Abstract

Science policy promotes open access to research data for purposes of transparency and reuse of data in the public interest. We expect demands for open data in scholarly publishing to accelerate, at least partly in response to the opacity of artificial intelligence (AI) algorithms. Open data should be findable, accessible, interoperable, and reusable (FAIR), and also trustworthy and verifiable. The current state of open data in scholarly publishing is in transition from ‘nice to have’ to ‘need to have.’ Research data are valuable, interpretable, and verifiable only in context of their origin, and with sufficient infrastructure to facilitate reuse. Making research data useful is expensive; benefits and costs are distributed unevenly. Open data also poses risks for provenance, intellectual property, misuse, and misappropriation in an era of trolls and hallucinating AI algorithms. Scholars and scholarly publishers must make evidentiary data more widely available to promote public trust in research. To make research processes more trustworthy, transparent, and verifiable, stakeholders need to make greater investments in data stewardship and knowledge infrastructures.

Keywords: research data, open access policy, verifiability, data reuse, data stewardship, knowledge infrastructure

1. Why Open Access to Data?

Policymakers expect open access to research data to improve scholarship and to serve the public interest by making the research process more transparent, improving the ability to verify methods and findings, increasing reproducibility, facilitating reuse of data, and democratizing participation in research. These are among the stated goals of the recent Nelson memo that applies to U.S. federal research funding and of similar policy statements in Europe, Asia, and elsewhere (European Union Publications Office, 2018; GO FAIR, 2020; National Institutes of Health, 2020; Nelson, 2022).

At the same time, the rapid rise of generative AI (artificial intelligence), large language models, and machine learning are undermining our ability to trust the authenticity of artifacts of knowledge—books, journal articles, news stories, data visualizations, and other media (Bly & Brand, 2023; Loukides, 2023; Milano et al., 2023). Current AI technologies such as ChatGPT can produce remarkably persuasive texts, but these tools also are known to ‘hallucinate,’ that is, “to produce false information contrary to the intent of the user and present it as if true and factual” (Kuta, 2023; Norlen & Barrett, 2023). Examples include citing imaginary journal articles as sources. The opacity of current AI algorithms is a primary driver for government and industry regulation, with calls to label AI-generated content, to disclose data sources used to train large language models, and to promote ‘algorithmic transparency’ by which AI tools explain their reasoning (Electronic Privacy Information Center, 2023; Exec. Order No. 14110, 2023; Rotenberg, 2022).

Tensions between the transparency goals of open science and the opacity of AI algorithms may be the tipping point that moves data sharing from ‘nice to have’ to ‘need to have’ in scholarly publishing. Access to research data offers a means to audit the published record and to verify AI-generated content. The FAIR principles (Findable, Accessible, Interoperable, Reusable) that form the basis for most open data policies are aspirational and focus largely on technical aspects of data release (Wilkinson et al., 2016). The FAIR principles do not provide guidance on the ability to interpret or trust research data, once released. As scholarly publishers seek new means to promote trust in their journals, access to data sets is an important mechanism to strengthen evidence for conclusions and to promote transparency of the research process.

While the goals of open research data policies are clear, the consequences for stakeholders are not. The types, uses, and reuses of data vary widely across disciplines, context, and over time (Borgman, 2015; Pasquetto et al., 2017). An important factor in the ability to reuse data is the degree to which communities agree on standard formats for data production. The astronomy community, for example, established a common structure for observational data from telescopes in the late 1970s. As a consequence, large corpora of astronomical observations exist that can be analyzed with available software tools (Borgman, Sands, et al., 2016; Scroggins & Boscoe, 2020). Qualitative data such as interview transcripts are more difficult to structure, although software tools to mark up and analyze qualitative data are now available widely. Regardless of format, considerable labor is required to contextualize data for reuse (Borgman et al., 2021; Pychlau & Wagner, 2023). The short- and long-term costs of providing open access to data are high regardless of data type and discipline. Economic models for open data are poorly understood, as are the myriad risks associated with data access (Lane et al., 2024).

2. Transitional Status of Research Data

Our objective here is to highlight key issues arising as open research data becomes essential to scientific and scholarly publication. We address concerns about data utility within and between communities, imbalance of costs and benefits, and risks associated with open access to data. The growth of AI is a subtext to these transitions, rather than a primary driver.

