20. Mai 2022

APE2022 Considers The Permanent Record in Academic Publishing

By Marydee Ojala

The 17th annual Academic Publishing Europe (https://www.ape2022.eu) chose as its theme “The Future of the Permanent Record”. Originally planned to be in-person in Berlin, APE went virtual on the Morrissier platform and spread itself over 3 days to accommodate speakers and delegates from many time zones. APE, the brainchild of Arnoud de Kemp, is now held under the auspices of the Berlin Institute for Scholarly Publishing (BISP), a non-profit organization formed in 2020 by the Walter de Gruyter Foundation. The Foundation’s primary focus is on continuing the APE conference, although it intends to have other functions as well, including training for early career researchers, publishers, and research funders.

A striking change in APE 2022, aside from its being virtual, was BISP’s Managing Director, Eric Merkel-Sobotta, as the host for all three days. The face of the in-person conference, founder de Kemp, was missing this year, due to circumstances beyond his control. As always, however, he was deeply involved in planning APE, curating the entire program. Merkel-Sobotta did an excellent job of keeping the conference on schedule and introducing sessions.

Quality, Equity, and Technology

In her welcome to the conference, Caroline Sutton, CEO of the STM association and Director of Open Research, Taylor & Francis Group, asked the audience to consider who has responsibility for the authenticity of the permanent record. She also asserted that publishers matter greatly in the quest for reliable data but admitted that the APE conference was likely to raise more questions than answers. That’s actually as it should be, since one function of a conference is to stimulate thinking and present different viewpoints.

The first keynote speech, by Maria Leptin, President of the European Research Council, focused on quality and equity in academic publishing. Although preprints have gained traction during the pandemic, not all are refereed. Plus, the preprint can vary significantly from the final published article. Leptin thinks that refereed preprints give trust and reliability to research. Open publishing programs can be valuable but raise questions about quality. The economic models of high quality selective publishers tend to exclude researchers in poorer countries, sometimes making it obligatory for them to publish in lower quality journals, regardless of the quality of their research. She thinks publishers should publish OA journals without imposing APCs and wonders what funders can do, which could be anywhere between everything to nothing.

Todd Carpenter lays out the variations in formats over time

The second keynote speech, by NISO’s executive director, Todd Carpenter, started with a brief history of the internet, noting that it was designed by and for researchers not just defense. Sharing research was a bedrock concept for Tim Berners-Lee. He observed that content for digital natives needs to be much more than a replication of old technology, particularly print, more than simply a PDF. Attributes that appeal to digital natives include multi-format, multimedia, interoperability, machine readable formats, adaptive design, accessible, transformable, atomize-able, high quality metadata, preservable, linkable, and trackable.

Looking towards the scholarly record of the future, Carpenter concentrated on changes in technology, including metadata, algorithms, and machine learning. These affect all aspects of what people find and how they read. Technology drives change but doesn’t determine the direction or eventual destination. People are motivated to change either because they are forced to or because they realize the change is for the better. In his view, seamless access is vital, as is building a notification system of changes that would identify retractions, final version, and dataset alterations.

Todd Carpenter showed how algorithms affect what can be found

CrossRef Product Manager, Martyn Rittman, stressed the importance of research integrity in light of the number of retractions. He noted the difficulties of connecting related research due to the proliferation of outputs and formats. Quality metadata holds some answers to these issues and is particularly relevant for OA publishing.

The Dark Side of Scholarly Publishing

Liz Marchant, Global Portfolio Director, Taylor & Francis Group, said the version of record is under attack, calling it the dark side of the scholarly publishing universe. When considering the version of record (VOR) that scholars should consider as the authoritative one, the existence of fraudulent papers raises its ugly head. Paper mills churning out worthless articles with flawed research findings undermine the public’s trust in science and scholarly publication. Christian Behl, Editor-in-Chief, Journal of Cellular Biochemistry and Institute for Pathobiochemistry, University Medical Center of Johannes-Gutenberg-University, concurred that trust in science is needed but that trust is undermined by the huge increase in false research from paper mills, with their fake data, non-traceable authors, and image manipulation.

