Software architecture: object vs. process approach

A software architecture is a collection of independently-written components that interoperate through a set of connectors to function as an integrated system. Computation (components) and interaction (connectors) are separated from each other and this is considered to facilitate maintenance and reuse of large and complex software systems. In this paper, we describe and compare two different approaches to software architecture definition, namely the object-oriented and process-oriented approaches. The former defines components as computational objects that interoperate through communication objects. The latter, on the other hand, assumes that both components and connectors can be defined by their behaviour in terms of processes. Object- and process-oriented architectures can be illustrated with existing compositional models. This paper gives examples of two models: the Regis environment and the Wright language-based on objects and processes respectively-as we consider them to illustrate the basic features of each type with regard to explicit representation of independent components