Date/Time/Location: Tuesday, 2018.11.20, 1:15-2:30pm, School of Information, UTA 5.522
TITLE: Constructing Digital Infrastructures to Support “Open Science”: Governance Strategies from Open Source Scientific Software Organizations
The “open science” movement aims to support sustainable, reproducible scientific research by constructing digital infrastructures for widespread sharing of research processes and products. Organizations involved in the initiative build complex technical networks of open source software (OSS) applications, data repositories, journal databases, and other information technologies. However, information technologies constitute just one part of efforts to construct digital infrastructures; robust development also requires effective governance strategies for coordination and collaboration among participants (e.g., designers, decision-makers, and users). Governance strategies—or sets of policies, standardized practices, management principles, and organizational structures—often prove to be equal to or more important than technical capabilities in determining the success of an infrastructure. Effective governance provides participants with the direction and oversight necessary to achieve desired agreements about open science technology adoption, best-practices for technology use, and other policies ensuring robust, sustainable infrastructure evolution. Despite the scientific research community’s increasing reliance on digital infrastructures—particularly, OSS infrastructures—few studies document and compare governance strategies and evaluate their influence on scientists’ practices.
This talk reports on the results of a qualitative, ethnographic study comparing governance strategies across multiple OSS projects in ecology and astronomy and identifies strategies that elicit engaged, sustained contributions to OSS infrastructures. I will review existing research on digital infrastructure development, paying specific attention to what we know about governance best-practices. I will then note that governance strategies for OSS organizations are somewhat unique in that they must achieve voluntary contributions from participants who (a) are not paid employees of OSS organizations and (b) may not directly benefit from OSS contributions in their primary roles in academic institutions. Specifically, I will highlight the leadership strategies and incentive structures OSS organizations employ and how these mechanisms influence engagement and resistance in OSS infrastructure development. I will conclude by discussing how other infrastructure development efforts might learn from OSS governance strategies and how the findings can inform university and government policies that promote open science.
Dan Sholler is a Postdoctoral Fellow with the rOpenSci project at the University of California, Berkeley. He received his Ph.D. from the University of Texas at Austin School of Information in 2017 and was a member of the iSchool’s Information Work Research Group.
He studies the creation, development, and governance of digital infrastructures in a variety of industries, focusing on the impact of infrastructure use on the day-to-day work of users. In his dissertation work, advised by Dr. Diane Bailey, Dan studied the implementation of a federal program aimed at constructing an infrastructure to support data-driven healthcare delivery. The qualitative, ethnographic study examined the sources and outcomes of caregivers’ resistance to the program, focusing on the local realities of electronic medical records use and the resulting national-level movement to stall the federal program.
In his current work, Dan is working alongside co-advisors Dr. Karthik Ram and Dr. Carl Boettiger (UC-Berkeley) and Dr. Daniel Katz (University of Illinois Urbana-Champaign) to examine the development of infrastructures to support “open science.” The open science movement intends to make the processes and products of scientific research freely available to the public. Infrastructure development in scientific research requires voluntary contributions from disparate members of the scientific community, necessitating unique governance strategies that elicit engagement and manage resistance. By comparing governance strategies within and across scientific disciplines, the study seeks to document how organizations develop and implement strategies, explain how scientists are responding, and develop the findings into actionable recommendations for open science organizations, universities, and policymakers. Dan is a founding member of the Berkeley School of Information’s Algorithmic Fairness and Opacity Working Group and the Berkeley Institute for Data Science’s Best Practices and Meta-Research Working Group. He is also a fellow at the Berkeley Center for Technology, Society & Policy.
1:15pm to 2:30pm