Multimodal Archives: A Toolkit for Collecting Robotics and Other Complex Material in a Research Ecosystem

Robotics presents a significant challenge for archivists and information professionals. The field produces a large volume of multimodal products and by-products such as documentation, code, machines, and prototypes that depend heavily on in situ application and shift rapidly with changing technologies and research trends. This is compounded by the sheer complexity of the social networks, group processes, and interpersonal dynamics that drive research and development and take place in the context of multiple and interconnected teams, laboratories, communities, and institutions across the globe.

To explore this interconnected landscape, in 2019 Carnegie Mellon University launched The Robotics Project, a multi-phase collaboration between the University Libraries and the School of Computer Science. Our interdisciplinary team investigated the research ecosystem of robotics drawing on wide-ranging expertise emerging in modern libraries, archives, and museums (LAM) while conveying models and approaches for collecting and preserving complex robotics material. The first phase of The Robotics Project (2019-2022), funded by the Alfred P. Sloan Foundation, aims to demonstrate that well-designed archival practices incorporating interdisciplinary methods can illuminate a theme as complex as a mode of scientific inquiry.

In this toolkit, we present the results of the first phase of our investigation. We introduce an approach to gathering crucial precustodial data from community members, stakeholders, and informants that supports future collection development, appraisal, accessioning, and subsequent preservation and description work. We apply this approach to robotics at Carnegie Mellon University, but we feel it has applications beyond robotics, especially those characterized by tangible products and collaborative processes, such as engineering and architecture.