Why I Don’t Believe in Smart Watches

Thoughts About Wearable Computing

These days, there’s a lot of talk about smart objects and wearable computers. We are led to regard them as the next big thing after smartphones, now approaching market saturation in many places.

However, what strikes me is how little we really know about them: what they will do, how we will use them, and why we should prefer them to other kinds of computing devices. Even their very shape is not yet completely certain. Okay, a few smart glasses (Google Glass) and smart watches (Samsung Galaxy Gear, Pebble…) have been unveiled, but there is still no mass market success to show us a winning formula, no “product to beat”.

In your company, dear reader, I would like to speculate about what the next contenders for the crown of personal computing will be. If we believe past history, there will be at stake a lot of money, power, silicon and changes in our daily lives.

Warning: This piece is written from the perspective of a user with limited technical knowledge. This view is relevant in the long term, if we assume that the main technical challenges for core functionality (battery life, miniaturization and sensors) are already within reach, and solutions are maturing fast. I believe that the biggest unknowns are related to ergonomics, use cases, and social acceptability.

Wearables Are the New Frontier in the Computing Industry

Yeah, that’s what they all say. Well, almost all.

Computers Are Now Transforming Other Objects

Computers have changed a lot over the years. Once you needed a building to house one ; during the past decades, they have transitioned from mainframes to desktops, laptops, and finally smartphones and tablets. Because the hardware can now fit in a very tight space, you could argue that a “computer” is not a form factor anymore. Computers can invest other objects, and transform them at the same time.

The first such transformation happened with the phone, because the handset business had already integrated the hardware and transactional framework required to operate wireless internet access. Due to this transformation of phones, we now employ and enjoy much more the smart part (computing) than the phone part of our smartphones.

Is there any reason to stop here ? Of course not. And because the trend is already well known, most big tech companies are actively preparing for the next revolution, lest they lose their current position. You could call that a self-fulfilling prophecy.

Personal Computers Can Be Redefined Again

In this piece, I’ll be looking at wearables from the angle of the personal computer, rather than mere appliances and sensors.

The very definition of a personal computer has proven to be elusive, much more changing than we once thought. At the core, a personal computer helps us to execute tasks reliant on data processing. This scope expands each time we manage to add new actions and new data to the digital world (e.g. photo, music, games, e-books…)

Using traditional PCs requires to sit in front of a screen for extended amounts of time, usually typing on a keyboard to do “computer stuff”, centered around documents and work. With mobile computing, the new personal computers increasingly blend into the fabric of our workflows, they are tools, albeit with amazing capabilities, that we use to make our daily life more convenient.

The next “personal computer” would not annihilate or make irrelevant the previous one, but it should capture a significant share of the time we spend doing computing tasks. It will probably not perform all of them, but that should be no surprise : after all, does anyone build spreadsheet on their smartphones?

To read what respected tech blogger Marco Arment wrote recently, smartphones are already the best always-with-you device, so why bother with additional mobile stuff? I believe other differentiators than pure mobility — notably ergonomics — will also play a key role, but I share his scepticism regarding smart watches. But you should already know, if you read the title of this article.

Not All Objects Can Be the Next Game-Changers

There are two categories of objects that I’d like to remove from the discussion.

Some Will Be Simple Vehicles for Embedded CPU

Pieces of clothes, such as pants, shirts, coats, belts, shoes, are highly susceptible to wear and tear, or repeated washing machine exposure. They are better suited as the medium for cheap embedded CPUs than expensive computing devices. Furthermore, those objects are often located in a place of the human body that just isn’t convenient for repeated manipulation — smart shoes anyone?

By embedded CPU, I mean specialized computing devices, allowing simple interaction with the wearer. The most basic example is the RFID chip, which enables identification and remote action. Houses and cars open automatically to their owner and to nobody else; patrons of a bar are immediately allowed to the VIP zone and served their preferred cocktail, etc. Such devices are (relatively) low cost, perform a specific duty, and their main data input methods are their sensors or passive radio reception. They are not likely to function as a personal computer, because of their very limited human interface.

