Wednesday, February 14, 2018

The OA Interviews: Ashley Farley of the Gates foundation

The Bill & Melinda Gates foundation (aka the Gates foundation) is a private foundation launched in 2000 by Bill and Melinda Gates. According to Wikipedia, it is the largest private foundation in the US, and currently holds $40.3 billion in assets. 

The primary aims of the foundation are to enhance healthcare and reduce extreme poverty on a global basis and, to expand educational opportunities and access to information technology in America. The foundation is controlled by its three trustees: Bill and Melinda Gates, and Warren Buffett.

What is most noteworthy about the Gates foundation in the context of open access is that in 2014 it announced the most radical OA policy to date. The policy applies both to Gates-funded publications and to associated data, with the stated aim of enabling “the unrestricted access and reuse of all peer-reviewed published research funded by the foundation, including any underlying data sets”.

The policy is striking both in its requirement that all research funded by the Gates foundation must be made freely available immediately on publication, and in its insistence that all funded papers must be published with a CC BY licence attached. The CC BY licence allows unrestricted reuse of a work, including for commercial purposes. At the time, Nature described it as the world’s strongest open access policy.

The policy came into effect in January 2015, although publishers and researchers were given a 2-year grace period during which a delay of 12 months was permitted before papers were made freely available.

In furtherance of its open access ambitions, in February 2017 the Gates foundation signed a year-long deal with the American Association for the Advancement of Science (AAAS), the world’s largest general scientific society, to allow Gates-funded researchers to publish their papers open access in the highly-prestigious journal Science (along with four sister AAAS journals) without having to pay an article-processing charge (APC).

To cover the costs, the foundation awarded AAAS $100,000, which OA advocates estimate represents a price of between $6,667 to $10,000 per article.

A month later (March 2017), the Gates foundation announced that it was partnering with F1000Research to allow funded researchers to publish their papers on its own online publication platform Gates Open Research. The platform includes novel services developed by F1000 to encourage faster and more transparent publication of research outputs. The Wellcome Trust launched a similar platform called Wellcome Open Research in 2016.

While there is no NDA in place, F1000 has requested Gates not to reveal how much it is paying to use the platform.

Below Ashley Farley, Gates foundation Associate Officer of Knowledge & Research Services, answers some questions about the foundation and its embrace of open access.


The Q&A begins …

RP: Can you say something about yourself, your background, and your role and responsibilities as Associate Officer of Knowledge & Research Services at the Gates foundation?

AF: Yes! And thank you for the opportunity to share my work. I have spent my career working in both public and academic libraries. I completed my Masters in Library and Information Sciences, where I first learned about open access through an internship at the Gates foundation. While I had had experience in scholarly communications, open access was never part of the conversation.

In my current role (2.5 years now), open access and open research is my primary focus. A large portion of my work has included the implementation of the open access policy, which encompasses the build and roll out of the Chronos platform and Gates Open Research.

I love working on advocacy of openness, both internally and externally. I assist our grantees in understanding, complying with and benefiting from the policy.

I have become very passionate about open research and spend part of my day staying current on related news and discussions within the community.

Aside from open research I support all foundation staff as a librarian, by ways of finding research, completing literature reviews and recommending best solutions for data curation needs.

RP: How much does the Gates foundation contribute each year towards funding science, and what are the main areas it is focused on?

AF: The foundation has 29 program strategies, that administer over 1,669 grants, totalling $3.9 billion. The foundation focuses on Global Health, Global Development, Global Growth & Opportunity, Global Policy & Advocacy, and U.S. Programs.

For a further break down of the program teams and their strategies please visit the “What We Do” section of the foundation’s website. Overall, the strategies can be quite varied, but all are important in ensuring that “all lives have equal value”, which is the message driving our work.

RP: The figure of $3.9 billion you cite: is that the sum of money granted each year?

AF: Yes, in 2015 it was $4.2 billion. The total value of the grants awarded since inception is $41.3 billion (through Q4 2016).

Open access policy

RP: Can you talk me briefly through the journey the Gates foundation took to arrive at its open access policy, and what the thinking behind it was.

AF: The foundation has had a Global Access Policy in all grant agreements since 2003. The spirit of this policy is to provide information generated by foundation funding to the individuals we’re trying to help. The open access policy is a natural extension of this.

The foundation was also closely watching the successes of other institutions driving open access through their work. During this time, journals were also discussing how to make publications and data available during a public health crisis – why stop there?

