Wearables: not the beginning of the end, the end of the beginning

Another day, another wearables company is leaving the market. The announcement that Fitbit will acquire smartwatch pioneer Pebble follows this year’s trend of shutdowns and consolidation in the wearables space. The market, much like the aging devices I still wear on my arms, has lost its shine. While it might be sunset for the golden days of Wearables 1.0, it is far from the end of wearables. We are merely at the end of the beginning, not at the beginning of the end.

When I joined Misfit Wearables as their data scientist in 2012, the already buzzing wearables field was just about to cross over into the mainstream. Fitbit had launched their second and third products, Jawbone’s UP had just been re-released, Nike’s Fuelband was bringing wearables further into the mainstream, and Pebble had raised over $10 million on Kickstarter for production of its first smartwatch, breaking records and drumming up demand for smartwatches. At Misfit, we had announced our first product, the Misfit Shine, and our own Indiegogo campaign was picking up steam.

Piggybacking on the advances of mobile technology, the stage was set for an explosion of connected wearables, smarthome, and Internet of Things devices. In late 2011, Apple’s iPhone 4S shipped with Bluetooth 4.0 (BLE), leading to its adoption as the de facto standard for wireless communication between devices and phones. This, plus the increasing availability of smaller, cheaper electronics allowed device makers to develop affordable devices that would work both with Android and iOS. Enter the gadget-filled age of Wearables 1.0.

Wearable technology itself is not new, but the ability to seamlessly connect devices to apps and cloud-based services has dramatically improved its utility and its popularity. Current wearables automatically offer a level of accountability that had previously required real commitment to obtain. This kind of access to data about one’s activities and habits can be a useful tool. Just as bathroom scales can provide feedback to help manage weight, wearables allow for a similar kind of insight about an individual’s patterns of activity and sleep.

In the past few years, devices and apps have experimented with ways to remind and encourage their users to be more active. Though some have found the gamification and social features compelling, wearables have been largely B.Y.O.Q.: Bring Your Own Question. If you have a question about yourself or your activity that data could help you solve, these devices can provide this data. But without a question or specific goal, wearables alone won’t make you fit or even provide much beyond a verbatim account of your activities.

As people realized that the devices they were wearing didn’t come with a built-in purpose, let alone a killer app, many abandoned them, unfulfilled. In 2014, Endeavor Partners shared their study that 30% of consumers had stopped wearing their device within 6 months. This was largely viewed as a critique of wearables, and a failure of companies to engage their users. My view was more optimistic: 70% were still wearing their devices after 6 months! Even without a clear value proposition, people continued to charge their wearables, put them on their wrists, and collect data. However, the fundamental question remains unanswered: what would make wearables universally useful?

To answer this question, I often compare wearables to mills for grinding wheat into flour. Though some consumers have ideas of things they’d like to bake, most don’t want flour, they want cookies. The same is true for data: most people don’t want the data itself, they want cookies: the useful products and services that are built with this data. The problem is that for wearables data, we still don’t know what a cookie looks like. This is not the fault of wearables companies, but rather a reflection of where we are as a field.

Wearables 2.0 will be about the products and services built on top of wearables data. Of these, the most important will be the ones that bring this data into the healthcare system. Right now, evidence-based medicine is missing a critical dimension. Connecting longitudinal data about individuals and populations to clinically relevant outcomes will fill this gap. Wearables are already being used in hundreds of clinical studies, and will ultimately play a fundamental role in the future of healthcare and medicine. It will still take some time to get there.

My Pebble watch has been one of my favorite wearables, and I’ll be sad to take it off of my wrist when it reaches the end of its line. Since September, I’ve also received notices about the shutdown of the Basis B1 watch, the Narrative Clip wearable camera, and the AngelSensor open source smartband. These wearables will all join the growing collection of my device graveyard, leaving behind orphaned apps and drawers full of chargers.

Though these clunky plastic step-counting wristbands may be on their way out, the data they created and the lessons they taught us will play an important role in shaping the next wave of devices. It may take awhile to get here, but Wearables 2.0 is on its way.

Cross-posted on Medium.com.