Nearly 20 years ago, Philip Bourne (2005), then serving both as editor-in-chief of PLoS Computational Biology and as codirector of the Protein Data Bank, asked, “what is the difference between an entry in a database and an article in a journal?” His answer was to characterize the difference “as a mix of perception and content.” Ideally, a scholar can search a data archive and link directly from a sequence or specimen to the paper describing the methods and findings. Conversely, the scholar can read a paper and link directly to the evidentiary materials on which the paper is based. The publication and data set have complementary, but not equivalent, value, as they serve different purposes in scholarly communication. Bourne proposed several means of strengthening the relationship between publications and data sets, largely with the goal of improving discoverability of individual units of content and relationships between them. Digital object identifiers (DOIs) are now assigned routinely to data sets, one proposed improvement, but adding more markup and structure to publications remains rare.

While most research communities now consider data products to be beneficial assets, investments in the knowledge infrastructures necessary to sustain access to data sets in ways that they maintain their utility is spotty at best (Borgman, 2015; Borgman & Bourne, 2022). However, data sets are becoming more readily discoverable due to advances in retrieval tools and data citation mechanisms (Borgman, 2016; Cousijn et al., 2019; DataCite, 2022; Fenner et al., 2019; Gregory et al., 2019, 2020, 2023; Groth et al., 2019; Koesten et al., 2021; Parsons et al., 2019; Zhang et al., 2023).

2.1. Utility of Research Data

The epistemology of data has a long history (Meadows, 2001; Meyns, 2019; Rosenberg, 2013). Research data are entities constructed by people for specific purposes, often based on implicit models and unstated assumptions. Most scientific data are tightly coupled with instrumentation used to generate them and with software used to generate, process, analyze, interpret, and display them. Some instruments and software are standardized within research communities, but many are highly customized technical methods developed for a narrowly defined inquiry. Customized data pipelines and other tools often are the ‘secret sauce’ of a research team.

While data sets produced by specialized tools are released and interpretable by others, data creators retain the advantage of intimate knowledge about how those data were created (Pasquetto et al., 2019). Researchers frequently use open data for comparisons, ‘ground truthing,’ exemplars, and background. However, to reuse a data set to pursue a new form of inquiry, researchers benefit from direct contact with the data creators.

Consider the field of child language acquisition, which thrives on data collected from spontaneous interactions between children and caregivers in naturally occurring situations, and on cross-linguistic comparisons of the language learning process. In 1984, researchers at Carnegie Mellon University launched a shared data resource called the Child Language Data Exchange System (CHILDES) to serve as a central repository for data concerning first language acquisition. It originally contained transcripts of audiotapes from the 1960s of child utterances in several language populations. In the intervening years, it has grown to include hundreds of corpora across 26-plus languages, in different media, and has been cited as a resource in thousands of publications (CHILDES, 2023).

The existence of CHILDES and the development of computational tools to manipulate the data it contains are considered significant factors in the steep growth of cross-linguistic research on child language during the late 20th century. Researchers without the time or resources to travel to different parts of the world to collect relevant longitudinal data could use this rich online resource instead.

Early users of the CHILDES database accessed human-transcribed text, without having access to the audio recordings on which those transcriptions were based. Metadata accompanying the transcripts indicated the ages of the children recorded. Descriptions of where and when a given recording was made were also typically provided. Beyond that, however, the data reuser had to make a leap of faith concerning the reliability of the transcripts, knowing that the opportunity for inaccuracies and human error in transcription was significant, and recognizing that the utterances of very young children are often difficult to interpret, even first-hand (Pierce, 1989).

The researcher who recorded the utterances in real time has the most certainty in coding and uses of these data. However, it is important to recognize that both data creator and reuser are vulnerable to implicit biases and motivated reasoning. Reproducibility of research results using qualitative data such as childhood utterances is an issue when multiple researchers are asking the same question of the data. More commonly, resources such as CHILDES are source material for myriad research questions, whether alone or in combination with other data sources. Researchers can address concerns about subjective judgments in coding and analyzing transcripts by methods of iterative coding and hypothesis testing (Glaser & Strauss, 1967; Strauss & Corbin, 1998).

2.2. Benefits and Costs of Open Access to Research Data

In the simplest economic terms, most of the benefits of open access to research data accrue to the reusers of data and most of the costs are borne by the data providers. Open data is a ‘commons problem,’ in the economic sense, in need of governance structures, and subject to ‘free riders’ (David, 2004, 2005; Hess & Ostrom, 2007; Ostrom, 1990).