Th policy grew out of an 11-person working group, across the foundation, and took nine months to develop. An in-depth landscape analysis was critical to the work. The policy recommendation was introduced to the Executive Leadership Team and was approved with a two-year transition period. This allowed for many important discussions to occur both internally and externally.

We worked with publishers that were not compliant and built Chronos to aid in making the policy come to life. More details on the policy’s journey can be found on SPARC’s site.

We are now part of the Open Research Funders Group, which aims to bring funders together to discuss open sharing of research outputs. Funder’s aligning in open policy will greatly benefit grantees and the overall research community.

RP: Ok, so the open access policy, which was introduced in 2015, grew out of the Global Access Policy and requires all Gates-funded research to be made open access immediately upon publication, and with a CC BY licence attached.

AF: Yes, the Global Access policy (adopted in 2003) is not an open access policy. The spirit of the Global Access policy is similar to the OA policy, but does not strictly mention publications or no-cost access. Thus, the OA policy was created to focus on access to peer-reviewed research that results from grant funding.

The OA policy also means access to underlying data. Not all grants result in publications. A feature of the Global Access policy is if a vaccine is developed, or something else created, it would be required to be sold to poorer countries at a discounted price. As the website puts it:

Global Access is a creative concept we came up with in 2003 that requires our grantees and partners to commit to making the products and information generated by foundation funding widely available at an affordable price, in sufficient volume, at a level of quality, and in a time frame that benefits the people we're trying to help. What role does Intellectual Property play in the foundation's approach to furthering Global Access? Intellectual Property provides a great opportunity to think creatively and strategically about how we can reach our ultimate beneficiaries. The careful and deliberate management of IP (patents, copyrights, trademarks, trade secrets, and rights in data) and the associated rights created or accessed through foundation-funded projects, is a critical component to achieving Global Access. Global Access commitments also apply to collaborations with for-profit entities. Whether it is a groundbreaking diagnostic tool or a new toilet that does not require a sewer connection or electricity, they are allowed to sell what they develop with foundation funding at a profit in the developed world, as long as the products are made available to the people who need them most.

So yes, the open access policy grew out of the underlying principles of the Global Access policy. And the OA policy has always included a CC BY licence. 

RP: How many research papers a year are produced as a result of funding by the Gates foundation? Do you have any stats on where these papers are published? Who are the main publishers that publish Gates research for instance?

AF: One of the best outcomes of the open access policy has been better tracking of the foundation’s research outputs. As the policy also applies to any sub-grantee (the policy covers any research funded in-part or in-whole from the foundation) it’s historically been difficult to track publications.

We built Chronos for several reasons (see below) and one of them is real-time publication tracking data. Now the data is much easier to track and more reliable, as we offer templates for authors to include their grant numbers in the acknowledgments.

So, since Chronos was launched in August 2016 there have been 1,311articles for which the foundation has covered the article processing charges. The top 5 journals are PLOS ONE, Journal of Infectious Diseases, Vaccine, Gates Open Research, and Clinical Infectious Diseases. The top 5 publishers are Elsevier, Springer Nature, Oxford University Press, Public Library of Science (PLOS), and Wiley.

I’m sure there are publications that we have missed but moving forward our data continues to get better.

RP: I note that list of publishers is top-heavy in legacy publishers. I assume some Gates-funded research has been published in subscription journals but not on an OA basis? If so, can you put some numbers on that? Likewise, with CC BY: how many Gates-funded articles do you estimate there are out there that do not have a CC BY licence attached?

AF: Any research papers arising from grants prior to January 1st, 2015 do not have to comply with the Open Access policy. These articles are not tracked in Chronos but should be archived in PubMed. We haven’t done any sort of analysis on this scale yet.

Green open access

RP: As you said, Gates-funded research publications must now have a CC BY licence attached. They must also be made OA immediately. Does this imply that the Gates foundation sees no role for green OA? If it does see a role for green OA what is that role?

AF: I wouldn’t say that the foundation doesn’t see value or a role for green open access. However, the policy requires immediate access, reuse and copyright arrangements that green open access does not necessarily provide.

I see green open access as a great way to provide open access to articles already published in closed journals. The practice of self-archiving isn’t as prevalent as ensuring the published version is open, with no embargo.

Not all journals agree with immediate green open access and often the policies are difficult to understand. I also think it’s very important for authors to retain their copyright.

RP: As you indicated, the Gates foundation also has an open data policy. Can you say something about this policy and why it is deemed necessary?

