Measuring the Impact of Open-source Projects

Abstract

Foundations and communities for open-source projects often want to determine the impact of their software projects. This impact can be understood in a variety of ways, and this research explores this subject by examining the interdependencies between an open-source project and other projects. In this context, the open-source project is dependent on components created upstream by the other projects. Conversely, software is used downstream by other projects. This thesis proposes an index called the V-index, through which impact of an open-source project, as used in downstream projects, can be measured. The V-index is developed using the open database libraires.io, which provides the requisite dependencies and, thus, a determination of the impact of open-source projects. Further, to explore how the V-index can be understood, project-specific open-source health metrics are identified as potentially easier targets for change than is project impact. A correlation matrix is formed among the identified metrics and the V-index is calculated to determine the corresponding relationships among them. Finally, the conclusions and implications of this research are drawn.