Dynamic and static views of software evolution

Abstract

In addition to managing day-to-day maintenance, information system managers need to be able to predict and plan the longer-term evolution of software systems on an objective, quantified basis. Currently this is a difficult task, since the dynamics of software evolution, and the characteristics of evolvable software are not clearly understood. In this paper we present an approach to understanding software evolution. The approach looks at software evolution from two different points of view. The dynamic viewpoint investigates how to model software evolution trends and the static viewpoint studies the characteristics of software artefacts to see what makes software systems more evolvable. The former will help engineers to foresee the actions to be taken in the evolution process, while the latter provides an objective, quantified basis to evaluate the software with respect to its ability to evolve and will help to produce more evolvable software systems.