Milti-party Information Systems development : the challenge of cross-boundary collaboration

Abstract

Information System (IS) development has always involved multiple stakeholders, but the IS literature has traditionally focused on collaboration between two groups: technical developers and users. With changes in technology, growth in IS outsourcing, and the increased strategic role of IS applications, the number and diversity of stakeholders on IS projects has increased significantly. People from different disciplines-business strategists, technologists, graphic designers, marketers-spanning multiple organizations and hierarchical levels need to find ways of working together. In this thesis I draw on data from an ethnographic field study of a multi-party IS development project and an R&D group in an Internet consulting organization to understand how people collaborate across a variety of boundaries and how such collaboration shapes the IS product. This thesis develops a theoretical framework for understanding the boundaries involved in IS development, and how they shape and are shaped by agents' practice. Multi-party collaboration can be understood through Bourdieu's practice theory lens as a struggle of agents situated in nested and intersecting industry, organization, profession, and project-based fields of practice. I analyze organizational discourse in these fields through a communicative genre lens, which focuses on socially recognized and habitually enacted types of communicative actions. The analysis shows how relational boundaries are represented, produced, reproduced, and transformed through the discourse in the various fields of practice.

(cont.) The enactment of communicative genres on an IS project shapes a collective reflection-in-action spiral, which involves iteratively reflecting on objects produced by others and either adding to or challenging past experiments in future actions. The relative input of different agents to the collective reflection-in-action spiral shapes the IS product-the object that results from the last experiment. Whose input is reflected on and preserved in future experimentation is at stake in the project field. Attaining such stakes shapes agents' relative positions in the field, that is, the boundary power dynamics. The framework contributes to both research and practice by increasing our understanding of current practices on collaborative multi-party IS development projects and by offering insights into the tradeoffs involved in such practices and opportunities for innovation and improvement.