Working software over comprehensive documentation \u2013 Rationales of agile teams for artefacts usage

Gerard Wagenaar,Sietse Overbeek,Sjaak Brinkkemper,Garm Lucassen,Kurt Schneider

doi:10.1186/s40411-018-0051-7

Abstract

Agile software development (ASD) promotes working software over comprehensive documentation. Still, recent research has shown agile teams to use quite a number of artefacts. Whereas some artefacts may be adopted because they are inherently included in an ASD method, an agile team decides itself on the usage of additional artefacts. However, explicit rationales for using them remain unclear. We start off to explore those rationales, and state our primary research question as: What are rationales for agile teams to use artefacts? Our research method was a multiple case study. In 19 agile teams we identified 55 artefacts and concluded that they in general confirm existing research results. We introduce five rationales underlying the usage of artefacts in ASD: (1) Adoption of ASD leads to agile artefacts, (2) team-internal communication leads to functional and technical design artefacts, (3) quality assurance leads to test-related artefacts, (4) agile teams impose governance on their own activities, and (5) external influences impose user-related material. With our contribution we substantiate the theoretical basis of the Agile Manifesto in general and contribute to the current research with regard to the usage of artefacts in ASD in particular. Agile teams themselves may from this research extract guidelines to use more or less comprehensive documentation.

Highlights

Fifteen years have passed since the Agile Manifesto (Beck et al 2001) was published
We add another piece in the puzzle for the hybrid agile / waterfall research arena, when we show that agile teams in practice use all kinds of artefacts, inherent to an Agile software development (ASD) method or not
We started our research from the observation that recent research reports that organizations blend waterfall and agile system development methods in all kinds of hybrid development varieties

Summary

Introduction

Fifteen years have passed since the Agile Manifesto (Beck et al 2001) was published. After half of this period, some 71⁄2 years after its inception, the research community had lavished attention on issues related to agile software development (Dybå & Dingsøyr, 2008). But five years later a research gap with regard to the implications of agile information system development on the coordination, collaboration and communication mechanisms within agile teams was still noted (Hummel, 2014). The Agile manifesto itself values “working software over comprehensive documentation” and emphasizes “The most efficient and effective method of conveying information to and within a development team is face-to-face conversation”.

Objectives

Methods

Results

Conclusion