Using version control data to evaluate the impact of software tools

Software tools can improve the quality and maintainability of software, but are expensive to acquire, deploy and maintain, especially in large organizations. We explore how to quantify the effects of a software tool once it has been deployed in a development environment. We present a simple methodology for tool evaluation that correlates tool usage statistics with estimates of developer effort, as derived from a project's change history (version control system). Our work complements controlled experiments on software tools, which usually take place outside the industrial setting, and tool assessment studies that predict the impact of software tools before deployment. Our analysis is inexpensive, non-intrusive and can be applied to an entire software project in its actual setting. A key part of our analysis is how to control confounding variables such as developer work-style and experience in order accurately to quantify the impact of a tool on developer effort. We demonstrate our method in a case study of a software tool called VE, a version-sensitive editor used in BellLabs. VE aids software developers in coping with the rampant use of preprocessor directives (such as if/ endif) in C source files. Our analysis found that developers were approximately 36% more productive when using VE than when using standard text editors.