CAREER: The layering problem in graphical user-interface design

Michigan State University, East Lansing MI

Investigators

Abstract

A two-pronged approach is proposed to add rigor to the current state of software-engineering knowledge. The research component focuses on the technical design of highly graphical interactive systems, specifically the design decisions required to separate a graphical user interface from an underlying computation. Currently, the nature of design knowledge in this area is heuristic and anecdotal. The research will clarify and extend such knowledge to produce a deductive, synthetic theory of composition in graphical interactive systems design. Being synthetic, this theory creates new knowledge, specifically the detailed modular design of separate, but composable, user-interface and functional-core components. This knowledge is synthesized from existing knowledge, specifically an object-oriented model of the objects with which a user will interact. Being deductive, this theory will be rigorous and repeatable in sharp contrast with its technology-laden predecessors. The theory exploits and generalizes prior results in model-based user-interface generation and hierarchical components. The education component exploits and facilitates rigorous software-engineering knowledge to enable precise measurement and improvement of learning in undergraduate software-engineering courses. The approach is founded on the use of cognitive objectives, which have proved useful for course improvement and curriculum design in other engineering disciplines.