I've got a new way to walk

"For decades," the website of the pedestrian-interest lobby America Walks informs us, "walking has been the forgotten mode of transportation." A few exceptions notwithstanding, this has generally been the case for historians and sociologists of technology. For my first post, I want to talk about how technology - specifically, a device, the Fitbit, and a web app, Walk Score - has been applied to call attention to how (more specifically, how much) people walk. This ties together a few of the topics that I'm looking forward to exploring on American Science over the coming weeks and months: the relationship between technology and daily life, the role of the structure of data in shaping how people see the world and what they do in it, and the science and technology of things as mundane as, well, walking.

Fitbits (http://www.fitbit.com/product)

Fitbit is the best-selling of a slew of activity trackers: wristband or clip-on devices that log a user's daily accumulation of calorie-burning, health-building movement. The device measures motion using accelerometers (and in newer versions, an altimeter), and, after passing that data through pattern-recognition software, computes, displays, and stores fitness statistics for the wearer. These statistics express various conventional measurements of exercise - calories burned, miles traveled - but also "floors" climbed and, most prominently, "steps" taken.  The Fitbit awards users with encouraging badges for daily accomplishments such as walking 10,000 steps.*

Walk Score rates cities, neighborhoods, and addresses by their "walkability." Based on the proximity of everyday destinations - grocery stores, restaurants, coffee shops, parks, and the like -it computes a rating on a 0 to 100 scale, from "Walker's Paradise" (90-100) down to "Car-dependent" (0-49). The site's walkability scores and heat maps are used predominantly by realtors to promote urban residential properties, but Walk Score data has also been used by economists studying real estate pricing, by public health researchers, and by urban planners. And presumably by a fair number of procrastinators - the app is pretty fun.

My neighborhood in Newark (not to brag). In this heat map,
the highest Walk Scores are represented in dark green and the lowest in red.

There's plenty of grist for STS critique in Fitbit and Walk Score: their bugs, what they fail to capture, the privacy concerns they raise, how they have been adapted and repurposed, for example by people who are mobility-impaired, or who value inaccessibility in an address.  (Readers: very interested to hear your comments on all of these!)

For the moment, though, I want to reflect on what exactly these devices set out to measure, and how they might spur their users to rethink what exactly walking is.  As a starting point, let's take the two sorts of walking outlined by CUNY researchers studying the relationship between "walkability" and walking behavior.  For the purpose of this study, one of whose aims was to quantify how much walkability affected the behavior of people who professed not to care much about walking, the researchers distinguished "instrumental walking," that is, walks intended "to get some place," from "health/relaxation walking."

This definition of "instrumental walking" seems to be a pretty good description of what Walk Score measures, given the app's emphasis on the proximity of common destinations. In fact, I suspect that the proximity of an address to a prime site of "health/relaxation" walking is generally correlated with a lower Walk Score (a suspicion borne out anecdotally in the case of my hometown of Pittsburgh).

Google map of the East End of Pittsburgh (right), with heat map overlay from walkscore.com (left).
Note the low "walkability" of parks and neighborhoods adjacent to them.

One might presume that Fitbit, in contrast, measures health-related walking.  It is called Fitbit, after all. But that's not exactly what it does - at least not in a way that maps neatly to the instrumental-walking vs. health-walking breakdown.  Rather, Fitbit is a machine for flattening out the distinction between "instrumental walking" and "health walking," and for generating a new sort of walking altogether.**  By focusing on those two numbers, steps and floors, instrumental walking not only counts as health walking, but health walking is also measured out in units that look more like instrumental walking.  The Fitbit doesn't make the rat race into an athletic endeavor so much as a kind of rat fun run.

At the same time, by quantifying fitness in terms of steps, the Fitbit also encourages people to fill the interstices of the day with exercise that doesn't seem to fit either of these categories of walking. Fitbit users discuss parking at the far side of a parking lot, or pacing while on the phone, or walking "up and down long subway platforms past bizarre-looking people or into deserted corners while waiting for the train" to up their daily step count.  The step-counting generates a new sort of walking, neither an instrumental trip nor a health/relaxation amble, but an opportunistic, accumulative process. In fact, compared with non-step-based exercise, it privileges this sort of walking; while the Fitbit can be made to keep track of diet and formal exercise, the user has to input meals or activities like swimming or stationary biking manually, and users earn those badges for automatically-tracked steps and floors only.

Both the Fitbit and Walk Score are technologies for substituting walking for dedicated transportation technologies and exercise activities that constitute more obtrusive interruptions in daily life. By tracking steps automatically, the Fitbit promotes a mode of exercise that doesn't involve going to a gym or even changing into running shoes. Moving to an an address with a high Walk Score, or taking action to boost that of a community, reduces the frequency of car or public transportation use, and the burden on those without access to a car (that's the idea, anyway).  In doing so, these tools partake in a general trend, from design to citizenship and governance, for using technology as a model for and a means of attaining the virtues of simplification.

What, if anything, is new about these ways to walk?  Not the quantification and "gamification" of walking; that's been going on for quite a while, as a new book on competitive walking in nineteenth-century America, recently reviewed in The New Yorker, demonstrates. Nor the idea of scientific and technological intervention to adjust habits and patterns of human locomotion, either at the level of the individual or society.

One might say something about Foucauldian self-monitoring, in the case of the Fitbit, or the primary application of Walk Score as another item for the endless rubric of real-estate listings. To me, though, what's most striking is the way in which both tools express distance and movement on a human scale - in the vague lifestyle-tagged heat maps of Walk Score, and especially in the steps and floors of Fitbit. They fit data about us to a scale based on human experience, and reconfigure that experience in the process.  Thinking of your day in terms of steps and flights of stairs is unthinkable, until a device is keeping track for you; with that device in hand, it's hard to think of anything else.



*As Fitbit's blog has it, the 10,000 steps rubric began as a marketing campaign of a Japanese pedometer manufacturer in the 1960s.

** Another factor that collapses these categories is dog ownership, which the authors of the CUNY study controlled for in some of their analyses.  They elected to classify "walking the dog" as "health/relaxation." Walking to the PetSmart for a bottle of carpet cleaner was presumably classified otherwise.