Colin Dietrich says it was a passion project that got away from him. Eight years ago, the Seattle-based climate scientist decided he wanted to systematically evaluate the quality of his bike commute. He attached a low-powered video recorder to his front fork to count cracks in the sidewalk. From there, things spiraled. He added accelerometers, a tablet for taking notes and a Wi-Fi network, and he’s accumulated gigabytes worth of data.
Kim Voros, a friend and planner at Alta Planning and Design, a Seattle-based transportation planning firm, got word of Dietrich's bike and thought it might be useful in gathering data for a project the city of Seattle was working on, an update to its Bicycle Master Plan. This summer, Alta’s interns took the DataCycle, more affectionately known as the "Frankenbike," on a series of rides to carefully map 40 miles of Seattle’s bike trails. The idea was to log potholes, unkempt vegetation and other sub-par conditions, and to take stock of public amenities along the route, as part of a citywide plan to improve its trails. With the data, Seattle's department of transportation has identified sections of trail that are the highest priority. They're releasing a trail upgrade plan in December.
Where did the idea for the DataCycle come from?
Partially due to my experiences advocating for Neighborhood Greenways in Seattle I realized the Herculean task government has in keeping up to date on their infrastructure. I knew that things that I saw everyday, that could be improved on the road, would never get to planners and engineers at the resolution of my daily experiences.
The process of creating the bike really was the process of a hobby getting away from me. I initially wanted to measure one thing in my commute for curiosity's sake, and it lead to several increases in computing power, more batteries and the addition of more sensors.
What’s it like to ride the DataCycle?
It's a frame from an old road bike, so it’s good, stability-wise, but, as with any good bike, fit and control are subtle things. The addition of most of the hardware has been unnoticeable; however, the computer on the handlebars makes it handle like you have some groceries on a front rack. More importantly, other riders and I have found that just like distracted drivers, it’s challenging to look at the screen and ride. We've used a second spotter rider in Seattle.
You’ve teamed up with Alta Planning and Design to work on Seattle’s Master Bike Plan. How is the city using the data?
Alta worked closely with the Seattle Department of Transportation to develop codes for attributes they wanted to inventory—things as simple as number of services along a route or number of bollards in the pathway. That has been very insightful for seeing what is needed at the policy level. Without a planning and policy perspective, some of the data would likely be useless—the era of data-driven decision making still requires initial problems to be described. Some of what we're still doing is digging further into the data, looking for relationships.
Is it scalable? How can other places use this?
The process can be applied anywhere, but it’s not exactly scalable. The data collected is in the context of a city's transportation system and its planning goals, as well as the individual rider. On the bike, a certain amount of calibration to the rider is required before use, so I wouldn't expect to see this sort of thing as an app, per say. It's more like a portable laboratory or survey instrument.
Alta has been looking for new applications of the DataCycle to aid other municipalities. I'm also working on an improved version, likely based off a different bicycle design.
What do you hope comes from it?
I hope that cities can efficiently improve their bicycle trail and route networks and reduce car trips. I think there are a lot of willing but wary riders out there that would ride their bikes if the actual ride felt like a Cadillac. Some places the pavement and traffic makes the route feel like an off-road stampede.
What’s been the most surprising thing you’ve found?
How many metal plates end up in a cyclist's path! And how complicated a simple question can become. We collected a lot more data than I initially expected.