It doesn’t mean that embedded computing will have no impact ; on the contrary, we probably stand on the edge of a radical transformation of our everyday environment, the so-called “Internet of Things”. Our children may well grow up in a world where doors speak, clothes complain when they need to be washed, cars suggest destinations to their passengers (no drivers required); a magical world. It might also become a tedious world harassing us with ads, or a paranoid’s nightmare spying on our every action, but in any case the impact is going to be huge. Actually, that’s a topic for whole other piece on Medium.

However, these wearables will only be pieces of a whole network, and none of them will be the hub, the expensive personal computer with which we have a rich, even holistic interaction.

Some are Just Not That Good

Many wearable items are strictly inferior to other existing choices.

For example, hats and caps are worse than glasses, mostly because they are much bigger and susceptible to be blown away by the wind (unacceptable!).

Jewellery raises interesting questions. A smart necklace (did I just write this?) would be equivalent to a smartphone mounted on a chain, which is not a compelling proposition. Plus, the appetite for wearing necklaces is not universally shared. In the same order of thought, I will not elaborate on earrings, even though there is some cool Sci-Fi to write about them.

Finger rings are functionally the same thing as a wristwatch, but even smaller, and with no significant upside apart from the ability to wear several at the same time.

So, maybe there was a reason for all the hype about glasses and watches.

It is very likely that these two devices will carry more computing punch in the future ; but for them to become more than specialized tools for professionals or hobbyists, they need to have more than marginal use cases. In this regard, watches and glasses have very different potentials.

The Smart Watch: a Specialized Device For Health and Fitness

When I started thinking about this piece, I was sold on smart watches — not just wristbands with a cheap screen, but the real deal: star-trek like, personal-computing watches. After all, watches have always been computers, counting time in different ways (calendar, hours, chronometer) and displaying it for serious, in the field uses. The formula has been tested and perfected over a very long period of time: this is why watches are so conveniently located, and equipped with a well-designed, user-friendly output interface — at least very good for indicating time. So they should be easy to bring up to the next level, right?

Yes, a Smart Watch Would Have Significant Upsides

Not Easily Lost: One of the big weaknesses of smartphones is our propensity to let them fall and break their huge screens, have them stolen, or just lose them. Objects attached to our bodies are much safer, which makes a lot of sense for valuable, mobile computing tools. In this area the watch is hard to beat: even glasses can fall or be stolen, unless you screw them to your skullbone.

Collect Biometric Data: There are already apps that count your heartbeats using the LED flash and camera of a smartphone; but you have to consciously and manually perform the whole operation. A sensor-equipped computer attached to the body could continuously capture loads of data about your body state, which can either save your life or be totally creepy, depending on who’s controlling it. This biometric data, such as pulse, blood pressure, temperature, etc, would enable to create whole new categories of applications in the fields of health, fitness and sports.

To be fair, smartphones have already access to a lot of data using their accelerometer and other sensors. Apps are already able to monitor how much you walk, how fast, how well you sleep. And they’re getting better: improvements such as the M7 chip further increase the capability of smartphones in this area. To outsmart them, a wristwatch will have to collect a more advanced set of data, of the kind needed for serious health uses, or high-level sports.

Easy to check at a glance: just twist your arm a bit, and you can see if you have any incoming mail, who’s calling you, or any other notifications. The gesture is almost hard-wired in our DNA. This is the most obvious, but also the most overrated of the features of a smart watch: would you really pay several hundreds of dollars, just so that you don’t have to take your phone out of your pocket to perform basically the same checks?

It could also be easily accepted

Appearances will make or break these products: watches are fashion accessories, and anyone paying hundreds of dollars for something to wear on his/her wrist expects it to deliver good looks. It plays to the advantage of companies with strong design skills (who could that be?), but also creates a specific challenge : how to build a radically new object, that still reminds you somehow of Dad’s good old watch? It wil also be necessary to accomodate the need for customized look — especially along gender lines, as men and women tend to carry different-looking watches. While it is a question to address seriously, there is nothing impossible here.

We already carry watches, and use them as primitive wearable computers. Recycling the same form factor for new, smarter tasks will create very limited user acceptance issues, if any. A company planning to launch a new form factor may look for a path of least resistance, an existing device ready to be smoothly transformed. The watch arguably embodies this path of least resistance for the future wearable personal computer; and without surprise, it is the path that everyone is expecting Apple, the most user-aware company in the industry, to choose.

