Dave Shumaker: Heather, congratulations on receiving the Miles Conrad Award. It’s been 4 1/2 years since our last interview, and I’m delighted to have the opportunity to speak with you again.

Heather Joseph: Thanks, Dave—has it been that long?

Shumaker: A lot has happened in that time. I’d like to start the conversation by discussing some of the themes in your lecture and then revisit a few points from our prior interview. To begin, your lecture was powerful advocacy for the role of information in achieving social justice. Could you elaborate on that linkage?

Joseph: I appreciate the opportunity to talk about this. The concept of open access to knowledge as a social justice issue is something I’ve recognized but hadn’t been able to articulate clearly until I started doing more work internationally, particularly with colleagues in Africa. The framing of access to knowledge as a permeating human right is foundational, particularly among people in the developing world. They recognize the centrality of the free flow of information in order that all individuals in a society can really achieve parity and equity. It struck me that this was missing from the way we talked about open access, even though it was part of the intention of the Budapest open access declaration [in 2002]. In the lecture, I wanted to take the opportunity to re-raise this as an important framework.

Shumaker: I noted that your lecture mentioned two aspects of this: information as a human right in itself and information as cross-cutting, enabling progress toward other goals.  

Joseph: That’s right. It’s been a privilege to take part in the United Nations work on sustainable development goals. Right before the pandemic, we met at the UN headquarters in New York to focus on the role of access to information, open access, and, in particular, open science as accelerating strategies for reaching all sustainable development goals. It was eye-opening to see how fundamental open access is to achieving these goals. The policy statement that came out of the meeting expresses the point well.

Shumaker: It’s interesting that so much of your work has been identified with open access, but your lecture wasn’t really about access. You made a point of that, and you used the phrase “open knowledge” in your title instead. Could you expand on the reasons for that wording?

Joseph: That’s an important question. At SPARC, we’ve been talking a lot about our vision of what “openness” is really supposed to achieve. We’ve expressed it as a vision of a world where everyone can contribute to sharing knowledge and benefit from accessing and using it. We realized that a lot of our advocacy has focused on the “access” component, but a knowledge-sharing system has to be multidirectional. It’s not acceptable that some participants are restricted to access. That’s not enough; it’s not robust, and it’s not equitable. We have to look at all the components of the knowledge-sharing system. A social justice framework is really helpful. In the various definitions of social justice, I see four underlying principles: access, participation, equity, and rights. These principles are inextricable from any social justice movement, and they help us understand the entirety of the system we’re trying to change. So, it’s not just that journals cost too much and we need a cheaper way (or a free way) to get access to them. It’s a much deeper issue than that.

Shumaker: Could you say more about what you mean by rights, in this context? Are you referring to intellectual property rights?

Joseph: Intellectual property rights are certainly part of it, and we often use the phrase “set the default to Open” for access. But when we think about the default mode for the other elements of the system, we don’t approach knowledge-sharing with the assumption that knowledge is ours unless or until we choose to sign over our ownership. Much of the current knowledge-sharing system, such as scholarly journals and textbooks, is predicated on authors signing over their rights to a third party. That creates lots of imbalances, and it has allowed the commoditization of knowledge. In the academic community, we formerly had more of a “circle of gifts” economy: I share my contribution, knowing that you’ll share yours, and we build on each other’s. That’s the foundation that I’d like to see returned to the system.

Shumaker: Has the uptake in open access in effect traded the “pay to read” business model for a “pay to publish” model? Could you expand on the economic models and economic barriers?

Joseph: The business model issue is really critical for achieving equity in the system. Commercial publishers in particular have latched onto the model of article-processing charges as a replacement for subscriptions. Unfortunately, some publishers, and some funders, have seen this as a zero-sum game, where the same amount of revenue that formerly came from subscriptions now comes from publication charges. The appeal of that model is that it seems simple, but what it does is replace a barrier to access with a barrier to participation. Article-processing fees often present insurmountable barriers to authors in developing countries as well as institutions in the U.S. that simply don’t have the funding to pay.

Issues of equity in business models are critically important. Article-processing charges are in the spotlight, and we want not just to reject models that entrench current power players and create additional inequities, but promote models that take that burden out of the system. The role of research funders is key. They increasingly recognize that the cost of research communication is part of the cost of doing business, because if the research they fund is never communicated, then its value is zero. But we need publisher fees that are tied to services, and not a simple, black-box, zero-sum replacement of subscriptions with article-processing fees. We need transparency about the costs for value-added services: organizing peer review, paying editors, contextualizing and prioritizing articles, and so on.

