IoT Reality: Smart Homes Not Smart Enough Yet
The Internet of Things needs standardization and easier programming options. Until then, the promise of IoT for consumers remains unfulfilled.
The Internet of Things is the most exciting thing to happen in the technology industry since the introduction of the iPhone in 2007. At the same time, it remains challenging.
The iPhone demonstrated the utility of having a connected computer available at all times and provided proof of the utility of cloud services at a time when on-premises computing was still the norm.
The Internet of Things promises similar benefits by adding computational power and sensing capabilities to previously non-communicative objects.
But that promise isn’t yet convincing, at least as far as consumers are concerned. Last June, Icontrol Networks, which offers white label IoT gateway and device management software to connected home service providers like Comcast and Time Warner, published its 2015 State of the Smart Home Report based on a survey of 1,600 consumers from the US and Canada. Icontrol found that about a third of respondents said they intended to purchase a connected home device in the next year.
In a phone interview, Corey Gates, CTO of Icontrol Networks and IEEE member, said smart home adoption has moved beyond early adopters, but he noted that home security remains the dominant application by far. “Peace of mind as the value proposition is the No. 1 thing people will spend money on,” he said.
The easiest path to a smart home, Gates said, leads to service providers like Comcast that offer professional installation. DIY device installation, networking, and programming is the more affordable option, but that approach can be rough.
My own experience with Scout Alarm and Nest devices offers an example of the some of the pitfalls and the benefits.
After years as an AT&T and ADT customer, I was fed up with high fees and went wireless over the holidays. I was paying about $50 per month for a landline I had maintained for my alarm system and about $30 per month for alarm monitoring. I cancelled my ADT and AT&T accounts in order to switch to a self-installed Scout Alarm system and Ooma VoIP calling. (I considered Piper, a competing alarm service run by Icontrol Networks, but went with Scout because Piper’s focus on security through video and motion sensing seemed ill-suited to a household with pets. Piper says its system can be tuned for pets.)
Scout Alarm was easy to install and comes with a well-designed app. It’s mostly a pleasure to use, and it’s more affordable than legacy systems. But its hardware isn’t very flexible.
Its access sensors and door panel, for example, work well on doors and door frames with flat surfaces, but they don’t sit well on the sort of decorative molding and trim used in old houses. ADT offers magnetic sensors that can be inserted into holes drilled in doors and frames, which ensures the contacts are in close proximity. Scout’s sensor requires exterior mounting and that isn’t always the best solution.
Worse still, Scout Alarm’s alarm is far too quiet. An alarm should be loud enough that an intruder finds it uncomfortable to remain in the house — during a break-in over a decade ago, I’m certain that my blaring ADT alarm kept the thief from lingering. Scout’s alarm is, to put it bluntly, pathetic. The problem has been discussed at length in Scout’s online forum, but the issue hasn’t yet been resolved.
In theory, this should not be a problem. In the Internet of Things, where everything is connected, I should be able to use a different siren. But reality hasn’t quite caught up with that idea. The best workaround appears to be buying a third-party siren, plugging the siren into a Belkin WeMo connected outlet, then activating the outlet through IFTTT, a cloud service that allows you to trigger events based on other IoT events. Using IFTTT, the activation of Scout’s alarm can turn the WeMo outlet on, activating the third-party siren. If that sounds a bit like a Rube Goldberg machine, that’s because it is.
[Read MIT’s Eyeriss GPU Could Transform IoT, AI.]
The workaround is far more complicated than it should be and may not happen instantly. That isn’t surprising given that free services like IFTTT don’t provide service-level agreements, but it’s not useful to have a siren delayed several minutes if there’s a break-in in progress. In an ideal world, I could just replace the Scout door panel with a louder one, but IoT hardware isn’t composable. You can’t just replace one component with another and expect the system to work.
Standardization, Gates concedes, is not quite there yet.
The extent to which IoT devices can be programmed is also an issue. For example, I’d hoped to be able to tie the motion sensor on my Nest thermostat into my Scout Alarm system. But Nest hasn’t exposed enough device capabilities to make that easily doable. It’s surprisingly hard to color outside the limited lines provided by IoT device makers. And that’s really where the value of IoT lies. A home isn’t smart if it can’t respond to complex events and conditions.
I have faith the situation will improve. Standards will be hammered out. Devices will play together nicely in time. And it will be easier to add intelligence to the system through programming. But I expect nature will provide me with a better security solution than technology — I’m planning to get a dog.