AF: Part of the policy requires that the underlying data to the publication be openly available. We do have guidelines for this, but nothing very prescriptive – I wouldn’t call it an open data policy. This is something that we are hoping to work on soon, so that we can provide more support to grantees in sharing their data.

The reason it is necessary is that data is the most important part of a research project or publication and is needed for reuse, reproducibility or reanalysis. Our funding is often for projects that are long term or are continually working on a specific issue, such as eradicating malaria, thus it’s important to have access to data outputs to inform and strengthen future work.

Sharing data outside of the foundation is also very important and increases our impact and the chances of solving some of the world’s toughest issues.

OA partnerships

RP: A year ago, the Gates foundation entered into a deal with the AAAS to allow its researchers to publish open access articles in a number of legacy subscription journals, including Science. I believe that this project has now expired or will do shortly. How successful (or otherwise) has the initiative been? Do you expect it to be renewed? Do you anticipate similar deals being done with other publishers?

AF: I can’t speak to this as we are in current renegotiation. We are working on producing a public report reflecting on the first year to release in the next couple of months.

I do not anticipate similar deals with other publishers.

Our main goal was two-fold: ensuring grantees can publish in journals important to their career and to explore possible different business models. I care most about the quality and impact of the research itself – not the container.

RP: A month after signing the AAAS deal the Gates foundation announced a partnership with F1000 to allow the funder to publish its own research. What is the logic behind Gates Open Research, what are the costs of running the platform, and how does it fit with other initiatives like the deal with AAAS?

AF: I am very excited about Gates Open Research, which is built and supported by F1000’s model and technology. Here are what I think are the benefits: 
  • It’s a dynamic way of publishing, it is more of a living document that is completely transparent. Authors (Gates grantees) are in full control of what research outputs they want to publish. I believe that using the foundation’s branding shows that we are interested in a non-traditional way of publishing. I believe that this will greatly improve the scientific outputs and solve known issues in current publishing. Negative results are encouraged, as well as, pre-registration reports.
  • Research outputs do not get “stuck” in the publishing process. Since the creation of Chronos we can see articles at the submission stage, giving us data on time-to-acceptance. Looking at the submissions over the past 18 months I noticed that almost 200 submissions have been in the “submitted” status for six-months to a year. This is research that is not having any impact in the larger community. There are many reasons for this – lack of available peer-reviewers, editing the manuscript, waiting for the right journal issue, resubmission to other journals, etc. The Gates Open Research model assures that the research is available quickly, followed by open peer-review.
  • We can better serve different demographics. The foundation supports many grantees in the global south and these authors have struggled to be represented in journals. Gates Open Research provides an established, accredited venue to showcase their research. There is also the option to provide editing and language support services to authors whose primary language is not English.

The costs of running Gates Open Research is fairly low compared to other subscriptions and APC costs (which can be as high as $5,200 for some journals). We pay a yearly fee for the platform and a basic publishing fee for each article, similar to F1000’s stand-alone journal.

RP: Can you say how much the yearly fee is, and how much the basic publishing fee is for each article (I assume these are two separate fees)?

AF: The maintenance fees are commercially sensitive information, and I will not be revealing them. I can let you know, however, that the APCs range from $150 to $1000 depending on the article length, which is quite competitive compared to average APCs elsewhere.

RP: When was the first paper published on Gates Open Research and how many has it published to date?

AF: The first paper, a study protocol, was published November 6th, 2017. Currently, there are 22 articles on Gates Open Research, 15 have passed peer-review and the remaining are in different stages of the peer-review process.

Wellcome Open Research published a great report on their experiences of the first year of publications. We hope to have similar results one-year in.

RP: Do you think that publishing their own research papers raises any conflict of interest issues for funders, or that it could have a negative or distorting impact on the scholarly publishing market? Might it rather have a beneficial impact on the market?

AF: I do not believe that funder-driven publishing platforms present conflict of interest issues or that they will have a negative impact on the market. As the platform is fully transparent readers can assess any potential conflicts of interest.

As publication is only available to Gates-funded authors, program staff have already vetted the research team to give them a grant. Gates Open Research lets authors decide what information is important to share – not the journal.

Eliminating the need to reach a novelty standard can help safeguard research from being manipulated to seem more ground-breaking. We have created a stellar advisory board to help inform the work and direction of the platform.


RP: Can you say more about Chronos and why the Gates foundation uses it? And how do you see it developing in the future?

