The Nonage of Home Automation

There are many gimmicky ideas that seem appealing at first blush, but fail to take hold – in some cases they remain a topic of conversation until some breakthrough technology makes them feasible, and in others they simply fade away without ever getting traction.  The idea of the automated home has been around for decades and bubbles up every so often, but hasn’t become a part of everyday life, in spite of the fact that the technological infrastructure has been in place for at least the last two decades.  Periodically, we pause to wonder why.

What follows are some loose notes from a few coffee-break conversations that identifies some plausible reasons that we do not presently live in “the home of the future” that has been anticipated since the middle of the twentieth century.

Lack of a Common Platform

Given that most of my conversation is with designers and developers, the kind of people who leap to figuring out how to solve a problem before considering whether the problem is worth solving, the most common excuse for the failure of home automation is lack of a common platform for home automation devises.

That is to say that there are gadgets that exist that enable you to control the thermostat from your mobile device, set the sprinklers to water the lawn,  turn lights on and off and control the stereo with voice commands, and all sorts of useful things – but each of them has an independent set of controls.   Many of the controls are still manual (you have to go to the garage and fiddle with a control panel to set the timer for the sprinklers), some are configurable, and others communicate with a mobile app – but there is no central “home directory” that controls them all.

So while it is possible to have an automated home, it requires the user to learn to use an array of controls in various channels and locations to manage each system separately.   Manufacturers stubbornly resist the notion of having a common platform into which each application can integrate to give the user a single method of controlling everything, so the “convenience” of automation is outweighed by the inconvenience of managing multiple systems.

This is an entirely plausible argument, with one counterpoint: it is possible to have a home built (or refitted) to have a central control panel with redundant controls through smart phone and voice commands.  I haven’t heard of many people doing this, and a few people I’ve spoken with who were having homes built remarked that it was a very expensive option.  So it isn’t a technical limitation so much as a financial one, which transitions neatly to the next topic …

Lack of Cost Benefit

While it is possible to build or refit a home with an automation system, it is very expensive to do so and customers do not consider it to be worth the expense.   This is a far more likely explanation than the lack of a technological infrastructure for the failure of home automation to take hold.

To my knowledge, there is only one automation device that has proven its worth: the “smart” thermostat that can be controlled via a smartphone application, and which also has some learning capabilities to recognize patterns and adjust itself accordingly.   While these devices are expensive relative to manual thermostats, those who purchase them say that it takes a few years to break even on the expense and start saving money.

This seems a reasonable claim, given that the heating and air conditioning units in existence today are power-hogs and that electricity remains one of the most significant utility bills.  In many areas, a 10% savings can mean $10 to $20 per month, a hundred or two dollars a year, so the claim that there is a two-year break-even seems valid.

However, the thermostat is likely the most expensive device in the home to operate: leaving a lamp or television set running 24/7 would not consume as much energy as the heating system (which can run round-the-clock during summer and winter months) and most appliances are used as-needed and switched off afterward as a matter of habit.   So it’s unlikely a device such as a lamp timer (which already exists) would have a financial benefit that would repay the cost of the device.

For this reason, the popularity of smart thermostats cannot be extended to other appliances – and given that appliances are being made more energy-efficient by means of engineering, the cost-benefit of an extrinsic means to managing their operation is becoming even less favorable as their intrinsic efficiency is improved.

It is also a mistake to take as premise that consumer behavior is motivated by cost-benefit analyses.   Where the cost is significant, people consider efficiency (a more fuel-efficient car, for example), but generally in a vague and general way, without a detailed analysis of the actual dollar amounts, and the result is a gut-feel reaction that something is thrifty or efficient. 

Even were someone to do the math, figuring out that a lamp timer would save them fifteen cents a month, an amount so paltry lacks the power to motivate them – few will consider that fifteen cents a month is about two dollars a year, and if they do this to eight lamps, the savings are $16 annually.  Even that seems to be a paltry amount for the cost of the timers, the effort of configuring them, and the inconvenience of overriding the controls and then resetting them when actual behavior does not match predicted behavior.

Lack of Consumer Demand

Of all explanations, a lack of consumer demand seems the most likely reason that home automation has not taken hold.   People are simply accustomed to the current technology and have developed habits (such as turning the lights on when entering a room and switching them off when leaving it) based on the current technology of independent devices that lack remote control or automation capabilities and therefore see no need for home automation at all.

