On The Investigation of Software Development and Evolution Practices

Abstract

Software engineering is a diverse field composed of different platforms and domains with a large variety of people involved in all stages of the development. Software platform refers to the underlying structure where the software is built on (e.g., Android) and software domains refer to systems developed for specific segments (e.g., healthcare). Understanding patterns of evolution of mobile and non-mobile software may support developers and companies within the open source software context. Revealing how development practices are adopted across domains may give new research directions and insights for industry professionals. In this Master Thesis, we investigate development and evolution practices using a mixed-method. Initially, we rely on the Frequent Itemset mining algorithm to analyze the co-evolution of code changes. We then interview cross-domain developers from the industry. Our findings show that different platforms and domains require practices to be applied in different ways.