Life on the small screen

Recently I’ve been living life on the small screen.

Specifically, that’s been both a smart watch (the Sony Smartwatch 2) and a fitness device (the Fitbit Flex). I’ve been looking at the interface experience on small screens, and using both in anger, day to day, for everything you’d expect to use them for.

In the case of the smart watch, I’ve been using it for all the possibilities currently (realistically) possible – gaming, communication, notifications, responding to messages, changing phone settings, weather updates, rejecting calls and – gasp – telling the time. For the fitness device I’ve been using it to set /use alarms, track sleep, track activity, report progress, record a food log, and set overview goals.

Both come with an app that sits on the phone and allows for higher levels of interaction and management, but I am not commenting on those here; they’re relatively okay interfaces, they have their faults but they work, for the most part. Given the interface real-estate and the interactivity available on a phone, that isn’t really surprising. What’s interested me most, is the challenges, successes and failures of using the small screen – or, in the case of the Fitbit, a few flashing lights – to communicate with and control such a device.

The results?

The results are – well, a mixed bag.

Starting with the smartwatch, I have to say that I’m convinced the challenges have been met, for the most part. Obviously this platform is well suited to output, and that works well. Whether it is displaying a notification of an SMS, an email or a photo, these interfaces can do the job relatively well. Navigating the content by zooming or scrolling works relatively well, and is relatively intuitive. If you know how to use a smart phone, you’re not going to have too much trouble figuring out how to use a smart watch.

There are also some clear advantages offered by this platform. For example I discovered that I missed far fewer calls, with the device-to-skin connection making a subtle vibrating device far easier to detect. Where I often miss phone calls whilst power marching (yeah, okay, ambling) through the streets of Sydney, my phone grumbling unnoticed in a pocket somewhere, I found I never failed to miss the vibrating wrist watch. Not only this, but the reaction was generally faster – glancing at the wrist is far faster than digging around in a pocket to extract the phone and see who’s calling, so I found I missed fewer calls I wanted to take, and spent less time slowing down to dig out a phone for a call I was happy to miss; a win-win.

These screens are also a boon when it comes to subtlety. Sitting in a meeting and glancing slyly at your watch is far less obvious than digging out a loudly rumbling mobile phone. Again this is ideal for this interface type, neat small notifications you can choose to ignore or respond to. You get slapped far less often when glancing at a watch during your kids recital, I’ve found.

Finally, the news was good when it came to extending to remote functions. One app for the Sony is a mobile camera remote control. Place your mobile somewhere nearby and you can use the watch to check the picture and press to capture the shot. It’s not a killer feature, but useful – and in terms of interface options, simple remotes are definitely a winner on the small screen.

When it comes to input options – for example, responding to an email or text – life gets a little trickier, though certainly not impossible. Some apps on the smart watch allow you to set up default responses – for example “running late, be there shortly” or “it happened again, send help!”. You can then select these responses and send them. This is not exactly smooth, and at times can be quite slow to select – but it still works. It’s starting to stress the platform somewhat though, in terms of simple human interaction – for example I found several times that I’d forgotten how this feature worked, and missed using it at the moment I needed it.

As of today, more complex input interaction isn’t really available – mostly, it must be said, because nobody has really worked out how this can work effectively, short of voice-interaction (which is missing from the devices I’m currently testing. And yes, you will look like an idiot, telling your wrist that you’re going to be there shortly.

Games are somewhat limited, and in my experience were either too fiddly, too limited or just too ‘not fun’ to bother with for more than a few minutes. This screen format is minuscule when it comes to an immersive experience, is slow and unresponsive in terms of play interaction, and is simply too limited to be fit for this purpose. Here, at least, the small screen fails.

So overall – at this stage – watch interfaces are, as far as I can see so far – working.


When it comes to the fitbit, it’s a slightly different story.

Whilst the app that comes with this is relatively powerful and fun to use (excluding the food log, which should be taken outside and given an interface wedgy), the interface here is a set of five lights. These are used to provide messages; your completion of your daily goals, your current goal percentage, and confirmation of entering and exiting sleep mode.

In theory, given the limited purpose of the device, that should be sufficient. In effect a single interaction is supported, a multi-tap on the interface. Tap a little longer and this becomes a toggle for entering/exiting sleep mode.

For this interface, you don’t exactly get a head start. You are provided with a link to a website in lieu of a user manual, and so you start somewhat on the back foot – with a flashing thing on your wrist, and no idea what the light patterns mean.

At one point you tap, and one light flashes at you. You tap again, and this time two lights flash, and dance back and forth. Another time they all flash a few times, then disappear. What does it mean?

It took me repeated visits to the website to figure that out, and to remember.

Several days after I’d started using it, I still found myself getting confused. Tapping to see my progress, I saw the two dancing lights. Urhh? I think that means it’s still in sleep mode. But who can tell.

Part of the problem, it appears, is the difference between ‘multi-tap’ – which is two, relatively fast taps, and ‘long tap’ – which is tapping several times over a second or two. Not exactly a huge difference, and something that I struggled with. Repeatedly I would accidentally set it into sleep mode when trying to validate how far I’d walked.

One success was shared with the smart watch, tactile response. The device comes with an alarm which silently wakes you at a set time (set on the phone app, since this would be impossible directly on the device). This was something I was keen to try, and it worked very well, at least for me (a renowned light-sleeper). The gentle vibration pattern woke me on time and first time. A success in terms of interaction.

But it failed miserably in terms of control. Not knowing what was required to turn the alarm off, I tried the multi-tap thing. It didn’t seem to have any affect, but since the alarm stopped after three vibration sets, not a problem.

Half an hour later however, three more vibration sets. Was the alarm still set? Had I snoozed it? Or changed it? I had no idea. But a bit later it did the same thing.

Perhaps my device is faulty, and more than likely this has been explained somewhere – I’m terrible for ignoring or misreading user manuals, so I’m not claiming this isn’t explained somewhere. But that’s not the point. Faced when a device that only seems to support one interaction, the tap (or multi, or repeated tap, or whatever it might be), how do you know what to do?

Years ago I had an early MP3 Player USB device, that was controlled via a small jog wheel. You could move up or down with the wheel, or press it like a button. Whilst this device supported music, file selection, volume and various other features, it was simple and easy to use – the wheel changed context depending on what was displayed, and I found myself understanding it without ever having to read the instructions. I loved it till the day I lost it on a train. Now that is usable life on the small screen.

Fitbit works, and once I fine-tune my understanding of the interface, I’m sure I’ll time it fine. But in terms of life on the tiny 5-dot interface, I would mark it a failure. It’s not obvious, it’s not easy, it is error prone and at times very confusing to use.


Small screens are only going to proliferate. So let’s make sure usable experiences proliferate with them.