Reproduce the import. Bring contributions into application and testing. Build reviewed consensus and return actionable recommendations to BAFU.
This public Sentier project connects partner evidence, Brightcon hackathon work, candidate improvements and the discussion around BAFU/UVEK. It spans releases and preserves the provider's canonical source alongside traceable, separately identified candidate layers.
Import & use · Findings ledger · Evidence & testing · Contribute · Consensus
Sentier importers parses the provider release. Inventory, methods and mappings hold the derived artifacts; sentier-brightway builds the pinned application baseline.
git clone https://github.com/sentier-dev/sentier-brightway.git
cd sentier-brightway
uv sync
uv run sentier-brightway coverage
uv run sentier-brightway files --out ../bafu-baselineRecord the release, input hashes, manifest pins and residual coverage. The build includes its pinned mappings and transformations; treat it as a derived build. Obtain raw exports under the provider's terms for regeneration. The redistribution basis for the full published inventory remains open in issue #1; read data and attribution before sharing builds.
| Step | What belongs here |
|---|---|
| Canonical import | Source release, identifiers, hashes and reproducible import path |
| Partner contributions | Findings, proposed mappings, models and scenarios with evidence |
| Application & testing | Separate candidates, named exchange changes, coverage and score comparisons |
| Reviewed consensus | Actual reviewer decisions, objections and their disposition |
| Recommendations to BAFU | Agreed findings, delivery evidence and provider responses |
A finding starts with the source release, dataset code, what you measured and how. Read the ledger, then open an issue. Reproduce observations against the provider's source files before concluding that BAFU's data is wrong. An importer defect belongs to the component that introduced it.
A contribution layer gets a record, isolated candidate build and validation plan. A reconstruction of an aggregated inventory is a new inventory claim; representation parity, inventory correction and forecasting require different tests.
- The hackathon copied-process contribution remains proposed. Its round-trip run passed across 766 families; units/mass balance, score parity and held-out forecasting validation remain open.
- The lci-bafu-catalog findings connect source observations with their discussion threads. Importer parity, aggregation, maintenance and method/build identity have separate evidence and follow-up requirements.
- The measured BAFU export audit passes identity/count checks and core-gas characterization coverage on the pinned 420,063-row registry; it does not claim raw-source or score parity.
- Shared tooling checks named row-count deltas, required characterization, exact artifact-hash metadata and portable summary identity. Passing a supplied audit does not establish source parity or partner agreement.
python3 scripts/community.py check
python3 -m unittest discover -s community/tests
python3 scripts/build_checks.py path/to/build-audit.jsonThe synthetic audit example is a format demonstration, not a measured BAFU build.
Open the BAFU / UVEK site. The responsive site shares Sentier's visual identity with the AGRIBALYSE review preview. Its public pages cover import, findings, application tests and consensus. The Documentation workflow checks pull requests and publishes updates from main to GitHub Pages. Build locally with uv run --no-project --with markdown==3.7 scripts/build_site.py; see site maintenance. Generated pages contain coordination material only, with no inventory downloads.
Original code and documentation are MIT. Source datasets retain their own terms, attribution and modification requirements. Keep restricted exports and confidential partner evidence local. No new reviewer consensus, sent recommendation or provider endorsement is implied by the workspace. See the workflow, recommendation tracker and terms.