Running Code on my Wrist


Original Pebble – The model I own

Filed Under: Things my Grandparents would never believe.

A few weeks ago, I purchased a Pebble Smart Watch. I didn’t have a lot of experience in Wearables prior to ordering; although ubiquitous computing and mobility has always been an area of interest for me. After a couple of weeks with my Pebble, I can safely say that I’m truly impressed. The device itself has a ways to go in terms of its capabilities, and the ecosystem of applications is pretty weak right now, but there is clearly a huge amount of potential here. I wanted to write a bit about some of my observations so far, a little project I’m working on, and then do some wildly irresponsible forecasting about the future.

Strength: The Power of Context

One of the most obvious advantages of the Smart Watch form factor is the fact that it fits right into the context of how you already behave. We’re all conditioned to check the time regularly (we’re all very important, after all), and the fact that it can leverage those habits makes it a great place to surface information. A quick check of your watch isn’t disruptive, and it doesn’t feel like it’s stealing “focus” from whatever is going on in your life.
Contrast that with the Phone which requires you to partially leave the current context (take your hands off the wheel, put down the dry-erase marker, etc.), reach into your pocket (or purse) for the right device, which then consumes your attention while you bring it to life and navigate to what you want. Obviously this doesn’t hold for interactions which require navigation or heavy-interaction with the watch, but I’ll get to that in the Project section.

Strength: Power in Pairs

The fact that my watch is paired to my phone is a strength: it can rely on my phone for “heavy” computation, and it has all the capabilities of my phone — it can connect to the internet, track my location, etc. The two augment one another and become something more useful than the sum of its parts.

Strength: Reducing Screen Time

The watch actually leads me to spend less time futzing with my electronics. The fact that it alerts me to super important stuff (eg: Incoming Call, Important Text) means that I don’t fall into the “let me check my phone very quickly” trap. When I use the watch for some microinteraction (looking for the next bus arrival time, checking in on Foursquare), I bypass the phone entirely.
This has been a positive for me. I find that when I pull out my phone “just to check something,” it leads me down the rabbit hole: while I have it out I’ll also check Facebook, maybe my email, etc. All those things lead to further interaction, and what started as a simple check-in leads to five minutes glued to the screen. I still spend plenty of time (read: too much time) with my phone, but it’s noticeably less.

Weakness: Two Hands Are Not Better Than One

Here’s something you might not have thought about: interacting with a Smart Watch in any way beyond just a glance requires two hands. I’d never thought about it, but the fact is, you have to hold arm wearing the watch steady while using your other hand to press buttons. This is actually a downgrade from the phone. Most mobile phones are easy to use with one hand: you can hold it in your palm and use your thumb to get through the basics. When I’m carrying my work bag, it’s often easier to use my phone in one hand than to hold my bag up high with my left hand while pressing buttons on the watch with my right.

Weakness: Hardware is Lacking

The current version of the Pebble has very few inputs for the wearer:

  • Four Buttons – Back, Up, Down, Select
  • Accelerometer

I find myself wishing it could do a bit more: What if the accelerometer were more on par with FitBit, and it could work as an activity/sleep tracker as well? What if it had a microphone I could use to talk to my phone’s digital assistant? What if it had an NFC chip and could be used to pay for my coffee? (I can actually pay for coffee at Starbucks by displaying a barcode on screen, but it’s awkward.) There are endless possibilities here if the Pebble was just a bit “smarter.”

In addition, battery life is a concern — both for the watch (it lasts about 3 or 4 days) and the Phone when the watch is connected.

As far as computational power is concerned, it’s pretty weak. But as I mentioned in “Strength: Power in Pairs,” it’s less of an issue because I can rely on the phone to do a lot of the heavy-lifting.

Weakness: Immature Ecosystem

The Pebble ecosystem of apps is still pretty immature. There are quite a few apps out there — some of which are super cool! — but it has a long way to go. In addition, many of the apps with advanced functionality require a jailbroken phone, which kind of limits their reach. The good news is that they have a pretty rocking SDK that is reasonably well documented, so hopefully this is a problem that won’t be around for long. I think every new tech product experiences this for a while, but it’s definitely a bummer right now.

Project: Wrist-Up-Display

There were a couple of things I observed in the first two days of using my Pebble:

  • Apps weren’t super convenient to navigate to or interact with (two handed thing again).
  • There weren’t many watch faces that improved on the at-a-glance experience you get with an old-fashioned watch. Most of them just told you the date and time.