So why am I so sceptical ?

Functionality Would Be Severely Limited by the Form Factor

For touch, screen size matters a lot

The limited screen size of a smartwatch makes it impractical for intensive media consumption such as video, web surfing, and for gaming; that is, unless we find a way (Pico projector? Virtual lens? Portable black hole?) to expand screen size without jeopardizing portability. The same can be said about text entry by touch.

There has been wild speculation about curved screens, pieces of glass made in a shape that would circle around your wrist. It is supposed to solve this issue of screen size. Well, I haven’t had the opportunity to hold one in my hands, but a simple look at my wrists tells me that the increase in surface, while useful, would not radically improve usability. In any case, one could not use the whole external surface of the strap, which, being a ring, would never be visible in its entirety by the user.

Voice Input Will Have to Improve a Lot

The limited physical input options (screen size, number of buttons) mean that any advanced use of these devices would have to rely massively on voice command and dictation. However, pure voice input seems unrealistic at the moment. My personal experience shows that frequent on-screen correction is required, especially when entering uncommon words. But this is still a young technology, and the issue might be solved in the future.

Still, even if it were technically feasible, would people actually use voice input? Existing solutions (Google Now, Siri) work great in some contexts, but few people seem to use them in a systematic way. It is caused either by their lack of reliability, or the awkwardness of dictating private messages in public. People have gotten used to the apparent monologues we have when talking on our cellphones in the open, but speaking to a machine brings the weirdness to a new level. And in some situations, we really have to remain silent.

I bet that once the performance of voice input becomes good enough, the uneasiness will be outweighed by practical advantages. However, it will remain a negative for a device to rely on voice, due to the loss of conveniency and privacy.

Wearing a Computer at Your Wrist Is Actually Not That Great.

There are many typical smartphone uses that are just not going to happen on a smart watch, if only because they require you to keep one arm continuously positioned: video calls, games, reading. Taking pictures and video seems cumbersome as well, due to the need to aim with the whole arm, without using the finer motricity of the wrist and fingers. Even audio calls would be slightly less convenient than with a cellphone, unless you systematically use a headset. And plugging in a headset at the wrist will cause cable dangling around at each movement of the hand (but maybe that’s just me).

In addition, even if carrying a watch doesn’t mobilize your hands, by definition one of them can never reach it: thus single handed use is mandatory. Noone is ever going to type single-handedly on a watch, no matter how cool it is. And because the watch is bound to move whenever the arm carrying it twitches, the recourse to extensions — such as a pico projector for increased screen space — will be very challenging.

Compared to a smartphone, in exchange for slightly increased convenience, the watch loses a lot of functionality.

So Don’t Expect Smart Watches to Be That Revolutionary

Apart from biometric data collection, this form factor has few breakthrough uses. Interaction could occur through buttons, a touch screen, and voice: this is only a variation of the smartphone paradigm, with a few additional features in exchange for many big losses in usability. To make this device a real game-changer, a groundbreaking and unexpected innovation would be required in the way we interact with it.

Because they won’t make the smartphone redundant, smart watches will have to coexist with it. Therefore, to keep the price low, they will probably be focused on their core use cases and differentiators, especially biological sensors. They will rely on the smartphone to provide wireless internet connexion (using bluetooth, or wifi tethering), and even heavier data processing. Therefore, I expect smart watches to remain specialized smartphone peripherals, rather than a distinct device competing with them for our money and daily mindshare.

Would you buy this kind of device if it were sold below, let’s say, $200? If you are serious about sports, or have a serious health condition, very likely (not to mention the gadget-loving crowd). And for these users, it could have a life-changing impact. But for many other people, the cost, set-up effort and cognitive load will be harder to justify. Therefore, we’re talking about a potential market that doesn’t compare to the volumes of tablets and smartphones sold worldwide.

So that’s my conclusion: although it has a real potential as a specialized device, the smart watch is not a smartphone successor.

Meanwhile, smart glasses might bring bigger change, and brutally so. This will be the subject of my next piece.