To acquire research data, researchers expend great effort, and often great sums of money, whether in the sciences, social sciences, humanities, medicine, engineering, arts, or other disciplines. The time spent acquiring data for a paper might be short, such as one night on a telescope or one day of recording child language, or very long, in the case of longitudinal or comparative studies. In either case, the time and labor devoted to analyzing, interpreting, and writing for publication are usually far greater than that spent in data acquisition (Borgman, 2015; Borgman, Golshan, et al., 2016).

Processing research data in ways that they can be released for use by others is another labor cost, and one that often requires very different skill sets. Archiving and publishing data involves transfer into standardized formats that can be read by available software and maintained in stable structures; includes documentation to facilitate interpretation and manipulation; contains sufficient metadata and provenance information to ensure their validity and authority; and gives credit to the originators (Baker & Mayernik, 2020; Borgman & Wofford, 2021; Koesten et al., 2020, 2021). As Bourne (2005) noted, authors are reluctant to invest more effort than necessary in structuring their publications or their data sets. Advances in information retrieval methods, including those based in AI, can improve the discovery of publications, data sets, and relationships between them (Gregory et al., 2019, 2020, 2023; Lane, 2020; Lane et al., 2020; Zhang et al., 2023).

One research project may yield one data set and one publication, in which case the relationship between data set and publication is clear. More often, one-to-many or many-to-many relationships exist between projects, publications, authors, universities, and funding sources, as data are mined and combined to address multiple research questions over time. Determining how to extract a data set associated with a specific paper is a nontrivial matter, as is maintaining the chain of provenance (Borgman, 2015).

Responsibility for the labor, costs, and expertise required to transform research data into portable data sets, to steward these data resources for long periods of time, or to provide access to those data resources is rarely specified by the policies of governments, funding agencies, journals, or universities. Rather, responsibility for these aspects of data stewardship is diffuse (Borgman & Brand, 2022). As explored in a prior article in this journal, “it takes a village to manage and share data” (Borgman & Bourne, 2022).

2.3. Risks of Open Access to Research Data

The risks of providing open access to research data have received little empirical attention to date, although they are a common theme in researchers’ private conversations (Borgman, 2015). Some researchers are concerned about other investigators using their data without investing the labor of obtaining grants, conducting the research, analyzing those data, and writing the paper. When reusers were labeled ‘data parasites,’ a biomedical community established a ‘data parasite award’ for those who reuse data sets most effectively (The Research Parasite Awards: Celebrating Rigorous Secondary Data Analysis, 2021; Wofford, 2022).

Some researchers view the loss of control over their data as a risk, and one that is compounded by lack of attribution and credit to the originator. Provenance trails facilitate credit, authority, ownership, licensing, and the ability to assess the integrity of research data. Better data citation mechanisms and practices are the usual response to attribution concerns. Over the course of the last decade or so, data citation practices have matured with the availability of formal schema, assignment of DOIs, and repository support for these mechanisms (Borgman, 2016; Cousijn et al., 2019; DataCite, 2022; Fenner et al., 2019; Groth et al., 2019; Parsons et al., 2019).

Researchers have been slow to adopt formal citation practices for many reasons, however. Cataloging data sets requires professional expertise, and only a small subset of available research data are described sufficiently for formal data citation. When authors do cite data sets, it is often via mentions in text or footnotes, both of which are difficult to extract for search and retrieval (Pepe et al., 2014; Uhlir, 2012). Recent developments in automated extraction of data set mentions are a promising, if partial, response to problems of attribution and provenance (Lane, 2019, 2020; Lane et al., 2020).

Intellectual property (IP) concerns in reusing data arose in the earliest days of electronic publishing. By the mid-1960s, the Text Encoding Initiative was maneuvering around the landmines of IP rights (The Text Encoding Initiative, 2024). By the early 2000s, the Google Books project and related scanning initiatives brought authors, publishers, libraries, and readers to legal battles, a contest that accelerated during the COVID-19 pandemic over laws governing access to digital content (Hachette Book Group, Inc. v. Internet Archive, 2023; Harris, 2020; Hernandez, 2023; Roberts, 2020; Streitfeld, 2023). A battle currently is brewing about AI-based large language models that mine intellectual property without attribution or compensation (Bly & Brand, 2023; Loukides, 2023; Milano et al., 2023).

