On the limitations of software process assessment and the recognition of a required re-orientation for global process improvement

  • Authors:

  • E. M. Gray;W. L. Smith

  • Affiliations:

  • I.S.D.M. Software Process Research Unit, Computer Studies Department, Glasgow Caledonian University, Cowcaddens Road, Glasgow G4 0BA, UK;I.S.D.M. Software Process Research Unit, Computer Studies Department, Glasgow Caledonian University, Cowcaddens Road, Glasgow G4 0BA, UK

  • Venue:
  • Software Quality Control
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

Through the years many techniques and tools have been tried and failed to deliver substantial global improvements to the software evolution process. The flavour of the early 1990s has been: attempted improvement using software process assessment methods such as CMM, BOOTSTRAP, etc. This paper represents a considerable and coherent critique of software process assessment as a process improvement driver, focusing on the concerns and perceived shortcomings present. A call is made to re-direct attention and resources towards understanding the true nature of software evolution and the software evolution process towards an adequate theory and practice for process improvement, all in order to save us from the computer. The important, possibly crucial role of feedback at many levels of the organizations that follow the processes that implement software development and maintenance is discussed. Utilizing some empirical findings in relation to software system evolution, it is suggested that the ignoring of feedback may be a major reason why the software process is so difficult to improve. It is argued that the clarification of such an important issue will help to achieve a significant step forward in process improvement, and through its exploitation, lead to significant improvements in quality, satisfaction and performance as well as more relevant education and training on process improvement.