Don’t archive you assets¹ — frontier them. Research is a living process.² Even when a research project is “finished”, is it really?

model-memo The things of concern¹ are materials, processes, measurements, and ingredients. Materials are output by processes and are subject to measurements. A process may take materials as ingredients. Things are recorded in three ways: as templates, as specs, and as runs. You can record a template for a thing – what might be the case. You can also record a spec for a thing – what is intended.

For each layer in the structured-content stack,¹ from least to most volatile (i.e. domain modeling ⟶ content design ⟶ interface design), draft successive model expressions,² from most to least ambiguous (as many expressions as needed to move confidently to the next stack layer). Structured-content-stack breakdown: Domain model (object types and relationships) Content content model (content types and attributes) content s

The key technical foundations for FAIRifying data are (1) ubiquitous persistent identifiers; (2) rich controlled metadata; and (3) granular programmatic access. These foundations provide a basis for FAIR data infrastructure. This note is inspired by Rory Macneil’s recent interview with Sharif Islam on the FAIR Data Podcast, published on 2022-12-21. In particular, I expand on the Q&A segment starting at PT14M10S.

How does a Research Software Engineer (RSE) — often responsible for developing infrastructure to manage and share digital research objects (data, models, code, notebooks, workflows, etc.) — get from “Yes, FAIR sounds great, but how?” to “I better understand what the FAIR principles really mean and how I can put them into practice.”? I hope the diagram below can help.

I’ve been trying to grok architecture patterns as presented by Percival and Gregory¹ to support domain-driven design and event-driven microservices with Python. I hope you find the diagram below useful.

The “one platform¹ to rule them all” is unlikely to be realized for scientific research in any domain. Rather, instead of small and numerous on-premises silos for data + code + compute, we are on track to achieve large and somewhat less numerous cloud-based silos.² What’s the alternative?

I have sought to identify and enumerate core FAIR-enabling services. I attempted a five-week experiment to expand on my tentative list, but I did not complete it. The list wasn’t compelling for me. I have been brewing an updated list of core FAIR-enabling services, which I hope to be less bombastic about.

Don’t. Identifiers should be opaque. If you’re given an owl:sameAs assertion from a party you trust, use that. If you need to mint surrogates because what you’re given aren’t Globally Unique, Persistent and Resolvable Identifiers (GUPRIs)¹, either house your inheritance as local parts/suffixes in your global namespace, assert datatype properties to record the historical correspondence, or both.

Is a metadata record “almost” expressed in the same language you used for your filter criteria?