Long-simmering tensions over misuses and misinterpretations of research data are also coming to the fore. Scientists working on climate change face skeptics who ‘cherry-pick’ their data to extract facts and figures out of context, claiming alternative conclusions. The COVID-19 pandemic brought scientists, medical and public health professionals, and policymakers into storms of skeptics attempting to reuse research data for contradictory claims. The trolls are everywhere, challenging paradigms from archaeology to zoology. Public engagement in scholarship has risks as well as benefits. Data creators, data reusers, publishers, and the public can benefit when evidentiary data are widely available, especially if documented sufficiently to ensure their validity and veracity.

Artificial intelligence raises a whole host of risks to scholarly publishing, open science, and open access to data. Among the risks in uses of open data that are apparent at the current state of AI development are known errors such as stating unverifiable ‘facts,’ citing nonexistent sources, interpreting data in ways incompatible with scientific reasoning, and combining unrelated data to draw conclusions that are not scientifically supportable. These are but a few of the many risks about AI, science, scholarship, data, and truth raised to date (Crawford, 2021; Gil, 2009; Hutson, 2018; Organisation for Economic Co-operation and Development, 2023; Williamson et al., 2021).

3. Conclusions

The ubiquity of AI-generated text will continue to diminish our ability to trust digital content, including published research, unless and until AI companies are required to disclose the provenance and verifiability of the content on which their large language models are trained and the reasoning algorithms employed. Already, the research community and the general public are demanding more transparency of the ‘evidence’ necessary to discern trustworthy content. For readers to trust digital content, they need information about the provenance of assertions. Experts in a domain need the ability to audit data and to inspect statistical analyses underlying those assertions.

Depositing data associated with publications is now accepted practice in most scholarly disciplines. Data release is enforced by policies of funding agencies and by publisher guidelines. In our vision of the future, open data will play large roles in democratizing and accelerating research and in supporting evidence-based knowledge. We acknowledge that different fields of research have different relationships to data, and fields vary in the degree of apparatus required to be interpretable and reusable. Data do not ‘speak for themselves’; rather, they are useful only in relation to their origins and context, and the knowledge of the user. Nonetheless, we expect to see rapid growth in data sharing requirements for publication in all fields of research. We hope, furthermore, that more open sharing of data brings more globally equitable participation in the research process.

Disclosure Statement

Christine L. Borgman and Amy Brand have no financial or non-financial disclosures to share for this article.

References

Baker, K. S., & Mayernik, M. S. (2020). Disentangling knowledge production and data production. Ecosphere, 11(7), Article e03191. https://doi.org/10.1002/ecs2.3191

Biden, J. R. (2023). FACT SHEET: President Biden issues executive order on safe, secure, and trustworthy artificial intelligence. White House. https://www.whitehouse.gov/briefing-room/statements-releases/2023/10/30/fact-sheet-president-biden-issues-executive-order-on-safe-secure-and-trustworthy-artificial-intelligence/

Bly, A., & Brand, A. (2023, August 7). AI is muddying the truth. We’ve known how to fix it for centuries. The Boston Globe. https://www.bostonglobe.com/2023/08/07/magazine/ai-is-muddying-truth-weve-known-how-fix-it-centuries/

Borgman, C. L. (2015). Big data, little data, no data: Scholarship in the networked world. MIT Press.

Borgman, C. L. (2016). Data citation as a bibliometric oxymoron. In C. R. Sugimoto (Ed.), Theories of informetrics and scholarly communication (pp. 93–116). De Gruyter Saur. https://doi.org/10.1515/9783110308464-008

Borgman, C. L., & Bourne, P. E. (2022). Why it takes a village to manage and share data. Harvard Data Science Review, 4(3). https://doi.org/10.1162/99608f92.42eec111

Borgman, C. L., & Brand, A. (2022). Data blind: Universities lag in capturing and exploiting data. Science, 378(6626), 1278–1281. https://doi.org/10.1126/science.add2734

Borgman, C. L., Golshan, M. S., Sands, A. E., Wallis, J. C., Cummings, R. L., Darch, P. T., & Randles, B. M. (2016). Data management in the long tail: Science, software, and service. International Journal of Digital Curation, 11(1), 128–149. https://doi.org/10.2218/ijdc.v11i1.428

