I Got Tired of Google Maps... So I Built My Own

Video Description

Ever wondered how navigation apps like Google Maps actually work? In this video, I break down the entire routing process in a simple, intuitive way — no heavy math and almost no coding. We’ll start with how computers see maps as graphs made of nodes and edges, how routes are calculated, and why apps optimize for time instead of distance. Then we’ll move from a toy example to real OpenStreetMap data — the same open-source map used by Uber, Amazon, and even Pokémon GO. With OSRM, the Open Source Routing Machine, I’ll show how we can build our own routing engine on top of raw map data, define custom rules, and even override the algorithm itself. This leads to the main project: a toll-road optimizer that chooses tolls only when they genuinely save time. To make it work, I had to classify French toll booths into entry, exit, and passthrough types, extract real pricing data, deal with international borders, fix dozens of edge cases manually, and watch OSRM hilariously escape toll roads through tiny forest paths. Once everything worked, the app could calculate the true “cost” of any road by combining travel time with the time you’d need to work to pay the toll — a surprisingly powerful and realistic model. And the best part? All of this was built with open data and open-source tools, proving once again that a single person can create something useful that giant corporations never bothered to build. Beyond routing, this video is also a reminder of why open, transparent, privacy-respecting tools matter today more than ever. Big tech doesn’t always act in our interests, but with open-source projects — from OpenStreetMap to self-hosted apps and local AI models — we can build systems that work for us. Thanks for watching, and if you enjoyed this deep dive into how code interacts with the real world, make sure to subscribe for more projects like this. 00:00 - Introduction 00:39 - A tiny bit of theory 01:31 - Wikipedia for maps 02:35 - Inside the digital map 03:32 - The shortest route isn’t the fastest 04:40 - Gaming the algorithm 06:00 - Be your own fate-maker 07:03 - Idiots in cars 08:02 - Finally, some practice 09:19 - The problem with Google Maps 09:51 - The theory of saving money 11:34 - How Google Maps knows about traffic jams 13:08 - Stuck inside the fortress 14:55 - Making greedy capitalists lose some money 16:32 - Roadside robbers 18:22 - AI is useless… again 19:14 - A mind-blowing solution 22:25 - Reality bites 25:09 - I created a monster 26:03 - Experiments with water 26:58 - Getting the pricing data 28:05 - We are the power 29:55 - Do not submit to the system Music: Hampus Naeselius - Always Beside Max Anson - Definition of Color Real Heroes - Latency Clarence Reed - The Drive Home Rachel Sandy - Impossible Theory Johan Landqvist - Open Fields Video Courtesies: What is an OpenStreetMap Mapping Party？(MapGive) OpenStreetMap： The map that saves lives ｜ CNBC International Navigation System Design at Uber： Maps + Interaction — Cady Wachsman Mexico City： Dijkstra Pathfinding Algorithm on Real Maps (DansBox) Trucks hitting bridge compilation part (1) (FULL ACCESS) Traffic Calming is Everywhere in the Netherlands (Not Just Bikes) Autoroute en flux libre Paris Normandie - Phasage des travaux (version de Mai 2024) (Autoroutes Sanef) Crossing with truck through the Puymorens tunnel in France direction Toulouse (WORK & TRAVEL) drive to mont blanc tunnel (Chris Fairbrother)