Generic Tools, Specific Languages

Adapting tools to a particular domain is expensive, and the adaptation is often not very deep. To address this challenge, Generic Tools, Specific Languages shifts the focus from building and adapting tools (windows, buttons, algorithms) to building and adapting languages to a domain. The thesis applies the approach to embedded software development: mbeddr is an extensible set of integrated languages for embedded software development built with JetBrains MPS language workbench. The evaluation of mbeddr suggests that it is a productive tool for embedded software development. The evaluation of the development of mbeddr itself suggests that MPS is a suitable platform for Generic Tools, Specific Languages, and that the approach in general is worthwhile. 1.1 O V E RV I E W A N D C O N T R I B U T I O N This section provides an overview of the thesis. It starts out by describing the contribution and the research methodology. It then discusses mbeddr, the primary artifact built during the dissertation and explains its context, embedded software engineering. A discussion of the technologies used for building mbeddr – language engineering, projectional editing and JetBrains MPS – follows. The section concludes with an overview of the results. Contribution Today’s software engineering tools are hard to adapt to specific domains. A major reason is that while platforms such as Eclipse support easy extension of the tool (views, editors, buttons, menus) they do not easily support the extension of the data formats or languages that underlie a tool. The contribution of this thesis is the introduction and evaluation of a new approach to developing domain-specific software engineering tools called Generic Tools, Specific Languages. It shifts the focus from adapting the engineering tool to adapting the underlying languages to solve the problem of tool adaptation. It relies on language workbenches (the generic tool) and recent advances in language engineering, particularly, user-friendly projectional editing. The problem of tool extensibility, the Generic Tools, Specific Languages approach and the way it is evaluated is discussed in Chapter 2. The specific research questions that arise from the approach (domain-specific extensibility, language modularity and projectional editing, tool scalability, implementation efforts and the learning curve) are introduced and motivated in Section 2.6. Methodology As part of the dissertation, the Generic Tools, Specific Languages approach has been applied to embedded software engineering, resulting in a