Responsibility for cleaning up this mess, said Bernhard Sabel, Editor-in-Chief, Restorative Neurology and Neuroscience and Medical Faculty, Otto-von-Guericke University of Magdeburg, lies with publishers. The number of fake publications is distressingly high, as much as 10 to 15% in some disciplines, and that, he believes, is criminal. Fraudulent papers are getting more sophisticated and harder to identify, noted Tim Kersjes, Research Integrity Manager, Springer Nature. He believes that taking time to educate editors will help fix the problem. In a very different approach, Alicia Wise, Executive Director, CLOCKSS, discussed digital preservation, saying that time is a theft of memory. When journals disappear from the web, the VOR is lost. Even when journals are digitally preserved, challenges in the core infrastructure cry out for better discovery tools.

Version of Record

Asking if the VOR is worth it for the research community was the question Lisa Janicke Hinchcliffe, University of Illinois librarian, posed. Traditionally, the VOR was published, after being copy edited and typeset. But what happens when there is no print version? Should researchers cite the gold OA VOR or an eprint archive? Her “landscape scan” of the VOR was comprehensive and far-reaching. She published an adapted and updated version of the talk on The Scholarly Kitchen, 14 February 2022 (https://scholarlykitchen.sspnet.org/2022/02/14/the-state-of-the-version-of-record/). The comments are well worth reading as they offer a thoughtful give and take on definitions of the VOR and its future for scholarly research. This seemed like an extension of the APE conference even though it was in a different venue.

A panel discussion following Hinchliffe’s talk added various perspectives on what the version of record (VOR) could (or should) entail. Hinchcliffe began by declaring that the VOR was central to scholarly research. However, a record of various versions would be useful since the current system detaches the final form from earlier versions. Moderator Anne Kitson, Managing Director and Senior Vice President, Cell Press and The Lancet, Elsevier, made a distinction between open and transparent, calling for more accountability.

Panelist Niamh O’Connor, Chief Publishing Officer, PLOS, did not think that the VOR was the right way to create trust and value. Publishing, she said, is a process, not an event, that is open by design. Although many changes have occurred over the past 20 years, as publishing has moved from the physical to the digital and research has transitioned toward open science, too much has not changed. Publishing should be regarded as a service not a product. The PLOS global equity model focuses on sustainability and represents a radical reinvention of scholarly publishing. Discussions around the VOR, she claimed, ignore how search actually works. Selectivity does not need to equate to higher costs. As a rhetorical question, she asked who gets to decide what’s valuable, the creator or the viewer/reader?

Bernd Pulverer, Chief Editor, EMBO Reports, and Head of Scientific Publications at EMBO Press, responded by saying he thinks the article of the future is the network but the VOR is the anchor. He contended that database curation should be pre-publication not post publication. Preprints can be versioned but the VOR is static. He acknowledged that the academic community tends toward the conservative, interested in quality control and reproducibility. The community’s interests were hijacked by impact factors, which he thinks led to corruption of the system of research assessment. However, he does see some improvements and urged incentivizing peer review, perhaps by giving academic credit.

Strongly disagreeing with his fellow panelists, Ulrich Dirnagl, Director, Quest Center for Transforming Biomedical Research and Dept. of Experimental Neurology, Berlin Institute of Health, Charité and MDC, declared the VOR dead. In his opinion, publishers are more innovative than scientists. Publishers created the concept of VOR, which he views as a distraction. The VOR does not guarantee trust and the deluge of output further erodes quality. Young researchers accept the idea of open reviews but then they get absorbed into the system and are not rewarded for going outside the system that publishers created.

