home · mobile · calendar · colloquia · 1995-1996 · 

Colloquium - Murphy

Lightweight Structural Summarization as an Aid to Software Evolution
Department of Computer Science and Engineering, University of Washington

All too often, changes to software systems lead to budget and schedule overruns. This occurs in part because software engineers lack up-to-date information about the structure of the systems they are changing. This talk describes a new approach to quickly and easily providing engineers with the desired structural information.

In this approach, structural information extracted from a system's artifacts is summarized in the context of a high-level model chosen by the engineer as suitable for reasoning about the planned modification. I will discuss two new techniques developed as part of this approach. The "software reflexion model" technique permits an engineer to use a high-level model as a lens through which to view the structure of the system's source code. The "lexical source model extraction" technique facilitates the scanning and analysis of system artifacts for structural information that is difficult or impossible to extract using existing techniques.

Each of these techniques is lightweight: the engineer is able to balance the cost and time of applying the technique with the completeness and accuracy of the desired summarization. In addition to describing the techniques, I will discuss the validation of the research, focusing on a case study of an experimental reengineering of Microsoft Excel.

Gail Murphy is a PhD student in the Department of Computer Science and Engineering at the University of Washington (Seattle, USA). She received an MS degree from the University of Washington in 1994, and a BSc (Honors) degree from the University of Alberta in 1987. She spent five years as a software designer at MPR Teltech Ltd. in Burnaby, B.C. where she was involved in the design, development and investigation of new technologies for telecommunications network management systems.

Department of Computer Science
University of Colorado Boulder
Boulder, CO 80309-0430 USA
May 5, 2012 (14:13)