Reflections on a large software development project

Abstract

This paper is an attempt to extract some lessons,with some subsequent and very personal reflections, from my involvement with the development of the ICL VME Operating System. They represent personal views and personal reflections and in no way necessarily reflect those of the company. I present them in the hope that they are useful inputs to the current plethora of noise surrounding software engineering and its application. In particular I am concerned that the experiences gained from the lengthy usage of an early support environment should yield some input to the present round of investments in Integrated Support Environments. The paper attempts to justify the notion that a framework fashioned for Process and Architectural support should be the starting point for the development of the core of such an Environment. “It takes longer than you expect, even when you take into account Hofstadter’s Law” [hofl.