Dissertation talk: Collaborative Tools and Strategies for Data-driven Development Engineering

Lecture: Dissertation Talk: CS | May 13 | 10-11 a.m. | 405 Soda Hall

 Jordan Freitas

 Electrical Engineering and Computer Sciences (EECS)

There is simultaneously growing interest in open data and concern for privacy in data-driven international development projects. Current practices typically attempt to balance the two by manually removing personally identifying information and publishing a view of the remaining data. Both practically and theoretically–as demonstrated by Cynthia Dwork and the differential privacy literature–this approach fails to satisfy the open data objective of reusability, and fails protect privacy of individuals in the data. This thesis explores how to improve both the utility of shared data and how well privacy is maintained with strategically designed tools and methods. We propose and evaluate these tools and strategies for collaborative data management to help navigate tensions between open data and data privacy in the context of international development engineering projects. We first share the results of interviews with those working most closely with data in one subfield of development engineering, and analyze the results in terms of implications for developing data management and sharing tools. From there, we propose design requirements for workflow sharing tools based on four motivating use cases in different areas of development engineering, and present our implementation of a tool to satisfy these requirements. We then provide an overview of privacy considerations and our improvement mechanisms. Both our workflow sharing tool and privacy strategies enable more fine-grained control over data and code sharing with an emphasis on usability. Finally, we situate this work politically and socially in the context of international development.

 jordanfreitas@berkeley.edu