Visualizing Software Instability (slides - PDF)

Presenter: Jennifer Bevan, Graduate Student, UC Santa Cruz

Abstract: As software evolves, incompatabilities between its operating environment and assumptions in its architecture become more pronounced. Maintenance becomes more costly in both time and effort as a result of this structural decay. By analyzing historical change patterns within dependence-related regions of the code, design decisions that persistently affect maintenance efforts can be identified and classified. Our proposed tool, IVA, will provide a visualization that identifies and classifies unstable code regions; the cognitive workload of the engineer can then be directed towards identifying the structural cause of these instabilities.

Bio: Jennifer Bevan received her B.S. in Electrical Engineering/Computer Science from U.C. Berkeley in 1993. She then worked for the Radio Science Systems Group at the Jet Propulsion Laboratory until 1999. She is currently a graduate student at U.C. Santa Cruz, and is working towards a Ph.D. in Software Engineering.