16 Association of Research Libraries Research Library Issues 298 — 2019 New Collaboration for New Education: Libraries in the Moore-Sloan Data Science Environments Jennifer Muilenburg, Research Data Services Librarian, University of Washington, and Visiting Program Officer, Association of Research Libraries Judy Ruttenberg, Director, Scholars and Scholarship, Association of Research Libraries In 2014, the Gordon and Betty Moore Foundation and the Alfred P. Sloan Foundation partnered to invest $37.8 million across three US universities to build what they called Data Science Environments in order to “demonstrate how an institution-wide commitment to data science can deliver dramatic gains in scientific productivity and lead to significant new discoveries.”1 The Moore-Sloan Data Science Environments (MSDSE) were a five-year experiment “to better understand how best to bring together interdisciplinary people within institutional environments in order to provide them with the resources, freedom, and interconnected networks necessary for science to flourish.” As that five-year experiment and its funding wind down, how can these and other research institutions—and their libraries—advance data science education? What did the MSDSE sites learn about the pedagogical relationship between data science methods and traditional disciplines, and intra-institutional partnerships? ARL staff spoke with key personnel at the three MSDSE sites: New York University (NYU), UC Berkeley, and the University of Washington (UW). Of course, these three institutions are not unique in dedicating resources to data science many other universities in Canada, Europe, and the United Kingdom have also launched degree and research programs focused on data science and analytics.2 This grant, however, structured as it was across three institutions, provided a unique opportunity to assess each university’s goals against the outcomes achieved. This article highlights not only their individual learning, but also draws overarching conclusions of value to all research libraries engaged with data science or looking for a pathway to doing so.