Proceedings of a Workshop—in Brief

Stakeholder Actions to Implement Open Scholarship

Proceedings of a Workshop—in Brief

Launched in early 2019, the Roundtable on Aligning Incentives for Open Scholarship at the National Academies of Sciences, Engineering, and Medicine (the National Academies) convenes critical stakeholders to discuss the effectiveness of current incentives for adopting open scholarship practices, current barriers of all types, and ways to move forward in order to align reward structures and institutional values in advancing open scholarship. In its first phase of activity between 2019 and May 2022, the Roundtable held six meetings and three public workshops to foster the exchange of ideas and joint strategic planning among key stakeholders.¹

As the first public activity in its second phase, the Roundtable held a workshop on December 5, 2022. Members of three groups formed in conjunction with the Roundtable—comprised of higher education institutions, professional societies, and funders to engage the broader communities—shared updates on their activities and plans. Representatives from federal agencies also discussed their plans and priorities for advancing open scholarship, both ongoing and in accordance with guidance on Ensuring Free, Immediate, and Equitable Access to Federally Funded Research issued in August 2022 by the Office of Science and Technology Policy.²

In opening the workshop, Roundtable co-chair Keith Yamamoto (University of California, San Francisco) emphasized the convening power of the Roundtable to catalyze, harmonize, and scale open scholarship across stakeholder groups. He lauded the creation of the three spin-off groups, noting that hearing from them and from the “critical vector” of federal agencies provides a way to understand the priorities of each group, make open scholarship more sustainable and rewarding, and address roadblocks. Greg Tananbaum (Open Research Funders Group), who serves as the coordinator of the Roundtable, moderated the four workshop sessions.

HIGHER EDUCATION PERSPECTIVES/HELIOS UPDATE

The Higher Education Leadership Initiative for Open Scholarship (HELIOS) is a cohort of institutions that want to take collective action to advance open scholarship,

__________________

explained Geeta Swamy (Duke University), the HELIOS strategic lead. At a high level, engagement in HELIOS has three core components: presidential commitment, campus engagement, and participation in a community of practice. Three presidents (Ronald Daniels, Johns Hopkins University; Roslyn Artis, Benedict College; and Michael Crow, Arizona State University) sent a Dear Colleague letter to invite other leaders to join with them to support open scholarship. At the time of the workshop, HELIOS had over 80 member institutions, both public and private. Four working groups have been formed: (1) Good Practices in Open Scholarship; (2) Institutional and Departmental Policy Language; (3) Shared Open Scholarship Infrastructure; and (4) Cross-Sector Alignment.

“Each of these groups is addressing complementary aspects of the open scholarship landscape,” Swamy said. “Collectively, we’re working to figure out how to do this across our institutions to make it easy for individual researchers, as well as how institutions can support them.” Issues to consider include how to align incentive structures; reward open activities, ensure efforts are durable and sustainable; and coordinate with agencies, funders, professional societies, and other stakeholders. “Our overarching goal is to ensure as a higher education community that we are moving in the same direction as these sectors to promote a more transparent, inclusive, and trustworthy research ecosystem,” she said. Co-leads for each work group made brief presentations, followed by discussion.

Good Practices in Open Scholarship

Athena Jackson (University of Houston) thanked the group’s co-leads and members and explained they are charged with developing a vision to make open scholarship easier for researchers at any stage in their research and in any context. “Good practices [related to open scholarship] are already happening, and we realized this is the time to explore how sustaining and scaling open practices look,” she said. “It has to be easier for individuals to know what they have to do and how to do it.” Guidance and infrastructure are needed to support a transition, she pointed out. There is also a question of equity because most researchers do not have the resources or support to prepare data and undertake other processes involved in open scholarship. “If we want everyone to do open scholarship, we must lower the barriers to these activities as much as possible,” she said.

Recognizing the wealth of resources becoming available, Jackson reported the group is discussing a “one-stop-shop support service” to link researchers to local resources and peer networks and to provide a single point of entry for support when researchers have questions. The aim is to move from “philosophical alignment to the challenging practicalities” and to address current and long-term needs, she explained. To illustrate the envisioned support service, Jackson shared a case study entitled Open Scholarship Support Service for Geophysics. In the use case, a faculty member who is unfamiliar with public access requirements connected with the National Aeronautics and Space Administration (NASA) funding goes to the service to build competencies, connect with experts, and create communities. She added the service could support other disciplines, and it especially help early-career researchers and all colleges and universities. Jackson concluded with four “big questions” for the group to consider: (1) How do we build and sustain a community for this? (2) How do we fund this work? (3) Is this a valuable service to your discipline, organization, or institution? (4) Does this service fill a particular need for your organization?

