Paper

Workflow for rebuilding the Wonton paper figures and PDF.

Dataset

The shared runtime lake is the dataset. Lake jobs are debug or release slices, not alternate evidence.

Inputs:

• LAKE_DB_PATH or the shared-lake default resolved by paper/build_figures.py
• dossiers/wonton-soup/paper/build_figures.py
• dossiers/wonton-soup/paper/main.typ
• dossiers/wonton-soup/paper/refs.bib

Build output goes under dossiers/wonton-soup/paper/artifacts/.

Build Figures

LAKE_DB_PATH=/path/to/shared/lake.duckdb \
uv run python dossiers/wonton-soup/paper/build_figures.py \
  --out-dir dossiers/wonton-soup/paper/artifacts

Expected artifacts:

• paper/artifacts/fig17-followup-provider-splits.svg
• paper/artifacts/fig16-ged-bimodality.svg
• paper/artifacts/fig18-followup-basins.svg

Compile

typst compile \
  --root . \
  dossiers/wonton-soup/paper/main.typ \
  dossiers/wonton-soup/paper/artifacts/main.pdf \
  --font-path addenda/typst-field-manual/assets/fonts

Optional Slices

Optional lake presets for debugging and release prep:

• dossiers/wonton-soup/analysis/lake/presets/79_wonton_paper_closeout_v1.json
• dossiers/wonton-soup/analysis/lake/presets/80_wonton_paper_program_v1.json

They do not replace the shared lake for paper analysis.

Gates

Before publishing:

1. Claims and run counts match the shared runtime lake.
2. Scheduler/controller claims come from the same dataset as the figures.
3. Figures are refreshed after new runs or backfilled analyses enter the lake.