Research proposal: a Bourdieusian ethnographic study of agile development methodologies

Overview of the Proposed Research Topic

During the relatively short history of software development, several development methods have been introduced, from the rather standard waterfall model derived from development methods that can be found in more ‘traditional’ industries (Royce, 1970), to solutions more adapted to the particular issues involved in software development such as user-centered design and agile development methodologies.

In order to deal with the particular constraints involved in the development of interactive applications, the latter tend to propose radical changes in the way project management is considered, combining a certain form of iterative development with a strong emphasis on communication. According to Cohen et al. (2004) agile development methods have underlying common values expressed in the Manifesto for Agile Development (Beck et al, 2001). An interesting aspect of this manifesto is that the focus is shifted from the “processes and tools” to “individuals and interactions”.

However, these new methods themselves are not exempt of problems, and a large number of software development projects encounter numerous difficulties regarding, among other things, cost estimation and time management, staff organisation and communication between the different actors involved in the project (see for instance Eckstein et al., 2006 or Taylor, 2001). In regards to these issues, a relevant, important question that will guide the proposed research is how well do recent software development methodologies such as agile development address the social dynamics involved in software development? Do the values they advocate corresponds to the social reality of today’s work practices in this particular sector?

The aim of this research is therefore to apply a powerful sociological analysis framework such as Bourdieu’s concepts of fields, practice, habitus and symbolic capital (Bourdieu, 2001) as a basis for the analysis of the social forces involved in the framework of particular methodologies used in software development, such as agile and user-centered development methods. This research would take place in the framework of a PhD undertaken by myself as the main researcher. The data collection method would mainly consist of a period of 6 months of participant observation in a software development company, which would be preceded by a series of short preparatory semi-structured interviews. In parallel to the qualitative data analysis, quantitative data would also be collected to support and elaborate on the findings.

Semi-structured interviews were preferred to other traditional preliminary methods such as surveys so that a direct contact would be made possible with the members of the potential companies in which the participant observation would take place, whereas the richness of the data provided by the use of participant observation as a data collection method makes it a better candidate than other methods, such as focus groups, or a series of interviews. Indeed, participant observation would allow a deep understanding of the actors’ behaviour and practices, hopefully as successful as Leslie Salzinger’s account of the very sexualized nature of work relationships between factory workers and their managers in Salzinger, 2003.

The main difficulty for pursuing this research would be to fine a suitable company for the period of participant observation. Indeed, one of the most important factors that need to be carefully taken into consideration is the representativeness of the chosen software development project, and to the same extent of the people involved in it, as an important aspect of this research being the possible generalisation of its findings in the framework of the design of a novel development method. From a more ethical point of view, it would also be necessary to ensure that the privacy of the participants is respected, by preparing if necessary a research contract.

Discussion: The Purpose of the Proposed Research

The main purpose of this project is to provide a useful analysis of the sociological aspects of software development, which could then be used for designing a new development methodology applicable in a real-world business context. Bourdieu’s theoretical and methodological framework provides a solid basis on which the particular issues involved in software development could be seen in a new light, hopefully providing new insights into the very particular work practices involved in this field. Indeed, software development projects involve different categories of individuals or groups, which can be considered as social agents having their own particular behaviours, interests, habits or point of views which have to be taken in consideration within the framework of an effective project development process, hence hinting the potential usefulness of a sociological analysis of such a process.
Thus, this research is particular in the fact that its aim is to inform the design of new development methodologies, which could ideally be in turn tested in a real work environment: Although this almost ‘theory-generative’ aspect could be emphasised, it could also be seen as an example of practical research (Taylor and Smith, 2008, p.30.)

Software development methodologies are a fertile topic of research and a lot of literature is available on the subject, taking the form of reviews of agile development methodologies as well as their impact on software teams such as “Agile Software Development: The People Factor” (Cockburn and Highsmith, 2001), moreover, sociological studies of certain aspects of some of these methodologies such as pair programming can be found in the work of Bryant et al (Bryant et al., 2007). An analysis of software teams based on game theory (Tenenberg, 2007) could be seen as rather similar in its approach to the research proposed here, although relying on a different theoretical framework. From a purer sociological point of view, the bourdieusian framework proposed for this research have been applied by Bourdieu himself and by other to the analysis of a wide range of different subjects, from the Kabyle population in Algeria (Bourdieu, 1979) to the French academic field (Bourdieu, 1990). More recently, his work has been applied to the analysis of the impact of new interaction technologies on society (Kapitzke, 2000).

However, there seem to be no application of bourdieusian theories to the analysis of software development practices, although it could be the way to gain a better understanding of these development methodologies and to address their pitfalls. Indeed, a survey from the British Computer Society found that only 13% of all IT projects were considered successful (Taylor, 2001). Although agile methods have been created to address this issue, they are not sheltered from problems entailed by their own “structure” and assumptions. For instance, individuals working in the framework of such a methodology tend to attach more importance to practices (such as, for instance, pair programming) instead of adapting the values of agile development to their particular work environment (Eckstein et al., 2006). Other issues that arise are difficulties to comprehend project constraints such as time, scope, quality or resources, or the lack of an effective “overall plan” for the project (Eckstein et al., 2006). This research would hopefully allow a deep and objective analysis of the different interactions between the agents involved in the process we are interested in, and could help to understand the reasons for the difficulties faced in practice regarding the implementation of agile development methods, and hopefully inform their improvement. The main aim of this study being to describe software development methods using the concepts provided by bourdieusian literature, it could also be seen as a development from the “generic towards the topical” (Hammersley and Atkinson, 1995, p.31) in its adaptation of a well established research framework.

