These Things Called Systems

The present paper proposes a way of reading telecommunications systems, software protocols and specifications in terms of collective imaginings of mobility. It reports on an ethnographic study of a large software system designed to control telecommunications infrastructure in rural Australia. Drawing on a notion of infrastructural work as involving an imagining of mobility, the paper develops two lines of inquiry. First, it asks how practices of imagining help us to understand the embodied practices and bewildering variety of artefacts circling around the development of a large distributed software system. These practices include the heavy use of citation, the interweaving of different tropes, metaphors and figures through the system, and the role of figures such as `system' and `process' in organizing work. Second, the present paper suggests that intensely invested issues in software production such as configurability, scalability, flexibility and distributing process imply that infrastructural design and implementation have a complicated relation to place. Via recent theoretical critiques of ethnography, it asks how notions such as locality and place can encompass mutually contextualizing movements and imaginings of movement.