Borgman, C. L., Sands, A. E., Darch, P. T., & Golshan, M. S. (2016). The durability and fragility of knowledge infrastructures: Lessons learned from astronomy. Proceedings of the Association for Information Science and Technology, 53(1), 1–10. http://dx.doi.org/10.1002/pra2.2016.14505301057

Borgman, C. L., & Wofford, M. F. (2021). From data processes to data products: Knowledge infrastructures in astronomy. Harvard Data Science Review, 3(3). https://doi.org/10.1162/99608f92.4e792052

Borgman, C. L., Wofford, M. F., Golshan, M. S., & Darch, P. T. (2021). Collaborative qualitative research at scale: Reflections on 20 years of acquiring global data and making data global. Journal of the Association for Information Science and Technology, 72(6), 667–682. https://doi.org/10.1002/asi.24439

Bourne, P. E. (2005). Will a biological database be different from a biological journal? PLoS Computational Biology, 1(3), Article e34. http://dx.doi.org/10.1371/journal.pcbi.0010034

Child Language Data Exchange System (CHILDES). (2023). [Home page.] https://childes.talkbank.org/

Cousijn, H., Feeney, P., Lowenberg, D., Presani, E., & Simons, N. (2019). Bringing citations and usage metrics together to make data count. Data Science Journal, 18, Article 9. https://doi.org/10.5334/dsj-2019-009

Crawford, K. (2021). Atlas of AI: Power, politics, and the planetary costs of artificial intelligence. Yale University Press.

DataCite. (2022). DataCite Metadata Schema. DataCite. http://schema.datacite.org/

David, P. A. (2004). Can ‘open science’ be protected from the evolving regime of intellectual property protections. Journal of Institutional and Theoretical Economics, 160(1), 9–34. https://www.jstor.org/stable/40752435

David, P. A. (2005).Creating the information commons for e-science: Toward institutional policies and guidelines for action. http://www.codataweb.org/UNESCOmtg/pres-pdavid.pdf

Electronic Privacy Information Center. (2023). AI policy. https://epic.org/issues/ai/ai-policy/

European Union Publications Office. (2018). Turning FAIR data into reality: Final report and action plan from the European Commission expert group on FAIR data. https://publications.europa.eu/en/publication-detail/-/publication/7769a148-f1f6-11e8-9982-01aa75ed71a1/language-en/format-PDF

Exec. Order No. 14110, 3 C.F.R. 75191 (2023). https://www.govinfo.gov/app/details/FR-2023-11-01/2023-24283

Fenner, M., Crosas, M., Grethe, J. S., Kennedy, D., Hermjakob, H., Rocca-Serra, P., Durand, G., Berjon, R., Karcher, S., Martone, M., & Clark, T. (2019). A data citation roadmap for scholarly data repositories. Scientific Data, 6(1), Article 28. https://doi.org/10.1038/s41597-019-0031-8

Gil, Y. (2009). From data to knowledge to discoveries: Artificial intelligence and scientific workflows. Scientific Programming, 17(3), 231–246. https://doi.org/10.3233/SPR-2009-0261

Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. Aldine Publishing.

GO FAIR. (2020, January 29). Universities push for greater global open access to research data. https://www.go-fair.org/2020/01/29/universities-push-for-greater-global-open-access-to-research-data/

Gregory, K. M., Groth, P., Cousijn, H., Scharnhorst, A., & Wyatt, S. (2019). Searching data: A review of observational data retrieval practices in selected disciplines. Journal of the Association for Information Science and Technology, 70(5), 419–432. https://doi.org/10.1002/asi.24165

Gregory, K. M., Groth, P., Scharnhorst, A., & Wyatt, S. (2020). Lost or found? Discovering data needed for research. Harvard Data Science Review, 2(2). https://doi.org/10.1162/99608f92.e38165eb

Gregory, K. M., Groth, P., Scharnhorst, A., & Wyatt, S. (2023). The mysterious user of research data: Knitting together science and technology studies with information and computer science. In K. Bijsterveld & A. Swinnen (Eds.), Interdisciplinarity in the scholarly life cycle: Learning by example in humanities and social science research (pp. 191–211). Springer International Publishing. https://doi.org/10.1007/978-3-031-11108-2_11

Groth, P., Cousijn, H., Clark, T., & Goble, C. A. (2019). FAIR data reuse – the path through data citation. Data Intelligence, 2(1–2), 78–86. https://doi.org/10.1162/dint_a_00030

Hachette Book Group, Inc. v. Internet Archive, No. 20-cv-4160 (JGK), 2023 WL 2623787 (S.D.N.Y. 2023).

