Is the Internet of Things all hype? Nicole Kobie explains why smart homes and other connected gizmos are struggling to capture the imagination – and purchasing pounds – of consumers
The promise of the Internet of Things (IoT) and smart homes is glorious: a life of leisure enabled by the tech equivalent of a house staff straight out of Downton Abbey. A revolution in home convenience unmatched since the heyday of consumer home appliances in the 1950s.
Imagine it: no need to call your family to tell them you’re heading home from the office: your car will ping them when you get behind the wheel – and take over the driving when you hit the motorway.
As you arrive, the lights turn a warm glow to welcome you and the door unlocks without any fumbling for keys. A digital voice says ‘hello’ as you step through the door, and asks if you’d like to play your favourite album. Dinner is suggested by your fridge – it tracks when food is about to go off, so it’s eaten while still fresh – and cooked by your oven in a recipe downloaded from the internet.
The dishwasher automatically turns on at 1.43am, precisely the time when energy prices are lowest and the noise is least likely to disrupt your sleep, which is in turn measured by the wearable on your wrist that awakens you the next morning at just the right time in your sleep cycle. That watch alerts the coffee machine that you’re awake so your morning cup is ready when you step into the kitchen, which is warmed to a perfect temperature by the smart thermostat that takes into account how sunny it is, and turns itself down in advance of the warming sun streaming through the windows.
Smart homes could be like having a cloud-based butler who attends to your every digital need – but there are downsides that could keep this tech convenience utopia from becoming a reality.
All the technologies mentioned exist in some shape or form already, though they may be a few years away from commercial availability. But while the technology may be in place, other challenges remain: many consumer IoT devices aren’t yet actually very useful, especially given their high price, while the classic IT issues of compatibility, security and privacy are exacerbated by the always-connected sensors collecting data on everything we do. In other words, smart fridges are expensive, smart cars are hackable, and connected home helpers extend surveillance into our living rooms, kitchens and bedrooms.
Such challenges aren’t expected to hold back the IoT. Industry analyst Gartner is predicting there will be 4.9 billion connected devices chatting to each other and to databases this year, leaping to 25 billion within five years. While much of that will be business uses, consumer IoT will make up half of those devices by 2020, Gartner believes.
But will the IoT succeed with consumers? Will our homes become truly smart, will we have a few intelligent connected devices, or will we merely have an extra handful of gadgets cluttering up our shelves? These are the challenges the IoT needs to overcome to find a place in our homes and our lives.
Product problems
Setting aside for a moment the other challenges the IoT must overcome, smart home kit has to perform one particular role above all: it has to do something truly useful.
Look at smartphones. They’re now ubiquitous, despite compatibility issues, potential security flaws and privacy concerns. But who cares if you have to choose between iOS and Android, if it means you have a map of the world with directions to everywhere in your pocket? So what if cybercriminals and governments are targeting handsets, if we can have access to all our music, games and the entire internet everywhere we go?
Convenience is one heck of a selling factor, but many IoT devices are niche compared to the modern Swiss Army knife that phones have become – and their high price means it’s costly to get the benefits they do offer.
Look at Belkin’s Wemo lights. Plug a Hub into the wall, screw in a smart bulb, and your light can now be controlled from a phone. A starter kit costs £60 – you need to really hate flicking the light switch to invest in that.
There are other issues aside from price. We spent a solid hour trying to make a Belkin Wemo smart switch work with a Motorola phone, but it was impossible. It worked with an iPhone and an alternative Android handset, but a yet-to-be-issued update was required for specific smartphones – a point not noted on the product’s packaging.
Other smart home gadgets weren’t as smart as we’d hoped. The Netatmo Welcome camera’s facial-recognition feature means it can tell when an unrecognised person enters a room. Image quality is fantastic and setup is simple, but the key selling point of automated recognition identified our microwave as a human face (possibly foreseeing an appliance uprising) and still gets confused by different hairstyles and glasses. Not so smart, then.
What’s the use?
