An Approach to Assessing the Health of Opensource Software Ecosystems

Abstract

With the development of open-source technology, open-source software ecosystems (OSSECO) have been formed due to various connections between open-source projects and developers. To measure stability and sustainability, the health of an OSSECO is proposed, like the health of ecosystems in nature. Unfortunately, there are not a set of unified and mature OSSECO health evaluation rules yet, nor have effective governance methods. Existing researches mainly analyze the health of a specific open-source ecosystem or design performance indicators related to OSSECO health. This paper combines the classic OSEHO model with the rapidly developing CHAOSS open-source community metrics, establishes an OSSECO health evaluation model that can provide specific scores based on Entropy Method, and develops an open-source ecosystem health evaluation system based on the model. To the best of our knowledge, we are the first to propose a qualitative and quantitative model to show the health status of open-source ecosystems. Meanwhile, we conduct the automatic evaluation of the health of an OSSECO. Finally, we analyze the software ecosystem health of 10 open-source projects on GitHub by the established evaluation system. The result can prove the effectiveness of the system and provide data support for developers to make governance decisions.