Seminar: Dan Suthers, "Tracing Interaction in Distributed Socio-Technical Systems" (4/2/2009)

Tracing Interaction in Distributed Socio-Technical Systems
Dan Suthers

Laboratory for Interactive Learning Technologies
University of Hawaii
http://lilt.ics.hawaii.edu/

Information technologies that power complex social systems are rapidly scaling up and becoming integral to daily life around the globe. Online social networks have seen explosive growth, and emerging cyberinfrastructures for collaboration and networking promise a revolution in research and education. However, we still do not know how such socio-technical networks actually work, or how our designs affect their functioning. A barrier is the nature of the data, which derives from synchronous, quasi-synchronous, and asynchronous interaction of hundreds or thousands of people distributed across different media and tools. Study of such networks requires tracing thousands of individual paths of activity as well as identifying the myriad of occasions where these paths intersect and affect each other. To drive the evolution of emerging technologies, we must be able to answer fundamental research questions concerning how our technological designs encourage synergistic transformations of people and ideas.

Work within our laboratory (the Laboratory for Interactive Learning Technologies) spans some of the diversity of the study of socio-technical systems. We have studied co-present and distributed interaction via various synchronous and asynchronous media and at scales including pairs, small groups and online communities. Our research methods have included experimental studies, ethnographic approaches, and adaptations of ethnomethodological methods. Although our work began in experimental and micro-analytic studies of small group interaction, we seek to scale up the study of mediated interactional accomplishments to larger groups and temporal frames, while continuing to apply a mixture of methods for discovery, hypothesis generation and hypothesis testing.

Because of the diversity in our own work, we have encountered the need for a common foundation for theoretical and methodological dialogue; a microcosm of the need that also exists in the field of socio-technical systems as a whole. Although prior efforts by others and us have met immediate and local research needs with specialized tools and mixed methods, we seek a globally integrative solution because a common analytic foundation will enable cross-disciplinary and cross-methodological collaborations that lead to improved understanding of and ability to design new, innovative socio-technical environments.

Responding to this need, we are developing a unifying framework for analysis of mediated interaction. The conceptual foundation of this framework that serves as the basis for construing interaction as a common object of study is a building block of interaction we call "uptake". The representational foundation of this framework is a data model--the contingency graph--that offers a common representational basis for diverse analytic methods applied to various media and interactional situations. This data model captures interactions that are distributed across space, time, and media, and supports multiple analytic methods for exposing sequential patterns of interaction and their dependencies on features of the technological infrastructure. We propose to develop a suite of software tools that operates on this data model to trace out the movement, confluences, and transformations of actors, artifacts and ideas.

The seminar presentation will describe the motivations for this work (as outlined above), describe the analytic framework in more detail, provide several examples of analyses inspired by the framework that we have already conducted with ad-hoc tools, and discuss work planned for the future, for which we are seeking funding and student and professional collaborators.