middle-passage 0.1.0

Protection-first forensic GIS scaffold for Middle Passage memorial research

middle-passage

"The ocean kept receipts. We are learning to read them."

Open-source forensic data toolkit for the Middle Passage Forensic Recovery Project.

This package is a protection-first research scaffold for integrating voyage records, oceanographic assumptions, bathymetry, and memorial GIS outputs. It models survey-priority zones, not certainties, and every output is designed to carry uncertainty, source references, and non-disturbance language.

License: Creative Commons CC BY 4.0
Purpose: Humanitarian. Scientific. Memorial.
Status: Phase 0 scaffold with synthetic-safe defaults.

Principles

• People, not cargo.
• Ancestors, not artifacts.
• Probability zones, not treasure maps.
• Public methods, careful precision.
• Protection before spectacle.

Install

After the first PyPI release:

pip install middle-passage

Until then, install from GitHub:

pip install "middle-passage @ git+https://github.com/Yufok1/middle-passage.git@main"

For local development:

pip install -e .

CLI

Generate a synthetic-safe sample GeoJSON:

middle-passage sample --output sample.geojson

Model survey-priority zones from a voyage CSV:

middle-passage model-deposits --voyages voyages.csv --output zones.geojson

If no voyage file is provided, the CLI uses a clearly marked synthetic sample for development only.

Data Sensitivity

The package supports data sensitivity labels:

• public: safe public metadata or generalized zones
• generalized: public map data with reduced precision
• restricted: high-precision working data for qualified review
• do_not_publish: sensitive records or coordinates that should not be released

Precise burial-site candidates should not be published casually. Release policy must be governed by descendant/community review, legal review, and protection risk.

Current Scope

This first package cut includes:

• flexible CSV voyage loading
• simple voyage filtering
• current-vector drift estimation
• conservative descent/scatter model
• survey-priority zone creation
• GeoJSON FeatureCollection export
• CLI and tests

It does not claim to identify remains. It creates reproducible, inspectable research objects for future expert validation.

Release

This package should be published from GitHub with PyPI Trusted Publishing. See docs/PYPI_RELEASE.md for the release checklist.