CSRS

Summary:  CSRS is a computer-supported software review system that enables declarative definition of review processes and provides instrumented data collection facilities.

Formal technical review (FTR) is a cornerstone of software quality assurance. However, the labor-intensive and manual nature of review, along with basic unresolved questions about its process and products, means that review is typically under-utilized or inefficiently applied within the software development process.

There are two primary goals for the CSRS research project:

  1. The design, implementation, and evaluation of a computer-supported software review system (CSRS) that enables declarative definition of review processes and provides instrumented facilities for gathering and analyzing review data. For this goal, our approach has been to incrementally design and evaluate the CSRS system over a period of several years, along with public release of the system for use by other research and industry sites.
  2. The use of CSRS as a laboratory testbed for studying differences between review methods under highly controlled conditions. For this goal, we have carried out one controlled experimental study thus far, involving a comparison of review methods designed to assess the impact of the review meeting on review cost and effectiveness. We further extended this research with a comparative analysis of this study with a similar study performed by Adam Porter at the University of Maryland. See the Research References section below for details.

Our results indicate that CSRS increases both the breadth and depth of information captured per person-hour of review time, and that its design captures interesting measures of review process, products, and effort.

Our study of review meetings found that meetings do not significantly increase defect detection effectiveness, but do increase overall review costs. Interestingly, our study also found that the student population studied greatly preferred the meeting-based method over the non-meeting-based method, and believed that the meeting-based method led to higher review quality, even though there was no empirical evidence to support this opinion!

Principal researcher(s): Danu Tjahjono

Publications: Citations and publications

Project Page: Not available

Status: Active development 1991 – 1996.