Software, particularly research software, impacts all parts of the modern scientific enterprise, especially the central, vital activity of data analysis and statistical inference. One would be hard pressed to name areas of contemporary research that are not reliant on research software. These include hundreds of open-source libraries and software packages, many of which are developed by researchers themselves. Research software is so essential that a large majority of researchers in the UK and United States say that they would no longer be able to continue their work if research software tools stopped working (Jiménez et al. 2017). Despite the growing importance of research software, much of it is cobbled together with little regard for high standards that are characteristic of other research activities. As a result, the research software ecosystem is fragile and the source of numerous problems that afflict modern computational science (Carver et al. 2018).