The Little Switch

In an episode of Friends, Monica becomes obsessed with a light switch that, according to Joey, does “nothing.” Monica, convinced that it must control something in the apartment (“They wouldn’t have put it there if it didn’t do something!”), begins an obsessive hunt to uncover what the switch does. Though it’s scarcely more than a silly subplot, Monica’s dilemma exposes an interesting household problem: Every home seems to have a mysterious light switch somewhere—near the back door, next to the porch light, at the basement stairs, along a row in the den. A light switch that does nothing.

The light switch is a lovely, ordinary thing. You can look at one and understand intuitively that the up position means on and the down position means off. The panel sits flush against the wall, elegant in its unobtrusiveness. The placement of light switches is so familiar that in the dark, you can feel around at the standard height (four feet from the ground) until you find the right panel to illuminate an unfamiliar bathroom. 

Light switches exemplify familiar design, which, according to Henry Petroski’s Small Things Considered: Why There Is No Perfect Design, is a name for things that are “so predictable in their form and function that we do not give them a second thought.” We haven’t given light switches much thought for nearly a century now. The toggle light switch was patented in 1917, replacing the push-button switch of the late 19th century. Since the toggle’s inception, it has remained the most ubiquitous switch in North America. The 1980s saw the introduction of the rocker, a flat-paneled switch that became popular domestically and throughout Europe and Asia. But the rocker was really just a facelift, a minor aesthetic evolution of the traditional design similar to the shift from push-button to toggle. The light switch has essentially never changed.

In that episode of Friends, Monica was confronting a problem inherent to light switches: mapping. A switch does not designate what light or outlet it actuates or if it even controls a circuit at all. According to human-centered design, a pervasive design philosophy in user experience engineering, this amounts to a total failure of task analysis—when a function does not communicate to the user what it does.

Think about the times you've gone to an unfamiliar home and had to guess which toggle turned on which light. You try one and it turns on the porch light. Another and you've activated the garbage disposal. There's a big design question here: Has the light switch's mapping problem not been solved because no one has come up with a better solution in the past century, or because we’ve accepted its design as satisfactory?

According to Don Norman, co-founder of the Nielsen Norman Group, it’s the latter. Norman, a cognitive scientist, is an influential advocate of human-centered design and the author of the user-interface (UI) textbook The Psychology of Everyday Things. Despite being published in 1988, it remains a seminal text for UI design programs (a friend of mine in the master’s program at the University of Washington’s Human Centered Design & Engineering department affection refers to the book by a nickname, POET.) By definition, human-centered design is focused on the needs of the users rather than secondary concerns, like aesthetics. This design philosophy pervades much of modern user interfaces.

In POET, Norman outlines his concerns in regards to his own home:

With six light switches mounted in a one-dimensional array, vertically on the wall, there is no way they can map naturally to the two-dimensional, horizontal placement of the lights in the ceiling. Why place the switches flat against the wall? Why not redo things? Why not place the switches horizontally, in exact analogy to the things being controlled, with a two-dimensional layout so that the switches can be placed on a floor plan of the building in exact correspondence to the areas that they control?

These aren’t rhetorical questions. Norman actually prototypes his idea of the solution, a panel protruding from the wall with a blueprint of the room and little switches in exactly the locations where the lights are.

Almost 25 years after he proposed it, Norman’s solution never took off. In his attempt to alleviate the mapping problem, Norman’s light switch unraveled all of the things that worked well about the original design. It was no longer intuitive, requiring the user to interpret a map with corresponding switches and breaking a century-long familiarity with the existing design. And not only was Norman’s switch obtrusive, but it was also ugly. Nevertheless, Norman’s concept was the only proposed home light switch solution I could find in the past century.

In Small Things Considered, Petroski argues that design always amounts to a decision-making process with clear trade-offs. As an example, Petroski discusses two-way light switches, which introduce a problem different from the mapping issue that bothered Norman: When two switches control the same light, up may no longer mean on and down may no longer mean off, depending on which switch was last used. The two-way relationship breaks the simplest physical affordance of the light switch. But Petroski argues that this isn’t necessarily a bad thing, raising the example of the two-way switch at the top and bottom of a staircase in his home. “This can be disorienting,” he writes, “but who would not gladly accept a moment’s confusion if it means being able to turn the staircase light on without having to go downstairs in the dark?”

In fact, how much of a problem is it that in an unfamiliar room you might have to try a few light switches before finding the right one? Norman used light switches as an example of failure in regards to human-centered design, but perhaps instead light switches  reveal that those design principles themselves are imperfect. While human-centered design may be an effective and prevailing philosophy, the inadequacy of Norman’s light switch—and that no other solution has taken off—illustrates that it may not be the one-size-fits-all design ideology he hopes to prove that it is.

