Defining Research Software: a controversial discussion

Software is essential in modern research; it plays vital roles at multiple stages of the research lifecycle. The term Research Software is widely used in the academic community but, what do we mean when we use these terms? Software and research? When you think of software, you may think of a digital object that is executed on a machine. Yet software is more than just this, it is a complex and evolving artifact. It may be a concept or a project designed to solve a puzzle by a team or a community that develops its functionalities and algorithms, which might not be digital objects. Furthermore, the software artifacts are digital objects, e.g., executables and source code files for different environments. These digital artifacts, which are used in a scholarly setting, might be important in the research process, but should all these be considered Research Software? This report is the result of a discussion examining the scope of the community definition of the FAIR principles for Research Software as part of the work in the FAIR for Research Software working group (FAIR4RS). We aim to clarify the scope of the FAIR principles by identifying which software artifacts the FAIR principles should apply to. This discussion portrayed a complex landscape of software uses in research and existing definitions that can help to better understand the complexity of different types of software in academia. Finally we determine the scope of the FAIR4RS with a short and concise definition of Research Software as a separate metaphor of software in research.