Data Obesity: The Latest Threat To Your Digital Fitness

Author

Owen Thomas

July 25, 2014

So many of our modern health problems stem from issues of abundance: fast food, car-friendly suburbs, and endless couch-borne entertainment.

The cure for these, according to the quantified-self cult of Silicon Valley, is more data! Just collect information about every step we take, every beat of our heart, every toss we take in our sleep, everything we eat and every drop we drink, and all will be revealed.

But what if the abundance of data, too, is part of the problem?

Networked weight scales at MyFitnessPal headquarters.

Your Data’s So Big …

Mike Abbott, a partner at Kleiner Perkins Caufield & Byers, a venture-capital firm, casually dropped the term “data obesity” in a talk he gave at a digital-health event hosted by MyFitnessPal in its San Francisco headquarters Wednesday. He used it to describe startups’ habit of gathering as much data as possible without clear forethought of how they were going to use it. Like ever-widening stadium seats or the extra stretch in our jeans, cheap cloud storage and computing accommodate this overindulgence in bits.

The problem with data gluttony is that noise can overwhelm signal. We don’t yet understand the complex interplay between nutrition, exercise, sleep, and stress. And many of the data types we capture are a poor approximation of what we’re actually trying to measure. (Take, for example, the fixation most wearable makers have with measuring steps as a proxy for activity.)

There are ethical problems, too, in this broad data capture. Abe Gong, a data scientist at Jawbone, the maker of the Up fitness band, also spoke at the MyFitnessPal event about the conundrum of all the data we’re gathering about the state of our bodies. He cited the stethoscope, a tool which has become metonymous with the medical profession: It allowed a doctor to hear things happening in a patient’s body that she couldn’t hear herself.

Jawbone’s Abe Gong suggests technological tools like the stethoscope introduce a power imbalance between doctors and patients.

That information asymmetry is happening on a far larger scale: While we are notionally in control of the data we track about ourselves with devices like the Jawbone Up and Fitbit, the truth is most of us are ill-equipped to process the scale of data we generate. Yes, we tap buttons to grant permission, but we don’t have much choice about it if we want to use an app. Some developers give you the option of downloading raw data files, but what would we do with them? In a practical sense, we don’t really control this so-called “self-reported data.”

The New Data Diet

Chris Tacy, a developer and entrepreneur, recently called for technologists to adhere to a new code: the “minimal actionable dataset.” In other words, only gather the data about your users that you really know you need.

It’s a radical intervention against data gluttony. Call it the informational equivalent of the paleo diet. When data storage was expensive and computational power likewise dear, software only gathered the bare minimum data required to get the job done.

It’s easy to make the argument for data abundance. If you don’t capture all possible data, how will you know what proves useful? But the ultimate restriction on data processing isn’t computation or storage. It’s our human bandwidth to analyze data and translate it into concrete action.

Everything In Moderation

For an example of a startup that practices lean data, I didn’t have to look farther than the host of Wednesday’s event, MyFitnessPal. MyFitnessPal is far from imperial in its approach to data collection. It lets you track the calories you eat and the calories you burn. (It only recently added step tracking as another data type.)

This simplicity is a big reason why I’ve stuck with MyFitnessPal for five years, accumulating a vast store of simple, easy-to-understand data. Correlation may not be causation, but I can see a pretty clear relationship between the rigor of my diet and my weight. MyFitnessPal also doesn’t have to struggle to explain what it does and how it benefits users the way startups with ambitions of tracking every last bit of health data do.

So is it time to put your apps and gadgets on a data diet? Perhaps. At the very least, we should be thoughtful about the data they ingest and the analytical exercises we perform. It’s just healthier that way.

Lead photo via Shutterstock; other photos by Owen Thomas for ReadWrite

Great ! Thanks for your subscription !

You will soon receive the first Content Loop Newsletter