Home automation seems to appeal to technophiles (who love gadgets so much that they will pay a high price and bear the inconvenience of using them) and narcissists (who believe that owning sophisticated things causes others to give them esteem, rather than be bored and offended by their constantly bragging about their possessions).  For most people, the idea of an automated home is an interesting concept, but not compelling enough to pursue or invest in having their home automated.  And this, ultimately, is the cause of the adoption or failure of any new technology: if people do not see it as worthwhile, most will not adopt it.

Efficiency Isn't Innovation

In general, the approach to improving things, products or processes, begins with analyzing the as-is situation and identifying areas in which problems could be fixed or improvements could be made.   That is to say that it begins rooted in present reality and ends with only minor changes.   This is different to, and likely preventative of, true innovation, which requires starting with a blank slate and imagining the possibilities that might exist, independent of what currently does exist.

It is a common, but fundamental error, to regard anything new through the lens of existing processes.  This results not in innovation, but efficiency improvements, as firms seek to streamline what they are presently doing rather than considering whether there might be an entirely new way ("new" being the "nova" in "innovation") to achieve the desired goals - or even to change the way in which the goals are defined if doing so is necessary to achieve a better outcome.

In many instances, efficiency improvements are merely automation.   In the early industrial era, automation merely replicated human motion with machines; and in the present era of information technology, automation merely replaces human thought processes with digital ones - but "merely replaces" means the that task remains the same, it is just performed by a different actor.

For example, a computerized accounting system automates the way in which invoices are processed, in that the very same thing is done with databases and spreadsheets that nineteenth-century clerks did with ledgers and quill pens.   The process is made faster, and less labor is required, but the process itself has not changed.

In that sense, replacing a worker with a machine or a clerk with a computer system is not innovative at all: it's doing the same thing more quickly and efficiently, but still doing the same thing.   To innovate requires asking: what goal are we attempting to achieve by doing things this way ... and is there a different way in which we might achieve it?"

Knowledge of existing business practices is not only unnecessary, but can be harmful.   That's not to say that they can be completely ignored - the inputs and outputs are likely still the same (though one might reconsider whether the inputs or outputs could be improved) - but all the "stuff" in the middle is entirely irrelevant.   So long as the goals of the process are achieved, the rituals by which they are pursued is irrelevant.

As a final note: innovation is not always necessary, and sometimes efficiency improvements are the best that can be done - let's not throw that concept away entirely.  But at the same time, let's not assume that the two are similar or can be accomplished in the same way.  To be innovative in the outcome requires being innovative in the process - and that holds true even when the process is one of defining processes.

Control, Convenience, and Control of Convenience

A snatch of conversation I couldn't help overhearing has got me thinking about two distinctly different approaches to the use of technology: control and convenience. My gut reaction is that the second is ultimately of greater value than the first, but that the notion of empowerment and control merits some consideration.

The bit of dialogue, or perhaps monologue, was about the use if cell phones to monitor houseplants - the person speaking was going on about how much he would like to be able to wire his home to his iPhone, install a sensor that would tell him when the lawn needed watering so he could get an instant message and press a button to feed the grass.

Aside of the obvious psychological issues, it seemed to me a half-baked idea: it the point of technology is to alleviate us of the necessity of annoying but necessary little tasks, wouldn't a better system use a hygrometer that communicates directly with a sprinkler system? That way, when the grass needs watering, it gets watered - without having to interrupt my day and require me even to push a button?

(Granted, the flaw in the notion is that a fully-automated system might put on the sprinklers at any given moment - but it should be simple enough to rig the system to water only at times when I'd be unlikely to be in the lawn, or have an override I could use on the rare occasions I'd be there - but that's entirely perpendicular to the thread.)

It struck me that these two perspectives, control and convenience, are quite common, and are largely matters of consumer preference. While I'd like to think that using technology to make a task easier is just a stepping stone on the path to making the task unnecessary, there's no denying that certain people prefer to have some modicum of control.

Perhaps it would be accurate to say that each person has differing preferences for different products: I'd be delighted by a system that transferred funds between checking and savings accounts to maximize interest earnings (a notion I've proposed to bankers, who took absolutely zero interest in the idea) but am entirely leery of a system that would automate the management of my investment portfolio (a system I have produced, in spite of the sense that I was doing something that would ultimately be unappealing and disadvantageous to consumers). So I want convenience in some tasks, control over others, and I don't sense that everyone would be in agreement about the level of control/convenience in every situation.

Ultimately, it's a product design decision that will lead to consumer preference of one solution or another - though ideally, a product would be designed to enable the customer to indicate and implement their own preferences for convenience or control.

While I still have the sense that control is still a step on the evolutionary path toward convenience, I also have to admit that not every consumer is comfortable making the transition from one to the other in every situation. Perhaps it's that I'm not yet ready to concede the point.