R2R is a relatively new meeting which, being in Europe in February, fits in well for conference junkies between APE in Berlin and the group of meetings in March and April (all now cancelled this year). It is now owned by the independent consultant, Mark Carden, who took it over from the Association of Subscription Agents when they were disbanded. Mark continues the policies of the ASA of bringing together the stakeholders in the scholarly communication ecosystem over questions of mutual interest advised by an international advisory board, a little like Charleston and Charleston subsets, (More on the governance Here.)
There were almost 200 delegates, of which a little under half were publishers. Although researchers were thin on the ground, about 20% of the delegates had doctorates. The conference appears to be on the up. From 2019 to 2020, the feedback “strongly agreeing” on relevance grew from 40% to 63% and on value from 32% to 56%. There was an emphasis on what publishers were doing, which will be reflected in this report, but the librarians in the audience made sure that their interests were not forgotten.
The program and slides for presentations are here. These have not been separately referenced, but you can be sure that they are there. A special feature of the conference is that the presentations and panels are interspersed by a debate on journal and article metrics moderated by Rick Anderson of the University of Utah and meetings of five workshops (see later in this report).
Most scholarly communication meetings at present are under the shadow of Plan S and its gradual development and R2R was no exception, but, as readers will see, there were other topics on the agenda. The keynote was on Research Ecosystem Dynamics – publication adaptation, evolution or extinction. The presenter was Dr. Jonathan Adams, now the chief scientist of the Institute of Scientific Research (ISI). His LinkedIn profile bares looking at as his doctorate is in Ecology, Systematics and Population Ecology, and he joined Clarivate (the owner of ISI) after some time at Digital Science (owned by Holtzbrinck) in a similar role. Here are some points made in a fairly dense argument:
1. His thesis was that we as a group of stakeholders are threatened by climate change which is also assaulting the research system. The movement towards Open is disruptive but so is the investment of new players, preeminently China.
2. Globalization (now sometimes a dirty word) is a significant feature of top science. US total output continues to rise but this is driven by increased international collaboration. In Western Europe, about two-thirds of ‘national’ papers have an international co-author. For leading universities, the collaboration share is higher. There are some good slides here.
3. As well as an acceleration of research, there are movements to control costs by enforcing quality measures since 1992 which are now worldwide and which have led to some (unintended?) consequences. In the UK there was a shift in engineering and social sciences towards deploying journal articles as evidence. The UK model of research assessment has spread globally because it was seen as effective.
4.The ecosystem is under threat from invasion by poor quality (and worse)information. Key actors are complicit in this, including institutions and some countries; many more are compromised, including editors and publishers. Many stakeholders are responsible for upholding research integrity. There is no single group that will fix the problem; there are no knowledge police. Without an adequate defense, the system of research publication that has maintained a validated knowledge corpus so effectively over 350 years will disintegrate.
Not surprisingly there were some questions:
-“Is Open Access a good thing?” was a question from an open access publisher. “Yes,” he said, and “But” was the answer, “There are some parasites.”
-“Who are the gatekeepers now?” asked a London medical librarian. “It used to be the role of publishers and their editors but research managers are important now.”
There were two very different presentations on Open Access Models and Impacts. From the Global South, Dr. Solomon Derese, from the Department of Chemistry at the University of Nairobi, Kenya, explained from his own experience the impact of Research4life and Open Access: Impact on access to e-resources in Africa. For those not familiar with this program, the STM Association since 2002 has been a prime mover in the collaboration described as a public-private partnership between WHO, FAO, UNEP, WIPO, ILO, Cornell and Yale Universities, STM and up to 200 individual publisher partners. Its goal is to reduce the scientific knowledge gap for low and middle-income countries by providing researchers in eligible institutions with free or low-cost access to academic and professional peer-reviewed research journals, books and other resources, as well as helping them to develop the skills required to play an equal role in the research communication ecosystem. Dr. Derese was positive.
Past African civilizations have been famous for their libraries with Timbuktu only one example. Researchers need access to world scholarship or they cannot contribute. Africa’s World publication output reached 1% in 1987 but subsequently declined to 0.7% in 1997. A WHO survey undertaken in 2000 showed that 56% of institutions in countries with a gross national income per capita (GNIpc) of under US $1,000 had no current subscription to any international journal. This is where Research4Life changed the situation. He showed an impressive slide demonstrating a sharp increase in research output in R4L Countries since its Introduction in 2002.
It could be argued, but was not on this occasion, that with complete open access no barriers will remain. Derese’s own list of barriers were lack of knowledge of the scheme, the need for more publishers taking part, and lack of access and high cost of access to the Internet. Problems with the cost of APC’s were mentioned in the related workshop, but not here.
Tasha Mellins-Cohen (Director of Publishing at the Microbiology Society, the European equivalent of the ASM and once known as the Society for General Microbiology) spoke under the heading of Open Access Models for Society Publishers: a framework for Institutional Publish & Read Deals. She has been organizing small and medium sized learned society publishers. Mainly in the UK, The Society Publishers’ Coalition (SocPC) is a group of likeminded, not-for-profit learned societies and membership charities who publish as part of their charitable objectives and who reinvest the surplus from their publishing into the disciplinary communities they serve.
This presentation, however, was explaining the problems her organization has with Plan S as an example. Sustainability is important, and it is not just about money. How they flip to open access has to be simple and scalable. The way their discipline relates to others means that SCOAP3, which has enabled subscription models being replaced by open access in the discrete areas of high energy physics and other physics subdisciplines and with great library support, cannot work for them. They have, however, managed some consortia deals which enable OA that is not free but for researchers it is fee free. The link to an article in Insights provided in the presentation does not work (But, read more Here.)
