For almost everyone and for most technology providers, the Internet of things is all about home automation. Gizmos that have mobile apps, Internet connectivity, and local direct access via Bluetooth, Wi-Fi, or other local radio network seem to be announced every day. But almost none makes much sense, suggesting that the current IoT mania may come to a crashing halt soon.
There's a lot of good in the Internet of things concept, which of course is broader and more sophisticated than what you see at Best Buy, Cnet, and so forth. Too bad that what is actually on the market falls short of the promise.
I got a firsthand feel for the home automation market's marginal value in recent months as I've helped a family member build and equip a new home, with me focusing on the technology side, from lighting to Internet connectivity to "smart" devices.
What became apparent quickly is how dumb many of these "smart" devices are. They're mainly solutions in search of a problem, most overly proprietary. That's been true for the last decade's worth of talk on home automaton -- how much have you read about refrigerators that will tell you when you're low on milk, never to see them become real? And if they did, would you even care?
Oort: The epitome of the silly Internet of things
A great example of such a who-cares product is the Oort home automation controller, a Bluetooth device that controls various devices in your home. Bluetooth makes little sense as the primary network for home automation given its limited range: The concept of walking near a device with your smartphone to control it is silly. I love Bluetooth, but it's really designed for close communications and location-based intelligence.
Oort promises a mesh network at some point (assuming it survives, of course), so multiple devices can communicate with you, to address that range issue. Why not use Wi-Fi instead? Sure, it eats up more power, but the devices you'll have at home are typically connected to a wall outlet or power lead.
The only product that currently works with Oort controller is a color-changing LED lightbulb. Outside of a teenager's bedroom or basement pool room, the need for such a device is tiny, and it shrinks further when you think of the home automation infrastructure you'd need in place for remote control.
Oort is the epitome of IoT silliness, but much of what is available is not much better. Take remote-controlled lightbulbs and switches you attach to lamps. Saving energy is good, but at $50 or so a pop, it's much cheaper -- and easier -- to get LED bulbs instead for your lamps, which use a tiny fraction of the power of regular bulbs.
Even better, replace your wall switch with a $25 vacancy sensor so that your main lights are turned off automatically. That's a smarter approach to home automation than individually remote-controlled bulbs. Yes, that won't help with a table lamp not connected to a wall switch, but chances are your main energy wastage comes from ceiling fixtures left on.
Also in the dubious side of the spectrum are Internet-connected garage door openers, door locks, and home alarms. Given that Internet-connected goods such as cameras regularly suffer from security breaches, do you really want your home's perimeter defense connected to the Internet? Security experts warn consistently that most IoT devices have far less security built in than, say, your home PC.
Even if you're willing to make the security gamble, most of these devices tie you into a specific platform. For example, several door locks work only with iPhones (due to Android's late support for the Bluetooth Low Emergy standard, which these devices require), so neighbors and family members with other devices can't use them. You need to provide a Bluetooth fob, which is pricey. You can't use a regular metal key on most, which is the most compatible way to share keys with occasional users. Worse, many IoT home alarms have no keypad, so a person must have a smartphone or another Bluetooth fob to access your home. That makes it more complicated than it should be.
These kinds of issues are why the current crop of home-automation devices will go nowhere. They haven't really worked through the needs or the workflow of the multiple ways they would be used in the real world should there be a plausible need.
Where home automation could make sense
Despite my skepticism of what's out there, I believe there is real promise in home automation, as there is in the broad Internet of things beyond the home.
The key is to get past the fascination with phone-based remote control that so many techies have.
A great example of home automation done right is what Nest is doing. Its thermostat monitors your usage and patterns, as well as local weather, then regularly optimizes your heating and cooling accordingly. Its smoke detectors tell you what room the smoke or carbon monoxide is coming from, so you know where the problem is, rather than try to figure it out when all the detectors are sounding their sirens. It shuts off your Nest-controlled furnace if carbon monoxide is detected, eliminating a possible source for the deadly gas.
Yes, you can remote-control the Next from your phone, but that action doesn't matter. What matters is a system of sensors that work together to optimize the heating, cooling, and detection functions, and doing so in a more human-friendly way than what we have today. For me, a Nest Protect smoke detector is worth its $65 premium simply because it tells a person which room has the low battery, fire, or carbon monoxide leak, rather than send a blind signal as standard devices do.
Remotely monitoring a Nest thermostat has its advantages, too. You can see if your device is in Auto-Away mode when someone is supposed to be there, such as the person feeding your pets while you are out. I suppose turning the heat back on from your smartphone when you land at the airport so that your home is comfortable when you return is a plus.
But if the Nest devices didn't have such Internet-based remote-control features -- and the risks that come from the Internet connectivity they require -- they'd still be useful. That's what other home-automation devices need to figure out: How to be useful within the home, even if not remotely controlled.
Here are some examples of what that could mean:
- Your security lights could have Bluetooth embedded so that your pet's collar tag overrides the motion sensor and your outside lights don't blaze on when your pet comes in and out through your pet door.
- Your vacancy sensor switches could communicate with other devices, so you don't need everything wired to the sensor to be turned off. Rather than buy a remote control for your table lamp, buy a sensor extender that receives a vacancy signal from the wall switch and turns off the lamp that the switch can't directly control. Or use the vacancy sensing thermostat as the hub.
- Your HVAC system or dryer could alert you when a filter needs to be changed and, in the case of the dryer, pop out the offending filter for replacement. That may not work for your HVAC system, as you shouldn't run it sans filter, but maybe it could adjust its settings to minimize the damage caused from running with a clogged filter in addition to alerting you.
- Auto-shutoff should be in more devices, from faucets to ovens. I know many folks hate sensor lights that can turn off a light because you sat still or are behind glass (such as when showering), but with an "are you still there?" option, more devices could be more efficient and safer than what we have today. I'm not sure they need to query each other, but doing so might make sense as a second layer of detection in areas where home intrusion is a big issue; if your security system is disabled, other devices would note there's activity when you've somehow indicated you're away.
- Your video doorbell or entry camera would recognize UPS, USPS, and FedEx insignia on delivery staff, and let you know you have a package at the door. Sure, you can sign up for some of those services' notifications of deliveries, but why pay for the privilege on separate services?
- Your appliances would auto-adjust their time displays for daylight saving time, like your computer does. Maybe a time server becomes common to routers, so other devices don't each need an Internet connection but can use a master time device instead already in the home.
These are not the sexiest examples of home automation, I grant you. But they are useful to most people. If the home automation segment of the IoT industry thought more about such practical applications, then built the shared protocols and multivendor hubs we need for true home automation to work, there'd actually be a market.