Designing in and for Complex Coordinative Spaces: Interagency Traffic Management and the Evolution of Public Service Work Infrastructure

Abstract

Major traffic incidents such as multi-car crashes can spawn countless human and machine processes, as public agencies activate to get to the scene, save lives, move debris, and when possible, get traffic safely moving again. How should this work be organized, and what can make it happen more effectively? Disciplines such as design studies and CSCW (computer-supported cooperative work) study how to create and situate technology systems into varied contexts of use. The idea of "work infrastructure," originally developed by Pipek and Wulf, offers a generative framework to understand how people appropriate and improve the technology systems they use to do their jobs, and offers a useful antidote to simple binaries between "designers" and "users." However, complex inter-organizational contexts of shared work and technology use, like the inter-agency management of major traffic incidents, are not as well understood by these fields. This is in part because of the complexity of gaining access to the sites of work and decision-making, and of analyzing so many different participants simultaneously. There are great opportunities for CSCW, organization studies, and design research to better understand how different public agencies with different arrangements of technology systems can come together to create shared, multi-organizational systems for managing complex work, such as that of traffic incident management. In this dissertation, I present one such project, the Virtual Coordination Center (VCC). From 2018-2023, designers and researchers from the University of Washington and software design consultancy Pariveda worked with Seattle-area government agencies to help them improve their coordination during and after major traffic incidents. A large component of this work was the development of the Virtual Coordination Center, a collaborative web-based system that all participating agencies could view and use to share information about unfolding incidents. As a participating designer/researcher on the VCC, I conducted numerous design and prototyping activities, ethnographic workplace observations, software quality assurance tasks, and many other forms of participation. In this dissertation, I similarly use a wide variety of qualitative analytical and empirical methods to analyze the VCC project. I employ a sociomaterial framing to identify key concepts for understanding how this project worked and how its phenomena connect to larger questions of design, work infrastructure, and coordination. The resulting analysis has yielded three prongs of conceptual findings: a significant extension and refinement of the dynamics of work infrastructure embeddings in inter-organizational work settings; an analytical framing of collaborative system project stakeholders as coordinative units; and an analysis of designers as stakeholders themselves, using discursive techniques to steward project dialogue and advance conceptual and design goals. In the Conclusion, I synthesize findings from the three different prongs of analysis, and in so doing, critique and extend the notion of ``work infrastructure" originally described by Pipek and Wulf to better account for inter-organizational system use. In short, while work infrastructure is characterized in terms of its complexity, pervasiveness, and flexibility, conceptual work around this idea doesn't fully take into account the longer-term infrastructural processes of an organization. The de facto work infrastructures that get embedded over time--those that are actually within the reach and understanding of real users--don't behave the same as the formally documented versions that organizations manage. These "holographic" work infrastructures are moving targets. Moreover, in inter-organizational work contexts where there is no single locus of analysis, design work will visibilize specific parts of cross-organizational influences and information-sharing, and make them susceptible to shaping and changing. I further examine the "point of infrastructure," a key moment in the development of work infrastructure that can lead to "design-in-use." I identify, based on the previous findings of the dissertation, processes that can inhibit or mask points of infrastructure, particularly in inter-organizational collaborative projects; suggesting that designers and designer-users may have to look more broadly and deeply for these potential points of infrastructure, and be willing to unpack apparent smooth functioning to do so. All of these conceptual findings are linked to specific recommendations for researchers in designers in similar spaces.