The debate centred on a motion: The venue of its publication tells us nothing useful about the quality of a paper. The rules were those popularized at Charleston conferences in the Hyde Park Debates. There were two polls at the start of the debate and at the end to see how convincingly the case was made. Under these rules, the motion was won. Among those speaking for the motion was regular Charleston attendee Toby Green, former CEO of OECD and now with a new initiative Coherent Digital, and speaking against was Pippa Smart, the editor-in-chief of Learned Publishing.
There were two individual presentations in the afternoon:
“Measuring Science Your Way” was the title chosen by Dr. Sabine Hossenfelder of the Frankfurt (Germany) Institute of Advanced Studies. Her argument was that scientists do not like assessment of scientific impact but we cannot do without them. Researchers, as well as those in administration and policy making, need a fast, reliable way to quantify scientific relevance. So, they use quantifiers that are readily available, which leads to more trouble with measures. They create perverse incentives. The popular H index is a good example: it encourages too much publishing and it militates against novelty. She then unveiled her own remedy. The site does not help that much. It is a non-profit venture. You design your own metric and apply it to a list of authors. No gaming is possible.
Secondly, Dr. Laura Fogg-Rogers from the University of the West of England spoke about Catch 22 – Women Peer Reviewers. Her thesis was that there is a significant under-representation of women in STM research which is damaging societal progress for democratic, utilitarian and equity reasons. There was plenty of evidence offered, for example, a Nature paper which showed that all male review panels rejected more submissions with female last authors than all female panels. The equivalent of blind auditions now common in orchestra selections was an answer but there was a basic unconscious structural bias, the remedies for which need to be covered in guidance for peer review. In the questions the previous speaker did not think most women would be happy with open peer review.
The second day began with a panel on Research metrics, which was mainly a UK concern, though some of the points made were relevant worldwide. The speakers were Sage publisher Caroline Porter (convener who made sure that the social sciences were not ignored) Euan Adie formerly of altmetrics.com, who remained mostly quiet, and David Carr of the Wellcome Trust and Professor James Wilsdon of Sheffield, currently Digital Science Professor of Research Policy, Department of Politics and Director, Research on Research Institute (RoRI), both of whom spoke a lot. There was some discussion about assessment of funders themselves. Carr told us that Wellcome, often in the lead as a funder, were now forcefully guiding the institutions that they fund, to really commit to approved policies such as DORA– judging by the article and not the journal the article is published in. It was agreed that incentives were important but difficult to align across national boundaries. The question was raised about lack of consultations with disciplinary representative bodies among the funder world before rather than after they mandated new policies.
Two expert presentations by publishers followed on Reproducibility and Reusability. Both can be accessed from the R2R site and are worth studying. Catriona Fennell, who directs publishing services at Elsevier, spoke on Applying the Reproducibility Manifesto. The challenges were many. Incentives are needed for researchers and more rigour needed in methods and statistics. It is difficult to validate reproducibility in peer review. Rebecca Grant of Springer Nature’s presentation was titled From Data Policy Towards FAIR Data For All: How Standardised Data Policies Can Improve Sharing. The emphasis was very much on benefits to researchers and the increase in numbers actually sharing. Almost all major publishers have endorsed FAIR. The current emphasis is having good policies (guidance) but Grant thinks that mandates are on their way.
The final two presentations were on Artificial Intelligence. The original speaker on Artificial Intelligence in Scholarly Information: a guide to the current landscape was Jim Longo, but he is no longer with HighWire and a colleague Olly Rickard gave a skeleton presentation ending with the summary: AI doesn’t replace humans; it deals with the huge scale and leaves the clever bit to us. Michael Upshall of UNSILO did give his presentation as expected: why has the take-up of AI been so limited in the Academic Work Flow? The answer helpfully began with some sensible definitions and some explanatory graphics. Scholarly publishing needs a scalable solution. OK, it was about publishing, but it was still useful. How to implement AI tools successfully is summarized here. Start with a business use case, for example, you may want to find related articles to my submission so that I can find an expert who can peer review it or you may want to find the most relevant journal for this manuscript article so you can submit it. Then you choose the most appropriate tool which could be manual classification or automatic concept extraction? Does it need technical skill and knowledge to implement? Or, alternatively, can humans contribute to the outcome? Then you move on to more technical questions.
Mark Allin, former president and CEO of Wiley, provided a magisterial summary. The conference has produced a lot and there has been a lot of consensus. There are reasons to be concerned. The Ecosystem is under threat and measurements are inadequate; women are under-represented in STEM research. Peer review is broken and Journals do not matter. There are, however, reasons to be optimistic. There are new models for learned Societies; research for life is strengthening research in the Global South. There is an agenda for greater diversity and transparency. Technology can enable better peer review, measurement, data sharing and improve reproducibility and, finally, journals still matter. Among the main themes are how robust, transparent and fair is the scholarly communications process? Are the key players aligned? On Impact – how to define, measure and maximize as a public good? And as far as Talent is concerned: how can global, female and early career researchers be better supported?
The workshops were mentioned earlier. For some they represented what was especially useful about the conference
They each produced a report – short but available on the site. Here are the topics which were discussed and reported: it looks as if the workshop with the most difficult mission did not send in its report – “Open Access Price Transparency- a demand from Plan S.”
Achieving an equitable transition to Open Access for low and middle-income countries; Improving Peer Review Support for Researchers; Transformative Agreement Collaboration – Identifying the problems; Recommendations for a sustainable and successful environment for the development and dissemination of scholarly research.
Finally, there was a keynote postscript from semi-retired Richard Charkin, former OUP academic publisher and the founder of Bloomsbury Academic, which nicely complemented the Adams keynote which began the meeting. He had similar concerns but his context was different- is unfettered open access an unfettered opportunity or a threat?