Ebook: Empirical Research Towards a Relevance Assessment of Software Clones

Redundancies in program source code - software clones - are a commonphenomenon. Although it is often claimed that software clones decreasethe maintainability of software systems and need to be managed,research in the last couple of years showed that not all clones canbe considered harmful. A sophisticated assessment of the relevanceof software clones and a cost-benefit analysis of clone management isneeded to gain a better understanding of cloning and whether it is trulya harmful phenomenon.This thesis introduces techniques to model, analyze, and evaluateversatile aspects of software clone evolution within the history of asystem. We present a mapping of non- identical clones across multipleversions of a system, that avoids possible ambiguities of previousapproaches. Though processing more data to determine the context ofeach clone to avoid an ambiguous mapping, the approach is shown tobe efficient and applicable to large systems for a retrospective analysisof software clone evolution.The approach has been used in several studies to gain insights into thephenomenon of cloning in open-source as well as industrial softwaresystems. Our results show that non-identical clones require moreattention regarding clone management compared to identical clones asthey are the dominating clone type for the main share of our subjectsystems. Using the evolution model to investigate costs and benefits ofrefactorings that remove clones, we conclude that clone removals couldnot reduce maintenance costs for most systems under study.