And these are some of the more worthy examples of home IoT. Others simply aren’t worth the cost or trouble for the small benefit they bring. Parrot’s FlowerPower plant monitor issues notifications for water levels, but only if you’re in Bluetooth range, as it doesn’t support Wi-Fi. It also can’t recognise the plant’s variety to decide the right moisture level – and you still have to do the watering yourself (for now, at least; a new version will water them too). For £50, you have to do all the work yourself. What’s the point of that? Unsurprisingly, our test plant shrivelled and died.
Of course, some consumers may find these products useful. Thanks to the IoT hype, developers are shoving connectivity into all sorts of products: Drop is a connected kitchen scale (£80); June is a smart bracelet that tracks sun exposure (£80); and Darma (£130) is a smart cushion that helps your posture. If it’s a thing in your house, someone has tried to connect it to the internet.
Those could all be useful to someone, and aren’t the silliest by far – see ‘The Internet of Useless Things’ (opposite) for proof of that – but many are niche devices that may appeal to a few, but never break through to the mainstream. Crowdfunding sites Kickstarter and IndieGoGo are full of such gadgets, but the most popular smart home products remain thermostats and security devices, the only proven product categories to date.
That’s a lesson learned the hard way by Indian entrepreneur Yash Kotak. While he now runs startup FundaMine, his last project was the Lumos switch, a smart controller for lights. He shut it down in 2014. “The value that Lumos switches provided did not justify the cost the target customer had to pay for it,” he said. “Because we couldn’t find a way around this, we shut down the project. Turning lights on with a smartphone is cool. But it does not add enough value. Party tricks won’t sell.”
He believes more IoT devices will fail for the same reason. “Other products are facing such problems. The cost of IoT hardware products is too high right now,” he said. “The value-pain equation for your product should make sense for your target customer. Unless value [is more than] pain, your product will not sell.”
That doesn’t mean startups should leave IoT devices to big players, though. “Just build something that’s not only cool but also provides value,” Kotak said, pointing to Nest and DropCam, two startups bought by Google.
Google was wise to choose the smart thermometer and security camera firm, noted Berg Insights analyst Lars Kurkinen. “In smart homes, it is especially smart thermostats and security products such as home alarms and home surveillance systems that are the big sellers today,” he said, and that’s little surprise given the utility offered by both.
The number of homes with smart thermostats made by the likes of Nest, Honeywell and Hive grew by 96% last year, with Berg Insights predicting such growth will continue because they’re genuinely useful for consumers by helping them save money. They’re also good for energy companies who have a remit to reduce power usage and also want more data on when and how energy is being used.
Those are much bigger selling points than a Bluetooth notification that a geranium has a 23% water level.
Working together
Even if devices do actually work, they often don’t work well together. The Internet of Things isn’t just supposed to let us talk to devices, but to let them talk to each other - indeed, the idea has previously been known as machine-to-machine (M2M).
The problem is most devices aren’t capable of sharing the data they collect and telling each other what to do about it. It’s an IT problem as old as the industry: interoperability. There are efforts to solve the problem. Tools such as If This, Then That (ifttt.com) make it easier, acting as a link between devices that don’t actually communicate. You can, for example, make a ‘recipe’ to turn up your heating if the outside temperature falls. Your smart temperature gauge uploads its measurements to cloud-based servers, IFTTT reads the data, and triggers your smart heating system.
However, manufacturers are also trying to build in interoperability at a more basic level, building standards to ensure devices can play well together. The problem is they’re each writing their own set of standards - from the AllSeen Alliance to the Open Interconnect Consortium to the Thread Group, all backed, respectively, by tech giants such as Microsoft, Intel and ARM, while the UK has its own data cataloguing standard called HyperCat. "There are too many interoperability initiatives competing in the market today,” noted Berg’s Lars Kurkinen. "This is not only a barrier for consumer adoption, it is also an issue for product manufacturers. Currently, product manufactures need to develop several different versions of the same product in order to ensure compatibil|ity with the widest possible range of smart home systems.”
There’s a (smart) light at the end of the tunnel, however, though it comes with an existing tech divide. Apple’s HomeKit and Google’s Brillo are software platforms that look to put the smartphone at the centre of consumer loT, making it easier for devices from disparate developers to connect and share data, and for users to manage them. “Apple’s and Google’s entry will accelerate the consolidation towards a few leading interoperability standards, which benefits the market as a whole,” said Kurkinen.
