The internet waves generated earlier this week over Strava’s user data heatmaps we’re both exciting and frustrating to those of us who have been waiting for the technology revolution to catch up with smart route mapping for cyclists. It was exciting because it is definitely a sign of momentum, but frustrating in that there is still a gulf between the reality of crowdsourcing optimal bike routes and their actual implementation.
BikePortland breaks down a very interesting use of Stava’s data set by the Oregon Department of Transportation. ODOT has signed a one year $20,000 deal to have access to Strava’s dataset in Oregon. They are studying the data with the intention of better understanding bike usage in order to improve infrastructure and set policy.
While this is certainly an inspiring and forward thinking use of Strava’s data, there is still much to be desired. Of course there is the issue that this data is primarily generated by the bike racing set rather than by bike commuters. But what is really missing is an actively updated bike mapping system with recommendations based on where the most cyclists travel.
The most valuable thing that can be done with a bicyclists gps tracks are to collect them all and input them into a map that collectively shows where cyclists tend to ride. The majority of trips by bike are short, local trips on routes that cyclists know well. These short trips tend to be on streets that cyclists know are safe and out of traffic. This type of location data is perfect to collect and build recommended bike routes.
What is the holdup?
There is still not a consistent way that cyclists of all stripes are motivated to share their GPS data. Perhaps Strava will become the focal point for this and move beyond inspiring KOMs. With a solid business beneath them and their app already in the hands of many cyclists, they are certainly in a good position to pursue this opportunity. But conversely, Strava appears to be a very competition focused company. We can only hope that this is just the marketing face of the company for their current clients and that internally they are aware of the opportunity to service bike commuter maps and are considering how they might go after that opportunity.
Back in 2010, Google Maps launched the Bike There feature to much fan fare among the cycling community. This was certainly a positive move forward and the technology behind these bike specific routes continues to improve. That said, the routes are often wanting and do not match the nuanced side-street diversion that locally savvy cyclists discover. I’ve wondered why Google hasn’t figured this problem out given that they seem to have all of the pieces in their ever growing technology arsenal. For all we know, they are already harvesting the data from our Android phones.
Who else could solve this?
Without the infrastructure of Google or the user base of Strava, it is clear that this is not easy data to collect in a meaningful way. And there have been a few attempts to collect this data on a more localized level. That all said, the trick seems to be to get a significant number of cyclists to want to opt into sharing their cycling data. Strava figured out that ego is a motivator for competitive cyclists to share their data.
So how can we motivate the everyday cyclist to share their data?
Last year Bike Spike successfully funded the launching of a bicycle GPS tracking devise in a KickStarter project. The idea was to have a devise hidden on your bike where if you bike was stolen, you’d be alerted on your phone and could track and retrieve it. While Bike Spike is still working out the kinks in bringing the product (I am still waiting on mine) it offers a promising way to gather bike specific travel data. Another company, Mobiloc is also working on a similar devise and is accepting preorders.
Bike Spike’s website rhetoric about an API for their data, got me thinking about a way to motivate bike commuters to share their data. I’ve been wondering how compelling of an argument it would be for bike commuters to share their location data in return for theft security in the form of a tracking devise.
Perhaps some partnership between a company like Bike Spike and a local government like the Oregon Department of Transportation could fund or partially subsidize getting discounted GPS tracking devises out among regular bike commuters. Reducing bike theft and getting access to quality bike data both for usage statistics and with the potential to build a better informed bike routing system certainly seems like a worthwhile goal for a local cycling or transportation focused organization.
If such a pilot program were to happen, the next challenge would be to figure out what to do with the data. Feeding it into Open Street Maps would likely be the best solution, though I imagine a partnership with Google Maps might serve the goal of enabling high quality bike routes. With either of these solutions, determining how to interpret the data and improve bike routing data will be an interesting challenge for some smart programmer to finally be able to get at.