Evaluating Software Architectures for Stability and Evolution

Architectural stability refers to the extent an architecture is flexible to endure evolutionary changes in stakeholders’ requirements and the environment, while leaving the architecture intact. We report an early result in an emerging discipline of software engineering, the economic-driven software engineering research. We propose a novel approach for evaluating the stability of software architectures using a synergy of real options theory and viewpoints. Specifically, we contribute to a novel model that exploits options theory to predict architectural stability. The model is predictive: it provides “insights” on the evolution of the software system based on valuing the extent an architecture can endure a set of likely evolutionary changes. The model builds on Black and Scholes financial options theory (Noble Prize wining) to value such extent. We show how we have derived the model: the analogy and assumptions made to reach the model, its formulation, and possible interpretations. We refer to this model as ArchOptions. ArchOptions requires the estimation of several parameters. The estimation necessarily involves many partieseach with their own perspective on the system defined by their valuation objectives, assessment “regime”, skills, responsibilities, knowledge, and expertise. The problem associated with how to guide the estimation in this setting, we term as a multiple perspectives valuation problem. We describe the problem from a value-based software engineering perspective. We suggest a viewpoints-oriented framework as a solution. The framework appears to promote flexibility in implementing the model through aligning the estimation needs with value objectives, valuation strategy, expertise, tools, and/or the development (evolution) process. The contribution of this framework demonstrates the ability to evolve robust approaches in software engineering to solve problems in an emerging discipline, the value-based software engineering.