Characteristics of selected open infrastructures

Introduction

IOI’s annual snapshot of data from Infra Finder ^[1] provides an authoritative source of information about the governance, business form and finances, policies,community engagement, and technical attributes of over 100 open infrastructures (OIs).

This year, we’ve paired our data highlights with prompts to encourage deeper exploration and real-world application by readers.

As a companion to this chapter, we make the full Infra Finder dataset^[2] (including previous years’ data) available in Zenodo under a Creative Commons Attribution 4.0 International license to support reuse. You can also browse selected data points via our interactive dashboard.^[3]   

Transforming data into actionable insights

Explore integrations to reveal networks of reliance

We ask OIs to list their integrations with other tools, services, and standards that appear in Infra Finder, revealing networks of reliance and interoperability. For example, Thoth Open Metadata describes a number of integrations that leverage its ability to create, store, and distribute robust book metadata.^[4] Thoth, in turn, relies on upstream services and standards such as Research Organization Registry (ROR) identifiers^[5] and digital object identifiers (DOIs) to enrich the metadata it collects.

This web of connections highlights not only the collaborative potential within the ecosystem but also its complexity and fragility. Exploring these maps of interreliance provokes a range of questions about the strengths (e.g., distributed costs and risks, modularity) and vulnerabilities (e.g., single points of failure, redundancy) relevant to the open infrastructure ecosystem.

• Which tools and services are serving as “hubs” in the open research ecosystem due to their high level of interoperability?
• What would the ripple effects be across the ecosystem if a widely used infrastructure solution were to cease operations? Would critical services experience disruptions? Are there sufficient redundant or alternative systems in place to fill potential gaps, or would the loss of a single tool leave entire segments of the community scrambling for solutions?

One of the notable advantages of open research tools is their flexibility. As a user, you can customize solutions to meet your needs and migrate your data to new solutions if those needs change. But this flexibility also comes with challenges. Understanding which tools to select and how to connect them to achieve your goals can be complicated. By looking at how these tools work with one another, we can ask:

• Where are the opportunities for vertical interoperability between OIs to support diverse research workflows?^[6]
• How can we make service “stacks”, or sets of tools that work together for a specific purpose, more visible and easier to implement?^[7]

In some cases, having too many choices can be a detriment, such as when interoperability is the goal. Multiple OIs in the same solution categories risk overly segmenting the market, leaving too few resources for any to operate sustainably. Merging overlapping infrastructure solutions or strategically sunsetting redundant tools may strengthen the ecosystem, making it easier for users to make choices and infrastructure organizations to build resilient business models.^[8]

• Where do such opportunities exist, and how can we ensure that consolidation happens in a way that preserves community values and enhances long- term sustainability?

Dig into solution categories to find opportunities to help users identify the right tools

Many of the OIs listed in Infra Finder help users manage and find information. The most common types of solutions have remained unchanged since last year and include discovery systems (22 solutions), standards, specifications, or protocols (15 solutions), repository software (11 solutions), repository services (11 solutions), and publishing systems (11 solutions).

• Which research or scholarly communication functions is Infra Finder missing? 
• Do the existing categories make it easy for people to figure out which OIs will work for different needs? Are there other ways of organizing them that might make this more straightforward?

Identify shared funding needs and opportunities for collective investment

Last year, we reported on the range of funding needs infrastructures cited in their Infra Finder entries.^[9] Funding for operational and community engagement needs dominated the responses. These are areas where available grant opportunities often fall short. This year, we modified our classification scheme for funding needs to mirror that of our grants dataset, narrowing the categories to research and development (including enhancing technical features or developing new services); operations (such as paying or adding staff); adoption (such as community engagement activities and recruiting new members); and other (activities such as events, and strategic/governance/ business planning).

This year, 62 of the 101 OIs described funding needs related to their ongoing work for which they need additional revenue. Based on IOI’s analysis of funding awarded to the OIs in Infra Finder, over 64 percent of direct awards between 2016 and 2024 funded research and development work, aligning with the most commonly cited need, but falling short of the demand for operational and adoption support. The gutting of research funding in the US that is underway in 2025 will surely exacerbate revenue gaps in all categories in the coming years.

You’ll find each OI’s freetext description of its funding needs in our dataset.

• Where funding needs overlap, are there opportunities to pool investment and effort?

Examine the geography and geopolitics of open infrastructure

The OIs in Infra Finder listed 12 different nations as their primary location of activity or incorporation. Slightly more than half of the infrastructures in Infra Finder are US- based. While research is global, OI governance, funding, and operations are often subject to national laws, politics, and funding.

• How does the distribution and concentration of infrastructures in different regions relate to questions of digital sovereignty?^[10]
• Which infrastructures might be at greatest critical risk due to geopolitical changes, politically motivated efforts to dismantle academic research, and unprecedented funding challenges in higher education?^[11]
• Have open access/public access policies around the world contributed to infrastructure development? Can we tie infrastructure in different regions directly to these policies?
• To what extent is the information included in Infra Finder useful outside of North America and Western Europe? What could be added to make it more relevant to communities around the world?

In our dataset, OIs based in North America were most likely to say that their primary source of funding comes from program service revenue (income from selling their products and services) rather than contributions (including donations, membership fees, and grant funding) or another source.

• Are variations in primary funding sources meaningful indicators of how infrastructures sustain themselves in different regions?