The Internet of Things… that know a little too much about you

Though I’ve been wearing a few armsful of wearable activity trackers for about 4 years now, my entry into the smarthome space has been pretty recent. I’d purchased a Nest thermostat more than 2 years ago, but kept finding excuses not to set it up. Was I ready to invite Nest, Google, and the rest of the world [1] into my home? With wearables, it’s hard to forget that you’re wearing one, and they’re easy to remove if you don’t want a particular activity to be tracked. The reverse is true for home monitoring devices–it’s easy to forget that they’re there, and they’re more difficult to remove [2].

Last July, I decided it was time to take a deep breath and invite the Internet of Things into my home. If these devices are going to be around for awhile (which they will be), it’s in my best interest to understand their strengths, limitations, and capabilities, and how they fit into broader data ecosystems. Even if it does feel a little creepy.

When I moved to Cambridge last summer, I finally set up my Nest thermostat, and ordered an Amazon Echo. The Nest didn’t do anything very interesting for the first few months, as it was only connected to the heater (not the air conditioner). The Echo took minutes to set up, and my partner and I engaged with it immediately. “Echo, set a timer for 5 minutes.” “Echo, play November Rain.” “Echo, tell us a joke.” “Echo, add popsicles to the shopping list.” “Echo, who is better, you or Wolfram Alpha?” “Echo, play Chet Faker. Echo, stop. Echo, play Chet Faker. Echo, stop. Echo, play Chet F-A-K-E-R. Echo, stop.” [3] And so on. In the app, we could listen to an audio recording of each of these requests. I understand why this exists, and that recordings only happen when the Echo hears the wake word (in our case, ‘Echo’, but it can also be set as ‘Alexa’ or ‘Amazon’). But it’s still a little unsettling to not only know that you have a microphone in your living room, but to also hear its recordings of your voice.

Neither my partner nor I were excited about using our Echo as a commerce platform, so when we set it up, we turned off the ability to buy things using the Echo. I also noticed that I started acting differently when discussing some topics, opting to move to another room away from the microphones in the Echo and our phones. It reminds me a bit of how Cubans in Havana acted when asked about the government: they’d take you into a windowless room and turn up the music before telling you their thoughts [4].

The Internet of Things-That-Don’t-Actually-Connect-to-the-Internet

Yesterday, we added two more devices to our home collection, the Google Home and a set of Lifx lightbulbs. I spent an hour trying, fruitlessly, to set up the Lifx bulbs. Despite my attempts with different light fixtures, bulbs, power cycling, wifi network settings, proximity to the router, and more power cycling, the Lifx bulbs kept getting stuck when trying to connect to the wifi network. I’ll try a few more things before giving up and going with the Philips Hue. So far, I’m not impressed.

The Google Home took a couple tries to set up, but eventually connected to the wifi. In the onboarding process, the app asked for access to my Chrome web and app activity, as well as contact, calendar, app, music, battery life, and sensor readings from my phone. Basically, my Google Home device wants full reign to use my data from my phone and all of my Google services. That’s both intriguing and terrifying. It’s one thing to have cookies and retargeted ads follow you around the web, like a videogame bad guy who’s hot on your trail. But do I really want to add my home into this mix? I’m pretty sure that Adblock Plus doesn’t have an extension for smart speakers yet (at least, I couldn’t find one). I don’t have a subscription to either my Amazon Echo or my Google Home. Will the current business models of subscription-based music and the voice-based  commerce platform suffice, or will personalized advertisements soon be following me off the computer and into my living room?

Privacy concerns aside, it’ll be interesting to see how good the personalization of my Google Home can get. I’ve been impressed with Google Now, and can imagine that having a voice interface in my living room could be useful. And I can also see a million ways for it to go wrong. For now, I’m treating my home like a fishbowl. Which may be the right default anyway, smarthome devices or not.

“Abort! Abort!” screamed my gut. “Yes, I’m in,” I selected. For research.

“You want to turn up the temperature? No can do until you reply to those emails.” Thanks, bossypants. I think I’ll go get another jacket.

Do I really want Google helping me keep track of people throughout the day?

“Anyone in speaking range can access this info.” What could possibly go wrong?

[1] I take privacy and security of my devices and data seriously, but assume that any data I produce anywhere may end up being used in unanticipated ways, including becoming publicly available.

[2] And this assumes we’re talking about your data, in your home. However, the devices in your home also collect data about other people in your home, who may not realize or consent to being monitored. Likewise, you’re being monitored in other peoples homes, often without your knowledge or consent.

