About

It started with signing up for a backyard ultra. The format is well-known : run a 6.7 km loop every hour, as many hours as you can take — last one standing wins. Before committing, I wanted to know what 6.7 km actually looked like inside my own runs. Not an average across the whole route, not a Strava segment drawn for someone else's KOM, but 6.7 real kilometres, isolated somewhere inside a cool training run I had already done.

I opened Strava, Garmin Connect, Komoot. They all happily show a track, compute averages, list official segments. None of them will let you simply say : « take 6.7 km from the middle of this one and give me only those numbers ». At best you draw a public segment for the community ; at worst you download the .gpx and open it in a text editor. Not a real answer.

I've been a web developer since 2007. So one evening, after a run, I opened an editor and started coding. A few weeks later, GPXchunk was displaying twelve measurements for any portion of any track. And since the tool was useful to me, I figured I might as well make it useful to others.

Thomas Winkin, based in Aywaille, Belgium. Web developer since I finished my studies in 2007 — nearly twenty years in the field, with a frontend specialisation in more recent years.

I've been a trail runner since 2020. A few markers, for context :

• I finished the Infernal Trail des Vosges, 100 km — the well-known « IT100 ».
• I'm currently training for the X-Traversée at Trail Verbier–St-Bernard by UTMB — 77 km / 5,000 m of vertical gain in the Valais Alps.
• The rest of the time : Sunday outings in the Belgian Ardennes, plenty of climbing, and a taste for analysing afterwards what happened during.

That combination — a bit of dev, a lot of field — drives the way GPXchunk is built. Every metric you see in the app maps back to a question I once asked myself on one of my own tracks.

Every mainstream sports tool follows the same model : you send your track to their servers, they store it, and they do what they want with it — analyse, leaderboard, aggregate, ad targeting, occasionally resell. In exchange you get numbers, charts, and once in a while a little medal.

GPXchunk does the opposite, by deep-rooted conviction : your file never leaves your browser. XML parsing, distance and elevation computations, map rendering — all of it happens inside the open tab. I have no server that receives the track, no database that stores it, no account that ties it to your identity. Closing the tab wipes it.

This is not a marketing stance : it's the direct consequence of a simple observation — I have no need to record people's information, so I don't record it. The code running in your tab is the same for everyone, and you can verify it at any moment in your browser's developer tools. The promise isn't « we don't look » ; it's « we never receive anything ».

To be completely clear about the lines I draw for myself :

• No data resale. Even aggregated, anonymised, derived — no. Your runs are not a product for me to sell.
• No account, no sports profile. The tool doesn't need to know who you are to compute elevation gain.
• No leaderboard, no ranking, no gamification. Strava already does that very well ; it's not the point here.
• No third-party tracker beyond the strict minimum. The site is free, funded by Google AdSense ; the exact list of third-party cookies lives in the privacy policy, and the consent banner lets you decline everything.

If one of these points were ever to change, it would not be tucked away inside terms-and-conditions : it would be an explicit note on this very page, or the end of the project.

You can write to me at info [at] gpxchunk [dot] com — to flag a .gpx file that breaks something, suggest a missing measurement, etc.

Out in the field, you'll find me on Strava — it's the only social network where I publish my tracks. GPXchunk's source code and infrastructure remain private otherwise ; that's consistent with the rest.