An Empirical Study on Reactive Programming

In recent years, interactive applications have increased in popularity. However, due to the lack of fitting programming abstractions in this domain, developing these applications is challenging. While multiple reactive programming languages have been proposed addressing these challenges, empirical studies evaluating the usability of these languages are still uncommon. With this study we present an empirical evaluation of the usability and intuitivity of one of these proposed languages: REScala, a Scala library for functional reactive programming. The goal of our study is to validate claimed advantages of REScala, ascertain what improvements to the language could be beneficial and how these improvements could be achieved. During the study we observed 9 participants working with the REScala library following the think aloud approach. Our results show that after overcoming initial challenges, participants showed a quick learning effect and, in the end, understood the basic concepts of REScala and wrote valid REScala code. We therefore conclude that REScala doesn't present more difficulties compared to learning any other new programming concept while providing multiple benefits for reactive programming, such as increased composability, enforced consistency and fault tolerance. Based on our results, we also provide a set of improvements, we observed helpful, in order to overcome initial challenges when learning REScala.