In our previous posts, we’ve shared more about key criteria we are exploring as we assess and map the open infrastructure landscape. We’ve also shared our plan for modeling a system of reporting key data and findings to support those looking to invest in, adopt, and assess open infrastructure solutions.

This post unpacks more about our efforts to identify an initial subset of projects in the open infrastructure landscape to interview and examine, and the characteristics and key questions we are prioritizing in our analysis.

Just as different projection methods introduce distortions into the maps of the world (e.g., the centering of Europe, the distortion of regions further from the equator), criteria catalogs and assessment guidelines for infrastructure projects come with inherent distortions and biases. At IOI, we are hoping to bring the conversation about these distortions and biases to the forefront before starting to map the larger open infrastructure landscape in order to make sure that community, people, and issues of justice and equity are centered and represented.

More on the projects (and how we selected them)

A few weeks ago, we shared our project plan for our Costs of Open Infrastructure investigation. This work pairs public data collection and information gathering with interviews with a select number of open infrastructure projects to gain more insight into their funding, costs, and structure.

These interviews are intended to augment our audit and analysis of publicly available data from funding and grant databases, annual reports, public filings (like 990s in the US) and more. This approach also helps us model and refine our approach to gathering and verifying information for decision makers on program histories, operations, challenges, and funding. Projects chosen represent a small snapshot of our larger project list that we are building out and categorizing in parallel to help situate these efforts in the broader ecosystem and understand the relationship to other tools, technologies, and services.

We set out to identify a sampling of 5-10 projects to investigate in more depth to gain additional insight into the “hidden” costs of operating open infrastructure projects and services, points of resource scarcity, and dependencies on other efforts and initiatives. Our aim is to learn more about the costs associated with supporting these projects, such as hosting and maintenance costs, staffing support, margins, vendor relationships / outsourcing, in-kind support, and breakdown of past and current resourcing. By framing the impact and benefits of open infrastructure with real costs, we can better understand the scale of the commitments required to sustain open infrastructure.

They include:

The projects above were selected based on the following key dimensions and criteria:

  • Projects that are open source and not-for-profit;
  • Projects that serve the research and scholarly communication communities;
  • Projects that are stakeholder governed;
  • Accessibility of documentation of financial support of the project ideally over a 1-3 year timespan;
  • Transparency and diversity of business models (e.g., membership-based support vs national subsidy);
  • Adoption levels and usage as a proxy for influence and impact;
  • Diversity in services represented;
  • Other environmental considerations include:
  • Transformative influence (e.g., does this shift away from a dated, closed, less ideal model?)
  • Degrees (if any) of external investment (e.g., government subsidy, industry support, in-kind donation of staff and services)
  • Affordability and accessibility by underserved audiences

Out of the ten projects listed above, we were able to interview 9 and will be sharing our analysis later this year to support those looking to adopt and invest in open infrastructure.

Key infrastructure dimensions we’re tracking

What properties are we looking for when we assess whether a project or service is a tool versus an “infrastructure”? A seemingly innocent question that quickly leads to the common hodgepodge of conflicting definitions deriving from a multitude of worldviews, cultures & economies, missions & goals, and dreams & realities across the globe. The answer to the question of what counts as open infrastructure, thus, will always depend on the particulars of who is asking and who is answering.

To acknowledge the complexity, we’re examining that question as one of degrees and relationality. How open and how "infrastructural" does a project enter into the relation with its scholarly community of users? Within that conceptual and methodological framework, we propose 3+1 analytic dimensions to asses open scholarly infrastructure projects:

  1. Infrastructureness. We propose to understand infrastructure as a property on a spectrum rather than a binary description. This means that we seek to identify and describe the infrastructure-like characteristics of a project across its technical, organizational, and social structure. Instead of asking “what is an infrastructure project?'' We are looking to identify those moments when projects, services, and tools start to act as infrastructure for the community.
  2. Scholarlyness. The relevance to scholarly activities avoiding preconceived notions of what "proper science" ought to look like and including research-related areas such as education or citizen science. The practices and kinds of labour that are involved in research and scholarship change and evolve with the scholarly communities, thus, we should also be anticipating and looking for a moving target when we think what counts as "scholarly" or not.
  3. Openness. We propose to understand openness as more than technical and legal dimensions of software projects. Openness also relates to questions of whose voices are heard or silenced, to the organizational models for technology development, and questions of governance and agency for the community. We center the idea of openness as a radical tool for change and betterment of our communities.
  4. Transformative influence. To foster and enable change, we want to actively seek out projects that shine light on or operate in the fringes of our preconceived notions of technical infrastructure, scholarship, and openness for the betterment of the absent and silenced voices.

The first three dimensions should be understood as orthogonal, meaning that a project can be more or less infrastructural, tightly or loosely connected to scholarship, and open or closed, while the transformative influence of a project can be traced across all of the previous three dimensions.

In what follows, we break down each dimension into sub-categories derived from the amazing work of others before us such as Leigh Star's dimensions of infrastructures and the Principles of Open Scholarly Infrastructure (POSI). Concepts from both frameworks have been remixed and restructured to fit into and further structure the 3+1 dimensions of open infrastructures.

Mapping the infrastructure landscape

These projects represent a small subset of a more comprehensive list we’ve been pulling together to refine and analyze. To start, we’ve pulled project and infrastructure provider lists from the Mapping the Scholarly Communication Landscape 2019 Census and bibliographic scan, the Scholarly Communication Technology Catalogue (SComCAT), the list of Open Access Publishing Tools from the Radical Open Access Collective, and the 400+ Tools and Innovations in Scholarly Communication compiled by Jeroen Bosman and Bianca Kramer of Utrecht University Library. Our thanks to all who contributed to these resources and made them available for reuse.

Over the coming months we’ll be further refining and building out this list, categorizing projects based on attributes including those listed above. We’ll be inviting more contributions to this list and other resources in the new year, as we recognize the issues that exist around our own familiarity bias that can affect the “completeness” of a resource like this that's foundational for our work.

Posted by Asura Enkhbayar & Kaitlin Thaney