15 ways LLMs could ruin scholarly communication – and what we can do about it

Abstract:  Despite the dreams of science-fiction fans worldwide, the thing being marketed as “artificial intelligence” is no more than high-powered predictive text. What it gets right is thanks to its input data created by billions of humans, and to an invisible and underpaid workforce of content moderators. What it gets wrong threatens privacy, exacerbates sexism, racism and other inequities, and may even be environmentally damaging. There are situations that are well enough defined that machine models can be useful, but scholarly communication by its nature is full of new and unique information, relying on precisely reported data, that algorithms based on probabilities can’t deal with. So as a community we need to come with ways to prevent machine-generated fake papers from poisoning the well of science – and we need to be healthily sceptical of vendors selling us machine-based solutions to problems that can still only be addressed by human intelligence.

Citizen science projects tend to attract white, affluent, well-educated volunteers ? here’s how we recruited a more diverse group to identify lead pipes in homes

“Recruiting participants for a citizen science project produced a more diverse group when people were signed up through partner organizations, such as schools and faith-based organizations, than when they joined on their own. We used this approach to recruit volunteers for Crowd the Tap, a citizen science initiative that crowdsources the locations of lead plumbing in homes.

We signed up 2,519 households through partner organizations, in addition to 497 households that signed up on their own. We recruited households from all 50 states, though the majority came from North Carolina. Our project was initially funded by the Environmental Protection Agency, which led to nationwide sampling, but additional funding from the North Carolina Water Resources Research Institute led to prioritizing sampling in North Carolina.

We recruited 2.2 times more Black participants and 2.3 times more Hispanic or Latino participants through partnerships than we did through individual sign-ups. This allowed us to assemble a group of volunteers that more accurately represented the U.S. population. In addition, 11.2 times more lower-income participants took part in Crowd the Tap through partner organizations than on their own….”

Frontiers | Access to scholarly publications in the global North and the global South—Copyright and the need for a paradigm shift under the right to science

“The suggestions made in this contribution ultimately propose that the right to science should serve to properly realize access to scholarly publications for all, in the global North and the global South, by moving toward a world where commercialized science, economic copyright, and commercial scientific publishers play a reduced role. As it were, a paradigmatic shift is required under the right to science.”

Exploring the Dimensions of Scientific Impact: A Comprehensive Bibliometric Analysis Investigating the Influence of Gender, Mobility, and Open Access

Abstract: The Science of Science field advances the measurement, evaluation, and prediction of scientific outcomes through the study of extensive scholarly data. For these purposes, bibliometrics is an appropriate approach that studies large volumes of scientific data using mathematical and statistical methods, and is widely used to assess the impact of papers and authors within a specific field or community. However, conducting bibliometric analyses poses several methodological, technical, and informational challenges (e.g., collecting and cleaning data, calculating indicators) which need to be addressed. This thesis aims to tackle some of these challenges and shed light on the factors influencing scientific impact, specifically focusing on open access publishing, international mobility, and influential factors on the h-index. This thesis tackles methodological contributions, such as author disambiguation and co-authorship network analysis, as they provide insights into methodological and informational challenges within bibliometric analysis. Another methodological challenge addressed in this research is the inference of gender for a significant number of authors to obtain gender-related insights. By employing gender inference techniques, the research explores gender as an influential factor in scientific impact, shedding light on potential gender inequalities within the scholarly community. The research employs a bibliometric approach and utilizes mainly Scopus, a comprehensive dataset encompassing various disciplines to make the following contributions:

• We explore the impact of publishing behavior, particularly the adoption of open access practices, on knowledge dissemination and scholarly communication. With this intention, we investigate the impact of journals flipping from closed access to open access publishing models [74]. Changes in publication volumes and citation impact are analyzed, demonstrating an overall increase in publication output and improved citation metrics following the transition to open access. However, the magnitude of changes varies across scientific disciplines. In another study [76], we utilize a dataset of articles published by Springer Nature and employ correlation and regression analyses to examine the relationship between authors’ country affiliations, publishing models, and citation impact. Utilizing machine learning approach, we estimate the publishing model of papers based on different factors. The findings reveal different patterns in authors’ choices of publishing models based on income levels, availability of Article Processing Charges waivers, and journal rank. The study highlights potential inequalities in access to open access publishing and its citation advantage.

