Continuous evolutionary one-step-ahead testing

  • Authors:
  • Mechelle Gittens;Hanan Lutfiyya;Mike Bauer

  • Affiliations:
  • University of Waterloo;University of Western Ontario;University of Western Ontario

  • Venue:
  • ACM SIGSOFT Software Engineering Notes
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

The traditional software development life cycle considers testing to be an activity that occurs between the implementation phase of development and software release [4]. With this approach any testing subsequent to release is done in reaction to failures reported by software users. The realities of software in operation however causes questions about this approach to arise. Adams [1] showed that organizations developing significant software applications often provide several fixes after their software has been released as the result of errors found in the field. This work also showed that the most serious and frequently recurring errors are usually found by users soon after a product has been released. These are referred to by Adams [1] as virulent errors. The negative effects of remaining defects implies that post-release activities should be proactive. These post-release activities must include continued testing by the vendor to find errors even after release. This paper proposes a solution to this requirement.