Individual context-free online community health indicators fail to identify open source software sustainability

Software is often abandoned or shut down, for one reason or another, and whilst research on academic open source software is sparse, there seems little reason to assume it is any different. While some reasons may be straightforward, e.g. a sole maintainer has moved on, or grant funding has ceased - some projects are able to withstand these barriers and may remain active and maintained despite adversity. This study monitored open source projects over the period of a year, measuring common performance indicators, using both subjective and qualitative measures (participant surveys), as well as using scripts to analyse indicators associated with these projects' online source control codebases. We find that these health indicators can not be used as cross project benchmarks, due to the significant variation in context for each project. They can, however, often be useful in signifying changes in a single project's health, providing they are not used to compare between different unrelated projects.