• We investigate the association between scholars’ mobility patterns, socio-demographic characteristics, and their scientific activity and impact. By utilizing network and regression analyses, along with various statistical techniques, we investigate the international mobility of researchers. Furthermore, we conduct a comparative analysis of scientific outcomes, considering factors such as publications, citations, and measures of co-authorship network centrality. The findings reveal gender inequalities in mobility across scientific fields and countries and positive correlations between mobility and scientific success.

• Centered on the prediction of scholars’ h-index as a metric of scientific impact, another one of our studies [77] employs machine learning techniques. We examine author, coauthorship, paper, and venue-specific characteristics, in addition to prior impact-based features. The results emphasize the significance of non-prior impact-based features, particularly for early-career scholars in the long term, while also revealing the limited influence of gender on h-index prediction. 

The findings of this research hold implications for researchers, academic institutions, and policymakers aiming to advance scientific knowledge and foster equitable practices. By unviii covering the influential factors that shape scientific impact and addressing potential gender disparities, this research contributes to the broader objective of promoting diversity, inclusivity, and excellence within the scholarly community. 

Request for Information (RFI) on NSF Public Access Plan 2.0: Ensuring Open, Immediate, and Equitable Access to National Science Foundation Funded Research Survey

“NSF is seeking public input from the science and engineering research and education community on implementing NSF Public Access Plan 2.0: Ensuring Open, Immediate, and Equitable Access to National Science Foundation Funded Research. This plan, described in the SUPPLEMENTARY SECTION of this Federal Register notice, represents an update to NSF’s current public access requirements in response to recent White House Office of Science and Technology Policy guidance….”

Federal Register :: Request for Information (RFI) on NSF Public Access Plan 2.0: Ensuring Open, Immediate, and Equitable Access to National Science Foundation Funded Research

“The National Science Foundation (NSF) is seeking public input from the science and engineering research and education community on implementing NSF Public Access Plan 2.0: Ensuring Open, Immediate, and Equitable Access to National Science Foundation Funded Research. This plan, described in SUPPLEMENTARY INFORMATION, represents an update to NSF current public access requirements in response to recent White House Office of Science and Technology Policy guidance. A primary consideration during the development of NSF’s plan has been potential equity impacts of public access requirements. NSF’s goal is to improve equity throughout the research life cycle, making data and opportunities available to all researchers, including those from marginalized communities and historically under-resourced institutions of higher education in the U.S. NSF is committed to considering the needs of the diverse US research community, including identifying possible unintended consequences that the plan and its implementation could produce.”

 

2nd Helsinki Initiative Webinar on Multilingualism in Scholarly Communication, Dec 1, 2023, 3pm (CET) | helsinki-initiative.org

Helsinki Initiative organizes a webinar series on Multilingualism in Scholarly Communication with speakers representing different expert communities and strands of work. This event includes three presentations ranging from language policies to the role of national language journals and translations in science. The event is free and open for everyone to participate without registration. A recording will be made available after the event. 

Date & time: 1 December 2023 15:00-16:30 CET

Zoom-link: https://tsv-fi.zoom.us/j/82347877173?pwd=gMmrAnbYpZa74tNR1lD13Qh5TGYM38.1 

Programme

Michael D. Gordin: The Inevitably of Translation in Scientific Communication

Science for millennia in Europe and Asia has been strongly shaped by frequent and repeated translations of scientific texts among multiple vehicular languages. As multiple vernacular languages started to displace Latin in European scientific communication in the seventeenth century, and then recede in the face of English and Russian — and eventually just English as the overwhelmingly dominant tongue — in the twentieth, the activity of translation has receded into the background but has remained no less constant. Despite some of the claims for DeepL and AI (English-)text generation, the activity of translation in scientific communication is unlikely to diminish in importance in the twenty-first century, although its modes and visibility will surely shift.