Institutional and Department Policy

LeKeisha Harris (University of Maryland Eastern Shore) explained the working group’s purpose is to look how to incorporate public access in hiring, reappointment, promotion, and tenure policies and guidelines. She noted some agreements are already in place, such as the Declaration on Research Assessment (DORA), Coalition for Advancing Research Assessment (CoARA), and the Leiden Manifesto for Research Metrics, among others. The working group is set up to interact with institutions of all types and sizes, Harris stressed.

From the perspective of a smaller institution, Harris said open scholarship presents opportunities for regular review of the promotion and tenure process, increased access to research, a larger audience for its research, and

increased opportunities for collaboration and engagement across disciplines. For HELIOS to be representative, there must be inclusive on-ramps and approaches to promotion and tenure, she said.

The Theory of Change developed to support open scholarship involves mutually reinforcing vectors including university administration, hiring committees, department chairs, professional societies, government agencies, and philanthropies to change institutional culture, Harris stated. The working group’s charter is to develop a collective action plan for embedding open scholarship considerations within reappointment, promotion, and tenure (RPT) guidelines, recognizing institution and disciplinary differences. It is not an easy task, she acknowledged. An RPT Joint Statement was sent to HELIOS members, but few signed on to it. After a survey to understand why, the statement was reworked to commit not to specific changes but to begin to dialogue about them, and this has gained more traction. The next step is a census of RPT activities to surface what institutions are now doing and continuing discussion if they do not want to sign the statement, along with ongoing collaboration with other sectors.

Shared Open Scholarship Infrastructure

Alicia Salaz (University of Oregon) provided an update on this working group’s efforts on shared infrastructure, which she defined as “owned and/or managed by a collective of institutions or organizations with similar goals.” The group’s charter calls for a focus on the critical enabling mechanics need to openly share data, code, and other outputs and identification of opportunities to leverage existing and/or develop new shared resourcing to accelerate the adoption of open scholarship practices. As Salaz said, “Neither requirements nor incentives alone can make open scholarship happen without infrastructure to implement it.”

The group has supported the recognition by federal agencies and OSTP about the importance of infrastructure, such as through persistent identifiers and licenses to support sharing and reuse. A robust infrastructure can accelerate research across disciplines, with less time spent on the mechanics, make research findable by machines, and leverage computing power to foster innovation, she noted. Spreading costs enables a more inclusive environment, she added.

Institutions are struggling, for example related to data storage, Salaz said. Questions revolve around who will build and maintain the needed data storage infrastructure for access and reuse across the broader scientific community, as well as educate about how to use it. The group sees productive opportunities for partnerships with the private sector, although has expressed concerns about closed proprietary siloed systems and long-term reliance on the commercial market to serve the long-term public interest. One approach may be for institutions to share infrastructure as an alternative to building everything in-house or buying commercial solutions. The right strategy may be different for different components of open scholarship, but academy-owned solutions are needed, she asserted.

A key hurdle identified is that institutions have a number of key leaders with different priorities but no common framework for alignment. They struggle to understand the landscape, and they are looking for guidance on how to evaluate the best options. Thus, the working group is producing a guide, to be completed in the first quarter of 2023, to help these leaders evaluate shared infrastructure options. “We are convinced that the systems we build in the coming years will shape the speed and nature of knowledge creation nationally for years to come,” Salaz concluded. “Will we lead the world or fall behind regions like Europe and Asia? The strategies our institutions pursue now do matter.”

Cross-Sector Alignment

To close out the set of HELIOS presentations, Kimberly Cox-York (Colorado State University) explained the cross-sector working group was set up to surface the status of the other working group and take the lead in engagement with other sectors. The group’s vision is to facilitate the meaningful exchange of ideas, knowledge, and action among the HELIOS community and its partners in open scholarship.

To this end, Cox-York shared four “offers and asks” that have surfaced in HELIOS. Related to incentives, she

said, HELIOS can offer practical ways to align higher education reward structures, while it asks other sectors to consistently include open practices in reward schemes and to articulate good practice guidance that higher education institutions can adapt or adopt. Related to infrastructure, HELIOS can offer the rubric mentioned by Salaz and asks for funding for shared infrastructure and recommendations for sustainable and accessible infrastructure for sharing outputs. Related to making open scholarship easier, HELIOS can offer its help to other sectors to develop a vision and timeline for a support service, which asking for the alignment of policies and support to make compliance easier. Aligning requirements is not easy, she said, but some things are in place that can be more widely implemented, such as requiring ORCID identifiers in all proposals and embedding persistent identifiers and metadata across the research lifecycle. She urged the encouragement of sharing of data, code, and other outputs; assessment of costs for data sharing and management; and compliance guidelines that factor in the real costs of robust data sharing.

Cox-York noted HELIOS can also provide other sectors with collective feedback from its members about different ideas or programs, as well as collaboration to share and co-develop resources. She invited other sectors to partner with HELIOS in policy development and priority setting. “The work done thus far is encouraging, and key sectors have come to the table,” she said. “With continued collaboration, it is possible to create a more efficient, sustainable and equitable research ecosystem.”

