Domain Engineering: A Practical Application in Analysis and Design of a Generative Query Language

New and difficult analysis and design problems have emerged with the evolution of the Domain Specific Language which replaces generality for expressiveness in a restricted domain. In an effort to solve them, diverse techniques have been introduced which, when combined, have resulted in confusing and problematic analysis and design flows. In this paper, we outline our ongoing domain engineering research in Domain Specific Language. We identify requirements, components and a reference architecture, which are put together in a conception and construction phase of a Generative Query Language called Carbon QL for domain of forest dynamics. We describe and evaluate our approach in a case study. The domain engineering phase produced domain and design models and led to a reference architecture that follows the Model Driven Engineering approach. The application engineering contributed effectively to the Carbon QL implementation based on an explicit knowledge representation (Car bontology), and a generative query engine that translates from Carbon QL to SQL using Hibernate as object-relational framework. Our preliminary results indicate that the effective configuration, componentization and reuse are possible, and optimize the development process for this domain.