Lucía Céspedes: Implicit and explicit linguistic policies for scholarly communications in Latin America

Drawing inspiration from categories developed by the current known as Latin American Thought on Science, Technology and Development, this presentation seeks to analyze whether explicit linguistic policies for scholarly communications exist in Latin America. In such case, are they aligned with researchers’ linguistic ideologies and representations, and do they foster or hinder the region’s intrinsic multilingualism?

Vincent Larivière: The importance of French language journals for Canadian research

Many Canadian scholarly journals in the social sciences and humanities are distinguished by use of French language. This presentation discusses, based on a new study co-authored with Simon van Bellen, the relevance of national French-language journals for Canadian researchers, their role in the dissemination of knowledge on national themes, as well as the viability and sustainability of these journals.

The event is hosted by Emanuel Kulczycki and Janne Pölönen.

For further information email helsinki-initiative@tsv.fi.

 

Maximizing Access and Minimizing Barriers to Research in Low- and Middle-Income Countries: Open Access and Health Equity | Calcified Tissue International

Abstract:  Access to published research has always been difficult for researchers and clinicians in low- and middle-income countries, because of the cost of and lack of access to the relevant publications. The dramatic recent increase in electronic research publications has resulted in a marked improvement in reader access to these publications through their mainly Open Access policies, however the costs of processing of submissions and publication have now become the burden of the researchers wishing to publish, rather than the readers. For many researchers working in LMIC, the Article Processing Charges (APC) are prohibitive, hampering the publication of research being conducted in and relevant to these countries. A number of grant funding agencies and international not-for-profit organizations are trying to address these issues by including funding for article publications in their grants, or by supporting publishing entities by subsiding the cost of publication, but more needs to be done by major journal publishers through markedly reducing the APC being charged to researchers in LMIC for open access facilities.

 

The neglect of equity and inclusion in open science policies of Europe and the Americas | SciELO Preprints

Abstract:  National, international, and organizational Open Science (OS) policies are being formulated to improve and accelerate research through increased transparency, collaboration, and better access to scientific knowledge. Yet, there is mounting concern that OS policies—which are predicated on narrow understandings of openness, accessibility, and objectivity—do not effectively capture the ethos of OS and particularly its goal of making science more collaborative, inclusive, and socially engaged. This study explores how OS is conceptualized in emerging OS policies and to what extent notions of equity, diversity, and inclusion (EDI) and public participation are reflected in policy guidelines and recommendations. We use a qualitative document research approach to critically analyze 52 OS policy documents published between January 2020 and December 2022 in Europe and the Americas. Our results show that OS policies overwhelmingly focus on making research outputs publicly accessible, neglecting to advance the two aspects of OS that hold the key to achieving an inclusive and inclusive scientific culture—namely, EDI and public participation. While these concepts are often mentioned and even embraced in OS policy documents, concrete guidance on how they can be promoted in practice is overwhelmingly lacking. Rather than advancing the openness of scientific findings first and promoting EDI and public participation efforts second, we argue that incentives and guidelines must be provided and implemented concurrently to advance the OS movement’s stated goal of making science open to all.

 

Addressing the challenges in scholarly communication: the role of critical open access literacy | PUBMET

Abstract:  After twenty years of the open access movement, there has been growth in the uptake of open access, and some progress has been made in achieving its original goals; however, results have been far from satisfactory, with more than half of the research literature still closed (COKI, 2022; Curry, 2018). Moreover, new concerns have arisen, such as questionable quality and the reliability of peer review, in particular predatory publishing. There are also threats to equity, including stratifications of publishing as a consequence of the exclusionary character of the author-pays model of open access, and new risks of bias and exclusion in the means of transparent evaluation (Ross-Hellauer et al., 2022). It is argued that these challenges are the result of the uncritical narratives of openness and their narrow focus on access alone which fail to address inequitable power dynamics, systemic problems and the structural barriers in scholarly publishing and knowledge production (Perry, 2020; ibid.).