[3] We never were able to get our Echo to play Chet Faker. That said, most of the time the Echo does a really good job of interpreting our requests. (What it does in reply, that’s a different story. Let’s just say that we hate the 30 second song previews.)

[4] It’ll be interesting to see how this changes in the months and years that come, both in Cuba and elsewhere. To what extent will governments use these devices to eavesdrop into living room conversations? How can companies continue to develop these types of technology in ways that respect privacy?

Cross-posted on Medium.com.

A good day for the Internet of Things

Allowing security research on consumer devices (including wearables and the IoT) is obviously necessary. This had been illegal under the Digital Millennium Copyright Act, until today. Hopefully this will translate into better security for our devices.

Read more about today’s decision by the Library of Congress, on the FTC’s blog: https://www.ftc.gov/news-events/blogs/techftc/2016/10/dmca-security-research-exemption-consumer-devices

With the stroke of a pen, the Librarian of Congress has authorized security researchers who are acting in good faith to conduct controlled research on consumer devices so long as the research does not violate other laws such as the Computer Fraud and Abuse Act (CFAA). This temporary exemption to the Digital Millennium Copyright Act (DMCA) begins today. The new temporary exemption is a big win for security researchers and for consumers who will benefit from increased security testing of the products they use.

The Digital Millennium Copyright Act (DMCA) makes it illegal to circumvent controls that prevent access to copyrighted material. The result is that under the DMCA, researchers can’t investigate and discover security vulnerabilities if doing so requires reverse engineering or circumventing controls such as obfuscated code. The Librarian of Congress can adopt exemptions to the DMCA’s anti-circumvention statute for various technologies. These exemptions have allowed individuals to unlock tablets and wearables, jailbreak mobile devices, circumvent brand-specific 3D ink restrictions on 3D printers, and more. Exemptions take away a legal hurdle and help protect conduct without fear of legal recourse. It is important to note that the rule requires a careful setup and testing environment in order to fall under the good faith security research exemption, and does not exempt researchers from other laws such as the CFAA.

Lazy Tommy Pumpkinhead, 2.0

Growing up, one of my very favorite books was Lazy Tommy Pumpkinhead (very sadly, out of print). It’s a story about a boy, Tommy, who lives in an ‘electric’ house, where robotic appliances do everything for him: waking him up, brushing his teeth, feeding him, etc. Until one day, when things go terribly wrong…

This book page is from the Lazy Tommy Pumpkinhead post on the blog Books for Breakfast.

Clearly, home automation and the broader Internet of Things (IoT) have many benefits to offer. But I often find myself thinking of this book when I hear about new smart home gadgets and appliances that seem like they’re not…quite there yet. The fantastic Twitter account @InternetOfShit does a great job of compiling these stories.

Taking this a step further, Simone Giertz, a Swedish inventor and roboticist, wins the title of Lazy Tommy Pumpkinhead 2.0. She has a YouTube channel of maybe-not-so-useful house robots, like this lipstick robot:

The Nerdist article is pretty great, as is this video of a typical morning in Simone’s home.

Internet of Things + Blockchain?

I’m intrigued by IDEO’s upcoming Internet of Things + Blockchain make-a-thon and fellowship, exploring “The Future of Privacy, Automation, and Exchange.” The option to have distributed and encrypted systems for data storage is becoming increasingly important, and a version of a blockchain for the IoT offers some attractive properties.

The event will take place on March 12 at IDEO’s San Francisco studio, and applications to participate are open until February 21. More on the event (and application) here: http://bitsblocks.ideofutures.com/iot-blockchain

IOT + BLOCKCHAIN

The Future of Privacy, Automation, and Exchange

The objects in our lives are quickly becoming smarter and more connected. We are surrounded by more sensors than ever, and activities are becoming increasingly automated based on the data collected by these sensors.

But what does it mean to have devices—from our cars, to our refrigerators, to our thermostats and beyond—connect not just to us as individuals, but also to each other? How will the world change when data and devices are able to freely connect, mix, and create new services and experiences, without direct guidance by a person? What’s the interface by which we will tap into these emergent networks that cross boundaries between the physical and the digital? What businesses become possible when any device can participate in the sharing economy?