Wiki-Editable Cycling Routing Map With Crowdsourced Street-Level Hazard Annotations Cyclists Can Actually Edit

mobile app real project ••• trending

An Ask HN 'what do you wish existed' commenter named the gap precisely: cycling-friendly turn-by-turn that lets the people riding the routes flag specific stretches as death traps. Today's options either don't model bike-specific safety at all or model it only via aggregated heatmaps and historical crash data the user can't correct. A Strava-heatmap line going down a six-lane stroad without a shoulder is technically the most-ridden way home from downtown — and also the way people get killed. The wedge is editability: an OpenStreetMap-style wiki layer on top of cycling routes where any cyclist can annotate a segment ('this looks fine on the map but the right hook risk at 4pm is brutal') and that annotation immediately influences routing for everyone else, including for the OP's specific city.

builder note

Forking the OSM data model (specifically cycleway:safety_note=*) gets the civic crowd on side and avoids reinventing tile infrastructure. The hardest part isn't tech, it's seeding density — a routing app with three annotations per city is useless. Strategy: launch in five cities with active cycling advocacy chapters, partner with BikeOttawa/Bike Pittsburgh-style nonprofits for the seed corpus, and only then expand. Avoid the 'global day one' trap that killed every previous community-routing attempt.

landscape (5 existing solutions)

Routing engines and incident maps both exist. Nobody has fused them. The wedge product is a routing app with a one-tap 'flag this segment' action that mints an OSM-compatible note, weighted by reporter reputation, with optional category (door zone, right-hook risk, blind corner, broken glass cluster). Pull from BikeMaps.org and ICBC/NHTSA crash data on day one to seed; let users add the annotations Strava and Komoot won't. The civic-good positioning makes this fundable as a non-profit or a B-corp without requiring monster scale.

Komoot / RidewithGPS / Strava routing Best-in-class route engines using OSM road tags and Strava heatmaps. No first-party way for a rider to mark a segment as 'never route through here, here's why' that influences future routing.

BikeMaps.org Crowdsourced bike incident reporting (collisions, near-misses, hazards). Closest to the ask, but it's a data-collection platform — the data does not feed into anyone's turn-by-turn routing.

OpenStreetMap cycling tags + OpenCycleMap OSM has rich cycling tagging (cycleway=, bicycle=designated, etc.) but contributing requires an OSM editor learning curve. There is no segment-level safety overlay editable from a route view.

CityMapper Multi-modal trip planning including bikes. The HN commenter explicitly names it as 'pretty good' but uneditable — the gap they want filled.

Local advocacy maps (BikeOttawa, Cycling Embassy of Great Britain, NYC bike maps) Excellent quality where they exist. Hyper-local, volunteer-maintained, and only cover a handful of cities. No federation, no shared schema.

sources (3)

hackernews https://news.ycombinator.com/item?id=45509339 "A city map for cyclists that is updated and reviewed by other cyclists. How to get from A to B in a given city, the safest way possible, but with maps that can be edited like a wiki. CityMapper is pretty good but there's no way for me to correct a route because one stretch of it is actually a death trap that no one should bike on." 2025-10-07

reddit https://old.reddit.com/r/cycling/comments/1mfwslt/i_was_near... "Down a fairly narrow 30mph country lane, about an hour into the ride, so 5:30am, round a blind corner towards me came a souped-up red Mercedes doing what must have been 60mph at least, missing me by no more than a foot." 2025-08-02

reddit https://old.reddit.com/r/cycling/comments/1kvl7sj/if_you_cou... "If you could wish an app into existence that would solve a key cycling problem for you, what would it be?" 2025-05-26

cyclingmappingopenstreetmapsafetywikicivic-tech