Smart home devices from Apple started to arrive this summer, though many remain US only. There’s the ecobee smart thermostat, iHome smart plugs and Elgato Eve, a wireless sensor to monitor air quality and temperature, with more products expected later this year. Google’s Brillo smart home platform was announced only this summer and products haven’t yet hit the shelves, but one of the first will be Google’s own OnHub, a smart router that will eventually act as the connectivity hub for smart homes, linking up devices and letting them interact.
Amazon is also making a play for the smart home. It has launched internet connected buttons that let you add items to your Amazon order from anywhere in the home, such as a Gillette button that sits beside your razors in the bathroom. There’s also the voice-activated assistant Echo. Essentially Cortana or Siri for your living room, you can ask the Amazon Echo’s voice assistant Alexa any question, tell it to stream a playlist, and – this being Amazon – make a shopping list. The Echo can also be set up as the centre of a smart home via the Alexa Skills Kit, a set of APIs Amazon released in June.
Will Google, Apple and Amazon dominate the IoT as much as they have the web? It’s hard to imagine they won’t, but even if they sort out interoperability within their own platforms, there’s still two more challenges none of those firms has yet to solve completely: security and privacy.
Keeping it secure
The headlines are horrifying. Researchers take over a connected car, turning off the brakes while a journalist careers down a motorway. Creeps crack internet-connected baby monitors, screaming at small children. And everything from smart TVs to lighting and toilets have been found to hold security flaws.
Many of the biggest such hacks have been research projects by so-called white-hat hackers working with the industry, rather than actual criminal activities. Still, that doesn’t mean IoT is insecure in theory and safe in reality. “They have identified more problems in terms of the breadth of IoT devices than there have been abuse cases, but that cannot downplay the impact when a single insecure device may be abused by criminals and impact tens or hundreds of thousands of consumers,” said Mark Stanislav, senior security consultant at Rapid7. “The problem is, however, that the pace of IoT growth makes it hard to imagine security researchers being able to sustainably outpace criminals.”
That’s why high-profile hacks such as those against connected cars are so important, he argued. Charlie Miller and Chris Valasek are infamous for hacking multiple smart cars, and such behaviour recently won the researchers a job with Uber. Earlier this year, they showed noticeable glee as they took over a Jeep while it was being driven down a motorway by a nervous journalist, messing with everything from the radio station and air conditioning to windshield wipers and transmission, cutting out the accelerator.
“This example shows what happens when connectivity enters our physical world and can directly impact our safety, rather than just imply inconvenience to us or reduce our privacy,” said Stanislav. “While any abuse of a person’s property or life is horrible, the research they performed will hopefully serve as a catalyst for change in the automotive industry and will prevent real harm coming to real people. That’s important.”
As terrifying as the stories are, it’s difficult to say if security flaws and high-profile hacks are enough to dissuade shoppers – it’s worth remembering that hacked affair website Ashley Madison has added users since it was attacked – though larger firms are less likely to get punished for missteps than smaller startups, said Stanislav.
“Whether it’s a stolen credit card, stolen personal information, or worse, we see consumers return to brands they trust, regardless of security history, which brings significant questions around what constitutes brand loyalty,” Stanislav said. “This is a challenge for lesser-known brands or startups, however. Companies in the IoT market segment or product space, which has publicly experienced security issues in the past, are facing consumers that are less willing to take risks on relatively unknown brands.”
Cramming connectivity into everything we own will naturally lead to security challenges, but they aren’t necessarily insurmountable. At the moment, not enough work is being done to build security – or privacy – into such devices. F-Secure security advisor Sean Sullivan believes it’s a matter of cost. “IoT devices will only be secured as much as they need to be to exist on the market,” he said. “It’s up to consumers to demand more than a low bar. Active exploitation in the wild might force improved security, or then it might just kill or suppress the home IoT market.”
Security need not be expensive or complicated. “A television with a built-in camera would be a great way to chat to my mother and son,” Sullivan noted. “But such a TV had better include a manual cover for that camera lens if it expects to find a place in my future home.”
