Jeffrey R. Spies
Principal
221B Consulting
Rick O. Gilmore
Professor of Psychology
Pennsylvania State University
Openness with minimal restrictions is ideal but sometimes not feasible. For example, openness can be impractical or impossible when identifiable data about human subjects must be protected. Furthermore, the research community has a long way to go until openness is both normative and adequately resourced. Regardless of the practical or cultural barriers to unmitigated openness, it is still critical to promote reuse, reproducibility, and replicability to increase the quality and efficiency of the research process and the breadth and depth of research impact. Research data enclaves, where access is restricted to a trusted network of individuals, can foster these values while providing an iterative step toward wider openness.
In this talk, we will discuss our experience in designing, running, and now updating the technology behind the free, open-source behavioral video repository and restricted-access, authorized research environment Databrary, which has been in operation at New York University since 2014 and now includes 540 authorized institutions. In particular, we will focus on the pathways we are developing to promote wider openness wherever possible and our current curation-first design principle. More broadly, we will explore the benefits and challenges of such environments and discuss “almost open” as a strategy for the open community.