain kinds has resulted in many scientists chasing these derivatives of science rather than scientific questions”. In your lecture, you also said this but then highlighted a record in The National Digital Library by a famous Indian scientist. Do you not think there is a kind of contradiction there? Is it not inevitable that individuals considered particularly talented will always be singled out?

A: Individuals will get singled out and individuals do matter. Just two points. First, chase the science, not the derivative (journals, awards etc). Second, when recognised, acknowledge the team and bat for the team. You are ‘special’ because of the team, in very large measure.  And, there are many like you, and better, who were not recognised as ‘special’. So, a wee bit of humility may not be such a bad thing.

Q: In connection with this you said that researchers need to get beyond the current obsession of where they publish. One reason why open access has struggled to make headway is that researchers are evaluated, rewarded and promoted based on the Impact Factor and/or prestige of the journals they publish in, and these journals are usually subscription journals. The 2012 San Francisco Declaration on Research Assessment (DORA) was intended to address this problem, with signatories asked to commit to “not use journal-based metrics, such as Journal Impact Factors, as a surrogate measure of the quality of individual research articles, to assess an individual scientist’s contributions, or in hiring, promotion, or funding decisions”. I believe you have personally signed DORA, and one of the directors of the National Center For Biological Sciences is on the DORA board, as previously was the Wellcome Trust DBT project head. But I am not aware that many Indian research institution have signed up to DORA. Do you envisage the Indian government directing them to commit to the principles of DORA?

A: As I said in my interview. There is nothing to stop institutions from changing from within. Else, change may well come from outside. 

Q: As you will know, the global research community has been struggling to cope with the rise of predatory publishing, a phenomenon some claim is a direct consequence of pay-to-publish gold open access. India has been a particular victim of this. It is also the location of a number of predatory publishers, and last year the US Federal Trade Commission obtained a summary judgement against one Indian predatory publisher, which has been told to pay a $50 million fine. What are your views on predatory publishing and its impact on the research community in India? How much of a problem is it and what is the Indian Government doing to address the problem?

A: A menace here and everywhere. The Government is addressing this and is constantly alert to addressing the extraordinarily innovative predatory publisher community as each instance emerges. Such hydras will keep coming. Strong internal standards, messaging and strong action when such publishers are detected is needed. Yes, pay-to-publish gold open access has a (negative) role here, perhaps. 


** Update 1st November 2019:

It has been suggested to me that my use of the words “ban” and “outlaw” in connection with hybrid OA in the above text is not strictly accurate. 

I also cited a Nature article from September 2018 which states “As written, Plan S would bar researchers from publishing in 85% of journals, including influential titles such as Nature and Science.”

Let me try to clarify.

Principle 8 of Plan S reads, “The Funders do not support the ‘hybrid’ model of publishing. However, as a transitional pathway towards full Open Access within a clearly defined timeframe, and only as part of transformative arrangements, Funders may contribute to financially supporting such arrangements”.

Subsequent to the publication of the Nature article, cOAlition S published its “Guidance on the Implementation of Plan S”. Under the funding section of Route Two to compliance in that guidance, it states, “cOAlition S funders will not financially support ‘hybrid’ Open Access publication fees in subscription venues.” Note the addition of the word financially

That section also says, “Authors [can] publish in a subscription journal and make either the final published version (Version of Record (VoR)) or the Author’s Accepted Manuscript (AAM) openly available in a repository.”

It might, therefore, be more accurate to say that cOAlition S moved from a position in which was outlawing hybrid OA to one in which it does not support it.

Technically, therefore, Plan S does not now ban authors from publishing in a hybrid journal, so long as they pay the APCs themselves and immediately post the AAM or VoR online with a CC BY licence attached. 

So, I accept that my use of the words “ban” and “outlaw” might be a little strong, and that the figure of 85% is now out of date (I have removed that from the text). 

However, I do believe that, in practice, Plan S will prevent researchers from publishing in hybrid journals. Here is why I say that:

Since Plan S funders won’t cover fees for hybrid journals researchers will be strongly discouraged from publishing in a hybrid journal. 

It is worth noting that the guidance document also says, “cOAlition S urges individual researchers, research institutions, other funders, and governments not to financially support ‘hybrid’ Open Access publishing when such fees are not part of transformative arrangements.”

Moreover, I really doubt that most researchers will read beyond Principle 8 (which is ambiguous) or, if they do, understand (or want to spend time understanding) the nuances of the guidance document.

Since it seems authors will also need to post copies of hybrid OA articles in a repository in tandem with publication, and with a CC BY licence attached, this will (in my view) be a further strong disincentive to publishing in a hybrid journal.


Victor Venema said...

Also researchers being addicted to the Impact Factor is not strictly accurate. Politics has set up a research environment where publishing in high impact factor journals is a rational choice. When this environment changes, I am sure researchers will change their behaviour and not continue it because they are addicted.

Richard Poynder said...

Ok, how about this: Researchers are addicted to publishing (or trying to publish) in journals perceived to be prestigious and/or important. For them, the impact factor provides a simple (but flawed and gamed) way of representing prestige and importance (or as most prefer to call it, quality). That is why they are addicted to the IF. If the IF is replaced by a different metric they will become addicted to that new metric?