Ingenieria Software

Information and Software Technology 46 (2004) 473–489 www.elsevier.com/locate/infsof

Communication and co-ordination practices in software engineering projects
´ Ian R. McChesneya,*, Seamus Gallagherb
School of Computing and Mathematics, Faculty of Engineering, University of Ulster at Jordanstown, Shore Road, Newtownabbey BT37 0QB, UK b Division of Management and Information Systems, Schoolof Management and Economics, The Queen’s University of Belfast, University Road, Belfast BT7 1NN, UK Received 6 May 2003; revised 9 October 2003; accepted 14 October 2003
a

Abstract In this paper we report on the investigation, description and analysis of communication and co-ordination practices in software engineering projects. We argue that existing models of the software process do notadequately address the situated, day-to-day practices in which software engineers collectively engage, yet it is through these practices that effective co-ordination is achieved. Drawing on concepts from organizational theory, we describe an approach for studying co-ordination activity in software engineering and the application of this approach to two real-world software projects. We describe keyco-ordination issues in these projects and discuss their implications for software engineering practice. q 2003 Elsevier B.V. All rights reserved.
Keywords: Software engineering; Communication; Co-ordination; Software process; Software project management; Human factors

1. Introduction A major challenge in managing the software process is achieving the effective co-ordination of its constituentactivities, products and people [7,12,17,18,37]. The field of software engineering has seen a number of strategies for improving co-ordination at the organizational, process and tool level. For example, at the organizational level, the Chief Programmer Team arrangement [3] drew on the structure of a surgical team to minimize communication overhead while maintaining the conceptual integrity of thesystem design. At the process level, structured methods such as Yourdon Systems Method [41] and the Unified Software Development Process [20] have sought to integrate best practice, procedures and techniques for common software engineering tasks, achieving co-ordination through knowledge sharing and standardization. At the tool level, software engineering environments e.g. [2,39] have providedtechnological support for co-ordinating complex
* Corresponding author. Tel.: þ44-28-9036-8881; fax: þ 44-28-90366859. E-mail addresses: ir.mcchesney@ulster.ac.uk (I.R. McChesney), s.gallagher@qub.ac.uk (S. Gallagher). 0950-5849/$ - see front matter q 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.infsof.2003.10.001

software builds and version control. The ubiquity of large and complexsoftware systems is evidence of the success of previous strategies. However, as demand for software systems increases, software technologies change, and functionality becomes more complex, so communication and co-ordination breakdowns continue to manifest themselves [13,21,37]. There remains considerable scope for improvement both in the practice of co-ordination in software engineering projects and ourunderstanding of it. In this paper we argue that improved co-ordination will be achieved through a better understanding of how software engineers actually conduct their day-to-day technical work. In particular, we argue that there is a gap between existing process-oriented methods for describing the software process and the situated, day-to-day activities in which software engineers engage whendeveloping systems. A better understanding of co-ordination strategies and practices is the first step in bridging this gap. To this end we present an analytical framework for investigating coordinated activity in software engineering and report on its application in two real-world projects. The remainder of the paper is organized as follows. The next section outlines the relevant literature and...