Three U.S.-based academic library directors discussed Making Research Data Accessible

In another panel, Rick Anderson, University Librarian at Brigham Young University; Michael Levine-Clark, Dean of Libraries, University of Denver; and Judith C. Russell, Dean of University Libraries, University of Florida, gave their views on how to ensure public access to research data. Academic libraries are critical to this function but each approach it slightly differently. Anderson described “data as an economic asset” as both an opportunity and a challenge. On the opportunities side, he mentioned the benefits from both sharing and citing, while challenges revolved around physical storage space, preservation, and potential for misuse.

Dot Coms to Watch

Identifying technology startups has long been a highlight of APE. This year’s contenders for the “next dot com to watch” were Cassyni, CiteAB, Nested Knowledge, Pilloxa, and aipatents. Eefke Smit, STM Director of Standards and Technology, introduced the speakers, each of whom had 8 minutes to introduce their company and its products. Andrew Preston explained how Cassyni is making research seminars available online. Andrew Chalmers reported that CiteAB focuses text mining to identify the right antibodies to use for experiments in the lab by combining AI & expert human review. Nested Knowledge, represented by Keith Kallmes, uses technology to enhance systematic reviews. Pilloxa, said Francesco Mazzotta, concentrates on digital health and patient engagement. Liat Belinson presented the search platform aipatents.com, which relies on Natural Language Processing rather than keywords to connect different types of data and determine relevancy.

The overall winner, for all four criteria—appealing concept, most likely to succeed, most attractive collaboration partner, and investment potential—was Nested Knowledge. The “dot coms to watch” is a very popular session during the in-person APE but highlighted some deficiencies in a virtual setting. Although the companies had intriguing stories to tell, the audience was very sparse, which became evident during the voting where the largest number of voter was just above 20. It’s hard to keep people’s attention for a three day conference when they are not in a conference room. Other demands on their attention, both person and work-related, pull them away from conference session, particularly on the third day.

One “graduate” of the “dot coms to watch” is Morrissier, which has strong ties to APE. Not only is the company headquartered in Berlin, but it has also attended and presented at most of the APE conferences over the past 8 years of its existence. Thus, it was a natural choice when the conference was forced to go virtual, as it had been the year before. The Morrissier platform had only one technical problem at the conference—ironically, it occurred during the panel discussion with Sami Benchekroun, co-founder and CEO of Morrissier, and Ijad Madisch, co-founder and CEO of ResearchGate, moderated by Outsell Inc.’s, David Worlock.

Having weathered the storms of starting up, Morrissier and ResearchGate have grown and prospered. Both entrepreneurs acknowledged the role of the German school system in shaping their interest in the information industry. Benchekroun told of his fond memories of a sleepover in his local library.

Madisch, as a result of his own research, saw a need for a network to exchange scientific ideas. His aim was to reinvent not disrupt. He advised startups to think about opportunities instead of risks. Don’t overthink, he added, but concentrate on solving the main problems, which for him was making scientists more productive to drive new scientific knowledge. He credits his company’s success to its people who bring fresh ideas.

For Benchekroun, “success starts with science.” His original goal was to create content management tools for academic conferences. Being a “first mover” is not really an advantage, since explaining and educating takes up time. Benchekroun believes that virtual conferences will continue and are particularly appealing to digital natives who have no problem with attending virtual conferences via their phones or tablets. More touchpoints will develop around posters and presentations with richer conversations about content. He also speculated about 3D exhibit halls.

Marydee Ojala
Editor in Chief
Online Searcher: Information
Discovery, Technology, Strategies
Information Today Inc., USA
Editor, ILI365 eNews
Information Today Ltd, UK

APE keeps its own version of a permanent record by archiving its conferences. Sub-themes at the conference included the appalling increase in fraudulent research papers and worthless publications, the technological challenges confronting the scholarly publishing industry, and the need for diversity, equity and inclusion at all levels. These are important issues that will doubtless enjoy more scrutiny at APE2023, which is scheduled for 10-11 January 2023 in Berlin. As in the past, whether in-person or virtual, it will be the first publishing conference of the year. I am definitely hoping it will be in-person, back in Berlin at the Leibniz Hall in the Brandenburg Academy of Sciences and Humanities building.