Research Overview

Modern systems built for the desktop, the web, and the cloud are increasingly being developed in several languages, share code and data with multiple parties of varying levels of authority and trust, and run in environments split across client and server. My goal is to design programming language tools — including type systems, program analysis, dynamic instrumentation, and integrated development environments — to improve the reliability, security, and performance of these systems. Designing large, secure, distributed applications is difficult in any programming language. To further complicate matters, popular scripting, or “dynamic,” languages like JavaScript, Python, and Ruby must allow dynamic code loading and other powerful features that make it easy to glue together disparate application components. My work, to date, has culminated in an expressive, statically typed dialect called Dependent JavaScript (Djs) that dramatically improves the state-of-the-art in static type checking for scripting languages, a long sought-after goal in the programming languages research community. Because dynamic languages like JavaScript are at the core of modern web development, and because traditional statically typed languages like Java and C# have incorporated more dynamic, reflective features, Djs serves as a foundation for many directions of future research that will lead to the design and implementation of more reliable systems.