Visualizing bike rides as physical small multiples.
Twelve maps show my monthly bike rides from July 2012 to June 2013. I tracked all my bicycle trips in Berlin, and visualized them on a postcard for each month.
Last year, I sent a little digital map showing traces of my bike rides in Berlin to my mom. She called me the same day, telling that my dad and she were looking at it for hours trying to understand every single path, and guessing from where to where I was going. This year, while wondering what to give her as a birthday present, the idea was born to create something similar but with the possibility to compare different months. While my mother is able to browse the web, send e-mails, or book some tickets, she is not necessarily the most advanced user of interactive visualizations. Thus, an appropriate way of showing the data seemed to be physical small multiples.
Having physical postcards allows seeing the evolution over the year when laying them out next to each other. Picking up one card allows to study a single month in detail. And by taking one card into each hand one can easily compare two months.
Each postcard is 17.5 × 9.2 cm, and printed on cardboard paper (280 g/m2).
With a postcard for every month, temporal patterns become visible and some stories start to emerge. For instance, I seem to cycle less in winter months for some strange reason.
It also becomes apparent when I moved into my new office with onformative (the diagonal M starting in April). Mostly though, the maps act as some kind of visual diary for myself.
Of course, there are privacy issues with these kind of personal tracking visualizations. Everybody now can see my bike rides, and for instance might be able to guess when I will be at specific locations. In the end, I figured the monthly aggregation and the low resolution of the pictures embedded here are sufficient to prevent precise revelations. But if you are going to create similar visualizations please be aware of the implications.
All bike rides have been recorded via my GPS enabled smartphone. Then, I exported the data as GPX, imported them into a Processing sketch, and mapped them onto Berlin with the help of our Unfolding map library.
<?xml version="1.0" encoding="UTF-8"?> <gpx> <trk> <name>My bike tour in June</name> <time>2013-06-01T14:10:26Z</time> <trkseg> <trkpt lat="52.592373000" lon="13.266748000"><ele>37.2</ele><time>2013-06-01T14:10:22Z</time></trkpt> <trkpt lat="52.593083000" lon="13.266756000"><ele>38.0</ele><time>2013-06-01T14:10:26Z</time></trkpt> <trkpt lat="52.592937000" lon="13.266863000"><ele>38.7</ele><time>2013-06-01T14:10:30Z</time></trkpt> <trkpt lat="52.592873000" lon="13.266758000"><ele>38.3</ele><time>2013-06-01T14:10:34Z</time></trkpt> <trkpt lat="52.592732000" lon="13.266869000"><ele>38.4</ele><time>2013-06-01T14:10:38Z</time></trkpt> ...
All starting, pausing and stopping was done manually, which resulted in the occasional weird looking track segment when I forgot to switch off recording after I arrived at my destination.
I recorded my bike rides with Runkeeper, a tracking app for running and other activities. But it could have been done with any tracking software which allows you to export the data as GPX (e.g. Strava, MapMyRide, etc). The nice thing about Runkeeper is that you can bulk-download all your activities in a single step (via the export data form).
If you are interested in creating your own maps and you happen to have GPX data, send me a message or leave a comment.
An alternative to small multiples is to show the evolution of my bike rides in an animation. Interesting thing here would be to animate by day but aggregate by month, which would result in a rolling month display (i.e. always 30 days).
Postcard idea was inspired by Jo Wood’s beautiful “We are the city” project which includes visualizations of a single Boris bike (from London’s bicycle sharing scheme) over the months. See the gorgeous urban signatures this created.
And thanks to everybody from onformative for the support.