These cha(lle)nges in the system of scholarly communication, coupled with recent advances in technology and the tectonic transformations in the information environment, require new (pedagogical) approaches and foci that would enable researchers and students to understand and navigate such a complex environment; for this, a holistic and integrative approach to scholarly communication and information literacy is needed (ACRL, 2013; Špiranec, 2015).

Scholarly communication, including open access, is impossible without information literacy (Hebrang Grgi?, 2016). As they are both concerned with (access to) information, open access and in particular critical information literacy largely share the same goals, ethical dimensions and values of (social and epistemic) justice, equity and democratisation. Indeed, they have been considered instrumental to achieving these, and even proclaimed a panacea and deus ex machina for the current scientific, social and political challenges and crises (Hebrang Grgi?, 2016; Kapitzke, 2003). However, literature on the intersections and interplay between open access and information literacy has been limited.

This paper seeks to intertwine these two concepts more strongly, for mutual exchange and benefit, by analysing their convergent aspects as well as the role information literacy has in the context of the complexities of the scholarly communication system and in achieving open access. The paper builds on the few empirical studies of the intersections between the two concepts (e.g. Hebrang Grgi? (2016) which confirms the importance of ‘open access literacy’), but goes beyond their functional approaches, advocating a more holistic and critical approach to open access to help reinvent it and make a more substantial progress in open access. It argues that the transformations in the information environment and the scholarly communication system require not only basic skills and competences at the core of information literacy but also specific skills of, for instance, managing scientific data and publishing in open access sources. More importantly, what is also required is critical consciousness of all the aspects of the research process including the context, power relationships, and the privileged positions in knowledge production, publishing and dissemination, and an ability to evaluate the quality and reliability of information.

The paper proposes critical open access literacy as a pedagogical methodology and strategy to confront the challenges and enable a critical understanding of the contemporary information environment and scholarly communication. In line with the tenets of critical information literacy, critical open access literacy is directed at an analysis and critique of the economic, social, political, legal and technological conditions, aspects and implications of open access and the overall scholarly communication system, as well as the power dynamics, tensions and flaws within them. This will empower students and researchers to navigate this environment successfully and potentially become champions of its transformation to make it more meaningful, reliable, equitable and democratic.

Who should pay for open-access publishing? APC alternatives emerge

“When publishers first introduced APCs, the expectation was that these fees would be relatively small and a temporary measure that would provide an incentive for publishers to move to open access. “It’s one of those things that looked like a good idea at the time,” says Johan Rooryck, executive director of cOAlition S, a group of research funders and organizations supporting the shift to immediate open-access academic publishing. The coalition developed Plan S, a 2018 agreement originating in Europe, whereby research funders mandate full open access for the work that emerges from their support….”

Manco Vega (2023) Open Science Policies as Regarded by the Communities of Researchers from the Basic Sciences in the Scientific Periphery | MediArXiv Preprints

Manco Vega, A. (2023, November 14). Open Science Policies as Regarded by the Communities of Researchers from the Basic Sciences in the Scientific Periphery. https://doi.org/10.1108/OIR-03-2023-0135

Abstract This paper explores the different open science policy effects on the knowledge generation process of researchers in basic sciences: biology, chemistry, and physics. This paper uses a qualitative methodology with a content analysis approach. It uses seventeen semi-directed interviews. The main perceived effect of open science is access to research inputs, with open access, open research data and code reuse as primary sources. Another issue is the increase of collaboration with other colleagues in terms of the ability to collaborate faster and the encouraging the exchange of ideas. However, this benefit does not translate to the division of labor in large transnational teams. Time spent on tasks like cleaning up data and code, scooping and other ethical issues are unfavorable aspects noted. Policymakers could use it to enhance current open science policies in the countries. It analyzes perspectives of basic sciences researchers from two countries about open science policies. The main conclusion is the fact that open science policies should focus on the research process itself -rather than research outputs- in order to effectively tackle inequalities in science.

