NSF Award
2104535
Over the course of the last thirty years, computer scientists studying programming languages have worked on two major problems: how to incorporate software verification into the typical software developer's work flow, and how to extend software verification to support reasoning about data usage, for example, preventing the misuse of memory and socket handles. The main novelties of this project are to rectify this problem of misuse of data: the research generalizes and combines program logics for reasoning about data-usage, and the power of verification tools used in modern programming languages, within a single practical general-purpose programming language. The project's merits are: (1) The development of a new general theory that others can use to study and adopt this new combination; (2) A new programming language, Tenli, that encompasses this new powerful combination; and (3) The design of new pedagogical materials for teaching resourceful software verification at both the undergraduate and graduate levels. A major broader impact of this work is the incorporation of a wide range of students in this project, e.g., undergraduate students from Augusta University, a research university in Georgia, Wesleyan College, a historical liberal arts women's college in Georgia, and Clark Atlanta, a private Methodist historically black university in Georgia, and the first cohort of graduate students at Augusta University.<br/><br/>This project combines two powerful verification methodologies: (1) type-based verification, and (2) data-usage tracking through graded modal types. Graded modal types are generalized to support reasoning about a wide range of data-usage tracking including reasoning about imperative data structures. Both a new theory of graded modal types based on the notion of adjoint logics and a new practical general-purpose dependently-typed programming language, Tenli, are developed based on this new theory. Furthermore, several case studies within Tenli are being conducted to gauge its practicality.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
