Characterizing Software Maintenance Categories Using the Linux Kernel

Software maintenance involves different categories of activities: corrective, adaptive, perfective, and preventive. However, research regarding these distinct activities is hampered by lack of empirical data that is labeled to identify the type of maintenance being performed. A promising dataset is provided by the more than 800 versions of the Linux kernel that have been released since 1994. The Linux release scheme differentiates between development versions (dominated by perfective maintenance) and production versions (mainly corrective maintenance). The structure of the codebase also distinguishes code involved in handling different architectures and devices (where additions reflect adaptive maintenance) from the core of the system’s kernel. By analyzing different parts of the code separately, we demonstrate that corrective maintenance does not necessarily lead to code deterioration, that adaptive maintenance may improve some quality metrics, and that growth is largely the result of continued development as part of perfective maintenance.