Beyond compliance: Curation as essential open science infrastructure

Abstract:  Presented October 19, 2023: “Developing New Approaches to Promote Equitable and Inclusive Implementation of Open Scholarship Policies.” Hosted by The National Academies of Sciences, Engineering, and Medicine’s Roundtable on Aligning Incentives for Open Scholarship.

 

RLUK ICIL | Stolen Tools: Anti-Racist and Democratic OA Publishing Tickets, Tue, Dec 5, 2023 at 2:00 PM | Eventbrite

“In a decolonial project to disrupt traditional paradigms of academic publishing, a collaborative initiative was forged between a group of health inequalities researchers. With the support of King’s College London, Libraries & Collections, this synergy birthed ‘Stolen Tools’, a pioneering open-access journal providing a platform for voices historically relegated to the periphery of academic discourse.

Unlike conventional publishing frameworks that primarily evaluate a manuscript’s content before acceptance, this model adopts a proactive stance by accepting authors of colour at the ideation stage. It augments their scholarly journey by pairing them with seasoned academic mentors – termed ‘critical friends’. This symbiotic relationship nurtures innovative academic research. Central to this initiative is a decolonial ethos aimed at dismantling the ivory tower’s barriers that have hindered many potential authors from contributing to the academic discourse.

This model envisions a more inclusive, anti-racist, and democratic academic publishing landscape whereby diverse narratives and epistemologies are accorded a rightful place within academic scholarship. This talk will outline their journey so far, how the journal team and King’s Libraries & Collections have worked together and their plans for a crowd sourcing solidarity sponsorship model….”

Reading Differently : Expanding Open Access Definitions Towards Greater Knowledge Equity | Pure @ Coventry University

Kiesewetter, H. R. (2023) Reading Differently: Expanding Open Access Definitions Towards Greater Knowledge Equity. https://pureportal.coventry.ac.uk/en/studentTheses/reading-differently

Abstract: This practice-based thesis is situated in the globalised sphere of digital knowledge production in the context of Open Access (OA) publishing. It is reading different accounts of the history of knowledge production and a broad variety of approaches to OA publishing – emerging in English and non-English speaking research cultures, in diverse economic, socio-political, and disciplinary contexts – together.As part of this reading, this thesis emphasises the dominant humanist tendencies in this discourse as well as the attempts to critique them. By doing so, it problematises persisting inequities in the field –what can be called a Eurocentric or neo-imperialist bias – and presents ways to create more diverse and equitable conditions for OA publishing today.This thesis puts forward that increasing participation in the processes and practices of scholarly knowledge creation (such as research, writing, and editing) and sharing (such as reading and publishing) – and seeing this as an inherent part of OA publishing – is key to facilitating fairer conditions for OA publishing. The focus of many prominent approaches to OA publishing has insteadbeen on extending access to research outputs (such as papers and books), thereby restricting OA publishing to the consumption of knowledge. To substantiate this claim, this thesis conceptualises critical OA publishing as a distinct OA tradition – reflective of a variety of strands within OA publishing – positioned within a longer history of “antagonist” theoretical and practical engagements with dominant (humanist) epistemologies. This genealogical positioning emphasises that critical OA advocates have always stressed that OA publishing should not only be about how readers consume texts, but also about who has access to, and controls the governance of, the means of knowledge production; it elucidates why this includes an attentiveness to the processes and practices of knowledge production as sites of struggle for knowledge equity and diversity; and it helps me to devise a novel interventionist (reading) methodology.

This methodology is one of the main outcomes of this thesis. It exemplifies and enacts how minimising the socio-cultural, behavioural, and linguistic barriers to participation in the processes and practices of knowledge production can advance knowledge equity and diversity. This methodology adds to critical experiments with writing, editing, and publishing conducted by critical OA advocates to facilitate fairer conditions for scholarship. It can be applied in various contexts of collaborative academic knowledge production (for example, research or writing). It has been devised based on the main insights from this thesis, it has been tested within two experimental online reading groups, The Re-Reading Room, and it is discussed in an experimental piece of writing.