The Snakes and Ladders of Smart Home

What has the hacking of Ashley Madison got in common with smart homes?  The answer is that both are likely to increase the number of divorces.  If that seems a strange statement, talk to the wife or partner of anyone who’s got a smart home system (it’s generally the husband who buys and installs them).  Most feel that it’s not improved their quality of life; it’s just added another level of frustration, because now they have a home which can go wrong.

Of course, that’s not the message the industry wants to get out. If you believe the analysts and the smart home manufacturers, your home is about to evolve from the thick bricks on the block to the Nobel Prize winning genius of housing.  Technology is finally about to transform the place you live in into a high IQ domicile that reacts to your mood and presence, keeps you safe and saves you energy.

It’s a great story that plays to some excellent futuristic videos, from global technology giants like AT&T, through boutique technology leaders like Nest to the successful crowdfunded visions of Oomi, Nuimi, and Blaze Automation.  In their vision, it’s slick, it’s sexy and it’s almost here.

Of course it’s not the first time that it has almost been here. The first example I came across was in a 1951 issue of Practical Wireless I found in my father’s loft, which sang the praises of six houses which Motorola had just built in Arizona.  The article claimed that they would be the model for all future American homes.  The lights were automated, as were the curtains.  Everywhere around the house there were timers, motors and relays to save the owners from the tedious task of leaving their chair for anything other than changing the channel on their brand new 12”, “Glare Control” Motorola TV.  Remote control didn’t arrive until 1961.

Ten years later, when the idea hadn’t caught on, the industry tried to revitalise the idea by changing its name.  It’s an old trick and this particular industry has played it more often than most.  In the 60’s they were Automated Homes.  The same trick tried to persuade us to invest in Robotic Homes in the ‘70s, Intelligent Homes in the ‘80s, Connected Homes in the ‘90s, Internet Homes in the first decade of a new millennium and finally, as we hit 2010, the marketing departments hit on the concept of Smart Homes.

To be fair, it’s an idea whose time potentially has come.  The technology is starting to work; in some cases it looks great and in a few instances it both looks good and works.  Tech geeks and journalists love it, which has helped power a lot of innovation.  It’s poised on the edge of mass market appeal, but there are a few fundamentals which desperately need addressing if it’s going to make that leap from concept to commodity.  The key ones are interoperability and device management.

Interoperability takes many forms.  At the most obvious level it’s about whether the wireless link is Bluetooth, ZigBee, Wi-Fi, Z-Wave, Thread, or something else.  That’s a battle which the media loves.  Over time, it will become irrelevant as increasingly powerful chips will support multiple radios and protocols.  Smart Homes will probably end up using a combination of them for different applications.  But it’s a major issue today.

A bigger interoperability question is how to commission and maintain the devices.  That’s about how to add, remove and replace the smart stuff in your home.  It’s not trivial.  It’s the elephant in the smart room which the industry is trying hard to ignore.  The variant of the old joke of “How many IoT engineers do you need to change a lightbulb?” does not bring smiles to the faces of most smart home companies. Nor do they like the question of how they will fund the numerous application and firmware upgrades that these products will need over their lifetime.   Their horizons are typically just a few years ahead, looking to the point where they hope to be acquired, not the ten to fifteen years which most consumers consider products like these should last.  Junko Yoshida at EE Times has suggested that wearable devices are just a hardware Ponzi scheme.  The same could be said for much of smart home.

To illustrate the current state of the industry we need to play a quick game of Smart Home Snakes and Ladders.  Plenty of users are already playing this game (and by doing so, frustrating their partners).  You throw the dice and buy your first devices, typically some smart light bulbs.  You connect them up with your smart phone and off you go.  The problem comes when you buy your next device, as in most cases it won’t work with the same app.  The light bulbs might use ZigBee, or Bluetooth, but the best-selling smoke alarms on the market don’t, so when your new smart toaster catches fire, your smart lights won’t be bright enough to flash and warn you.

Assuming your house hasn’t burnt down, you may still fall in love with the fact that you can control your lights from your smartphone and suggest your wife gets some more when she’s down at the store.  Which she does.  But although they’ll be advertised as smart bulbs, just like the ones you bought before, they’ll probably use different protocols, so you’ll all end up with multiple apps on your phones to turn the lights on and off.

