Creating a Minimal Viable Identifier (minid) infrastructure for uniquely and unambiguously identifying research data products.
Biomedical researchers generate many digital objects of different types, sizes, and quality. Current practice provides for publishing “final” data products into repositories, obtaining digital object identifiers (DOIs) for those products, and recording DOIs in registries such as BioCADDIE. But what about the many intermediate, often transient, data products that are created during a research project? They are surely not all worthy of preservation and indexing. But imagine if every such data item had a unique identifier that anyone could resolve to access some basic metadata. Researchers, and their colleagues, would then be able to refer unambiguously to any data item, embed references to any data item in many different records (e.g., audit logs, provenance records), and know how to access data where access is possible. Moving to a culture where we pass around identifiers to data and know what they mean can have a transformative effect.