AF: Chronos manages the whole publishing process, connecting grants to publication, and overseeing invoice payment on the foundation’s behalf. The goal was to provide an easy, effective way to govern the policy.

We did not want the policy to be an administrative burden on grantees or program staff. For the first time we have real time data concerning publication outputs of grants and can track their impact via Altmetric.

While we built Chronos and were very involved in its inception, it is a stand-alone company and aims to serve other funders and institutions. I can see it becoming the main hub for researchers to publish under their various grants more easily than they can now.

Publisher submission systems differ widely and understanding policies can be difficult – Chronos solves both of these issues. For the future Chronos is working on automatically depositing articles in different repositories.

For small teams, such as ours (the foundation’s open access team is myself and my amazing manger, Jennifer Hansen), the Chronos Support team has been invaluable in assisting grantees with author forms and invoice payment.

RP: Last year the publisher InTech posted an announcement headed “Bill and Melinda Gates Foundation Joins InTechOpen’s List of Open Access Funders”. This surprised some (including me) as InTech has a checkered history and was on Beall’s list of “Potential, possible, or probable predatory scholarly open-access publishers”. The use of the word “joins” would seem to imply that the Gates foundation actively supported the announcement. Did it? Did InTech consult with the Gates foundation before publishing the announcement? If not, did the Gates foundation welcome InTech’s announcement?

AF: It did not. InTech wrote me to ask if their publications were compliant with our policy and if the foundation could be included on the list of funders who support the APCs of their grantees. They are compliant with our policy and we do have grantees that have published with InTech.

Many publishers include the foundation on a list of funders that support open access publishing and costs. Most publishers have been amazing advocates of our policy, even if they aren’t compliant.

I suspect InTech are using this to add credibility to their publication, but I have not heard any complaints from grantees who have published with them. I cannot say that for other potential predatory publishers.

Predatory publishers

RP: What, if anything, does the Gates foundation do to try and prevent funded authors from publishing their work with publishers that are, or might be perceived to be, predatory? Does it have a list of approved publishers/journals?

AF: The foundation does not want to editorialize where grantees publish outside of assuring compliancy of the open access policy. Chronos contains a database of over 26,000 journals, detailing their compliancy and aid in manuscript submission for users.

However, as built, Chronos does not contain Beall’s list. While we know of the bias of Beall’s list, predatory publishers do exist. I do not think the issue is unmanageable, but we have had grantees publish in suspected journals in the past. This has happened due to lack of information and the ability to publish in more well-known journals, and the need for quick publication for career advancement.

The main solution that I believe should be supported is more education on the topic. I would love to see the journals from the now defunct Beall’s added to Chronos with a layer of warning before users can submit. My thinking is to integrate Think, Check, Submit so that authors can use all of the available information to make an informed decision.

As a librarian I feel that quality control of journals can be a group effort of the community as well as the researcher doing basic data checks (valid phone numbers, addresses, board members, etc.) This is a common discussion point within the community and I’m excited to see what projects emerge to tackle the issue moving forward.

CC BY and APCs

RP: There has been some pushback against the use of CC BY licences for published research, most recently I think with the Declaration of Mexico in Favour of The Latin American Non-Commercial Open Access Ecosystem. What in your view are the pros and cons of making research available with a CC BY licence?

AF: The CC BY licence is a critical component of our policy. I believe it is the best licence to cultivate reuse and innovation, as researchers do not have to worry about infringing upon copyright. The foundation views the commercial space as another opportunity for innovation and discovery.

I don’t think commercial use should be feared. I have seen the examples in the past of commercial entities repackaging research materials and selling them online. These examples are rare and should be dealt with on an individual basis.

I feel that the pros outweigh the cons, by simplifying the process, empowering researchers, and opening up the research for future technologies (such as machine learning, Artificial Intelligence, and translations).

RP: Many people have come to feel that APCs are undesirable, not least because for some it means that paywalls are simply swapped for publication walls (since those researchers who are not fortunate enough to be funded by Gates, and other funders like Wellcome who pay the APCs for authors, may not be able to afford to publish in OA journals). Are APCs here to stay? Or do we need to move beyond them?

AF: I do not think that APCs are the solution and I hope that we move beyond them. APCs are not reflective of the true cost of publishing. We receive email requests to pay these fees for non-grantees daily and so I see the financial burden on authors quite frequently.

I’m not sure what the solution is, but I do believe that there are better models and that the market for APCs could and should change.

