A Separation-Based UI Architecture with a DSL for Role Specialization.

摘要

This dissertation was aimed at creating a new software development methodology centered on the natural role specialization characteristic of people. Based on existing work in the field, we propose a unique separation-based UI architecture focused on specialized roles, namely user experience (UX) professionals and computer programmers. A distinguishing characteristic of our approach is the use of a domain specific language (DSL) to bridge the gap between the two roles' knowledge domains, to accomplish higher-level abstraction, and to free the user-interface (UI) designer from having to learn and master a programming language. Furthermore, we rely on model-driven engineering (MDE) techniques of code generation to further enhance the benefits of using a DSL. In this dissertation we describe a complete methodology that implements our architecture, reaps the benefits of role specialization, takes advantage of DSL-based abstraction and automated code generation, and provides step-by-step practical guidance in a variety of software development applications that contain UI components. To illustrate our approach, an NetCDF file subset extractor application built using the proposed methodology is also presented. The proposed method is compared with related approaches and the results of applying it are evaluated using a usability study, an analysis of automated code generation efficiency, and a comparison with several tools that could have been used to create a similar software application to our NetCDF subset extractor program. Background details on domain modeling, model driven engineering, and user experience and user interfaces, together with pointers to several directions of future work are also included in the dissertation.