Discussion

In answer to a question about options for shared infrastructure, Salaz responded her working group has discussed several models, such as co-development by institutions, pooling resources into a third-party nonprofit, and publicly owned infrastructure. She said a variety of models can provide leaders with guidance for different use cases. A participant praised the remarkable progress to date but commented that open access is only one part of what is needed for diversity, equity, and inclusion (DEI) on campuses. Discussing where and for how long to store data, version control, and other issues related to where “data and publications come together,” Jackson said addressing these issues amplifies the work of archivists and special collections librarians whose expertise is thinking about enduring principles across the humanities and sciences.

Discussing whether open scholarship has become a compliance-driven in some institutions, Salaz posited there is value both to the carrot and the stick. Some communities are committed to the potential of shared infrastructure for the advancement of science, while compliance has also driven some activity. Cox-York commented that her institution has tried to surface the reasons why open scholarship is important and articulate its benefits and minimize a stress on mandates. Suggesting that open scholarship is an enabling strategy for more equitable structures and systems, another participant commented that repositories can operate at the nexus of articles and data, and they are key pieces of infrastructure. She pointed to the U.S Repository Network as a resource.

The trend in support of open scholarship was raised. Jackson noted that newer scholars are often more collaborative and open in their approach. A funding participant commented that openness is a necessary step toward collaboration, but not sufficient in itself. She called for the setting of standards, for example on the extent of data that should be shared for publication. Compliance monitoring will not be sufficient without this understanding, she said. Another commented that early-career researchers (ECRs) receive conflicting advice about how to present their open scholarship for hiring, promotion, and tenure decisions. Some ECRs say their open access work is not being recognized, or even asked about, in tenure and promotion decisions. New metrics are needed through a partnership between higher education and funders, she suggested. Swamy suggested aligning incentives in all aspects of research, which is hard if sponsors expect the status quo. Related to repositories, she urged thinking about how data will be used and shared across multiple repositories.

SOCIETY AND ASSOCIATION PERSPECTIVES/ALL4OS UPDATE

Another group that has spun off from the Roundtable is the Alliance 4 Open Scholarship (ALL4OS), which

brings together professional societies. In introducing three society representatives involved in the alliance, Tananbaum noted that professional societies can support open scholarship by promoting discipline-specific norms, standing up repositories and other infrastructure, and training and supporting researchers in good open scholarship practice.

Origins and Impacts of the Alliance 4 Open Science

Brooks Hanson (American Geophysical Union [AGU]) began by noting that the professional societies’ work aligns well with the points raised in the HELIOS presentations. He stressed the larger need for open scholarship: solving the key global challenges of the 21st century requires convergence science built on massive amounts of data and integration with modeling, in connection with society. Open scholarship with integrity is a must to enable convergence science, he said.

Although much of the structure of science is not aligned with convergence science, disciplinary societies can work together to support open scholarship and culture through their efforts to set standards, publish, connect, award and recognize, and other means, Hanson continued. Societies focus on integrity and quality, as well as on equity and inclusion, so some of the groundwork has been laid. For example, many societies are involved in coalitions on open scholarship, transition of their journals to open access, establishment and expansion of preprints, and promotion of open science to support DEI and expand participation.

However, Hanson said, these efforts have been largely disconnected. For this reason, All4OS started in 2021 to amplify and share efforts, learn from each other, and catalyze more activity. A joint statement released in 2022 expresses the societies’ commitment to support and sustain infrastructure, including discipline-specific guidance where appropriate; provide training and resources; devise recommendations related to incentives; and promote the open scholarship ecosystem. Each of the ten societies (more are welcome to join) has a high-level representative to follow through with actions and has agreed to meet and reach out to other groups. Hanson echoed the challenges identified in the previous session, particularly to ensure quality and integrity and to support existing and future infrastructure.

Reporting on his discipline, Hanson said that earth science journals have agreed to stop allowing data in supplements (which are not findable), but instead require authors to place data in repositories. A number of general and domain-specific repositories are used, most frequently Zenodo, but Hanson observed these repositories vary in their quality control and curation. A Data Help Desk can now guide earth science researchers in their choice of a data repository, and this idea could be applied more broadly. Another effort is to support computational notebooks for end-to-end scholarly publishing. Related to awards and recognition, Hanson reported that AGU has inserted a criterion related to open science in the nomination process for AGU Fellows. This kind of culture change is hard to do, he acknowledged, and he urged NASEM to adopt similar language.

