Build scalable, audit-ready Python automation that makes research data
Findable, Accessible, Interoperable, and Reusable — from the instrument to the institutional repository.

🌐 Visit the site → www.scientific-research-data-management.org

Scientific Research Data Management is a practical, engineering-first knowledge base for turning the FAIR Guiding Principles into working software. It is written for the people who actually operate research data infrastructure — research data managers, academic IT teams, Python automation engineers, and open-science advocates — and every guide pairs architectural reasoning with complete, runnable code.

Where most FAIR resources stop at policy and principle, this site treats compliance as something you engineer: a persistent identifier reserved before a payload is durable, a license resolved before a dataset is published, provenance written before an asset is citable. The result is a library of deterministic, reproducible workflows that hold up to institutional review — not manual curation that drifts out of date.

The site is organised into four in-depth sections, each branching into focused topic guides and hands-on walkthroughs:

• Core Architecture & FAIR Mapping — layered pipeline topology, metadata schema mapping and crosswalks, resilient API routing with circuit breakers, and how each FAIR sub-principle becomes an enforceable checkpoint at write time.
• Data Ingestion & Metadata Enrichment — parsing electronic lab notebook exports, pandas enrichment pipelines, Pydantic V2 schema validation, async batch processing, message-broker streaming, and pipeline orchestration.
• Open Science Infrastructure Planning — data-governance frameworks, funder-mandate alignment (NIH DMS, Horizon Europe), institutional repository strategy, and open-license configuration.
• Persistent Identifiers & Repository Integration — minting DOIs with DataCite, automating deposits to Zenodo, Figshare, and Dataverse, Crossref metadata deposit, DOI/ARK/Handle selection, and identifier reconciliation.

• Production Python, not pseudocode. Every implementation uses Python 3.10+ with full type hints and the Pydantic V2 API — idempotent consumers, dead-letter queues, exponential backoff, content-addressed deduplication, and two-phase identifier minting you can lift into a real pipeline.
• Grounded in the real standards. Dublin Core, schema.org, the SPDX License List, ORCID, the DataCite Metadata Schema, Crossref, ARK, and the Handle System are cited by name and mapped field-by-field.
• Diagrams that explain the hard parts. Every guide carries a hand-authored, theme-aware diagram of the pipeline, state machine, or decision it describes.
• Structured and discoverable. Pages ship valid JSON-LD (TechArticle, BreadcrumbList, HowTo, FAQPage) so the guidance is machine-readable as well as human-readable.

This is a fully static site with no build-time external dependencies at runtime:

• Eleventy (11ty) static site generator, with Markdown content and Nunjucks templates.
• KaTeX for server-rendered math and Prism for syntax highlighting.
• Self-hosted assets and a service worker for fast, offline-capable delivery.
• Deployed on Cloudflare.

npm install
npm run serve      # local dev server with live reload
npm run build      # production build into _site/

Issues and suggestions that improve the accuracy or clarity of the guides are welcome — open an issue with the page and the specific correction or gap.

Released under the MIT License.