Modeling variability in the context of software product-lines has been around for about 25 years in the research community. It started with Feature Modeling and soon enough was extended to handle many different concerns. Beyond being used for a mere description and documentation of variability, variability models are more and more leveraged to produce other artifacts, such as configurators, code, or test cases. This paper overviews several classification dimensions of variability modeling and explores how do they fit with such artifact production purposes.
Model-Driven Engineering for Software Product Lines