Missing Clean Code Acceptance and Support in Practice - An Empirical Study

Abstract

Code quality is important for maintainability and sustainability of software systems, and is addressed by guidelines and peer reviews. There are well-defined methods and standards for code quality for example the Clean Code approach [1], and they are integrated with current software development process models such as SCRUM. However, developers have to be encouraged to invest effort due to the trade-off between software quality measures and projects' time pressure. Unfortunately, there are no proven hypotheses about the reasons for missing acceptance of such methods and standards by developers. To investigate hindering aspects for code quality acceptance in practice, empirical studies might help to determine the root problems and to establish solutions. However, there is a lack of empirical studies concerning this topic. Therefore, we performed an empirical study with practitioners to investigate problems and challenges while enforcing Clean Code. Semi-structured interviews were used to consult experienced practitioners from different companies and backgrounds. The transcripts were analyzed with open coding. The results showed that a lack of even one of the people-related aspects such as motivation, knowledge, experience and code quality awareness, hinders the enforcement of code quality. Surprisingly, tools might establish hindering factors due to high initial effort. As conclusions we propose implications for academic research, organization, practitioners and tool providers.