Harris, E. A. (2020, June 1). Publishers sue Internet Archive over free e-books. The New York Times. https://www.nytimes.com/2020/06/01/books/internet-archive-emergency-library-coronavirus.html

Hernandez, J. (2023, March 26). A judge sided with publishers in a lawsuit over the Internet Archive’s online library. National Public Radio. https://www.npr.org/2023/03/26/1166101459/internet-archive-lawsuit-books-library-publishers

Hess, C., & Ostrom, E. (2007). Understanding knowledge as a commons: From theory to practice. MIT Press.

Hutson, M. (2018). Artificial intelligence faces reproducibility crisis. Science, 359(6377), 725–726. https://doi.org/10.1126/science.359.6377.725

Koesten, L., Gregory, K. M., Groth, P., & Simperl, E. (2021). Talking datasets – Understanding data sensemaking behaviours. International Journal of Human-Computer Studies, 146, Article 102562. https://doi.org/10.1016/j.ijhcs.2020.102562

Koesten, L., Vougiouklis, P., Simperl, E., & Groth, P. (2020). Dataset reuse: Toward translating principles to practice. Patterns, 1(8), Article 100136. https://doi.org/10.1016/j.patter.2020.100136

Kuta, S. (2023, December 15). “Hallucinate” is Dictionary.com’s Word of the Year for 2023. Smithsonian Magazine. https://www.smithsonianmag.com/smart-news/hallucinate-is-dictionarycoms-word-of-the-year-for-2023-180983443/

Lane, J. I. (2020). Democratizing our data: A manifesto. MIT Press.

Lane, J. I. (2019). The Coleridge Initiative. https://coleridgeinitiative.org/richcontext

Lane, J. I., Mulvany, I., & Nathan, P. (Eds.). (2020). Rich search and discovery for research datasets: Building the next generation of scholarly infrastructure. Sage.

Lane, J. I., Spector, A., & Stebbins, M. (2024). An invisible hand for creating public value from data. Harvard Data Science Review, (Special Issue 4). https://doi.org/10.1162/99608f92.03719804

Loukides, M. K. (2023). What are ChatGPT and its friends? Opportunities, costs, and risks for large language models. O’Reilly Media.

Meadows, A. J. (2001). Understanding information. K. G. Saur.

Meyns, C. (2019). ‘Data’ in the Royal Society’s Philosophical Transactions, 1665–1886. Notes and Records, 74(3), 507–528. https://doi.org/10.1098/rsnr.2019.0024

Milano, S., McGrane, J. A., & Leonelli, S. (2023). Large language models challenge the future of higher education. Nature Machine Intelligence, 5, 333–334. https://doi.org/10.1038/s42256-023-00644-2

National Institutes of Health. (2020). Final NIH policy for data management and sharing (NOT-OD-21-013:). Office of the Director. https://grants.nih.gov/grants/guide/notice-files/NOT-OD-21-013.html

Nelson, A. (2022). Ensuring free, immediate, and equitable access to federally funded research [Memo]. White House Office of Science and Technology Policy, Executive Office of the President of the United States. https://www.whitehouse.gov/wp-content/uploads/2022/08/08-2022-OSTP-Public-access-Memo.pdf

Norlen, N., & Barrett, G. (2023, December 12). Dictionary.com’s 2023 Word of the Year is…. https://content.dictionary.com/word-of-the-year-2023/

Organisation for Economic Co-operation and Development. (2023). Artificial intelligence in science: Challenges, opportunities and the future of research. OECD Publishing. https://www.oecd-ilibrary.org/science-and-technology/artificial-intelligence-in-science_a8d820bd-en

Ostrom, E. (1990). Governing the commons: The evolution of institutions for collective action. Cambridge University Press.

