Type systems affect programming productivity in several ways. First, the efficiency of a type checker determines how fast programmers can receive typing feedback. This is a real problem when typing software product lines, which typically encode huge numbers of programs. Second, the precision of a type system determines how accurately type errors are reported and thus how quickly type errors can be fixed. Third, the expressiveness of a type system determines the constraints on writing programs and thus affects the available programming idioms. In this talk I will present the concept of variational typing and its underlying principles. I will show how variational typing can improve efficiency, precision, and expressiveness of traditional type systems.
Sheng Chen is a PhD candidate at Oregon State University, working with Dr. Martin Erwig. Sheng's research interests are programming languages and software engineering with a focus on type systems, static analyses, functional programming, and software product lines. His current research is concerned with supporting programming productivity through improved static analyses. He also investigates a computational model of variational computing as a basis for accelerating computations through reuse.