[paulmach]

Map creator here if you have questions.

The data is aggregated into a quad tree based on number of GPS points in each pixel. Tiles are then served on the fly using Go and C using CGO.

Cloudfront tackles most of the load, but the load balanced i2.xlarge instances can do about 300 tiles a second.

[dougmccune]

Would Strava be OK with people using this data to trace paths for Open Street Map?

I noticed there are quite a few paths through Golden Gate Park in SF (as just one of many examples) that are pretty clear running/biking paths that are not in any of the typical basemap providers' maps and not currently in OSM. If you reached out to the OSM community and gave the go ahead, this would be a great way to improve the trail data in OSM.

[paulmach]

We have a self-hosted version of the iD editor that already includes this data as a tracing layer. http://strava.github.io/iD/#background=Bing&map=16.97/-122.5...

I've also built another tool, Slide, to help use this data for faster editing, see http://labs.strava.com/slide/

[femto]

Better still if it could be released an an official layer for Open Street Map. Any chance of that happening given the licensing conditions of the various data sets?

Unlike a traced layer, the heatmap is self-updating and objective, so I think has a value of its own.

[micro_cam]

Is this something you are going to release for people to include in mashups? I'd love to get it in hillmap.com

Also how accurate is the bike vs run stuff? I'm seeing a a surprising amount of people biking activity in wilderness areas.

[dalek2point3]

This map is cool, but it doesnt help me use it for what I imagine is a pretty common use-case, finding a popular route from Place A to B. Would it be possible to drop placemarkers between two places and do routing based on Strava popularity? Right now Google bike directions are pretty bad where I live, and it would be nice to know what route the "insiders" are taking to do a popular route (MIT to Concord is one Im thinking of, but I can imagine other alternatives)

[jstultz]

You'd probably be interested in this feature, which does exactly what you're asking for:

http://www.strava.com/routes

[twelvechairs]

Great map! Fantastic work. Two questions/points:

- Where's the source data from?

- You should think about producing an overlay with bike and run in different colours to see where these differ.

[paulmach]

The data is from http://www.strava.com

An overlay would be nice, but there is a version for both running and riding.

[chatman]

Why Google Maps and not OpenStreetMap as the base layer?

[paulmach]

The overlay tiles zoom weird in leaflet. They get treated like a base layer so when you zoom, they get scaled 2x and the new tiles load on top. That doesn't look right when the tiles are transparent. There must be an option to fix this but couldn't find it.

I couldn't get the styles I wanted with mapbox so I just defaulted to google.

[bbarn]

Geographic search, and click through to routes.. and this will be a really big deal. Love the concept, keep it coming!

[afx2in]

agreed. i'd love to be able to see what routes people run near my home so i can get new ideas.

[samstave]

Per my other comment -- can you do a version which show avg speed of the route goers? It would be awesome to see the fastest path via bike or run between points... if you noted each stop the people made, it could be the diff between a smooth, fun ride vs a stop-heavy route.

[sztanko]

Thank you for this map! Two questions: - what implementation of quadtree are you using? - will you ever consider opening the data for download as shapefiles/any other vector format? Thanks again, great job!

[paulmach]

My own quad tree, they're not that hard to write :)

We're not considering opening the data. But we are working with cities to help improve their bike infrastructure

http://bikeportland.org/2014/05/01/odot-embarks-on-big-data-...

[3rd3]

How do you calculate exacly the coloring of the heat map? Is the scaling of it locally adjusted or is it a global/absolute scale?

[paulmach]

Each pixel has a count. Each tile gets a 90 percentile value. Every corner gets the average of its 4 tiles. Every pixel in the currently requested tile gets a value [0, 1] based on the bi-linearly interpolated "max". It's then colored based on that final value.

Maybe unnecessarily complicated, but it does provide really good local normalization. You also get cool looking shading like in this location http://labs.strava.com/heatmap/#16/-121.76487/38.49134/gray/...

[ddunlop]

Enjoyed the heatmap when it was on raceshape and like it here as well!

Is this now the complete dataset from strava Run/Bike activities?

[paulmach]

It has every public ride and run with activity id < 120,000,000

It's not updated on every upload, but pretty much the complete dataset.

[pacofvf]

why are you serving on the fly? Are we seeing live data?