Production-grade engineering reference for music royalty distribution and metadata reconciliation. Live at www.music-metadata.org.
Modern royalty systems are distributed financial networks where ISRC and ISWC identifiers, DDEX ERN payloads, DSP usage feeds, and publishing splits all collide. When the data contracts break, money leaks — orphaned royalties, duplicate claims, misrouted statements, and audit findings that take months to unwind. This site is the field manual for keeping those pipelines deterministic, auditable, and correct at catalog scale.
- Label operations teams who need predictable reconciliation windows and clean audit trails.
- Royalty managers tracing every payout back to a single source usage event.
- Music technology developers building rights registries, matching engines, and settlement systems.
- Python ETL engineers shipping resilient, idempotent ingestion and reconciliation pipelines.
Fifty in-depth guides, each a production blueprint with copy-runnable Python (lxml, Pydantic v2, Polars, DuckDB, pyarrow, httpx), realistic royalty-domain data, and hand-drawn architecture diagrams. The material is organised into two connected areas:
Canonical identifiers and the standards that keep money flowing to the right people:
- ISRC → ISWC mapping workflows — tiered matching, conflict resolution, collision strategies.
- DDEX ERN 4.2 implementation guide — namespace-aware parsing and normalization.
- CWR registration & work matching — registering works with PROs and matching returned ISWCs.
- Metadata taxonomy best practices, cross-platform catalog matching, and fallback routing logic.
- Royalty audit trail reference — immutable logs, SHA-256 batch hashing, and end-to-end lineage.
- Security boundaries and troubleshooting metadata failures.
The resilient ETL layer that feeds reconciliation:
- DSP API polling strategies and webhook event ingestion.
- Automated CSV parsing and async batch processing for high-volume streams.
- Schema validation with Pydantic, incremental sync & change data capture, and error handling & retries.
- Data lake architecture and troubleshooting ingestion failures.
A fast, dependency-light static site — no tracking, no external runtime, fully self-contained.
- Eleventy static site generator (Nunjucks + Markdown)
- Hand-authored inline SVG diagrams and a custom "Fiery Ocean" design system
- Structured data (JSON-LD) on every page, with canonical URLs and Open Graph metadata
- Deployed on Cloudflare Pages
npm install
npm run serve # local dev server with live reload
npm run build # production build into _site/
npm run deploy # build + deploy to Cloudflare Pages- 🌐 Website: https://www.music-metadata.org
- 📦 Source: https://github.com/music-metadata
© Music Metadata — an open engineering reference for the music royalty community.