All4OS Member Survey

Yvette Seger (Federation of American Societies for Experimental Biology [FASEB]) welcomed continued engagement with HELIOS and other stakeholders as All4OS moves forward and grows. She reported on a survey of the 10 All4OS members, beginning with three high-level observations. First, all are pursuing a robust range of open scholarship activities, including developing ontologies, making content available and reusable, training, developing business models, publishing guidance, and training data for ML/AI. Second, members are engaged in discipline-specific initiatives focused on data sharing, with an emphasis on training and good practices. FASEB, for example, has launched an initiative called DataWorks! that is focused on data sharing and reuse in the biological and biomedical sciences. Third, there is an opportunity to collectively explore open scholarship incentives. As examples, she noted AGU’s inclusion of open science language in its award structure (see above) and FASEB’s efforts to promote credit for data sharing. She noted interest in exploring open scholarship-specific awards, prizes, challenges, and micro-grants.

Alliance members offer a range of training opportunities and resources to their members, Seger continued,

including help desks, train-the-trainer sessions, and webinars. Societies are mission-driven, she said, and there is the opportunity to link open scholarship with other societal impacts, such as DEI and accessibility. Common reservations expressed in the survey relate to resourcing; coordination within organizations, across disciplines, and across professional societies; and the velocity of change. While compliance requirements from federal agencies serve as a driver, it is also important not just to react to mandates, the survey respondents said. Seger looked to HELIOS’s working groups as a possible model for All4OS. She also said the next step is to figure out what already exists and where the gaps are to move forward.

Cross-Sector Collaboration Possibilities

Edward Liebow (American Anthropological Association [AAA]) briefly described the history of AAA and noted the field’s sprawling intellectual terrain across archaeological, linguistic, biological, cultural, and other domains. “We believe our future is open, and it is baked into our strategic plan, but the pathway will be filled with experimentation, monitoring and evaluation, and alliances,” he said. Societies should define their “lane” in advancing open scholarship, realizing their role is different from that of a university, funder, or agency. He noted societies’ “superpowers” to convene, set standards, and provide voice and illumination. These powers can benefit their members and the broader cause of open scholarship.

Liebow offered three areas of cross-sector possibilities: incentives, training, and infrastructure. Related to incentives for change, Liebow observed that department chairs are often the agents of change within the field, but not necessarily at their home institutions. When changes are proposed, he commented, department chairs suggest talking to their provost or dean, while the provost or dean says it is within the purview of department. There is work to do in aligning institutions, and societies can be an ally, he opined. He added that when AAA brings together department chairs from a range of institutions, it is clear that one size does not fit all, for example related to promotion and tenure guidelines.

A second area for cross-sector collaboration is training. The training needs to be discipline-specific, given different data formats and ethical considerations, but the model can be replicable and scalable. The third area for cross-sectoral collaboration is infrastructure. Liebow described several efforts in anthropology, including an Open Anthropology Research Repository and a Book Forum. As noted by Hanson, general repositories lack the curation and content area expertise that domain-specific ones do, so AAA is continuing to develop this area.

Discussion

Tananbaum asked for perspectives on the view he hears from some societies that view open scholarship may threaten their business models. Hanson reflected that the AGU membership and leadership have become more supportive over the past few years, and most recognize open scholarship as part of the future. Liebow concurred but stressed the need for sustainable resources in the publishing enterprise and for well-stewarded repositories. Seger added that individual scientists, especially early-career scientists, want and expect open scholarship in their societies.

Several participants pointed out that open data can create opportunities, such as NOAA’s data used by commercial companies and activities conducted by community scientists. Including a data expert in peer review, in addition to disciplinary experts, would send a powerful message to authors, a participant suggested. Asked about training methods, Liebow said stand-alone courses at conferences, online modules, and curricula for graduate and undergraduate students are all needed. Related to repositories, a participant agreed that domain repositories can be most useful, but posited that they can be built on top of general repositories as a more economic option.

In terms of culture change, a participant observed that department chairs may not have a big role in institutional policy making, but they have significant leeway on how faculty are evaluated, and they are often risk averse. He asked how societies can help change the cultures in their disciplines. Liebow said that AAA created a roster of outside reviewers who visit departments

when they are up for their reviews. It is a way to hear what other departments are doing on a host of issues, including changes in promotion and tenure guidelines. Hanson urged researchers to take the lead on and promulgate strong science practices, which can feed into changes in culture. Seger suggested societies can work with agencies and other partners to shape policies and processes.

FEDERAL PERSPECTIVES

Tananbaum introduced the next session in what he termed “the sector-by-sector journey through the world of open scholarship” with representatives from four federal funding agencies.

National Institutes of Health (NIH)

Jerry Sheehan (National Library of Medicine [NLM]) spoke on priorities and next steps at NIH related to open science. As background, he noted that NIH support for open science began in 2003 with a policy for data sharing when and where possible. Other policies over the past 20 years include those covering Public Access (first released in 2008), Genomic Data Sharing (2013), and Clinical Trial Registration and Reporting (2017). Many of the NIH’s next steps will be inspired by the OSTP memo of August 2022, which Sheehan said is highly supported by Acting NIH Director Lawrence Tabak and by NLM, which has a role in providing supportive infrastructure. “Our motivations are very much around improving transparency, accessibility, and supporting and advancing science in ways that are more open and accessible,” he said. He noted these goals benefit not only other scientists, but also entrepreneurs, health care provides, students, educators, the public, and others.