There are examples surfacing suggesting how to change this model, such as John Willinsky’s latest pre-print titled “If funders and libraries subscribed to open access: The case of eLife, PLOS, and BioOne”. Note I mention this merely to inform, not to endorse such a model from the foundation’s perspective.

RP: What are the main challenges open access faces today, and how do you think these challenges will/can be overcome?

AF: One of the main challenges that I see holding open access practices back are the incentives surrounding career advancement. There have been many examples of the “publish or perish” culture damaging research practices. It’s up to institutions to change and alleviate this stressor from researchers.

As a funder we can establish the priority in what is looked for in the grant making process to incentivize openness and sharing. I know this will be a major shift in policy, behaviour, and expectations, but I think this work will be quite important.

RP: What role do you see in the future for legacy publishers?

AF: I would love to see legacy publishers really adapt to current technologies and improve their systems. I think there is a large role that they could play in innovating within the open science space. Data sharing and curation is a monumental task that publishers could help support. We need to move beyond posting PDFs of figures and data to creating dynamic data sharing opportunities.

I believe publishers should also democratize more of their data and open up citations (see the Initiative for Open Citations for instance).

Scientific research doesn’t need more of the same, it needs compelling tools reflective of current technologies to more effectively find, share, and build upon discoveries.

Wednesday, January 24, 2018

Q&A with FinELib, the consortium of Finnish Universities, Research Institutes and Public Libraries

On 17th January FinELib, a consortium of Finnish Universities, Research Institutes and Public Libraries, announced that it has signed an agreement with Elsevier to provide access to around 1,850 journals on Elsevier’s ScienceDirect platform.

Valued at 27 M euros, the three-year contract applies to 13 Finnish universities, 11 research institutions and 11 universities of applied sciences.

In addition, to support Finland’s goal of transitioning to open access publishing, Elsevier and FinELib have initiated an open access pilot program intended to encourage Finnish researchers to publish their articles open access.

The open access pilot, which offers researchers a 50% discount if they publish OA in Elsevier journals, will be available for all corresponding authors in organisations that are parties to the agreement and covers 1,500 subscription journals and over 100 fully open access journals.

Observers were quick to point out that the agreement is far from comprehensive. As Ulrich Herb put it on his blog, since Elsevier publishes a great many more journals (2,967) than are included in the deal, “Finnish researchers can neither read all Elsevier journals nor publish open access in all of them at reduced prices.”

Unsurprisingly, therefore, the news was greeted with little enthusiasm by Finnish researchers, who also expressed irritation at the shortage of detail about pricing (despite the fact that no NDA has been signed). 

And they were quick to point out that, instead of facilitating a transition to OA, the deal will likely embed hybrid OA into the Finnish publishing landscape, helping perpetuate rather than replace the subscription system. After all, most of the journals to which the discount can be applied are hybrid (subscription) journals. 

Publishing at no cost?

Concern increased when researchers discovered that the University of Helsinki is already paying 50% of the APC when faculty publish OA in Elsevier journals. Combined with the FinELib deal this would seem to allow researchers at the university to publish in Elsevier’s hybrid OA journals at no cost, giving the legacy publisher a big advantage over pure OA journals. As Leo Lahti tweeted, “On top of 50% Elsevier OA (hybrid & full) discount @helsinkiuni pays remaining 50%: Elsevier is now free (and cheaper than full OA) for a researcher.”

When I emailed the University to check that its researchers would indeed be able to publish in Elsevier hybrid journals at no cost the response I received did not address the question directly but pointed out that researchers themselves rarely pay APCs themselves.

In the meantime, the University had posted a long explanation on its blog that does seem to confirm (amongst other things) that faculty will be able to publish in the designated Elsevier journals at no cost: “Open publishing at the University of Helsinki in the Elsevier publications is free of charge for researchers as the university pays for the second half of the APC.” (Google Translate).

Explaining the rationale for this, the post added that it is “a pragmatic solution aimed at getting centralised information on take-up as a result of the Elsevier agreement and how much it is increasing the University’s costs.”

Strangely, the post goes on to re-affirm the University’s position that hybrid publishing is not recommended.

Disappointment is all the greater in light of the fact that during the negotiations Finish researchers created a #nodealnoreview website where people could commit to boycotting the publisher if it failed to offer a satisfactory deal. More than 2,700 researchers signed up, although a boycott never went ahead (more details here), and it seems unlikely it will go ahead now despite unhappiness over the deal. [Please see the correction to this in the comments below].

