Finding Lost Features in Cloned Systems

Copying and adapting a system, also known as clone-and-own, is a common reuse approach that requires little initial effort. However, the drawbacks of clones are increasing maintenance costs as bug fixes and updates must be propagated. To reduce these costs, migrating cloned legacy systems towards a software product line promises to enable systematic reuse and customization. For both, managing and migrating cloned systems, it remains a challenge to identify and map features in the systems. In this paper, we i) propose a semi-automatic process to identify and map features between legacy systems, ii) suggest a corresponding visualization approach, and iii) assess our process on a case study. The results indicate that our process is suitable to identify features and present commonalities and variability in cloned systems. Our process can be used to enable traceability, prepare refactorings, and extract software product lines.