As a top priority, the new NIH Policy on Data Management and Sharing is scheduled for implementation for intramural and extramural research in January 2023.³ Guidance has been developed to address ethical, legal, and practical considerations, as well as FAQs and other educational efforts. NIH has been engaging across scientific communities, and NLM, in particular, is working with the library and information management communities so they can help researchers on their campuses. The NLM-funded National Center for Data Services provides additional support and training for the library community.

Sheehan highlighted several other activities related to open science. NIH is developing its plan to respond to the memo issued by OSTP in August 2022, following the deadlines specified in the memo (see footnote 1). He noted that the plan would be followed by policy development and implementation and include opportunities for public comment. For scholarly publications, NIH will consider how to adapt its Public Access Policy, which has worked well to date. Since the policy went into effect in 2008, 1.4 million NIH-supported articles have been placed in PubMed Central. About one-third have been submitted with no requested embargo period, many published in open access journals, and the goal is to increase that amount. In addition, NIH launched a Preprint Pilot as part of its COVID-19 response. About 3,600 preprints on NIH-supported research have been made available through Pubmed Central, with citation information in PubMed, and are highly used. Finally, NIH is making progress on implementation of a Data Science Strategic Plan, which is coordinated through its Office of Data Science Strategy that works across all Institutes and Centers.

National Aeronautics and Space Administration (NASA)

As Chelle Gentemann reported, NASA designated 2023 as the Year of Open Science to highlight the importance of open science for the agency. The Transform to Open Science (TOPS) Mission is a new initiative that allows NASA to lean into open science as a way to address big challenges and discover the next big breakthroughs, she said. An inclusive scientific ecosystem that breaks down silos and barriers, as well as greater participation by people with diverse experiences, are needed to meet these challenges. “What can we do to prepare for 21st century science? Open science sets the path,” said Gentemann (Figure 1). NASA is working on incentives and support for changing culture, which she noted others have brought up throughout the workshop.

__________________

NASA’s Open-Source Science Initiative is funded at $20 million annually in four areas: policy, infrastructure, funding elements and outreach. A new policy for grantees entitled SPD-41A brings together existing NASA and federal guidance on open data, software, and publications. The agency is supporting development of core services related to data cataloging, publications search, and science platforms. Elements that can receive NASA funding include: (1) workshops, symposia, and conferences; (2) tools, frameworks, and libraries; (3) software; (4) training; (5) high-priority science; and (6) software platforms. TOPS NASA Outreach provides $40 million over five years with the strategic goals to support 20,000 researchers receiving open science certification through NASA, to double participation from historically excluded groups, and to enable five major discoveries through open science.

Priorities for the Year of Open Science include the release of an open science curriculum, engagement with historically underrepresented groups, and development of more open science incentives. She invited others to join in marking 2023 as the “Year of Open Science for Everyone.” This might involve developing a strategic plan for open science; improving the transparency, integrity, and equity of reviews; accounting for open science activities in evaluations of all types; and engaging with underrepresented communities in the advancement of open science.

National Endowment for the Humanities (NEH)

The National Endowment for the Humanities funds research, education, preservation, and public programming in the humanities, explained its CIO, Brett Bobley. The research component is covered by the August 2022 OSTP “Nelson memo”; because the agency funds research at less than $100 million annually, it was not bound by the OSTP Holdren memo of 2013 on public access. That was a lost opportunity, Bobley said, because NEH is “playing catch up” in this area. He noted that one advantage of such federal memos is that they bring researchers and agencies together. He and the NEH director of research programs are now on the OSTP Subcommittee on Open Science.

Related to how it addresses open scholarship requirements, NEH is determining how many journal articles it funds each year, which journals publish NEH-funded research, which repositories could be used, and what a dataset means in the context of various humanities disciplines. The relatively small number of journal articles (about 150–200 annually) is because of limited funding and because humanities scholars tend to publish through other vehicles, such as monographs. NEH will engage with scholarly societies and other funders to get feedback on these issues in putting together its policy. He noted that OSTP has given new agencies a deadline of August 20, 2023, to submit a draft plan. He said he is looking for concordance with the rules of other agencies, such as the National Science Foundation, to help universities and individual grantees.

