Teaching JavaScript to expert Java developers

Abstract

In recent years, Single Page Applications (SPAs) emerged as a de-facto standard for modern, user-friendly web sites. While their advantages are manifold, SPAs massively impact the distribution of code and responsibility among an application: Where before, the web front-end of a Java application was essentially an orchestration of servlets and JavaServer Pages - driven from, and developed as part of the server - it is now an independent application in its own right. As a result of these developments, more and more developers are required to implement features to run in the browser, written in JavaScript. For companies as well as for individuals, the shift towards browser-centred engineering raises issues of developer education. Despite of their similar name and syntax, JavaScript and Java are highly different languages that require specific patterns and engineering practises. For a developer with many years of experience in Java or similar languages, learning JavaScript means more than just learning a new language: It requires a fundamental change in how to think about and approach programming problems - a so-called mind shift. The issue of knowledge transfer across programming paradigms and languages has been intensely researched in the field's transition from procedural to object-oriented programming, and diverse strategies have been proposed. The goal of this thesis is to show if, and how, existing strategies and experiences reflect in today's expert developer education, in the context of teaching JavaScript to expert Java developers. For this purpose, we conduct a qualitative content analysis of three real-world examples, each representing a popular format of eduction: (i) talks at developer-centred tech conferences, (ii) non-academic professional literature, and (iii) company-internal trainings. The on-hand thesis provides a detailed discussion of the phenomenon of (skill) transfer, which serves as the theoretical framework of our work. We present our research strategy based on Krippendorff's standard model for content analysis and discuss the results of our study on the level of individual cases as well as on an aggregate level.