Parsons, M. A., Duerr, R. E., & Jones, M. B. (2019). The history and future of data citation in practice. Data Science Journal, 18(1), 52. https://doi.org/10.5334/dsj-2019-052

Pasquetto, I. V., Borgman, C. L., & Wofford, M. F. (2019). Uses and reuses of scientific data: The data creators’ advantage. Harvard Data Science Review, 1(2). https://doi.org/10.1162/99608f92.fc14bf2d

Pasquetto, I. V., Randles, B. M., & Borgman, C. L. (2017). On the reuse of scientific data. Data Science Journal, 16, 8. https://doi.org/10.5334/dsj-2017-008

Pepe, A., Goodman, A., Muench, A., Crosas, M., & Erdmann, C. (2014). How do astronomers share data? Reliability and Persistence of datasets linked in AAS publications and a qualitative study of data practices among US astronomers. PLOS ONE, 9(8), Article e104798. https://doi.org/10.1371/journal.pone.0104798

Pierce, A. E. (1989). On the emergence of syntax: A crosslinguistic study [Doctoral thesis, Massachusetts Institute of Technology]. https://dspace.mit.edu/handle/1721.1/13993

Pychlau, S., & Wagner, D. T. (2023). The data of others: New and old faces of archival research. In H. Cooper, M. N. Coutanche, L. M. McMullen, A. T. Panter, D. Rindskopf, & K. J. Sher (Eds.), APA handbook of research methods in psychology: Data analysis and research publication (2nd ed., pp. 481–500). American Psychological Association. https://doi.org/10.1037/0000320-022

Roberts, J. J. (2020, November 28). As libraries fight for access to e-books, a new copyright champion emerges. Fortune. https://fortune.com/2020/11/28/digital-publishing-copyright-champion-lila-bailey-internet-archive/

Rosenberg, D. (2013). Data before the fact. In L. Gitelman (Ed.), “Raw data” is an oxymoron (pp. 15–40). MIT Press.

Rotenberg, M. (2022). Artificial intelligence and the right to algorithmic transparency. In E. Stefanini, L. Liguori, M. Ienca, O. Pollicino, & R. Andorno (Eds.), The Cambridge handbook of information technology, life sciences and human rights (pp. 153–165). Cambridge University Press. https://doi.org/10.1017/9781108775038.015

Scroggins, M. J., & Boscoe, B. M. (2020). Once FITS, always FITS? Astronomical infrastructure in transition. IEEE Annals of the History of Computing, 42(2), 42–54. https://doi.org/10.1109/MAHC.2020.2986745

Strauss, A., & Corbin, J. M. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory. SAGE Publications.

Streitfeld, D. (2023, August 13). The dream was universal access to knowledge. The result was a fiasco. The New York Times. https://www.nytimes.com/2023/08/13/business/media/internet-archive-emergency-lending-library.html

The Research Parasite Awards: Celebrating rigorous secondary data analysis. (2021). http://researchparasite.com/

The Text Encoding Initiative. (2024). [Home page.] http://www.tei-c.org/

Uhlir, P. F. (Ed.). (2012). For attribution — Developing data attribution and citation practices and standards: Summary of an international workshop. The National Academies Press.

Wilkinson, M. D., Dumontier, M., Aalbersberg, Ij. J., Appleton, G., Axton, M., Baak, A., Blomberg, N., Boiten, J.-W., da Silva Santos, L. B., Bourne, P. E., Bouwman, J., Brookes, A. J., Clark, T., Crosas, M., Dillo, I., Dumon, O., Edmunds, S., Evelo, C. T., Finkers, R., … Mons, B. (2016). The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data, 3, Article 160018. http://dx.doi.org/10.1038/sdata.2016.18

Williamson, H. F., Brettschneider, J., Caccamo, M., Davey, R. P., Goble, C. A., Kersey, P. J., May, S., Morris, R. J., Ostler, R., Pridmore, T., Rawlings, C., Studholme, D., Tsaftaris, S. A., & Leonelli, S. (2021). Data management challenges for artificial intelligence in plant and agricultural research. F1000Research, 10, 324. https://doi.org/10.12688/f1000research.52204.1

Wofford, M. (2022). Parasitic knowledge infrastructures: Data reuse by anthropogenic climate change skeptics. Proceedings of the Association for Information Science and Technology, 59(1), 837–839. https://doi.org/10.1002/pra2.743

Zhang, P., Gregory, K., Yoon, A., & Palmer, C. (2023). Conceptualizing data behavior: Bridging data-centric and user-centric approaches. Proceedings of the Association for Information Science and Technology, 60(1), 856–860. https://doi.org/10.1002/pra2.878

©2024 Christine L. Borgman and Amy Brand. This article is licensed under a Creative Commons Attribution (CC BY 4.0) International license, except where otherwise indicated with respect to particular material included in the article.