National Science Foundation (NSF)

Alan Tomkins (NSF) concurred with the value of the Nelson memo and the interagency coordination that ensues. He asserted there is not a one-size-fits-all solution. In explaining the NSF’s Public Access Policy, he clarified it is not a mandate for open access via article processing charges (APCs), or any other publication

model. He noted the Nelson memo of 2022 continues the approach reflected in the Holdren memo of 2013 of allowing the Author Accepted Manuscript (AAM) to comply with the requirement to make publications available, rather than a journal’s Version of Record (VoR). Nor does the Nelson memo require sharing all data—for example, protected information or data is an example of data that does not have to be shared. That said, as the Nelson memo points out, lowering barriers of access to science accelerates discovery and amplifies its impact; promotes scientific integrity; and ensures equity.

In terms of current practice, NSF requires publication access via either an author-accepted manuscript or the publication’s version of record. It encourages, but does not mandate, data sharing. To comply with the new public access requirement, which Tomkins referred to as “NSF Public Access 2.0 Planning,” NSF is using the opportunity to promote equity in a variety of ways, increase outreach and engagement, and move from data-management to data-sharing plans. NSF will fund research associated with public access and open science. Last year, for example, NSF sponsored the FAIROS-RCN competition⁴ to support research coordination networks consistent with the FAIR principles. It awarded $38 million to the University of Michigan⁵ to expand open access infrastructure for the social, behavioral, and economic sciences. It is important to understand the science and gather evidence about open science and public access, he said. He welcomed resources from other agencies to join NSF in supporting this type of research.

At a recent webinar and through other means, NSF is gathering stakeholder input, Catherine Patterson (AAAS Science and Technology Fellow serving at NSF) reported.⁶ She noted broad agreement in support of public access and the belief that the trend toward more open scholarship is likely. All agree that publishing is critical to the research ecosystem and that researchers should have equitable access to publish their work. Most are looking for ways to maximize the potential of cross-disciplinary shared data, but in a way that avoids duplicative efforts. Communities have told NSF that they want to help develop new mechanisms and ecosystems for sharing research, Patterson related.

Equity is a driving factor, she continued. Concerns include that overreliance on “Gold Open Access” may disadvantage many researchers and marginalized groups who do not have the resources to pay the associated costs. Concerns about the technical, cultural, and economic challenges of sharing data were also shared. Stakeholders also pointed out that publishing and data norms vary widely across disciplines. Patterson concluded that NSF is interested in understanding the interests, concerns, anticipated impacts, and resource needs of their research communities around open access, and how different stakeholders would define success.

Discussion

When asked how other sectors can work with federal agencies to implement the Nelson memo, Tomkins stressed the need for funding agencies to hear ideas and obtain buy-in from grantees. Sheehan invited non-federal participants to discuss what they need and share their experiences with open science policies to date. He said it is important to engage broadly across agencies and also with particular scientific communities. Bobley said data sharing for many humanities scholars is common. About 20 years ago, NEH put in their grant guidelines that all things being equal, preference is given to projects that make their results open to the public, although some data, such as related to indigenous culture, has special considerations. NEH wants to centralize its messaging on open scholarship, aligning it with others but in a humanities context. Sheehan commented that NIH is trying to move in the direction of both managing and sharing data. Gentemann said NASA’s 2023 funding announcement will require proposals to include an open science development plan to cover all materials created during the NASA-funded research.

When asked about the feasibility of openly publishing grant proposals, Tomkins said NSF can share awarded grants through a FOIA request, but NSF policy prohibits sharing of proposals on its own initiative. It does share

__________________

award titles and abstracts on its website.⁷ NIH makes information about awards available, taken from the proposal, but not the full proposals, Sheehan said. Under discussion is whether to share data management plans associated with awards, he added. Bobley noted that at a recent conference, the topic of sharing of proposals was discussed and whether seeing full proposals, with confidential information redacted, would benefit the field. Gentemann pointed out that applicants can choose to share their grants on a site called Open Grants (ogrant.gov). She stressed this is especially important for early career scientists who do not have access to more senior successful scientists who pass on their knowledge.

Funding open access requirements, such as related to data storage, was discussed. The fear, said one participant, is that it becomes an unfunded mandate. Tomkins said NSF allows grant funding to be used, but he acknowledged that it means less funding for other areas. Overall, he posited that the economic impact is relatively small. Another participant urged a cost-benefit approach. “The cost of open science compared to the total investment for research is small,” he said. “As a benefit, it provides tremendous leverage related to cooperation, public perception, and sharing.” He also predicted that new models will evolve and reach an equilibrium in the next five years. He suggested this as a potentially useful study for the National Academies.

FUNDER PERSPECTIVES

In the final panel of the workshop, the progress of open scholarship grantmaking at scale was discussed through presentations from representatives of the Open Research Funders Group, Templeton Foundation, and Willian and Flora Hewlett Foundation.

