Over the past five months, the IOI team has been working on a study aimed at understanding the landscape of community infrastructure for research software and identifying the gaps, challenges, and opportunities for advancing its sustainability and resilience. This research, funded by the Sloan Foundation, aimed at addressing a number of core objectives as listed below:
- Identifying existing connections between different research software infrastructure services and initiatives and where new connections can be built to advance sustainability and resilience of the ecosystem;
- Identifying areas in the ecosystem where there is overcrowding/excessive functional overlap and where there are gaps, blockers, vulnerabilities, and/or critical dependencies that need addressing;
- Indications of what helps infrastructure services and initiatives succeed in adding value to the research software community.
After months of intensive research, the IOI team hosted a virtual open presentation on January 23, 2025. The aim of the presentation was to share the preliminary findings of the research study and also provide a platform to receive questions, feedback, and recommendations from the community. Watch the recording of the open presentation via the link below.
A recording of the community infrastructure for research webinar
Key findings
In terms of findings, we grouped the key takeaways from the study in three broad categories namely:
Hidden versus visible
One of the key findings is that while there are very visible initiatives and ample documentation within the research software ecosystem, there are a lot of issues that also exist but are not as visible. This observation doesn't apply just to organizations or just to research software work itself, but throughout the whole landscape. So, a couple of examples of that are where there are a lot of resources, a lot of really good resources, but not necessarily canonical.
“Recognition and reward came up in almost all of our interviews, that there's a lot of that work happening. Again, is it explicit? Is it sort of hidden? Is it canonical? [...] The landscape (of research software infrastructure) is sort of in the middle here,” remarked Jennifer.
Fragmented versus coherent
On this spectrum, we noted that from the interviews and data analysed, there is a lot of fragmentation within the research software ecosystem. It is important to note that there is also a spread and some areas may be closer to one end of the spectrum or another. Generally, there was recognition that research software should be a first-class research object and that recognition and support are needed for functions across the software lifecycle. Most interviewees also acknowledged that it is challenging to secure funding for software maintenance. The respondents expressed a lot of interest in collaboration and in increasing collaboration. However, it is important to be cognizant of some of the very practical obstacles to that. For instance, channelling funding across borders is a real practical obstacle to more formal kinds of collaboration.
One of the areas where we saw more disagreement is the characterization of some of the work as siloed. There is a fine line between what work is characterized as siloed or niche, and often, those two types of work are not necessarily mutually exclusive.
Fragile versus robust
Through the desk research and the interviews, it is apparent that the research software infrastructure ecosystem is in a period of emergence currently. However, while this emergence comes with a lot of potential, there are also some perils to navigate. The defunding of science, the sort of fragility of academia (certainly in the U.S.), and variations in academic cultures make it challenging to align practices.
“And there, again, there's information missing. So, just having some more cost transparency, that's useful information for funders to have. And it's useful to kind of know, financially, where things are in terms of fragility and stability. There's a lot of potential for shared services. This is, again, something that we talked about a lot in our interviews. Shared services, shared programs, really kind of working together a little bit more formally. And so, for IOI, in the context of this project, that raises the question of whether and to what extent and how infrastructures can or should support multiple disciplines,” said Jennifer.
Through the research, the issue of volunteerism came up frequently. This is important because there seems to be an established tradition and practice of volunteerism. an enormous amount of work is done through volunteering in open-source work, in publishing, and academia. One of the salient points from the interviews is about the human infrastructure and the need to care for people who do this work in the system. Heavy dependence on volunteers also heightens the risk of burnout. On the other hand, with paid staff, there's administration, management, and overhead involved in that. Hence, there is a need to find an equitable balance within the research software infrastructure ecosystem for the ecosystem to be resilient and sustainable.
Recommendations
- Surface pertinent information about costs, participation, usage, etc. - One of the big challenges is that there is critical information missing on research software infrastructures. Lack of common standards and fragmentation of practices affects our ability to see what the practices are, how they're influencing growth, or how they're influencing scientists and what they're able to do with research software. We need to give time and attention at the field level to identify and subsequently gather the needed data to fill the information gaps.
- Strengthen the scaffolding - Scaffolding can be defined as elements that, with appropriate instantiation, might become backbone (social, technical, administrative) infrastructure supporting the field. There is a need to shift the dominant energy from creating to integrating and maintaining. There is a need to encourage and enable consolidation, specialization, mergers, and handoffs. As long as new creation is valued and rewarded over stabilization, the field will lack the coherence needed to structurally support the activities of creating, sharing, and reusing research software.
- Grow the market - One of the challenges we have noticed is that research software infrastructures are leaning on the same funding sources, and those funding sources may not last. We've seen this in other fields as well. There is a need to figure out how to identify not just the users of research software, but how those users connect to customers. It is also important to understand how that user connects to the dollars that are necessary to keep the research infrastructures running. This is not about profit but rather about keeping things running and having a system that's dependable.
- Invest in coordination - It is important for actors within the research software infrastructure to realize that one entity can't change things alone. Progressive change is an advocacy-based process, and advocacy-based processes require coordination which also requires funding behind the coordination.
Nest steps in the research
Next week, we will be releasing the final report on The State of Open Research Software Infrastructure. This report will cover the items shared in the presentation but in more detail. We are really interested to hear from the community in terms of what you think is feasible and where you’d like to focus.
If you would like to keep updated on the release of the report and also to learn more about the future of the research, please subscribe to our newsletter.
******
Key resources of note on the research