Meeting summary
The group began by discussing ongoing work in the School of Computing, including hiring activities and administrative matters. The main focus of the meeting was on open data sharing and management, with Vahid presenting on the FAIR principles and their importance for research data findability, accessibility, interoperability, and reusability. The discussion covered various aspects of data management, including repository selection, privacy considerations, and the role of funding agencies in promoting good data practices, with Vahid offering to hold a separate session on version control tools and announcing the next meeting would focus on reproducible analysis workflows.
The group discussed ongoing work in the School of Computing, including Vahid's role in chairing a search committee conducting 22 Zoom interviews for hiring. They then transitioned to their third meeting about open data sharing and management, with Brad noting he had watched the recording of the previous session which was helpful for those new to pre-registration. The discussion concluded with a brief mention of attendance tracking requirements for December sessions.
The group discussed the value and uncertain future of a research funding program, with Brad noting it currently relies on personal support rather than institutional funding. Vahid and Kyle shared positive experiences with previous teaching fellowships. The main meeting focused on open data sharing and management, with Vahid presenting on the FAIR principles (Findable, Accessible, Interoperable, Reusable) and the importance of aligning research workflows with funding agency requirements from NIH and NSF. The discussion covered data repository selection and privacy considerations, highlighting that while major funders are increasingly mandating good data management practices, the digital ecosystem still presents challenges for research data integration.
Vahid presented on the FAIR data principles, explaining their importance for making research data findable, accessible, interoperable, and reusable. He discussed how to measure the fairness of data using tools like FUJI and FairShake, and highlighted that compliance with FAIR principles is increasingly required by funders like the NIH and NSF. Vahid also covered policy requirements and funder rules, emphasizing the importance of data management plans and open licensing for metadata.
Vahid discussed data management plans (DMPs) and tools like Argos and DSVizard, which help create structured, machine-readable DMPs. He explained the importance of having a data management plan, including details on data collection, format, and volume, and noted that these plans can be made public with a DOI for use in papers or grants. Kyle shared an example of a teaching repository, the Virtual Inorganic Pedagogical Electronic Resource (VIPER), which uses CC-BY-NC-SA licensing for teaching materials and aims to expand to include research data. Vahid and Kyle discussed the potential for creating similar repositories in their respective fields, with Vahid expressing interest in submitting grants to establish such a resource.
Vahid discussed various data repositories, including institutional options like the University of Michigan and discipline-agnostic platforms like Zenodo, which offers unlimited free storage up to 50GB per dataset and provides DOIs. He highlighted the importance of checking data sensitivity and controlling access for repositories containing personally identifiable information. Vahid also mentioned OSF as a comprehensive platform for managing the entire research lifecycle, integrating tools like Dropbox and GitHub, and supporting pre-registration.
Vahid presented on data repositories across different fields, highlighting specific repositories for STEM disciplines and tools like Papers with Code, which has been replaced by a new section on Hugging Face. He also discussed RE3Data.org and Fairsharing.org as sophisticated tools for finding and understanding repositories, noting that Fairsharing.org provides detailed information about metadata standards and repository policies. Kyle mentioned that Papers with Code had redirected to GitHub and expressed interest in exploring similar tools, while Vahid suggested identifying gaps in existing tools to potentially implement new solutions.