FAIR Code Sharing in Open Access Repositories

The first Open Science Indicators Dataset released in December 2022 by the Public Library of Science (PLOS) to define and measure Open Science practices, understand the current adoption state and track progress over time, focuses on three practices - preprint hosting; data shared in data repositories; and code sharing - comparing +61,000 PLOS articles with a set of 6,000 publicly available papers from PubMed. The Dataset confirms that code sharing, whether in a repository or 'online', is less common than data sharing. Nevertheless, the percentage of articles sharing code, both in the PLOS articles and the comparator set, has consistently risen over the past four years. This has been done in disparate ways, with many sharing on GitHub, resulting in code lacking the persistent identifier necessary to ensure its long-term discoverability. With increased pressure on researchers to make all of their outputs FAIR, the benefits to sharing code via established repository infrastructures include ensuring that robust and appropriate metadata is always attributed, unlocking further potential for discovery and making it truly citable. We will showcase examples of code sharing, outside of and within existing data repositories, and discuss the benefits of sharing code and related outputs in a data repository.