Recent years have seen a proliferation of policies coming from research funders, universities, and publishers intended to prod scientists toward more proactive archiving, citation, and data sharing practices. The mere presence of a policy, however, doesn’t guarantee compliance, so the Foundation has looked to support technologies that make it easier to adopt best practices in research data and software management. The grant funds work by the Collaborative Knowledge Foundation and the California Digital Library to develop DataSeer, an open sourced rule-based platform that would be able to automatically identify data referenced explicitly or implicitly in a grant proposal, data management plan, or draft article and suggest appropriate repositories for deposit of that data. By providing nudges and suggestions at specific targeted moments in the research lifecycle (like the creation of a data management plan, submission of a grant proposal, or submission of a manuscript), DataSeer has the potential to substantively improve proactive data sharing and archiving by researchers, while reducing the costs of compliance checking for funders, libraries, and publishers.

The International Geo Sample Number (IGSN) grew out of an initial need to foster better citation of geological samples. As of mid-2018, there are over 6.5 million individual physical specimens represented within the IGSN, and a network of 25 IGSN “allocating agents” across five continents. After a number of years of growth within geoscience, the IGSN is confronted with increased interest from other disciplines; for example, the IGSN has already been used to register IDs for biological specimens and archaeological artifacts. Rather than encourage the development of a number of different discipline-specific registries, Lehnert and an international team of collaborators plan to redesign IGSN to support physical samples and specimens from across the sciences. Funds from this grant support technical development of the IGSN platform and a series of working meetings to bring together current IGSN registrars, other stakeholders, and persistent identifier (PID) experts to strategically plan the organizational and technical future of the initiative.

This grant provides three years of support for gatherings of WikiCite, a project within the Wikipedia ecosystem designed to both improve citation within Wikipedia and to expand the Wikidata project in ways useful to the scientific community. Grant funds will support a dedicated annual WikiCite meeting, as well as a series of smaller satellite meetings at other Wikimedia events. The organizers will also maintain a strong presence at other scholarly communication meetings, bringing the energy and technical sophistication of the Wikimedia community to bear on innovation in scholarly communication more broadly.

This grant funds a project led by Jonathan Crabtree, Director of Cyberinfrastructure at the University of North Carolina’s Odum Institute, to improve and expand the capabilities of the Dataverse open source data repository platform. Odum is responsible for executing and implementing the Replication and Verification Policy for the American Journal of Political Science (AJPS) and uses Dataverse as the underlying platform where authors who publish in AJPS can upload their data and software code to ensure results may be replicated. Because Dataverse was originally designed for data and not software, however, the process can be unwieldy and time consuming. Crabtree and his team plan to use the Jupyter computing platform and the open source software “containerization” toolkit Docker to create a “Confirmable Reproducible Research (CoRe2) environment” for Dataverse that would combine the ability to containerize computational research with communication and workflow tools. This would greatly speed and partially automate the process of verifying that submitted research results can be verified using the code and data uploaded. Grant funds will provide support for this project for three years.  years.

This grant funds a series of four projects by Mercи Crosas, Chief Data Science and Technology Officer at Harvard’s Institute for Quantitative Social Science to expand and improve software handling capabilities of the Dataverse open source data repository platform. First Crosas will integrate Dataverse with Encapsulator, an open source tool that allows the creation of a computational “time capsule” that preserves the exact computational environment used to conduct a piece of data analysis. Second, Crosas will create links between Dataverse and Code Ocean, a computational reproducibility platform that was spun out of Cornell Technion’s incubator program. Third, Crosas will develop a set of metadata versioning and exploration tools that will increase incentives for data curation by returning richer usage statistics to data providers and publishers. Finally, Crosas will model and pilot a fee-based curation service that would allow the sustainable scaling of data and code curation in Dataverse. This work, like all other development on and organizational innovation within the Dataverse community, will be freely available and useful to the dozens of other institutions running the software to power their own data archives.