Traditionally, ‘Bourdieusian studies’ have been relying heavily on fieldwork, although not exclusively, as quantitative data collection and the discourse analysis of newspaper interviews or articles also plays an important role in the data collection process of such studies. From a more practical point of view, interviews and participant observations seem to be particularly well adapted to the context of the study, as interviews would be hopefully relatively easy to arrange with software development companies, which often have links with the academic world. Another interesting aspect regarding data collection which could be explored in this particular context is the exploitation of the data gathered through the use of new communication technologies such as instant messaging, blogs, professional emails and online forums. Indeed, the ease of access and use of such data, which requires little to no transcription work, makes it an interesting candidate for discourse analysis and the study of communication between different actors, in the fashion of Hutchin and Klausen’s observation of the use of technology and of ‘task-related’ communication in an airline cockpit (Hutchin and Klausen, 1996).

The relatively wide objective of this research may require some adjustments if unpredictable practical problems have to be faced, which is relatively likely, as illustrated by Caroline Holland’s description of the constraints involved in her own field study (The Open University, 2006). Furthermore, the choice of the setting of the research is a crucial point for narrowing the focus of this study, which will be determinant for its potential for generalisation. Indeed, the rapid evolution of information technologies and of software development could harm the validity of the research’s findings in the future, an aspect which will have to be taken into consideration; however, the refining of the proposed development method and theory through additional research could be seen as a way to perpetuate the initial thrust of the proposed study by allowing the application of its findings to a wider context.

References

Beck, K., Beedle, M., van Bennekum, A., Cockburn, A., Cunningham, W., Fowler, M., Grenning, J., Highsmith, J., Hunt, A., Jeffries, J., Kern, J., Marick, B., Martin, R.C, Mellor, S., Schwaber, K., Sutherland, J. and Thomas, D. (2001) Manifesto for Agile Development.

Bourdieu, P. (1990) Homo Academicus, (French Edition), Paris, Les Éditions de Minuit, 1984. (English Edition) Polity press.

Bourdieu, P. (2001) Practical Reason, Polity press.

Bourdieu, P. and Wacquant, L. (1992) An Invitation to Reflexive Sociology, Chicago University Press 1992.

Bryant S., Romero P. and du Boulay B. (2007) Pair programming and the mysterious role of the navigator, International Journal of Human-Computer Studies.

Cockburn, A. and Highsmith, J. (2001) Agile software development, the people factor, Computer Publication.

Cohen, D. and Lindvall, M., and Costa, P. (2004) An introduction to agile methods, Advances in Computers, Elsevier Science (USA).

Coplien, J.O. and Harrison, NB. (2004) Organizational Patterns of Agile Software Development, Prentice-Hall.

Eckstein, J. (2006) Typical Pitfalls in Agile Development, Javapolis, Antwerp, Belgium.

Hammersley M. and Atkinson, P. (1995) Ethnography: Principles in Practice, London, Routledge.

Hutchins, E. and Klausen, T. (1996) Distributed cognition in an airline cockpit in Taylor, S. (2002) Ethnographic Research: A Reader, London, Open University Press in association with Sage.

Iannacci, F. (2002) The Social Epistemology of Open-Source Networks, Department of Information Systems, London School of Economics and Political Science.

Kapitzke C. (2000) Information Technology as Cultural Capital: Shifting the Boundaries of Power, Education and Information Technologies, Volume 5 Issue 1.

Lawrie, T. and Gacek, C. (2002) Issues of Dependability in Open Source Software Development, Software Engineering Notes vol 27 no 3.

Parliamentary Office of Science and Technology (2003) Government IT projects, POST report 200.

Royce, W. W. (1970) Managing the development of large software systems, Proceedings of IEEE WESCON.

Salzinger, L. (2002) Genders in Production: Making Workers in Mexico’s Global Factories in Taylor, S. (2002) Ethnographic Research: A Reader, London, Open University Press in association with Sage.

Sorensen, R. (1995) A Comparison of Software Development Methodologies, Software Technology Support Center.

Taylor, A. (2001) IT projects sink or swim, British Computer Society
Review.

Taylor, S. and Smith, M.J. (2008) D844 Study Guide, Milton Keynes, The Open University.

Tenenberg, J. (2008) An institutional analysis of software teams, International Journal of Human-Computer Studies 66(7), pp. 484-494.

Related posts:

1. Research proposal : A Foucauldian analysis of the evolution of the discourse about software development methodologies
2. What does the apparition of agile and user-centric development methodologies mean from an interdisciplinary point of view?
3. A short and biased history of software development methodologies
4. Towards explicitly sociologically-oriented software development methodologies: using social science to inform software design
5. Software as Discourse