A framework for finding where a new city can actually get entitled.

Geodesic is a research repo for one question:

Where can a master-planned, mixed-use city get a legally durable yes before the project has spent real money?

This is not a "where is land cheap?" project and not a generic rezoning guide. It is an elimination engine for killing weak jurisdictions early, using state law, county structure, and observable political signals.

Nothing here is legal advice. Every statutory claim stays provisional until local land-use counsel reviews it.

Why this exists

Most site-selection work starts too late and falls in love with acreage too early.

Geodesic starts one layer up:

• Can this state even create a durable entitlement instrument at this scale?
• Can a county say yes without handing veto power to another body?
• If politics turn hostile later, is the remedy an injunction or just damages?

The core durability standard is the breach test:

If a later actor revokes the instrument, can the project force performance, or only get paid after the damage is done?

That framing is the backbone of the repo. Full context lives in framework/00-the-question.md.

At a glance

Area	What it does
framework/	The operating method: question, heuristics, state screen, county screen, signal inventory, entitlement paths
ops/	Hand-off-ready game plans per step: tasks tagged by execution mode (automated / AI-assisted / human / phone), data-source registry, tooling build triggers
results/	Applied outputs: state files, county files, scorecard
research/	Dated research notes and raw supporting material
archive/	Original draft material preserved for reference
ROADMAP.md	Current execution plan and log

The method

Geodesic works as a funnel, not a thesis. Each stage exists to kill candidates cheaply before the next, more expensive stage begins.

flowchart LR
    A["🏛️ State law screen<br/><small>vesting · referendum reach<br/>mini-NEPA · boundaries</small>"]
    B["🗺️ County screen<br/><small>topology · capacity<br/>friendliness grade</small>"]
    C["📊 Signal inventory<br/><small>power × cost<br/>calibrated by pilot</small>"]
    D["⚖️ Path selection<br/><small>entitle · annex<br/>incorporate · district</small>"]
    E["🎯 Ranked<br/>jurisdictions"]

    A -->|"survivors"| B -->|"survivors"| C --> D --> E

    A -.->|"kill"| K1[" ✕ "]
    B -.->|"kill"| K2[" ✕ "]
    D -.->|"kill"| K3[" ✕ "]

    style A fill:#16324f,stroke:#5b8c5a,color:#f3f6fa
    style B fill:#16324f,stroke:#5b8c5a,color:#f3f6fa
    style C fill:#16324f,stroke:#5b8c5a,color:#f3f6fa
    style D fill:#16324f,stroke:#5b8c5a,color:#f3f6fa
    style E fill:#5b8c5a,stroke:#16324f,color:#f3f6fa
    style K1 fill:none,stroke:#c4677a,color:#c4677a
    style K2 fill:none,stroke:#c4677a,color:#c4677a
    style K3 fill:none,stroke:#c4677a,color:#c4677a

Loading

The repo is organized to match that sequence:

1. 00 - The Question — what we're actually asking, and the breach test
2. 01 - Heuristics — kill-rate ordering, infrastructure multiplier, scale thresholds
3. 02 - State Screen — four statute-readable variables that kill entire states for $0
4. 03 - County Screen — the phase-gate funnel and county topology
5. 04 - Signal Inventory — every signal scored on power × cost
6. 05 - Entitlement Paths — the path is a choice variable, not an input
7. 06 - Project Spec — the envelope being screened for (draft; placeholder numbers)
8. 07 - Federal Layer — §404, ESA, NEPA handles: minimize federal hooks

Current posture

Category	Current state
Framework	v1.3 — signal inventory (04) calibrated through Validate/Commit + Sweetwater Depth 1; adds overlay topology / contiguous-component testing to prevent aggregate-acreage false passes
Stage	Pilot complete through Stage 4 synthesis; Depth 1 underway for Sweetwater
Pilot states	Wyoming, South Dakota, North Carolina
Pilot counties	Sweetwater County WY, Fall River County SD, Jones County NC
Next move	Principal scale/phasing input, principal-facing upgrade/results report, then widen the funnel
Automation	Deferred except for proven Tier-A/overlay helpers; build order lives in ops/step-3

See the active plan in ROADMAP.md.

Reading paths

If you are here to understand the idea:

• Start with 00 - The Question
• Then read 02 - State Screen
• Finish with 05 - Entitlement Paths

If you are here to execute the workflow:

• Read the full framework/ sequence in order
• Open the output templates in results/states/_template.md and results/counties/_template.md
• Track progress in results/scorecard.md

If you want the research trace:

• Start with research/2026-06-09-entitlement-durability-synthesis.md
• Then inspect the raw notes in research/raw/

Design principles

• Kill cheap. Use the lowest-cost signal that can eliminate a jurisdiction.
• State law first. Many hard constraints sit above the county level.
• Durability over friendliness. A warm meeting is not a durable instrument.
• Optionality always. Keep multiple live jurisdictions in play.
• Earn automation. Do manual work first; code only what proves useful.

What this repo is not

• Not a legal opinion
• Not a policy advocacy project
• Not a land acquisition playbook
• Not a general-purpose urbanism wiki

It is a decision system for reducing entitlement risk before capital gets trapped.

Status note

This repo was restructured on 2026-06-09 into framework/, results/, research/, and archive/. The current version is meant to be readable, inspectable, and easy to extend as the first pilot runs.