Global Utilities

Seminars - Abstract

Department of Computer Science & Computer Engineering

Topic:   Software Evolution
Speaker:   Professor Lehman, School of Computing Science, Middlesex University.
Date:   19-04-2004
Time:   3:00 PM
Venue:   HS2 223
Abstract:   Interest in the software evolution phenomenon and its exploitation is spreading rapidly. Though first discussed by Lehman and Belady in the early 1970s, its reality and importance has only recently been more widely acknowledged. It is, however, now receiving widespread attention in, for example, a number of international networks and other groupings. The phenomenon results from an intrinsic need and desire to continually adapt software systems to a changing world, improve performance and functionality, ensure continuing reliability and economic viability. There are two separate, though related, aspects to its study. On the one hand it is of considerable interest to understand what software evolution is, why it occurs, why it is inevitable and what impact it has in practice. On the other hand industry needs to know how to recognise a need or opportunity and then plan, manage, control and, above all, implement evolution predictably, reliably, economically and on-time. Industrial and academic interest has, in the main, focussed on the latter aspects, the means whereby evolution is achieved. Full achievement of this goal requires, however, sound understanding of the nature and characteristics of the phenomenon. Sound answers to these questions will speed up process improvement to help meet the challenges posed by the ever increasing dependence of 21st century society on computer systems that function satisfactorily despite an ever changing world. The talk introduces and discusses aspects of software and the software process that provide a basis for answers to the study of these questions and of research addressing them. The talk will introduce one underlying sources of continuing evolution, identify its true nature, present examples of evolutionary behaviour, briefly discuss aspects of the modelling and interpretation of that behaviour, illustrate the dynamics of the evolution process and indicate how the results lead to a theory of software evolution. This provides background for an outline of practical implications, management and technical guidelines and rules that follow from the more theoretical discussion outlined above. These, in turn, suggest improvements to the software evolution process to achieve more reliable, timely, economic evolution that maintains software systems satisfactory in a dynamic, always changing, world.
Content Approved by: Head of School
Page maintained by: Applications Programmer
Last Updated: 14 October, 2009

Seminar Information