Complicate the open vs. closed binary

At first glance, it might seem simple to categorize tools and platforms as either “open” or “closed.” However, a deeper examination reveals a spectrum of practices and characteristics that blur this binary distinction. What do we mean by open or closed? Is it about openly licensed code, transparent governance, community engagement, no-cost usage, or something else entirely?

The profiles of OIs in Infra Finder further reveal the diversity of practices in the open infrastructure world, and shed light on what characteristics are erased in the use of simple criteria. Non-profit organizations that do not make their code openly available, despite operating with community-oriented missions and values. For-profit companies that openly license and extensively document their code, allowing public contributions and fostering collaboration, yet maintaining control over governance and strategic direction. Projects with strong community engagement — through forums, user feedback loops, and active social media presence — yet lacking formal community governance structures that define the ground rules for participatory decision-making. The false binary of open versus closed can obscure important nuances. Infra Finder data can help expose these complexities, fostering a more informed and critical understanding of openness. You can read more about exposing false binaries in a later section of this report, “Shorthand falls short: Why open infrastructure defies simple labels.”

Consider engaging with the following questions:

• What can Infra Finder data tell us about how to describe “open”?
• Use data from Infra Finder to complete an evaluation of an OI using the FOREST Framework (specifically looking at the value of “Openness”).^[12] Does looking more deeply give you a different impression or perspective on the OI’s openness?
• Are varied “flavors” of openness more common among OIs at certain stages of maturity, based in different regions of the world, or within different solution categories?

More information about the data

This year, we invited 77 new OIs to participate in Infra Finder and also accepted self-nominations via our expression of interest form. We asked our existing 57 infrastructures to update their information as needed. We received 44 new entries through our outreach and the expression of interest form, as well as updates (or confirmation that no data had changed) from about half of the initial 57 infrastructures.^[13]The total number of OIs included in our dataset now stands at 101.

This year, we focused specifically on recruiting and adding OIs for research data sharing and analysis and digital preservation and archiving in response to trends and needs in the open science ecosystem. ^[14]

Because participation in Infra Finder is voluntary, and largely by invitation, this dataset remains modest in size and scope and is not a random sample. However, the breadth of information captured provides a robust basis for exploring research questions about the open research infrastructure ecosystem, especially when paired with other quantitative and qualitative data sources.

We made minor modifications to our data collection instrument this year, including consolidating several solution categories and introducing nine new solution categories. ^[15] The new categories are

• Digital preservation service
• Data management planning tool or service
• Research software community
• Digital preservation tool
• Web archiving system
• Software preservation service
• Format conversion tool or service
• Computing library
• Computing framework

An infrastructure was eligible for inclusion if, at the time of the invitation, it was fully operational and in active use as a service, protocol, standard, or software that the academic ecosystem needs in order to perform its functions throughout the research lifecycle. Infrastructures also had to meet one or more of the following eligibility criteria:

• Meets the definition of open source software (OSS);
• Primarily or exclusively distributes openly licensed (open access) content;
• Is free to use by anyone (free of charge or other restrictions);
• Is community governed and is transparent in its operations and finances;
• Is operated by a non-profit or non-commercial entity.

Participating OIs were sent a partially completed data form and asked to correct or add information. IOI’s team validated the responses against publicly available documentation and resolved discrepancies with the respondents.

As a companion to this chapter, we make the full Infra Finder dataset^[16] (including previous years’ data) available in Zenodo under a Creative Commons Attribution 4.0 International license to support reuse. You can also browse selected data points via our interactive dashboard.^[17]

Readers are encouraged to explore Infra Finder and the associated data, whether to further investigate some of the research prompts we have presented or to support their own work. If that describes you, please let us know by completing our feedback form.

Feedback

Footnotes

1. https://infrafinder.investinopen.org/
2. https://zenodo.org/records/14647337
3. https://investinopen.org/data-room/characteristics-of-open-infrastructure-dashboard/
4. https://copim.pubpub.org/pub/growing-network-of-open-infrastructures-federated-services-with-thoth#nrwp0dvip5u
5. https://infrafinder.investinopen.org/solutions/research-organization-registry
6. Read more about vertical interoperability and open source research tools at https://upstream.force11.org/the-time-is-now-vertical-interoperability-between-research-tools-an-essential-enabler-for-the-fairification-of-data/
7. Read more about the concept of service stacks at https://mindthegap.pubpub.org/
8. https://educopia.org/wp-content/uploads/2024/09/Census2019_EducopiaPublications.pdf
9. https://investinopen.org/state-of-open-infrastructure-2024/sooi-characteristics-2024/#governance-business-form-and-finances
10. For more on digital sovereignty, see https://investinopen.org/state-of-open-infrastructure-2024/sooi-future-signals-2024/.
11. For more on the critical risks to open infrastructure provoked by the US political environment, see https://katinamagazine.org/content/article/open-knowledge/2025/US-funding-cuts-imperil-open-infrastructure-globally
12. https://www.nextgenlibpub.org/forest-framework
13. Recognizing that changes can happen at any time, in addition to IOI’s annual outreach campaign, OIs can make as-needed data updates throughout the year. These changes are reviewed by the IOI team and each individual service page of Infra Finder shows when data was last updated.
14. https://investinopen.org/blog/help-shape-the-future-of-infra-finder-explore-new-features-and-share-your-feedback/
15. Read our announcement and find definitions for each solution category in our Infra Finder documentation.
16. https://zenodo.org/records/14647337
17. https://investinopen.org/data-room/characteristics-of-open-infrastructure-dashboard/