Advanced Information Systems Engineering / Lecture Notes in Computer Science Bd.4495 (PDF)

1 Introduction (p. 17)

Agile software development methods have emerged rapidly since the mid 1990’s. The roots of agile methods are placed far beyond the last decade, however [1]. Industry has been keen on adopting agile solutions in recent years. Large software corporations such as Microsoft and SAP have announced publicly of their plans to adopt agile methods. Information systems build in embedded devices such as cars, telecom systems or consumer electronics systems present no exception to this.

Large companies such as Philips, Nokia, British Telecom, to name a few, are either already adopting or plan to adopt agile software solutions as their means to tackle software related challenges. A recent Forrester survey1 suggested that one out of seven software companies already use agile methods. Agile methods have been criticized for the lack of solid, scientifically valid empirical data to back up their claims [2, 3].

Yet, the situation is far from unique in the fields of software and information systems development. Fenton [4] argued that while software professional often seek for rational basis for making a decision about which development method they should adopt, the basis for such rationalization is completely missing. Fenton went as far as claiming that the "methods introduced continue to be based on more on faith than on an empirical data". Based on recent industrial attention, we can suggest that the lack of data has not slowed down the adoption of agile methods.

There is no agreed definition of agile methods in information systems or software engineering fields. In industrial engineering science the situation is quite different. They have proposed 17 competing definitions of the concept of agile manufacturing [5]. In software engineering, the concept of agile software is most often related to the elements presented in the Agile Manifesto (http://www.agilemanifesto.org). To some extent, we can assert that