Skip to content
HN On Hacker News ↗

OpenClimbing turns a climbing guide into shared infrastructure

▲ 40 points 24 comments by zbycz 1w ago HN discussion ↗

Pangram verdict · v3.3

We believe that this entire text is AI.

100 %

AI likelihood · overall

AI
0% human-written 100% AI-generated
SEGMENTS · HUMAN 0 of 1
SEGMENTS · AI 1 of 1
WORD COUNT 815
PEAK AI % 100% · §1
Analyzed
Aug 18
backend: pangram/v3.3
Segments scanned
1 windows
avg 815 words each
Distribution
0 / 100%
human / AI fraction
Verdict
AI
Pangram v3.3

Article text · 815 words · 1 segments analyzed

Human AI-generated
§1 AI · 100%

A climbing guide can easily become a private database with a nice map in front of it.OpenClimbing is trying the reverse.When somebody adds or fixes a route, the project wants that information to remain useful even if OpenClimbing disappears. Climbing data lives in OpenStreetMap. Topo photos can live in Wikimedia Commons. The GPL-3.0 application becomes a specialized interface over shared resources.3 4Version 2.0, released on August 11, pushes that idea much further.2Edit the guide from the cragOpenClimbing no longer just displays points on a map.It can show areas, crags and routes, convert grades, filter sectors and draw climbing lines directly over photographs. Version 2.0 improves the route in the other direction too: users can edit information, upload a photo to Wikimedia Commons and draw the route over that image from the application.1The launch post reports almost 40,000 routes, more than 6,000 routes drawn on photos, over 1,000 climbing areas and 31 countries. Those are the project's own August 11 numbers, not an external audit.1The interesting part is who gets the result.In a conventional app, adding a route primarily improves the product that stores the contribution. Here, route data is written into OpenStreetMap using community schemas, while imagery can be published on Commons.3 4The application can disappear. The contribution can still be reused elsewhere.Open data still needs a specialist interfaceOpenStreetMap can describe an absurd number of things. That does not make a generic map editor the ideal place to document a 6b with three bolts, a shared anchor and a crack that becomes miserable after rain.OpenClimbing therefore builds a domain-specific interface.It presents route hierarchies, preferred grading systems, photos, rock information and relationships between large climbing areas, individual crags and routes.1 4Version 2 also adds a proper logbook. Climbers can record ascents, styles and partners and view statistics.1Not every piece of user data is therefore public in the same way. The shared geography belongs in OSM and Commons, while personal log features need application storage. “Open data” should not be inflated into “literally everything lives in OSM.”The design principle is narrower: information describing the shared physical place should return to shared infrastructure whenever possible.Offline PDF is a surprisingly important featureCrags are extremely competent at removing mobile reception.OpenClimbing 2.0 can export an area or crag as a PDF guide containing the map, routes, grades and route lines over photographs.1That sounds almost retro next to an interactive web map. It is exactly why it works.A field tool cannot assume its best network conditions will follow the user into the field. A PDF can sit on a tablet, be printed, sent to a partner or stored before the trip.Offline use here is not a secondary accessibility option. It matches the environment where the information is needed.The terrain can tell you where the shade will beSome of the v2 changes show what happens when a map becomes deeply specific to a practice.OpenClimbing added sun and terrain shadows to estimate whether a crag will be in sun or shade. Its release history includes a custom DEM ray-marching shader for casting those terrain shadows.2A generic POI application is unlikely to care.The same goes for Fontainebleau grades, sit starts for bouldering, climbing-specific filters and lines drawn over topo photos.1 2The project even includes an experimental in-browser bolt detector to help when preparing photo topos. The author is careful to call it experimental.1The architecture is therefore not “generic UI over generic data.” It is common data underneath, opinionated tooling on top.Open does not mean correctA shared database still needs maintenance.Somebody has to map the route, confirm its position, choose the right OSM relationship, upload appropriately licensed imagery and correct mistakes. That community labor is much less visible than the PDF button.The reported 40,000 routes do not mean every crag is complete, current or locally verified. Reusable data can still be wrong.It does have one practical property: correction does not belong to one company.A club, climber or another developer can improve the same underlying object, and several interfaces can benefit from the correction.A guide can be a view instead of a siloThat is what makes OpenClimbing more interesting than another sports app.The product treats the guide as a specialist view over a shared corpus. Its value comes from editing ergonomics, photo topos, filters, shade, PDF export and personal tracking rather than from locking the essential climbing information inside one service.The separation transfers to many field tools.An application does not need to own the map, image and description to have a strong identity. It can be useful precisely because it knows what to do with information other people can reuse.At a crag with no signal, that philosophy eventually becomes a PDF.AuthorHugo MarchalGame engine dismantlerHugo believes there is no better way to understand a game than rebuilding its engine by hand, ideally while sacrificing a weekend. He writes his own tools because he never found anyone else's weird enough. His sweaters are on sale; his projects never are.