Uncovering Variability Models for Software Ecosystems from Multi-Repository Structures

Abstract

Variability is a significant source of complexity in many large-scale software systems. Software variability must be managed in order to effectively tame the arising complexity. Consequently, variability management processes are at the heart of current software product line engineering practices. However, legacy software systems exist that have not been developed with such practices. Moreover, an increasing amount of software is developed in large, fragmented communities, also referred to as software ecosystems. Variability in such systems is often not explicitly managed and causes significant difficulties during software maintenance and evolution. Methods and tools for uncovering and explicitly managing this variability have been subject to ongoing research. This paper presents our research in progress of empirically studying the application and combination of such methods in the context of real-world industrial case study in the health care domain.