Like other national licensing consortia, FinELib has been negotiating with a number of legacy publishers, but it is Elsevier that has proved most obdurate in all cases. This has seen the publisher in a standoff with national consortia on a number of occasions, both over access to ScienceDirect and over agreement on a satisfactory formula for transitioning to OA. 

In the process, the publisher has on occasion cut off access to its journals (or threatened to), and consortia have on occasions threatened not to renew their licensing contract and/or researchers have threatened to boycott the publisher and/or resign from editorial boards – e.g. in Germany, the Netherlands, Taiwan and South Korea.

Invariably, however, these countries have eventually capitulated and signed an agreement with the publisher, always it seems on terms they are unhappy with. As Kim Eun Sung, a librarian at Sogang University in Seoul explained to Science recently, “Even though we are not entirely happy with the proposed rate increase, we must still consider the importance of ScienceDirect journals for our professors’ research”.

Finland is, therefore, just the latest to capitulate, agreeing to a deal that Cambridge mathematician Timothy Gowers has characterised as a disappointing Dutch-style deal. “It looks to me as though it’s probably roughly what they had before but with some discounts on APCs thrown in.”

The one country that continues to hold out is Germany, which turned down an offer from Elsevier last March, and subsequently outfaced the publisher when Elsevier cut off access to its service (access that was later renewed without a deal being agreed). 

And the negotiators of Project DEAL (as it is called) continue to insist that they will not back away from their demands for a nationwide Publish & Read contract with Elsevier, one they expect to include fair pricing, open access for all the Elsevier papers authored by researchers at German institutions (“publish”), and perpetual full-text access to Elsevier’s complete e‑journal portfolio (“read”). 

Real sticking point

Pricing aside, the real sticking point in Europe is the extent and degree of the open access element in the deals being negotiated, and the way in which this will facilitate a transition to OA.

In other parts of the world, the focus appears to be exclusively on the cost of access to ScienceDirect, with less (or no) attention given to open access. One Taiwanese researcher I contacted was intrigued when I told him that in Europe open access is now viewed as integral to Big Deal negotiations with legacy publishers. He added he did not think this was an issue in the Taiwan negotations, with the entire focus on licensing content.

How the German negotiations will end remains unclear, but some researchers in the Netherlands have come to conclude that their national negotiators should have been firmer with Elsevier. It would seem Finnish researchers feel the same way. After all, as FinELib concedes below, the aim of its negotiation with Elsevier was to achieve “100% OA without additional costs”. This has clearly not been achieved

Meanwhile, the UK – which has been paying for gold OA for some five years now – is having serious doubts about its approach. Research Councils UK (RCUK) agreed to fund gold OA (including hybrid OA) for researchers on a temporary basis. For its pains, it has seen APC costs repeatedly increase, little sign of a meaningful transition to OA, and a continuing need to pay ever-rising subscription costs. (See also here).

The problem negotiators (invariably librarians) face is twofold. First, large legacy publishers have acquired so much content and so much power that they are able to call all the shots during the renewal process. Second, most researchers are still not committed to open access, continuing to prefer to publish in traditional subscriptions journals. They also have little exposure to the costs of scholarly publishing.

In short, the research community is conflicted, divided and in a weak position vis-à-vis publishers and open access, whereas publishers are unanimous in their determination to see the money tree that scholarly publishing has become continue to flourish and grow in ways that enrich them.

And while OA advocates consistently berate libraries for not simply cancelling subscriptions and putting the money towards open access, librarians point out that they simply do not have the power to do this – as is well explained in this blog post.

The upshot: the research community continues to dangle helplessly on publishers’ hooks and finds itself having to fork out more and more money each time a contract comes up for renewal. Unless the Germans can demonstrate that it is possible to wriggle off the hook it is hard to see how things will change in the near future.

Doubtless, we will eventually see near universal open access, but as things stand it seems that this will be almost entirely on terms dictated by publishers, and at considerable cost to the research community (and thus to the taxpayer).

For further discussion of the FinELib deal with Elsevier please read the Q&A below. 

The interview begins …

RP: Can you say what FinELib is and who it represents?

FL: FinELib is a consortium of Finnish universities, research institutions and public libraries. Its mission is to secure and improve the availability of electronic resources.

RP: FinELib recently announced that it has signed a 3-year agreement with Elsevier for its Science Direct Freedom Collection. How long did these negotiations take, and in what ways did FinELib have to adjust its aspirations in the process of negotiation (as inevitably happens)? That is, what did it not achieve that it had initially hoped to achieve?

