Agile Methods: The Gap between Theory and Practice

Abstract

Since the software crisis of the 1960’s, numerous methodologies have been developed to impose a disciplined process upon software development. Today, these methodologies are noted for being unsuccessful and unpopular due to their increasingly bureaucratic nature. Many researchers and academics are calling for these heavyweight methodologies to be replaced by agile methods. However, there is no consensus as to what constitutes an agile method. An Agile Manifesto was put forward in 2001, but many variations, such as XP, SCRUM and Crystal exist. Each adheres to some principles of the Agile Manifesto and disregards others. My research proposes that these principles lack grounding in theory, and lack a respect for the concept of agility outside the field of Information Systems Development (ISD). This study aims to develop a comprehensive framework of ISD agility, to determine if this framework is adhered to in practice and to determine if such adherence is rewarded. The framework proposes that it is insufficient to just accept agile methods as superior to all others. In actual fact, an ISD team have to identify whether they need to be agile, and to compare this to their agile capabilities before deciding how agile their eventual method should be. Furthermore this study proposes that an agile method is not just accepted and used. Rather it may be selected from a portfolio of methods, it may be constructed from parts of methods, or indeed it may be the product of the ISD team’s deviation from a different method altogether. Finally, this study recognises that agility does not simply come from a method. In actual fact, a cross-disciplinary literature review suggests that it is important to classify sources of agility, which could be the people on team, the way they are organised, the technology they use or the external environment with which they interact. A three phase research method is adopted, incorporating a set of pilot interviews, a large-scale survey and finally, a set of case studies. The survey is intended to produce generalisable results while the case studies are carried out to obtaining much needed qualitative information in an emerging field where little is currently known.