Distributed Infrastructuring and Innovation: an ethnographic enquiry into collaborative modes of work in an internet of things ecosystem

Abstract

Emerging low-power wireless networks are being used for a range of data collection systems such as asset tracking, environmental monitoring, smart agriculture and smart city facilities. The relatively low costs of hardware components, modular network architectures and open standards are allowing a diversity of new actors to engage with the construction of ‘internet of things’ (IoT) networks and applications. Various branches of research within management studies, critical theory, design theory, feminism and science and technology studies (STS) have explored collaborative modes of technology development among heterogeneous groups of actors and addressed questions of how and why users become involved in technology development. There is however scant empirical and theoretical work on the involvement of ‘users’ and other non-conventional actors in contemporary data-oriented infrastructures such as the IoT. Conjointly, most policy roadmaps concerning the rise of pervasive data networks rely primarily on industry-oriented analyses and quantitative forecasts and hence remain blind to the involvement of non-corporate actors in the shaping of technological futures. Building on an STS-inflected framework, this study contributes to bridging this gap with a micro-level enquiry into collaborative work practices in the realm of the IoT. This thesis explores the case of The Things Network, an initiative with the mission to build low-power wireless networks in a decentralised fashion with a strong reliance on geographically dispersed contributors. The initiative is far removed from traditional top-down infrastructure implementation strategies and faces a range of ambivalences related to organisation, growth and sustainability. The study is concerned with the questions of what types of work, social organisations and artefacts are subsumed in the emerging ecosystem? why/how contributors organise and operate local networks? whether and how control is exerted by the project owners? and how the uneven actions of users and other non-conventional actors are implicated in the generation of technical improvements and outcomes? The methodology comprised a multi-site ethnographic exploration over two and a half years with the practitioners contributing variously to the construction of data networks and the development of IoT solutions within the initiative. An ecological analysis is developed, drawing on theories and concepts from infrastructure studies and the social shaping of technology framework. The evolution of the initiative is traced throughout the stages of inception, early scaling up and global expansion. Through casting low-power networks as ‘data infrastructure’, the analysis foregrounds the challenges and dilemmas associated with scaling up in the context of decentralisation. The concept of ‘distributed infrastructuring’ is proposed as a means to capture the orchestration of the piecemeal work of disparate and dispersed actors operating autonomously with a common network architecture. The findings suggest that this mode of infrastructuring is symptomatic of an industry trend towards an increasing fragmentation and distribution of professional development activities among a range of actors. We conclude that policy and practice would benefit from a nuanced recognition of the diversity of contributions, positionalities and preferences in the broad landscape of data-driven technologies.