Open Research Funders Group

The Philanthropic Leaders Cohort comprises more than 60 funders and kicked off in June 2022, explained Erin McKiernan. The members, including the John Templeton Foundation and William and Flora Hewlett Foundation who are in this panel, have great diversity in terms of size and funding priorities, but all share the view that open scholarship enables their missions and increases the impact of grant dollars. Two main work streams have developed: the Open Scholarship Policy Development Working Group, which helps funders develop and implement open policies, and the Cross-Sector Funding Working Group, which is formulating “a set of funder priorities, asks, and offers” to share with other sectors.

The Policy Development group is building a community of practice through which funders come together to learn about open scholarship, develop policies that meet their missions and goals, and implement strategies to take action. A structured curriculum has been piloted, involving 6 funders in the first cohort in 2021 and 19 in the second in 2022 (osf.io/v2t4y). All of the 2021 cohort have drafted and are at various stages of moving the policies through their organizations, she reported. The 2022 cohort had completed two of the units at the time of the workshop and were beginning to draft policies. “We have a formula that works,” she said. “The driving force is hopefully making policy development as easy as possible for funders.” Peer-to-peer connections have been particularly important, she added.

The Cross-Sector group’s goals to engage, build, and operationalize open scholarship intersect with many of the points made throughout the workshop, McKiernan commented. She stressed the need to align policies, language, and guidance across sectors for a better chance of success. A survey of the Cross-Sector group reinforced priorities around making open scholarship as easy as possible for individuals and organizations, figuring out mechanisms to measure impact, building infrastructure, training researchers, and exploring how to embed equity in the implementation of open scholarship (Figure 2).

__________________

Funders expressed a great deal of interest in alignment with professional societies, higher education, and federal agencies. She noted that funders are in a good position to convene, support the effort, and harmonize language and best practices. Survey respondents were interested in learning from experts and peers about messaging, infrastructure, policy development, and funding mechanisms. They want to engage in dialogue across sectors to learn and promote big ideas around open scholarship. She noted that many are interested in providing sponsorship and financial support; developing baseline and tailored policies and incentives; and supporting measurement, evaluation and learning to assess the impact of open scholarship interventions.

John Templeton Foundation

Cassaundra Amato explained that the John Templeton Foundation distributes about $150 million in grant payments each year with an endowment of about $3 billion at the end of 2022. The foundation is now exploring incorporating “Open Scholarship Plus” into its goals, mission, and operations. Based on his writings, Sir John Templeton would likely have been very interested in open science, Amato said, in particular his belief in the self-correcting nature of science, competition of ideas, and democratization of ideas. Twenty-five years ago, for example, he wrote about how the Internet will open access to information to more people.

The foundation is working through development and implementation of its open scholarship policy. Points to resolve include how to make grantees aware of the requirements, how the requirements fit in an application, how they should be reviewed, and financial issues. A working group that consists of all internal stakeholders across the foundation (e.g., legal, operations, communications) is addressing these issues, and Amato noted the vital assistance of Tananbaum and McKiernan from ORFG in this comprehensive undertaking. An “Open 101” seminar, developed by ORFG, will be offered to the entire foundation, because the policy would affect all operations.

“Open by Default” will likely be the expectation, Amato said, recognizing there will be exceptions. There has to be responsiveness to the needs and goals of particular programs with different needs, for example for grants that support philosophy compared with genetics. Program staff would need the flexibility and discretion to make decisions on exceptions. They are working out how to make this a low burden both for the foundation and for grantees. They want the policy to have durability, so it does not need to continually be changed, and to be in line with industry best practices and clarity. The foundation will support the costs so that it is a funded mandate for grant recipients. Most important, she concluded, is the need for “grace for failure” as the policy unrolls and staff, reviewers, and grantees learn what compliance entails.

William and Flora Hewlett Foundation

Angela DeBarger extended the discussion of open scholarship to teaching and education, in addition to research. In describing the foundation and the diverse program areas and sectors it supports, she underscored that openness strengthens all the foundation’s guiding principles. “We are committed to acting on evidence, and that means it’s important to have high-quality research available not only to us, but also to the communities of practice where we work,” she said. The link between academia and action is core to all of the foundation’s strategies, she added. The foundation openly licenses its own work and requires open licensing for projects it supports.

While of interest in all areas, she said the foundation is especially interested in how open scholarship connects with better teaching and learning. Quoting Kathleen Fitzpatrick, DeBarger pondered how to close the disconnect between universities and communities through collectivity and generosity and a commitment to open scholarship as a public good.⁸

Open education is about democratizing learning, DeBarger continued. “It is about making sure that education is accessible, affordable, and centered on care. Every student should have access to the knowledge and information they need to learn, and open educational

__________________

resources [OER] help make that possible,” she said. OER is gaining traction and has been shown to benefit students through higher accumulations of academic credits, reduced withdrawal rates, and reduced gaps between white and non-white students.⁹

However, open education is only effective if it is current and connected with students’ interests and experiences, she stressed. She stressed that just as open scholarship is not just about publishing open-access articles, open education is not just about access to resources. She said it is important that learners have greater agency as creators and co-constructors of knowledge.

