Imagine a future of driverless cars, robotic soldiers, and computerized replacements for doctors. This is the stuff of science fiction and yet we increasingly see evidence of these technologies and their precursors in the news of real life. Google, various universities, and more than one automaker have undertaken efforts to develop driverless cars that are capable of taking you for a spin on a real world roadway. Unmanned aerial vehicles, of the remotely piloted and automated drone variety, perform reconnaissance missions and fire live munitions at real targets and it can’t be long until we witness pervasive use of similar units on land.
In his book The Glass Cage: Automation and Us, Nicholas Carr tells a cautionary tale about these expanding roles of automation in our lives. As computer speed, data storage volume, communication bandwidth, and database and programming sophistication have advanced, new roles are within reach of being automated. Some of these new roles for automation should raise serious concerns. For example, take the driverless car:
Automation in wartime, airlines, the medical profession, and many other areas of human endeavor raise analogous philosophical, legal, and human concerns that we need to consider carefully.
Carr’s book largely focuses these “big questions” of automation. Home automation may raise fewer philosophical and legal concerns but it has a human impact because of our significant interaction with our homes.
One of the few, if not the only, statements that Carr makes specifically about home automation follows:
“The technologies of home automation, which allow things like lighting, heating, cooking, and entertainment to be meticulously programmed, impose a Taylorist mentality on domestic life. They subtly encourage people to adapt themselves to established routines and schedules, making homes more like workplaces.”
While I think this outcome of home automation is conceivable, I don’t see evidence that it is our fate to reach this regimented end nor do the current trends in home automation support this vision of the future. The best examples of home automation technology do quite the opposite of what Carr suggests: they adapt to the humans in the household and free them from routines and schedules. Let’s take the home thermostat as an example.
Though there is regional variation and differences between survey methods, it is likely that less than 50% of U.S. households currently adjust their thermostats, either manually or via program, to reduce energy use (heating or cooling) at night or when they are away (Peffer, Printoni, et. Al., 2011). This approach gives the household complete freedom from schedule but it comes at an economic and environmental cost. Heating and cooling has a big impact on our utility bills, accounting for 42% of the average home energy expenditure in the U.S. (EIA). Along with that economic cost, comes the significant environmental impact of the associated electricity and heating fuel consumption. This approach to thermostat management is simply wasteful.
The alternatives to such wasteful avoidance of the thermostat have been to manually adjust it with each departure, arrival, slumber, and morning awakening or to program a representative daily schedule to reflect those events. Only about 19% of U.S. households program their thermostats (ENERGY STAR), so the majority are managing temperature settings manually. Ironically, these two approaches, which are the status quo for the majority of U.S. homes, result in the regimented outcome Carr foretells more certainly than newer forms of automation.
I used to manage my thermostat manually. That meant visiting it more than 4 times a day to responsibly control the temperature in my home. It was the right thing to do but I was a slave to the thermostat. Upon moving into the first home of my own, before even moving in any belongings, I stayed up late into the first night installing a programmable thermostat. For the next few years I had to adapt to the schedule I programmed in that thermostat. At 6:00am it was time to wake-up, 8:00am time to go to work, 5:00pm time to come home and 10:30pm time for bed. The thermostat regimented my daily schedule and any exceptions either required a manual adjustment or accepting the program to the detriment of comfort, finances or the environment.
Now there is a new class of thermostats that land squarely in the realm of automation. These thermostats can make changes to the temperature setpoint in the home independently based on a variety of factors like occupancy, time of day, and forecast. This automation offers opportunities for the household to be responsible and escape the regimentation of current approaches to managing a thermostat.
For over a year now I’ve enjoyed the benefits of one such thermostat, the Nest. The Nest offers advantages regardless of a household’s current approach to managing a thermostat.
But the Nest can do even more. I use a hybrid of the learned schedule, for nighttime, and geofencing to control my Nest. Geofencing links the cellphones in my household to the Nest so that they can inform the thermostat when everyone has left or when anyone returns home. As a result, our Nest is always set properly: when we’re home it is comfortable and when we’re away it turns down. We don’t have to interact with it at all. We can even enjoy energy savings when we’re away without coming home to a cold house. We aren’t slaves to managing it, we aren’t regimented to its schedule, and we’re being financially and environmentally responsible.
Another of Carr’s general concerns about automation is that it substitutes for human engagement and we become complacent and inattentive. Unfortunately, in our homes we are already complacent about many of our most significant choices.
Automation and smarter home systems could help occupants be less complacent and make better operational choices at home. Like the thermostats described above, smart irrigation systems are another good example. These systems monitor weather, historical and forecasted precipitation, and appropriate watering times to reduce irrigation water consumption.
Carr also laments automation’s propensity for deskilling, reducing the skills of the humans that interact with the system. For example, the autopilot on a plane has the potential to reduce the pilot’s skills by taking away the opportunity, and the necessity, for the pilot to continually amass experience behind the stick of an aircraft.
Unlike the pilot, the driver, or the medical professional, the tasks we carry out in our homes like adjusting thermostats, turning on and off lights and appliances, and watering our lawns require a much lower level of skills. Handing over some of the responsibility for these systems to automation will not detract significantly from our primary professional skills or the kind of complex skills that are required for an activity like driving (e.g. navigation, decision making, eye-hand coordination, etc.).
Furthermore, we cannot lose skills we don’t have. Studies have shown that there are significant misconceptions or gaps in knowledge with regard to operational decisions in the home that would impede the ability to manage those systems with “skill”. Some of these misconceptions and knowledge gaps include:
Well-designed automation can reduce the number of misguided decisions made, and support the occupant in making more informed decisions (e.g. the Nest displays the time it will take to reach the desired temperature, making it clear that setting it higher will not get to the desired temperature sooner).
Though home automation may not be as big of a philosophical challenge as automation in other aspects of life that does not mean that we should proceed without caution. As designers, engineers, programmers, integrators and users of automation we should heed Carr’s advice and proceed mindfully.
Home automation in excess has the potential to turn our homes into scripted, flashing, alarming, and overly complicated systems that steal our attention from more important and/or skillful endeavors (e.g. time with family, hobbies, etc.). Furthermore, home automation systems themselves use energy, so we must assure that our designs save more energy than they use.
Like any tool, the automated smart home can remove some of the mundane decisions and effort that face us daily. However, it is our responsibility to redirect that effort into greater engagement in our homes to make them safer, more convenient, and more efficient places for our families to reside.