Skip to content

Roadmap

Reviewed: 2026-09

The main question is how to make product data collection more scalable, collaborative, and reusable. Relab focuses on open, FAIR product-level disassembly data. Automation stays human-verified: software can suggest, people confirm.

The live work list is tracked in the public GitHub issues.

  • improve the data collection guides and keep the rest of the docs current
  • strengthen tests, especially around component hierarchies
  • simplify CSV and JSON export from live records toward dataset releases
  • improve admin and reference-data maintenance workflows
  • make camera-assisted capture easier to operate in repeated lab workflows
  • publish curated, versioned dataset releases separate from the live database, with stable identifiers and documented scope and licensing
  • improve API guidance and exports for analysis scripts, reproducible research, and LCA work
  • explore human-verified assistance for label reading, component suggestion, and quality control, with provenance records for assisted values
  • make reference-data search multilingual and meaning-based: embedding search over categories and product types next to the current full-text and fuzzy matching (accent-insensitive since 2026-09)
  • explore interoperability with semantic vocabularies and Digital Product Passport identifiers