This month, Norman is publishing a revised third edition of The Philosophy of Everyday Things—since renamed The Design of Everyday Things—but the chapter featuring light switches remains almost completely unchanged. Among his updates, Norman notes that in the quarter-decade that has passed since the first edition, the light switch section has received “a very popular reception.” He’s also added a coda to the chapter about the future of light switches, envisioning touch-sensitive screens as a likely evolution.

Norman predicts that the shift to touchscreen switches would also be accompanied by an update to the way houses are wired. He writes, “Someday we’ll get rid of the hard-wired switches, which require excessive runs of electrical cable, add to the cost and difficulties of home construction, and make remodeling of electrical circuits extremely difficult and time consuming.”  Norman imagines a home powered by a wireless network, bypassing the need for traditional behind-the-wall wiring. But he also questions whether a touch screen could ever replace the affordances of a physical switch. For example, he asks if a touchscreen offers a satisfactory replacement when the user is holding groceries or a cup of coffee. (Norman says “touchscreens are fine if the hands are free, but worthless otherwise.”) But simpler difficulties come to mind: bathrooms, for example, where wet hands could fail to register the capacitance necessary to activate a touchscreen switch.

Is the digitization of everything inevitable? Certainly when costs dip low enough, it's possible to imagine everybody’s lights might be controlled by their smartphone. In fact, there are several companies who are already producing smartphone-controlled devices around the house, including light switches. In late July, Belkin released the WeMo Light Switch, a product that is halfway to what Norman predicts. Its appearance resembles that of a rocker switch, with an added light to indicate power. It’s essentially the design you’re familiar with, with the added ability to turn the light on and off from your iPhone or Android phone.

I asked Ohad Zeira, the director of product management for WeMo, if his devices offer an opportunity to reinvent the light switch as we know it, to solve the mapping problem once and for all. Zeira admitted it was not a problem they had spent much time thinking about, since their goal was to create a light switch that improved upon a function without breaking it.

“The user experience of a switch is so familiar. When you change a physical interface, it needs to have a specific benefit,” he said. “We maintained a very specific test: no matter what we do, the user needs to be able to come home with a grocery bag under his/her arm and turn on the light. If we failed that, then we failed the day-one functionality of a light switch.” In this respect, WeMo is really just a remote control for your existing switches, not the digital paradigm shift Norman predicts.

Currently, the WeMo Light Switch sells for $50—not prohibitively expensive, but a likely prospect only for techies and home automation enthusiasts who prefer it over a regular $6 light switch. Belkin sent me a unit of the WeMo Light Switch to test. I’m all thumbs when it comes to what someone might call “handyman skills,” but even I had no trouble installing it. (Admittedly, I had to go buy a screwdriver at Target because I didn’t already own one.) 

I had a few friends over for dinner that evening, and they watched in awe as I turned my bedroom light on and off from an app on my iPhone.

“What else does it do?”

I explained that I could control the light from my phone and set up simple rules based on a service called If This Than That. I repeated an example that Ohad Zeira had mentioned to me: that a child who arrived home from school could press the switch twice and it could be set up to send his/her mom an email notification.

“Right, but what are you going to use it for?”

And I realized I had no idea.

The WeMo Light Switch is just the first of many forays Belkin and other appliance manufacturers will make into “smart” household objects, and I could see how those might all connect together in the future. But at least for now, I struggled to think of a personal use for it. 

Later, I set my light switch to tweet every time it turned on and off, just because I could. I set a timed rule for the light to turn on in the morning, as a way of getting me out of bed. I lacked the imagination to make the switch exciting, but it was interesting to think so intensely about a non-thinking part of my life. 

Before I went to bed, I looked closely at the old light switch. I had put in a plastic bag along with its faceplate and screws and left it on my nightstand. It occurred to me that I had just replaced something that still worked, something that had worked for the past 100 years, something that could work for another 100. What is the lifespan of my WeMo Light Switch? Even if the manufacturing quality matched that of a regular light switch (which seems unlikely given the WeMo’s sophistication), how long before the software would become obsolete? Could any software last a century?

Obsolescence has consequences. So much home technology—thermostats, alarm systems, and so forth—becomes quickly outdated. If Belkin were to shutter its WeMo division, it’s likely that my next iPhone wouldn’t be able to run the app, and therefore my lights along with it. If the standards for wifi networks change, my light switch may no longer be able to connect to the cloud. And when smartphones inevitably give way to another computing trend we can’t foresee, then the remote switch itself will cease to function. But even without a smartphone app and a wifi network, at the very least the WeMo would function as a regular old light switch. The 100-year-old part would still work, enduring and illuminating.

An ongoing series about the hidden lives of ordinary things