Some effort is being made to build security into the IoT, and that includes BuilditSecure.ly, a group co-founded by Stanislav that helps developers get in touch with security researchers who will do pro bono work to ‘sanity check’ devices before they’re sold to consumers. “Many IoT vendors don’t have a dedicated security team or even person on staff, which hurts their ability to leverage these initiatives,” he said. “Especially for smaller organisations, security expertise is often an afterthought, leading to many of the issues we see come to market.”
If you’re shopping for a smart home device now, be aware of that. “Participation in these nascent years of IoT does come with some risk, no matter how large or small the brand,” said Stanislav. However, it must also be noted that such criminal activity follows the money, and there will need to be a clear profit stream before serious work is undertaken by hackers to undermine your IoT devices.
Privacy concerns
Do people care about privacy? The surveys suggest they do: research by security firm TRUSTe in 2014 suggested only 22% of people think the benefits of smart devices outweigh privacy concerns, and one in eight want to know more about the information being collected – and want to control their own data. That could suggest people won’t let smart devices into their homes, though it might just be that the perceived benefits of smart devices aren’t yet apparent enough to be worth any risk, however small. Either way, smart devices will need to convince consumers they aren’t a privacy risk.
Such privacy concerns certainly aren’t unwarranted. Your smartphone has the potential to know exactly where you go, what you do and what you buy, but smart home gadgetry can extend that data collection to inside your home, revealing what you eat, when you’re in the house, and other intimate details about your home life.
“The biggest privacy concerns with the IoT are, as far as I’m concerned, that the kinds of data gathered are new, and directly connected with what we do in our ordinary lives,” said Dr Paul Bernal, a lecturer in IT and privacy at the University of East Anglia. “We’re not used to the idea that people might know this kind of information – and haven’t even really considered what they might do with it.”
Bernal said car manufacturers, for example, could use driving data to improve how vehicles work. While we might approve that use of our data, if insurance companies could also access it, they could use that information to increase premiums. What if a smart home device reveals to your employer that you’re not working from home as you’re supposed to be? Or if a service is hacked and reveals an embarrassing habit? Anyone who signed up for affair website Ashley Madison will know that feeling.
"I don’t, however, think this is on everyone’s minds yet - we still like cool gadgets more than anything else, and the privacy implications haven’t yet become clear to everyone,” Bernal added. “The Ashley Madison hack may be another step towards this understanding - people need to be shocked into understanding, I think.”
On top of selling the data on to third parties such as insurers or it being hacked by criminals, the loT could also give government spies another way to snoop on us, Bernal noted. "I’m afraid the loT is almost certain to become another layer in the surveillance infrastructure - much of what makes it work is about surveillance,” he said. “It monitors. It watches. It analyses. That’s the point of it, really - the question is whether we put appropriate controls in place to make that surveillance more beneficial than damaging.”
Knocking down hurdles
Despite such challenges, tech companies clearly believe the loT and smart homes are the next big thing - everyone from Apple and Google to Amazon and Microsoft is getting behind the hype. Whether consumers buy into it is another question. “I’m not sure the loT will be quite as big as the tech companies expect, but I think the jury’s out on that,” said Bernal. “It may become something we just accept as part of the way things work.”
Rather than an overnight loT revolution, then, it’s more likely that smart home tech will slowly creep into our lives, as the challenges outlined here are whittled away - or we become so used to the conveniences that we ignore the downsides.
While some smart home tech will arrive via individual devices, F-Secure’s Sean Sullivan predicted that new-build houses will be where it really takes hold. "Currently, there are folks upgrading their existing home and that’s a niche,” he said. “I think general consumers just want stuff that works and real growth will come once home-builders start to offer fully integrated systems.”
Either way, the change will be gradual, said Lars Kurkinen. “It starts out with bits and pieces of smart technology that consumers take into use when they find it personally useful,” he said. "Over time, all the connected devices in the home will be able to communicate with one another to create a truly intelligent home that serves the user in a wide variety of ways, many of which we have not yet imagined.”
That smart home utopia of a world full of conveniences may well happen, but it might be a long process to get there. The foundations will need to be sound, without having too many competing platforms and standards. It will need to be secure, so we’re not invaded by undesirable elements. And we’ll need transparent control of our privacy, so we know just what we’re sharing with the Googles of this world. In short, let’s get our cloud-based butler just right before giving him the keys to our cars and homes.