Metro SD

A multimodal trip planner for Greater Santo Domingo — including the informal transit that no existing app covers.

Interactive product mockup · Live demo

Santo Domingo has a metro, a cable car, and roughly 3.5 million people. It also has conchos and guaguas — shared taxis and minibuses that carry the majority of daily trips, with no fixed stops, no timetables, and no published routes. Just known corridors and word of mouth.

Existing apps cover the metro badly and the informal network not at all. I wanted to know how bad the data gap actually was, so I measured it rather than guessed.

So the country knows how to publish transit data — Santiago de los Caballeros has a complete GTFS feed with 168 routes and 7,391 stops, built by its city council with OMSA and INTRANT. The capital simply has nothing equivalent.

That gap is the product thesis: the hard part isn't routing, it's that the data doesn't exist.

Three screens: pick a destination, see the route drawn on the map, then follow the journey live with the current stop highlighted and your position advancing along the path.

The middle screen is the point of the whole project. A trip to Los Alcarrizos is walk → Metro Line 2 → walk → concho M-27B → walk, priced end to end at RD$65. The concho leg shows the real route code, the real fare, the operating union and the number of vehicles on the line — because that data exists officially and nobody has ever put it in front of a passenger.

Nothing here is invented, and where something is approximate the interface says so.

• 39 metro stations across Lines 1 and 2, with coordinates and ordering pulled from OpenStreetMap route=subway relations tagged operator=OPRET
• 189 urban concho and guagua routes with fare, operating hours, union and fleet size, converted from INTRANT's official CSV published on datos.gob.do
• Walking and concho legs routed along real streets via Valhalla on OpenStreetMap data
• Metro legs drawn as straight lines between stations — it runs underground, so following the road network would be wrong
• Base map: Esri Light Gray Canvas

The one honest gap: the drawn concho path is the most plausible road route between two points. The real geometry of those 189 routes does not exist in any public source — which is precisely the hole this project exists to fill.

Deliberately dependency-free. No build step, no framework, no package manager — open prototype/index.html and it runs. For a mockup whose job is to be shown to people on whatever device is in the room, that portability is a feature.

A few things worth pointing out:

Route drawing animates stroke-dashoffset on the SVG path Leaflet generates, so the line draws itself without a third-party animation library.

Live position interpolates along the routed polyline proportionally to distance, not to vertex count — otherwise the marker crawls through dense curves and teleports across straight segments.

Map centering projects the target point, offsets it by half the bottom sheet's height, and re-projects. Centering naïvely puts the current stop behind the sheet, where nobody can see it.

Journey animation carries a session token. Skipping ahead or leaving the screen mid-animation would otherwise leave two requestAnimationFrame loops fighting over the same marker.

prototype/          the mockup — open index.html, nothing to install
  index.html        three screens
  data.js           metro network + demo itineraries
  rutas.js          generated from the INTRANT CSV — do not edit by hand
  app.js            Leaflet map, routing, journey simulation
data/               source data (INTRANT CSV, OSM metro export)
tools/              build-rutas.mjs — regenerates rutas.js from the CSV
docs/               screenshots

Regenerate the route data after an INTRANT update:

node tools/build-rutas.mjs

Serve it locally:

python3 -m http.server 8000 --directory prototype

Handy for demos: #route=alcarrizos opens an itinerary directly, #route=colonial&live starts the journey.

This is a product mockup, not a shipping app. It exists to make a case: that a trip planner for Santo Domingo is worth building, that the informal network is the part that matters, and that the data problem is solvable — starting from official sources that are already public.