What I really wanted was the equivalent of a Heads-Up-Display, but on my Wrist: I wanted a way to present some small bit of contextually relevant information in a way that didn’t take me out of my “flow.” I decided it would be fun to try my hand at building a “Wrist-Up-Display.” I call it a “WUD” because it’s so fun and ridiculous to say. In the process, I got to dust off all the C programming cobwebs, which was an unexpected…benefit? Maybe? Anyway, it’s still a work in progress, but the general idea behind the first iteration of the WUD (hehe) is this:

  • I define a set of rules based around the time, date, day my location, etc.
  • When a rule is activated, it displays the corresponding information for that rule on the watch face

Using this approach, I don’t have to open an app or touch a button at all: I can just look at my wrist. For example, here’s a real rule I have on my watch today:

  • Conditions:
    • If it’s before 10am
    • And it’s Monday through Friday
    • And I’m within 600 feet of my bus stop
  • Result:
    • Display the time until the arrival of the next bus (route 545) at my stop using the OneBusAway APIs

Screenshot of my watch. Now if only it could hold a seat for me…

If I look at my wrist when those conditions are met, I see something like the picture to the right.

I actually managed to prototype this pretty quickly — a couple of evenings after work was all it took to get it up and running. I have quite a long way to go before it’s ship-ready (improve polling between the watch and phone to improve battery life, build UI for defining rules, etc.) but it was super cool to build a useful proof of concept in a matter of days.

Long term, a product like this could be much smarter. In addition to explicit rules, it could learn about the wearer and make some inferences. If I often put my phone on silent while I’m at the theater, it could learn that and silence notifications on the watch as well. Google Now has explored a lot of ideas similar to this, and you can imagine many of those on your wrist. For example, if I have a flight confirmation in my inbox, it could display my gate number, departure time, and seat so I don’t have to keep referencing my boarding pass.

The Future

I don’t think we’ll be doing all our computing on our watches in the future — in fact I don’t think we’ll be doing all our computing on any particular device. I think we’ll see our computing more spread out among “smart devices.” Much in the same way that our mobile phones have stolen jobs from our laptops (which stole jobs from Desktops, which stole jobs from pen and paper, etc…). Phones were really the first wave of “ubiquitous” computing, and in the next decade I think we’ll continue on that path toward truly ubiquitous computing: we’ll see an array of smart devices that can deliver highly contextual, highly personalized information to you in the right way at the right time. In some cases (while walking to the bus stop) that could be your watch, in others it could be a HUD on your windshield (blindspot warnings while driving), or your glasses (based on your location and calendar, turn left to find the office you’re looking for).

These devices have quite a long way to go before they transition from geeky fringe accessories to widely adopted tools, but I think we will get there much sooner than we expect. Adoption rates for personal technology are increasing rapidly — it took only 2.5 years for Smart Phones to go from 10% penetration to 40%. Compare that with adoption of the computer, which took nearly 15 years to make that same 10% to 40% move. One only needs to look at the popularity of FitBit and FuelBand for proof that people are willing to let these devices into their lives if they provide value, and this includes people who wouldn’t normally be considered early adopters. Combined with the growth of computing power, I don’t think it’s unreasonable to say that some classes of wearables could be commonplace among many demographics in 10 years or less.

The biggest question in my mind is: what wearable device will be the catalyst for wide adoption? It looks like Health and Fitness is the driving force. According to ABI Research, 90 million wearables are expected to ship in 2014, driven largely by Health and Fitness devices. FitBit and FuelBand have had a lot of luck getting people on board, though research suggests very high abandonment rates — up to 50% in the first 6 months! There’s clearly an engagement problem to be solved there.

There are rumors that just about every tech company is building a smart watch, Microsoft has published research related to a Smart Bra, and we’re all well aware of Google Glass. All of these devices have nontrivial obstacles to overcome:

  • They have to demonstrate value to those who aren’t tech-obsessed. Personal Fitness devices have done well here. We have to get away from the “just put a weather report on it!” fad we have now for “smart” devices.
  • Improve engagement by providing ongoing value. For example, they could provide health recommendations and insights rather than just raw data about calories burned and steps counted
  • Continue to improve the underlying technology in a way that brings their size down and capabilities up.
  • Focus on design — these aren’t just devices, they are accessories.

I also think there’s tremendous opportunity for the company that builds a family of connected devices — since you won’t interact with the world via one device, companies need to span multiple contexts and form factors.

Sent from my Pebble. Just kidding. For now.

No comments on "Running Code on my Wrist"

Leave a Reply

Your email address will not be published. Required fields are marked *