Issues in the adoption of object-oriented paradigm

Abstract

The author raises some of the issues that need to be addressed before the object-oriented (OO) paradigm can deliver on its promise. It is argued that, since OO technology lends itself naturally to encapsulation, large-scale systems should really be designed and implemented as a composition of small-scale subsystems. This has implications for the maintainability of the system and, in particular, its extensibility. While attempts are being made to define an OO analysis and design methodology, the work in this area is at a very primitive stage. Another problem is that one seems to be adapting the life-cycle models from the traditional software engineering paradigm. It appears that the OO paradigm is sufficiently different to require its own software development process. The author examines whether the OO software development process is sufficiently different to require a different set of tools and environment from those supporting traditional development. >