SHF: Small: Programming with Semantic Revision Requests

University Of Colorado At Boulder, Boulder CO

Investigators

Abstract

Software development today is a social process, and the workflow of programmers is increasingly centered around discussing program revisions - that is, artifacts that document small, incremental program changes. Meanwhile, traditional research on testing, debugging, and analyzing programs focuses on one program version at a time, ignoring any preceding or succeeding versions of the same software artifact. The project's novelties are a foundational examination of program revisions and the investigation of automated techniques that cooperatively assist software-engineering teams with reasoning about them. The project's impacts are to enhance the way engineering teams collaboratively evolve their software and thus enable them to more effectively create secure and reliable software on which society depends. The project's core contribution is defining the notion of a "semantic revision request" that enables treating program revisions as first-class entities for debugging and analysis tools. A semantic view of program revisions enables programmers to run or analyze the program revision (both before and after a change simultaneously), not merely each of the two program versions, as supported by today's conventional tools. By focusing on debugging and analyzing program revisions - not their individual program versions - the resulting tools come closer to capturing the intent of the programmer as they develop and commit program changes. The intended broader impact from a technical perspective is that these executable and analyzable specifications will supplement, or even replace, informal descriptions about how software systems evolve, by providing an unambiguous way to communicate high-level programmer intent. 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.