Philadelphia — design & roadmap
Planning artifacts for porting the NYC school-outcomes platform to Philadelphia (School District of Philadelphia + Philly's charter sector, ~300 schools total). Same architecture as the NYC build: a TypeScript Next.js front-end backed by Prisma/Postgres, fed by a sidecar Python analysis pipeline, audited by a read-only Python validation harness. The methodology (peer groups, residual-z outliers, comparability regimes, four-check verification) is lifted from KX's spec wholesale — only the source layer changes.
Contents
| File | What it is |
|---|---|
00_overview.md | Why Philly, scope, ownership, in-repo vs new-repo decision. |
01_inventory.md | Philadelphia data-source inventory — what's publicly downloadable. |
02_gap_analysis.md | Philly source layer vs. NYC source layer, axis by axis. |
03_integration_plan.md | P0–P5 capability sequence; analogue of KX's C0–C5. |
04_contribution_plan.md | Branch/PR workflow for the Philly build. |
05_validation_spec.md | The four-check (i)–(iv) taxonomy adapted to Philly sources. |
sources.yaml | Source registry stub — same schema as verify/sources.yaml. |
Conventions inherited from docs/design/
- Validation taxonomy: (i) completeness · (ii) base case (DB vs exact source) · (iii) spot check (DB vs official per-school report) · (iv) computed values. Same definitions as
verify/METHODOLOGY.md. - Branch-per-capability PRs; never commit directly to
main. - Additive + feature-flagged changes; the NYC site keeps working at every commit.
- Python sidecar for analysis; read-only Python for verification. TS for the app.
Suggested reading order
00_overview.md → 01_inventory.md → 02_gap_analysis.md → 03_integration_plan.md → 05_validation_spec.md → 04_contribution_plan.md → sources.yaml.
Status
Draft. No Philadelphia code or data has been ingested yet; this folder is
planning + scaffolding only. Implementation begins after the open decisions in
00_overview.md §3 are settled.