Software Sourcing in the Age of Open: Leveraging the Unknown Workforce

P.J. Agerfalk, B. Fitzgerald, K. Stol (2015) Software Outsourcing in the Age of Open: Leveraging the Unknown Workforce. Springer.

From Chapter 1:

Outsourcing of the software development activity has been on the increase in recent years according to both US1 and European2 reports. However, in many cases outsourc- ing of software development, often referred to as global software engineering, has not delivered on its promise (e.g. Lings et al. 2007; Nakatsu and Iacovou 2009; Tiwana and Keil 2009; Ó Conchúir et al. 2009). The success of the open source software move- ment, which seems to overcome many of the challenges associated with global soft- ware engineering, has been an inspiration for a number of specific forms of software sourcing, including opensourcing (Ågerfalk and Fitzgerald 2008), innersourcing (Stol and Fitzgerald 2015) and crowdsourcing (Stol and Fitzgerald 2014a). By integrating the findings from these earlier studies, this book explores and compares these forms of open source-inspired sourcing. These novel approaches to software sourcing differ from traditional outsourcing in significant ways and little is known about how they can work in practice. Therefore, our goal is to provide research-based advice to managers and developers facing software sourcing decisions.

The conventional wisdom of software engineering suggests that given the inherent complexity of software, it should be developed using tightly coordinated, centralised teams, following a rigorous development process. In recent times, the open source phenomenon has attracted considerable attention as an agile, practice-led initiative that appears to address the three core aspects of the so-called ‘software crisis’: namely, high cost of development, long development time-scale and poor quality of final software product. In terms of development cost, open source products are usually freely available for public download. From the point of view of development time-scale, the collaborative, parallel efforts of globally-distributed co-developers has allowed many open source products to be developed much more

quickly than conventional software. Finally, in terms of quality, many open source products are recognized for their high standards of reliability, efficiency and robust- ness, and the open source phenomenon has produced several market leaders in their respective areas—Linux and Apache spring to mind. Indeed, these are known as ‘category killers,’ so called because their success removes any incentive to develop competing products. The open source model also seems to harness the most scarce resource of all—talented software developers, many of whom exhibit a long- standing commitment to their chosen projects. It is further suggested that the result- ing peer review model helps ensure the quality of the software produced (Feller and Fitzgerald 2002).

Previous research has argued for the importance of studying both the customer and supplier side of the outsourcing relationship (Koh et al. 2004). Nevertheless, most research on outsourcing tend to adopt the perspective of the customer (Gonzalez et al. 2006; Lacity et al. 2010). Interestingly, research on open source has all too often focused inwards on investigating the characteristics of the development process and projects, that is on the supplier side of the relationship, and far less has been conducted on the customer side, in the sense of investigating the consequences of the open source phenomenon for organizations, for example. However, in the three specific forms of software sourcing discussed in this book, namely opensourcing, innersourcing and crowdsourcing, the distinction between customer and supplier, or company and community as it may be termed in an opensourcing context, is a very important aspect that needs to be taken into account.

Furthermore, these three forms of sourcing differ from conventional outsourcing on a number of aspects. For example, in the case of traditional outsourcing, inner- sourcing and crowdsourcing, the locus of control is firmly with the customer company, and the community/supplier motivation is largely extrinsic for these three categories. However, in the case of opensourcing the locus of control is firmly with the supplier community and the motivation is largely intrinsic. On the other hand, if we consider the nature of the community workforce, in the case of outsourcing, this cohort is well known and is chosen on the basis of certifiably proven past performance. They also typically have some narrow but deep knowledge that is less economical to leverage within the customer company. On the other hand, in the case of opensourcing and crowdsourcing, the community workforce is largely unknown but possesses broad and deep knowledge thereby affording an opportunity for greater innovation for the customer company.