The state of the open source ecosystem & how to fund away ‘software collapse’
By Georgia Iacovou | June 2024
<aside> <img src="/icons/reorder_yellow.svg" alt="/icons/reorder_yellow.svg" width="40px" /> Contents
</aside>
The open source science software landscape is vast and ecological. It consists of small, obscure tools that act as the connective tissue for other tools; it also contains ubiquitous pieces of critical infrastructure. In many ways, much of this is invisible to non-practitioners, making it difficult to see and understand the pain-points, and ultimately make good funding decisions.
To give a clearer picture on how the ecosystem is responding to current funding models, Section One of this report outlines the incentives for building open source software, and upholding open source principles. There is a distinct tension between those who maintain software as a hobby, and those who are paid to do so by industry or philanthropic organisations.
Section Two focusses on the ways in which software is, or isn’t, recognised as a valid research output — and therefore how it is valued by institutions and funders. There’s a pervasive culture of ascribing value to software only once it’s been cited in research papers — and this can be a cumbersome mechanic for research teams.
Finding time for these kinds of activities, which exist outside of actually building software, is a common challenge. Section Three explores the importance of supporting non-technical roles such as community management and marketing — as well as the reduction of technical debt, which contributes to the overall health of a piece of software.
Finally, Section Four outlines a few recommendations for funders looking to re-strategise and design funding programmes that more effectively support the long term maintenance of research software.
<aside> ⚫ Notes on research methodology used: this research took place between May and June of 2024. Research consisted of desk research, and hour-long interviews with twelve professionals working in open science, including software engineers, community managers, and other funders. The viewpoints, opinions, and experiences expressed in this report, even when not directly quoted, are those of the interviewees, and not of the author.
</aside>
<aside> <img src="https://prod-files-secure.s3.us-west-2.amazonaws.com/03d67217-d698-454e-a3ee-c894b1349b33/6043f5ca-03d0-44f2-a168-f353029c9332/quote_marks.png" alt="https://prod-files-secure.s3.us-west-2.amazonaws.com/03d67217-d698-454e-a3ee-c894b1349b33/6043f5ca-03d0-44f2-a168-f353029c9332/quote_marks.png" width="40px" /> …if you have been doing computational science for a few years, there is a good chance that you have experienced software collapse, and probably it was not a pleasant experience
Konrad Hinsen, Dealing with Software Collapse
</aside>