If you stick to a single manufacturer you’re probably OK, although that’s not always the case.  However, most manufacturers have a limited range of products.  Smart lighting vendors generally don’t make smart thermostats, so if you want one of those, you’ll probably be tempted by an offer from your energy provider.  The problem is that because they’re subsidised, quite a lot of energy suppliers will stop supporting them if you change supplier (which most of the world outside the US can and does do).  Which makes it difficult to keep on saving money without dumbing down your home.

One of the most popular smart home products is the smart door lock.  You can open it with your phone, and you can send the app and credentials to other phones, so that your kids, nanny, cleaner, plumber, burglar, friends, etc. can get in when they need to.  All they need is for you to send them the app and the credentials.  Which is great until you upgrade your phone and they disappear.  Or your wife gets a Windows phone and discovers there’s not an app for it.  Particularly if it’s pouring with rain and you’ve just boarded a long haul flight.

But these are teething problems.  Put in enough effort and your home will start to work almost as well as it did before you started rolling the smart home die.  As long as you don’t succumb to any of those adverts for cheaper or faster broadband.  In most of the world, people no longer go out and buy Wi-Fi routers.  They get them free when they change broadband provider.  The new router will be faster, shinier, have more flashing lights, and also have a new SSID and WPA2 password.  Those last two acronyms won’t mean anything to the majority of users.  They’ll mean much more to your smart home devices, because they’ll look at them and deduce that they’ve moved to another house and stop working.  Every smart device which relies on the router will lose its internet connection, with the result that your house starts dumbing down again.  (For a real life example, have a look at Nick Bilton’s report in the New York Times on setting up a smart home.)

And so it goes on.  As phone vendors push out upgrades, apps will break.  When you replace the lock because the software designers didn’t understand the real issues of locks, which is that they need to cope with warping door frames, and the new one needs different apps, your kids won’t be able to get in.  When Windows 13 arrives (unlucky for some) and the house starts behaving differently, you’ll begin to have fond memories of that old light switch which worked first time, every time for fifty years.  And don’t think of moving house.  Not only will you have to start again, but you’ll need to write a manual for the house you’re selling and provide network support for the new owners.  At some point along that path you’ll either give up, or discover that your wife’s left you. The problem is that the industry has designed most smart home products as a game of snakes and ladders, but have forgotten to provide the ladders.

In time things will sort themselves out.  A lot of the tech is starting to go into commercial buildings – hotels and offices, where many of the problems will get ironed out.  But the consumer side of the business is being hyped as if it’s already mature, not least because of the success of Nest, which was acquired by Google for a mouth-watering $3.2 billion – a figure which probably had nothing at all to do with their thermostat.  But it means that most start-ups aren’t really thinking about consumer needs or solving these underlying problems.  Instead they’re fixated on the magical pot of gold at the end of the acquisition rainbow.  For almost all of them it’s likely to be every bit as imaginary.

The reality is that the consumer market is probably not the right one in which to develop the technology, as every customer is a potential point of support.  That’s a cost which will break start-ups and is the reason why commercial installations are more interesting.  They will look at the cost benefits, deploy at scale and have facilities managers who will provide consolidated feedback.  They’ll also provide enough data to help develop useful analytics.

At a conference in London this week I heard representatives from smart home companies talking about “DIY products” and “next generational aspirations”, which suggest that even they are starting to question their own PR.  There is a lot happening which will move us to the point where smart homes may work, but at the moment the caveat emptor must be to remember the other meaning of “smart”, which is “to cause or be the cause of pain” and “to feel or endure distress”.  Until we see some ladders on the playing board, it’s wise to remember that every way is down.

I knew there would be more “Gotchas” and I was right.  At the end of 2015, Nest rolled out an upgrade to their smart thermostats.  It had a bug in it and a few weeks later the thermostats apparently drained the battery or turned off, leaving many North American homes freezing.  So I need to add another ladder to the board.  And a few months later, Hive – British Gas’ smart thermostat exhibited the opposite behaviour, setting consumer’s thermostats to 32°C  (90°F).

And an article from Crain’s Chicago Business suggests it’s still a project for dedicated hobbyists.  Even with the benefit of Homekit.