What’s Wrong with Requirements Specification? An Analysis of the Fundamental Failings of Conventional Thinking about Software Requirements, and Some Suggestions for Getting it Right

Abstract

We know many of our IT projects fail and disappoint. The poor state of requirements methods and practice is frequently stated as a factor for IT project failure. In this paper, I discuss what I believe is the fundamental cause: we think like programmers, not engineers and managers. We do not concentrate on value delivery, but instead on functions, on use-cases and on code delivery. Further, management is not taking its responsibility to make things better. In this paper, ten practical key principles are proposed, which aim to improve the quality of requirements specification.