Specs

SSR-friendly React app shell and file-based routing

UI runtime

Build tool and dev server

Typed codebase — strict type-checking is being rolled out (some files still opt out)

Styling via semantic tokens in src/styles.css

Server-state cache and loaders

Accessible component building blocks

Icon set used throughout the UI

Form state and schema validation

Rich-text editor for blog posts in the admin

Sanitizes rich-text HTML before render

Interactive maps of stations and routes

Database, email/password auth, row-level security, and media storage

Serverless edge runtime for SSR and server functions

Package manager and JavaScript runtime

Base map tiles for the interactive station/route maps

Historical station and route data compiled by hand from archival maps, timetables, and secondary literature — see /resources for the reading list

The data-consolidation pipeline that turns archival sources into the published dataset and Supabase

Reads and tile-crops very high-resolution historic map scans for close reading and georeferencing

Numerical arrays for image and geometry processing

Writes georeferenced GeoTIFFs of historic maps from ground-control points

Queries surviving and disused railway geometry to trace historic corridors onto the map

Our own pure-Python georeferencing (map pixels to coordinates from graticule control points; affine + thin-plate-spline, leave-one-out validated) and a shortest-path OSM corridor matcher, with a coarse-to-fine tool for reading map scans

AI-assisted product development environment

AI software architect and pair-programmer — planning, data modeling, implementation, and code review — and reading, transcribing, and georeferencing historic map scans

Unit and component test runner

End-to-end test runner (auth, blog, and admin flows)

Lint and format