Shumaker: Let’s turn to a question that combines equitable participation with another current hot-button issue: the problem of information quality. In one of the NISO Plus conference keynotes, Dr. Margaret Sraku-Lartey discussed the importance of Indigenous knowledge, and she also mentioned the challenge of assuring information quality as mechanisms to share Indigenous knowledge become more available. How do you see the interplay of Indigenous knowledge, open knowledge, and quality assurance? Does the traditional peer-review system need to change?

Joseph: That’s a huge issue. Now, our system is limited to the communication of knowledge in the form of articles and books. But those aren’t the only possible forms. For example, we need to recognize storytelling as a legitimate form of knowledge-sharing and scholarship. We’ve been working with Whose Knowledge? and our partners at the Knowledge Equity Lab, looking at the kinds and characteristics of Indigenous knowledge. As the knowledge-sharing system opens up to new forms of knowledge and new participants, we’ll need to learn from those participants. It’s a difficult problem, but the more inclusive we are, the more we’ll all learn.

Shumaker: Your point about the forms that knowledge takes leads into a topic from our past interview, when we talked about repositories. You brought up the importance of interoperability, and we talked about the linkage of data with articles. Since then, repositories have become even more complex, adding linkage to code, adopting new standards, and more. How do you see the evolution of repositories and their prospects?

Joseph: Repositories are taking on a much more central and robust role in the knowledge-sharing environment. Europe and Latin America are further along [than the U.S.] in developing a robust network of them, developing the community’s understanding of their role, and investing in them. One of the differences between 2016 and now is the evolution of open science in general. At that time, we hadn’t talked so much about the full workload of science using open processes and methods throughout the research lifecycle. Since then, the more the scientific community has started to apply open science ideas, the more the role of repositories has grown. There’s been a behavior change in science that has elevated them. They’ve evolved from a place to put versions of papers, and a dataset or two, to critical infrastructure where science actually takes place.

Shumaker: Another point you made about repositories was that they might develop around different scholarly disciplines, rather than being institution-oriented, where institutions collect all of the research from their researchers in various disciplines.

Joseph: It’s been borne out that we don’t need a repository on every campus. We can utilize repositories as shared infrastructure. We’re moving more slowly toward that model here in the U.S., but adoption will probably accelerate in the next few years. The experience we’ve had with COVID-19 data-sharing will help drive that. An example is the CORD-19 database. As we study the lessons learned [from COVID-19 research], it will be interesting to see which mechanisms, practices, and elements of infrastructure helped accelerate scientific progress. Then we can apply those lessons across other scientific disciplines—and it will be interesting to see what new strategies will emerge and how the thinking about repositories will change.

Shumaker: Regarding lessons learned during the pandemic, in our earlier interview, you suggested that traditional peer review might be supplanted or supplemented by reliance on preprints, in which authors would publish their drafts on preprint servers and get rapid feedback from their peer scholars. Another one of the NISO Plus keynoters, Dr. Zeynep Tufekci, described exactly that experience with a paper she co-authored about information in the pandemic. So your prediction seems to be coming true!

Joseph: Well, it’s exactly the vision Harold Varmus, Pat Brown, and Dave Lipman were talking about in the E-biomed proposal [in 1999]. It’s amazing to see it coming into being. COVID-19 is an example of the explosion in the use of preprint servers. As for concerns about the quality of preprints, during the pandemic, we’ve seen how fast misinformation can be corrected by the community on preprint servers, versus an article published in a journal, such as the mistaken one linking the MMR vaccine with autism, which took years to get withdrawn. There were preprints posted with misinformation that were corrected or taken down within 48 hours. The community self-corrected incredibly quickly. Meanwhile, Dr. Fauci mentions preprints all the time. I heard him the day before yesterday discussing the variants of the virus and referring to a preprint in medRxiv. My heart sang with gladness!

Another element of quality is educating the press about preprints. ASAPbio is an organization that promotes preprints, and they’ve been working on this: explaining the difference between a preprint and a peer-reviewed article, how the community can comment on them, and what the expectation of validity should be. There are many more resources to help raise the quality bar in the preprint environment, and these have been greatly accelerated by COVID-19 research. Preprints have become part of the landscape in biomedicine, and the trend is going to grow because the community needs the immediacy of preprints—not just in a public health crisis, but in other contexts as well.

Shumaker: As we close our conversation, it’s been interesting to see how well your vision from our previous interview has stood up over time. I look forward to progress in realizing your vision of open knowledge for social justice as well.

Joseph: So do I. It’s been a pleasure talking with you.