Tiiu Särkijärvi // January 26 2017
IoT: From Comedy to Commonplace
In recent months IoT (the Internet of Things) has been more visible than ever. We’ve seen a behemoth botnet of smart showerheads and other connected devices enable an unprecedented 1TB DDoS attack. We’ve also been gifted with the Ludela smart candle and the CHIP smart cookie oven, and this instant classic of a headline: “English man spends 11 hours trying to make cup of tea with Wi-Fi kettle.”
Botnets, ‘smart’ candles and 11-hour cups of tea. What as designers and engineers, can we do to our redeem ourselves? I’d say that we’re a crucial stage with IoT – right now, connected devices meant for everyday use are still the purvey of early adopters, but there is imminent potential for smart appliances and devices to make the leap from niche market to mainstream adoption.
However, there are five things we need to get right before IoT has a place in the majority of homes: Reliability, banality, interoperability, security, and – most importantly – limits.
Without considering these factors, IoT will never make the shift from comedy to ubiquity. And as much as I’d love an everlasting feed of Internet of Shit posts to ROFL until the end of my days, I’ll admit I’d be disappointed if we didn’t explore the benefits that well-designed and well-made connected devices could bring to our lives.
5 // Reliability: Fail better
This feels like a no-brainer. A basic requirement, especially since connected devices act as household appliances, home security solutions or installed in your car. Reliability is a foundational element to user experience design. In the UX design ‘hierarchy of needs,’ if the device or service fails at being always available or accurate, any efforts on convenience, desirability or even usability are wasted.
And yet reliability is more than ensuring that a device doesn’t fail or fails less often. It means preparing for the Kobayashi Maru – given your service or device will one day fail, what how will it fail gracefully?
In Calm Technology, cyborg anthropologist and writer Amber Case describes how owners of Philips Hue smart lightbulbs were left bathed in an irritating pink glow when an update to the lightbulb’s firmware proved buggy. Hue lightbulbs lets user turn their lights on and off with the Hue app; left unable to do this, it was only through a tweet from Philips that it became widely known that Hue light bulbs also turn on and off with boring old analogue light switches too.
Case sees it as necessary that when the ‘smart’ functions of a connected device fail, the device should revert to being functional as a ‘dumb’ device. So instead of failing outright, smart locks, light bulbs, refrigerators, bathroom scales should be designed to ‘revert’ to being as useful as their analogue counterparts.
Ultimately, an IoT-enabled home should be at least as comfortable and safe as one without; an update or a buggy Bluetooth connection shouldn’t impede you from unlocking your door or turning on your lights.
4// Banality: Consider the Segway
I’d argue that the opposite of banality is novelty. While originality solves problems simply, elegantly and sustainably, novelty solves problems with spectacular complexity.
So if originality is a bicycle, novelty is a Segway.
The bicycle and the Segway both solve the same problem: getting around a city without a car or horse-drawn carriage. Both cars and horse-drawn carriages have a massive environmental impact on the communities where they’re used. Today, we’re familiar with fossil-fuel driven climate change and urban air pollution, in times past the world’s first cyclists had to dodge small hills of horse ‘product’ on the streets of New York and London. However, while the bicycle is still used as a means of transport, a hobby and the foundation for several Olympic sports, the Segway has become a punchline.
True, the Segway is an engineering marvel, able to balance a human being leaning forward on only two wheels while moving forward. (Unless it runs out of batteries, of course, but that’s beside the point.)
I think it’s the Segway’s failure is grounded in being a novel solution to a real problem: an expensive, over-engineered, and over-hyped product that was designed to create its own market and bring its investors profit. Transport in cities is an unfolding health and environmental crisis, and the idealism of the Segway’s creators is sincere, but however noble their intentions, Segway forgot to consider the ordinary and the banal.
This was especially obvious in how the Segway’s creators seemingly failed to settle fundamental question of whether, in urban environments, the Segway would be ridden on the sidewalk like a mobility scooter or on the street with cars. In leaving this question open before going to market, the Segway has only really been successful in niche applications like tourism or indoor mobility. Prioritising the novel over the ordinary and the banal is ultimately counterproductive: the Segway will likely will never have the mass transport appeal of the bicycle. Even when the device itself is aiming at a lofty goal, if the solution ignores the banal and the obvious, it’s already redundant.
The sins of the Segway live on in devices like the QUBO smart bin: people who are conscientious enough to recycle, empty their bins regularly, and keep them clean aren’t likely to see the benefit in a connected QUBO smart bin, but those who don’t regularly recycle, empty their bins regularly or keep them clean won’t either.
3// Interoperability: The 11-Hour Cup of Tea
Let’s get back to the headline “English man spends 11 hours trying to make cup of tea with Wi-Fi kettle.” It’s click-baiting, revelling in its own absurdity: it should not take anyone with access to running water and a kettle 11 hours to make a cup of tea. Especially someone from a country notoriously full of people that would liken tea to a utility and not a mere refreshment.
The article itself, however, details the efforts of Mark Rittman, a writer and technologist, to get his WiFi-enabled kettle working.
“A key problem seemed to be that Rittman’s kettle didn’t come with software that would easily allow integration with other devices in his home, including Amazon Echo, which, like Apple’s Siri, allows users to tell connected smart devices what to do. So Rittman was trying to build the integration functionality himself.”
As technologists, we cannot ask users to do our jobs for us. While us dorks might spend a day trying to solve interoperability issues out of curiosity (or sheer stubbornness), the future of home IoT depends on persuading people outside the silicon tower that they won’t ever have to tinker their day away.
There are lots of ways to approach this: for example, American giant Target is helping customers get an overview of smart home devices through their openhouse initiative, and there are devices and services specifically designed to facilitate interoperability, like Samsung SmartThings and Cozify.