DeBarger posed the question of how students, especially those marginalized in society, can participate in open scholarship through intentional ways. Faculty, centers for teaching and learning, and libraries on campus; professional societies; and others all have a role to play, and she is encouraged by the discussions around incentives at the workshop and in other collaborations. She pointed to a group of state and provincial system leaders called Driving OER Sustainability for Student Success (DOERS³) that represents about 770 colleges and universities that might consider open scholarship in its work on OER. The Hewlett Foundation is also partnering with the American Association of Colleges and Universities to build capacity through a year-long Institute on Open Educational Resources. The campus teams that are participating develop and enact strategic action plans, and, she suggested, this is another prime opportunity of a group that might consider how open scholarship can achieve their goals.

“My main point is, are there ways we can more effectively collaborate to develop more holistic approaches to integrate open practices and bring more coherence to the implementation of open approaches,” she asked. She concluded that the foundation’s work is connected to the broader open movement through support of infrastructure, policy and practice, and equity.

Discussion

Referring to the requirements to share all work outputs, a participant noted that some of these are outputs that institutions seek to monetize. Amato said the John Templeton Foundation is discussing this issue and they are looking for ideas. McKiernan said, wearing her researcher’s hat, that it may be easier for a researcher to think holistically about all outputs, rather than a fractured approach for different pieces. DeBarger said open education involves multiple types of partnerships, including with for-profit publishers. It is still a place of learning, she said. Tananbaum added that OFRG has put together a “clause bank” of sample language that funders use in their policies that might be useful for this issue. Aligning Science Across Parkinson’s recently published a Blueprint of its experience.¹⁰

Regarding the connection between teaching and research, McKiernan said policies related to open education are increasing. The transition from “open science” to “open scholarship” is also a more all-encompassing concept in that direction, she added. In addition to financial support, DeBarger said funders can convene and create opportunities to learn and share. She noted she engages with grantees to bring new ideas to them and make connections. Amato said beyond requirements, some fields have “long tails” before anything is published. Funders can help grantees find ways to share their findings faster. McKiernan added that funders play a public-signaling role. Institutions pay attention to what funders say they value.

CONCLUDING REMARKS

Roundtable co-chair Keith Yamamoto said he was inspired by the workshop as a way to catalyze, harmonize, and scale open scholarship. In addition to the challenge to fashion policies to take on entrenched cultures, he acknowledged, “There is a big multidimensional, multisector set of issues, and the sectors themselves are very heterogeneous.” He noted

__________________

that the groups “have taken that heterogeneity and turned it into a power by coming together, sharing ideas, finding common ground, finding where differences can be informative, and moving things forward.” He concluded by thanking outgoing co-chair Thomas Kalil (Schmidt Futures) for his leadership and noted Kalil was a driving force in framing the Roundtable and thinking methodically about the steps needed for systems change.

DISCLAIMER This Proceedings of a Workshop—in Brief was prepared by Paula Whitacre as a factual summary of what occurred at the workshop. The planning committee’s role was limited to planning the workshop. The statements made are those of the rapporteur or individual workshop participants and do not necessarily represent the views of all workshop participants; the planning committee; or the National Academies of Sciences, Engineering, and Medicine.

REVIEWERS To ensure that it meets institutional standards for quality and objectivity, this Proceedings of a Workshop—in Brief was reviewed in draft form by Kristen Ratan, Strategies for Open Science and Ekemini Riley, Aligning Science Across Parkinson’s. The review comments and draft manuscript remain confidential to protect the integrity of the process.

PLANNING COMMITTEE Geeta Swamy, Duke University, (Chair); Randolph Hall, University of Southern California; and Brooks Hanson, American Geophysical Union.

STAFF Thomas Arrison, director, Board on Research Data and Information (BRDI); Greg Tananbaum, Roundtable secretariat; George Strawn, scholar, BRDI; Ester Sztein, deputy director, BRDI (until April 2023); Emi Kameyama, program officer, BRDI; and Sara Pietrzak, senior program assistant.

SPONSORS This workshop was supported by the Robert Wood Johnson Foundation and Open Research Funders Group. Over its first 3 years, the Roundtable has also received support from the Arcadia Fund, Arnold Ventures, the Leona M. and Harry B. Helmsley Charitable Trust, the National Library of Medicine, the Open Research Funders Group, the Open Society Foundations, Schmidt Futures, and the Wellcome Trust.

For additional information regarding the workshop, visit: www.nas.edu/brdi.

SUGGESTED CITATION National Academies of Sciences, Engineering, and Medicine. 2023. Stakeholder Actions to Implement Open Scholarship: Proceedings of a Workshop—in Brief. Washington, DC: The National Academies Press. https://doi.org/10.17226/27133.

Policy and Global Affairs Copyright 2023 by the National Academy of Sciences. All rights reserved.