Why Nobody's Wearing Wearables

Bloomberg View , March 03, 2016

Technology watchers have been proclaiming wearables the next big thing since at least the 2013 debut of Google Glass. The rapid sales growth of fitness bands and smartwatches has intensified that conviction. But to fulfill its potential, wearable technology has to be good for more than tracking workouts or getting notifications on a wristwatch.

Defined loosely as tiny computers worn somewhere on the body, wearable tech still needs its graphical user interface, its browser, its broadband, its VisiCalc, its Google, its Amazon: the enabling technologies and unique benefits that make it essential and easy to use. Early adopters of Apple’s headline-grabbing smartwatch are, after all, using itmostly for telling time and getting notifications -- not exactly world-changing applications.

“Where we are with wearables is about where we were with the Internet in 1993,” says Amanda Parkes, the chief of technology and research at Manufacture New York and a visiting scientist at the MIT Media Lab.

Think of wearable development as a series of generations. Today’s first generation, exemplified by fitness trackers but including smartwatches as they’re actually used, are worn as accessories or patches. They do just a few things and tend to aim at niche markets, from athletes to anxious parents. As Fitbit’s plummeting stock price reflects, basic fitness trackers are now easily knocked-off commodities, but first-generation wearables include less fully exploited markets. Instead of constantly checking on your sleeping newborn, for instance, you can now let an Owlet smart sock monitor the baby’s heartbeat and breathing. The TempTraq thermometer patch continuously records body temperature, allowing parents to use smart phones to check on feverish kids. (Although TempTraq markets the 24-hour patch to parents, it also works for adults.)

Second-generation wearables would integrate sensors and processing into garments, making the technology less obtrusive and more versatile. Clothes could both collect and react to new kinds of data.

At the moment, however, wearable chips and circuits, as amazing as they are, are still too large, delicate and hard to assemble for manufacturing second-generation ideas at scale, especially when you add in consumers’ expectations that clothes will be durable, washable and fairly cheap. “Fashion factories and technology factories are vastly different enterprises,” notes Liza Kindred, founder of the New York fashion and technology consulting firm Third Wave Fashion. Neither knows how to make smart clothes.

So we’re still in the superhero phase of second-generation technology. If you’re a Tony Stark or Bruce Wayne with lots of money -- or an artist or fashion designer planning a one-time show -- you can get a bespoke outfit to do something remarkable, such as protecting personal space with an exoskeleton that responds to encroachments. But a prototype, however impressive, isn’t a commercial product.

“It’s doing damage to the whole industry that so many companies are putting out prototypes and calling them products,” says Kindred, citing Google’s Project Jacquard as an example.

Or take the smart sports bra. The design house Chromat showed one at New York Fashion Week last September. It used Intel’s wearables-oriented Curie chip to sense and adjust to body temperature and sweat, opening and closing vents to cool or warm the wearer. The smart bra drew plenty of attention, but you couldn’t actually buy one -- a message that got lost in the hype.

Wearables have Moore’s Law on their side, since chips get smaller and cheaper with each new generation, but developing second-generation versions still requires other basic ingredients, such as easily installed connectors and better battery solutions. Smartwatches have helped spur development, but component quality is still a problem. “If you’re not deep inside the world” of wearable development, says Parkes, “you don’t understand why it’s necessary.”

Even promising ideas for second-generation wearables face a chicken-and-egg problem that makes it hard to go beyond proof of concept.

Consider the leggings that researchers at King’s College London recently developed to measure muscle fatigue in runners. With sensors embroidered at fixed points, the leggings don’t require a runner to know anything about the equipment. By collecting data in real-world conditions rather than in a lab, smart leggings could improve training and reduce injuries. But, again, they’re still a research prototype, tested on only two people.

Runners come in many different proportions. Just because athletes wear the same size doesn’t mean the sensors will hit them in the same place. Assuming these smart garments could be manufactured at scale, would they actually work in the field? How would they hold up? Would people want them? If you don’t know, how can you justify investing in a new manufacturing process to make them?

There’s the rub: Second-generation wearables need trial-and-error testing to find out if they’ll work in everyday life. That means large numbers of consumers using them under real-world conditions. And that in turn requires manufacturing an entirely new product at scale -- a costly investment, which itself demands learning over time. Why do it if you aren’t sure the new idea will catch on? The chicken-and-egg problem explains why prototypes that grab attention don’t show up in stores.

Like the leggings, and unlike today’s fitness trackers, advanced wearables offer the possibility of collecting entirely new kinds of information, telling us things we didn’t already know. Their great potential, says Parkes, is “prolonged and continuous data collection and mapping. What you’re trying to do is sense change.”

That brings us to a possible third generation, in which wearables gracefully and unobtrusively collect new kinds of data that is then analyzed and integrated with other information to provide new insights.

Imagine applying the idea behind TempTraq -- continuously recording body temperature -- to healthy people in everyday life. Normal temperatures vary from person to person, but by how much? How do they fluctuate by time of day? By activity? By surrounding environment? By subjective impression? (Is it just me, or is it hot in here?) How does temperature affect sleep? Can you predict the onset of an illness by a subtle spike? Once we have the ability to collect such data, the challenge is to figure out what it means.

Intimate and ubiquitous, third-generation wearable tech could give us a deeper understanding of our bodies and our environments. “We don’t really know yet how it’s going to change culture and behavior, but it has that potential, because it’s tied into the most human capacities,” says Parkes. Before we can get there, however, wearables need to become cheap, reliable, comfortable, understandable and good-looking enough for large numbers of ordinary people to buy them.