This is progress, but interoperability needs to considered non-negotiable both in designing devices and in planning IoT business models and revenue streams. It’s unproductive to try and lock users into a particular ecosystems within their own homes, or to expect them to troubleshoot the integration of a new smart device.
After all, the headline “English man spends 11 hours trying to make cup of tea with Wi-Fi kettle” is deceptive. Rittman is no ordinary English man, and he doesn’t spend all day trying to make a cup of tea so much as engineer a way to overcome an operational barrier between his home devices.
An ordinary English man would have given up after half an hour, driven back to the shop, and traded in his WiFi kettle for a decent Russell Hobbs with a 30-second rapid boil and 1-cup power-saving setting.
2// Security: Bad Vibrations
The Guardian definitely has form when it comes to blisteringly sarcastic IoT-related headlines:
Ouch. Someone call the burn unit. While it’s a giggle-inducing subject, the breach itself is no laughing matter. The extensive vulnerabilities of the We-Vibe 4 were exposed at DEFCON 2016 earlier this year by Kiwi hackers goldfisk and follower. Even more frightening than the easy hack of a very personal connected device was just how much data the device makers were collecting about their users.
As quoted in Motherboard, the duo highlighted the fact that we’ve become too comfortable with the idea the all and any user data is fair game:
“We want to question that assumption and say you know if you you’re making [intimate] devices that are controlled by mobile apps, maybe you should consider whether you should be collecting that information in the first place. If the information isn’t collected, then its not vulnerable to either security or data releases and legal enforcement.”
While sex toys are an embarrassing topic for us, it’s worth remembering that they’re illegal in some countries, where users can face real punishments like imprisonment or torture for possessing them.
Some might argue that these users are putting themselves in danger by buying devices which are illegal under the laws of their own sovereign country. However, that’s a weak defense for potentially exposing your user base to real danger because you might want their data for ‘research’ or ‘product improvement.’
Similarly, even though users might have consented for their data to be used for analytics, I doubt that they really understood that this meant livestreaming temperature and intensity settings in real time across an internet connection.
That’s really the crux of the matter: when the makers of connected devices get greedy for data they could unwittingly put their entire user base in danger. The recent DDoS attacks enabled by a massive botnet of connected devices exposed how easy it is for hackers to gain access to home IoT. We must now also consider what it is they’ll also have access to – which includes any and all user data. As Amber Case reminds us in Calm Technology, we must ‘respect user privacy. Your product is a service to them not to you.’
1 // Limits: It’s cool to be dumb sometimes
The proliferation of connected devices is no longer limited to smart appliances or other household gadgets. It seems that our furniture can also become smart – observe, the itBed smart bed. According to a recent review (bluntly titled, ‘Don’t Buy A Smart Bed’) it’s apparent that the itBed smart bed suffers from many of the same qualities that plague the smart devices we’ve already covered: it’s a novelty justified with a noble purpose, promising users better sleep through capturing data that maps their most intimate habits – if they can get it to work properly.
The interesting question when it comes to the itBed – or any smart bed – is not necessarily one of execution but of existence entirely. Does anyone need a smart bed? Is a smart bed ever going to be more useful than a high-quality mattress and bed frame when it comes to better quality sleep?
Let’s consider the noble purpose behind the ItBed. A recent report featured in Time magazine showed that more than a third of Americans don’t get enough sleep, and that the resulting fatigue costs the American economy an estimated $411 billion in lost productivity every year. The itBed claims to help its users to get a better night’s sleep though analysing their sleep habits as well as mapping how well they sleep on the mattress’s various settings. But is the root cause of insomnia or disrupted sleep really the down to the minutiae of your mattress?
There can be many reasons why someone is unable to get enough sleep – from shift work to caffeine intake. If you’re prone to binge-watching ‘House of Cards’ on Netflix until 2 AM and subsequently catch only six hours of restless sleep, do you really wake up unaware that you’ve slept badly? Will adjusting your mattress to be firmer really stop you from needing to know, right now, what Frank Underwood will do next?
No. It won’t, or at least it wouldn’t in isolation. There’s a lot a home IoT system could do to facilitate better sleep, like changing the atmosphere of your home to make it easier to ‘shut down’ after a long day.
Your home IoT could lower the lights after sunset and remove blue light from touchscreens and televisions, as well as limiting the volume of your home sound system. It could also create a better wake-up routine as well – you could be woken gently as your lights, home heating and kettle all gradually switch themselves on as efficiently as possible to ensure that your home is awake when you are. With the right smart devices around your home, maybe you don’t need smart furniture.
A better IoT is possible – but it has to solve problems, not make new ones
Maybe there’s only so many things that need to be hooked up to the Internet of Things – and this is going to be the biggest challenge yet when it comes to IoT. When all you have is a hammer, every problem looks like a nail; when you’re a technologist it’s tempting to see every problem as an opportunity for more technology.
After all, we love the stuff, and tinkering with tech is more than work for us. The devices we build, however, should just work. A smart toaster should reliably make toast – if its ‘smart’ functions are disrupted, it should still be able to make toast, like a regular ol’ toaster. A smart toaster should connect without protest to your home network, making toast in minutes, not hours. A smart toaster shouldn’t be an over-engineered monstrosity that aims to take over your kitchen to such an extent that you’ll eat nothing but toast three meals a day. Finally, a smart toaster shouldn’t easily sell out your data or get infected with malware, and maybe – just maybe – we don’t really need smart toasters. Maybe a smart toaster is as useful as a smart mattress. Maybe we actually need better designed smart thermostats and home security systems instead.
And that’s the real challenge that lays ahead for IoT. We’ll need to begin to create reliable, everyday IoT experiences that add real value to the lives of mainstream customers. Not to do so would be, quite frankly, a bit shit.