FL: FinELib started negotiations for access to the SD Freedom collection in 2016 with the goal of achieving reasonable pricing and advancing open access. At the end of 2016, an agreement was made for one year to give more time for the negotiations.

This year we reached a deal which includes an open access element (50% discount on APCs). We see this as a positive step. But, evidently, we have yet to reach the ultimate goal, which is 100% OA without additional costs. Ultimate goals cannot always be reached in as short a time as one would want to.

RP: I understand there was no NDA. Why then have so few details been released? Do you plan to release more detail in terms of what exactly has been bought for how much? If not, why not?

FL: We have published the core details of the agreement: The agreement period is 3 years, at a total cost of about 27 million euros. This gives 35 organisations access to the SD Freedom collection, and a 50% discount on APCs in Elsevier-owned hybrid and OA journals. We are looking into if and which details in addition to these can be published.

RP: If there is no NDA why do you need to establish what other details can be published? FinELib is surely free to publish whatever details it wants if there is no NDA?

FL: It is not that simple. The confidentiality of an agreement is not based on whether or not there is an NDA. At least in Finland the principle of loyalty between contracting parties needs also to be taken into account. For that reason, we prefer to make public information that according to the administrative court is public information. Public information is public regardless of an NDA.

RP: You say the cost of the deal is 27 million euros, but it is not clear what exactly this buys and how it relates to previous contracts. We know it covers access to 1,850 Elsevier journals and discounted APCs for 1,600 journals, 1,500 hybrid OA and 100 fully OA. But what researchers are asking is whether this figure represents an increase in the amount of money paid to Elsevier. If it does, how much is the increase compared to past deals, and is that increase the kind of percentage increase that renewal of Big Deals usually incurs, or is there an additional charge for the OA element? If the latter, how much of the extra is accounted for by the OA element? In other words, does the deal mean that Finland has agreed to pay APCs on top of subscriptions, and thus is paying more than for subscription access alone?

FL: With 27 million euros consortium members will have access to 1,850 journals in the SD Freedom collection for three years.

In addition, corresponding authors affiliated to the organisations that are party to this agreement have the possibility of publishing their articles with a 50% discount on the APCs. The discount is available in 1,564 hybrid journals and 104 full OA journals owned by Elsevier (please note that the SD Freedom collection and the latter list of over 1,600 journals are two different things). At the moment the discount is not available for the society owned journals published by Elsevier.

Each corresponding author who wants their article published open access needs to take care over the discounted APC payment. The FinELib consortium does not do this centrally.

RP: Does that open up the possibility that some authors might end up paying Elsevier the full APC because they are not aware of the deal?

FL: No. We have agreed on a process to check and correct any possible cases where an eligible author has not received the discount.

Two different lists

RP: I am not sure what you mean when you say that the Freedom Collection is not the same thing as the list of 1,600 journals for which an APC discount is available. Perhaps you could clarify?

FL: We are talking about two different journal lists: 1) those in the SD Freedom collection that are available for access (1,850 journals) 2) those for which a discounted APC is available (1,670 journals). These two journal lists are not the same. There are links to both lists on the FinELib website.

RP: Ulrich Herb has said the following: “It should also be noted that Elsevier publishes almost twice as many journals, 2,969: Therefore, Finnish researchers can neither read all Elsevier journals nor publish open access in all of them at reduced prices.” Is his statement correct?

FL: It is true that the FinELib consortium agreement does not include access to all Elsevier titles, only to SD Freedom titles. This has been the case in previous agreements. Elsevier is surely able to provide you the exact number of journals it publishes.

The open access discount agreed between FinELib and Elsevier is applicable in journals owned by Elsevier, not those owned by societies. Societies are welcome to join the agreement if they so wish.

RP: I note Elsevier’s web site says that in order to qualify for access to the Freedom Collection a university must first have a Complete agreement. As it says, “The Freedom Collection Journals is available to academic institutions only who have a current ScienceDirect Complete agreement. Qualifying customers will have access to all non-subscribed Elsevier journal content at a significantly reduced rate.”

This would seem to imply that all the universities/institutions in the FinELib agreement must already have a basic licensing contract with Elsevier (which would I think be more expensive than a Freedom contract). As such, perhaps, the 27 million euros cited in the FinELib press release is just a small part of what Finnish universities will be paying to Elsevier for access to its journals? Is that correct?

FL: We are not familiar with the information that you are referring to. FinELib members who are party to the FinELib agreement have no obligation to subscribe to anything else.

