The materials science working group focuses on the pragmatic aspects of sharing and publishing materials science data and analysis processes. The group will bring together materials scientists, computer scientists, repository owners, publishers, and others that are interested in the challenges, technologies, and approaches associated with computational materials science. The group is particularly focused on methods to publish data and analysis processes in ways that are easily sharable and reproducible. Thus, the group will cover aspects such as data management, analysis, computational modalities, curation, persistent identifiers, citation, reproducibility, distribution, archival, etc.
The goals of the group will be to discuss these broad aspects, with the aim of defining use cases to meet current and future needs, to prototype integrations with external systems, to review and evaluate software developed by our partner collaborators. The group has a strong focus (and input) into the the Whole Tale and Materials Data Facility (MDF) projects and will aim to use these technologies as well as define future trajectories.
Questions to consider:
- What data is used? Where is it from? Where is it stored?
- What types of data can be shared easily right now?
- How is data shared?
- What platforms/repositories exists?
- Is there any notion of interoperability?
- What packaging formats are used? E.g., BagIT
- What types of analyses are used (HPC, HTC, data analytics?)
- Where are these analyses run
- Are common registries of tools used? What registries exist?
- How are computational processes shared/disseminated
- What packaging formats would be useful? E.g., RO
- What types of identifiers are required?
- What curation workflows are needed?
- What metadata should be associated?
- What access control/embargo is required?
- What in papers could be reproduced?
- How would you like to reproduce?
- What other working groups from other projects have overlap?
- Is there potential for engagement?