| The requirements process is considered to be one of the most important parts of information systems development because it is through this process that system functionality is determined. However, much research and practice has shown it to be one of the most error-prone aspects of software development and implementation. Past research directed at this problem has focused on technical solutions. However, Howcroft and Wilson (2003) call for a "deeper understanding of the inherent conflict within the systems development process", arguing that "far from being an intellectual diversion, this constitutes a prerequisite for improving the rather poor record of systems development" (p.21). This critical ethnographic study in two information systems projects in a single organization aims to address this question by studying communication in the requirements process though the lens of Habermas' Theory of Communicative Action.; The research identifies several constraints on communication including bounds on who was allowed to participate and on the extent to which participants could provide input or challenge others. Several conditions that allowed these constraints are explored: the distribution and construction of knowledge, the construction of the requirements process as a technical process, project team control over the process, the organizational motivations that were embedded in the process, the imposition of a method that brought with it assumptions not consonant with the organizational context and existing organizational relations of power which privileged some voices and marginalized others. Based on these results, a Habermas-inspired requirements approach is proposed.; At one level, the results provide insight into the constraints and conditions that are at play in requirements processes. At another level, these findings are a cautionary tale of the difficulty of engaging full and equal participation. At yet another level, the findings provide a theoretical basis on which to evaluate requirements processes. Finally, these findings set the stage for a reconsideration of the requirements process, proposing one which is more equal and open, and makes possible the building of information systems which more closely meet the needs of users. Both researchers and practitioners should, therefore, find these contributions valuable. |