RP: It would seem that over 80% of the journals in the agreement are hybrid OA. I am told that many funders, including the Academy of Finland, do not recommend the use of hybrid OA. Is the deal therefore out of sync with the national consensus on OA?

FL: The Academy of Finland accepts for the moment hybrid open access, as long as it is on a temporary basis. I quote “The Academy is keen to emphasise that hybrid open access is only a temporary solution and part of the transition towards full open access publishing.”

We need all kinds of open access. We need as many publications as possible to be open in every kind of way – in green, in hybrid, in gold – to have more articles open to everybody. 

The reason we need to negotiate with publishers is that our researchers still publish in these journals, that hasn’t changed so far. That’s why it’s important we increase the amount of OA articles in hybrid journals and eventually they need to flip to fully open journals. 

Transition or stasis?

RP: Some Finnish researchers have expressed concern about the passage in the press release that says: “To support Finland’s goal of transitioning to open access publishing, Elsevier and FinELib have initiated an Open Access pilot program that stimulates Finnish researchers to publish their articles open access in Elsevier journals.” Given that the overwhelming number of journals in the deal are hybrid they assume this will only further encourage the use of hybrid OA rather than facilitate a transition to OA. Would you agree? What exactly is the transition strategy?

FL: The ultimate goal of the consortium is to have 100% OA without additional costs. This agreement adds to the number of open access articles. When the amount of open access articles is big enough in these journals, the journals need to flip.

The aim of open access agreements is also to improve and evaluate the open access publishing processes. They need to make it as easy as possible for researchers. During the negotiations with publishers, we have learned that this is not the case yet.

RP: How does the agreement conform to the plan outlined in the Finnish Open science and research roadmap 2014–2017?

FL: Open access in FinELib agreements supports many of the open science and research roadmap’s open access goals. Open access articles increase the societal impact of research, consortium deals create best practices for open access processes and improve the clarity of user rights (cc licences).

RP: How does the agreement meet the H2020 goal of 100% OA by 2020? Some Finnish researchers feel that in light of the deal the EU plan may now be unrealistic in the context of Finland? Would you agree?

FL: We fully support the 100% OA goal and want to see it happen as soon as possible. Negotiations with publishers are one way to move towards this goal, step by step.

RP: Researchers organised a collection of signatures to support the negotiations FinELib has been having with publishers, and also threatened to boycott Elsevier if a satisfactory deal was not reached. Did either of these initiatives help the negotiations? Or were they detrimental (or simply irrelevant) to FinELib’s discussions with Elsevier?

FL: It is extremely important that the open access goal is a common one for the whole research community. As negotiators, we attach great value to the support expressed by researchers.

The statements and the boycott threat have shown that open access cannot be kept separate from licensing negotiations. Many publishers are still reluctant to discuss open access without strong demands from different actors in the scientific community.

We do acknowledge that there are researchers who would have liked to see the licensing agreement being replaced by open access alone, and so are now disappointed. On the other hand, concern was expressed about potential loss of access to a large collection of scholarly journals.

Publish & Read?

RP: You said, “When the amount of open access articles is big enough in these journals, the journals need to flip.” How can you be sure that Elsevier will flip them, and at what point should they flip? You also said, “open access cannot be kept separate from licensing negotiations”. This reminds me that even if it flipped all its journals tomorrow Elsevier would continue to have a huge backfile of papers that it would expect to continue to sell subscriptions to. Would it not have made more sense for FinELib to have insisted on the Publish & Read model that DEAL are holding out for?

FL: We cannot make business decisions on behalf of Elsevier or other publishers. What we can do is to make every effort to ensure that the publishing landscape changes so that 100 % OA becomes the standard way of publishing. FinELib and DEAL share the same goals to make all publications open.

RP: For the moment, however, Finland has not achieved the kind of agreement that the DEAL negotiators in Germany are insisting on. What would you say to those who feel FinELib should have been as firm as the German negotiators?

FL: We wish DEAL every success in their open access negotiations. FinELib and DEAL share the same ultimate goals. Open access is a global issue and all efforts to increase it make a difference.

RP: What do you feel has been learned as a result of the negotiation with Elsevier, and what are the lessons for the future?

FL: Open access still needs a lot of work on all levels to make it the standard way of working. We need to improve processes, communication and data gathering.

In the research community we need to continue the discussions about how to move towards 100% open access. We need to be persistent in making open access the default in scholarly publishing.

RP: Thank you for taking the time to answer my questions.