by Bill Nye
I’m sure you have encountered tasks that seemed to be impossibly complicated—until you tilted your head, took a moment to think, and got the job done. We all have. Most of the time it is something small but instructive, like trying to figure out that vexing chalk drawing of an ellipse on the blackboard. Every once in a while, though, we encounter a truly great challenge. Then those smaller ones offer inspiration about how to proceed. They show us how constraints can make problems seem less overwhelming and can help guide us toward the most workable solutions. That’s the situation the whole world finds itself in right now. We need to draw on those crucial life experiences. Instead of despairing or evading, we need to embrace constraints, face our problems, and get to work.
CHAPTER 12
Upside-Down Pyramid of Design
Young people often ask me what advice I would give them to help them succeed. My answer reflects lessons I learned in my early days working in engineering. If you have any nerdish proclivities whatsoever and like to tinker—which, at some level, practically everyone does—I encourage you to get in at the beginning of a project. “Be part of the start” is one of my beloved aphorisms. When you join a conversation at the inception of an idea, whether it’s solving an existing problem, starting a new company, or designing a new product, you can influence the development process and know that you did everything you could to be part of something good. Of course, it’s a risk. You might be part of something not so good, or something that simply sucks. But it’s better to take that risk and do your best than it is to work on a mediocre something or other that leaves you feeling like an old Ford Pinto or Chevy Vega. (They were crummy cars.) But I’m getting ahead of myself.
At this point, I was working at Sundstrand Data Control with a very sharp, if somewhat curmudgeonly, designer named Jack Morrow. He was amazing. He was in his fifties, which seemed to me quite old back then; now I realize, of course, that it is actually the very prime of youth. Despite his “advanced” years, Jack was a helluva skier, quite an athlete. His mind never stopped moving, and neither did he; as he talked, he continually removed and replaced his reading glasses. While the managers kept themselves busy shuffling memos and moving around the boxes on their company organization charts, Jack was doing more than thinking about how to work through constraints. He was mapping out a whole organizational scheme of his own.
Jack must have sized me up as a fellow who could use a little sage advice, so he talked with me often about the importance of getting the design right. “If the design is bad,” he’d say, “no matter how well everyone else does their job, the result is never going to be any good” (or at least “never as good as it could be”). At Sundstrand, I was on a team that was designing drilling equipment, heavy stuff, and Jack strongly encouraged me—no, he pretty much insisted—that we make sure that everything really fit before we asked the machinists to cut metal. He, ah, drilled the point in my head: Work it through and be absolutely certain it works correctly on paper before you commit someone else’s time, someone else’s skill, and someone else’s materials to a finished part or assembly.
What Jack understood is that it is easy enough to get things almost right. Even for designers and engineers working on big-budget, large-scale machinery, there’s a constant temptation to stop at the point when things are 90 percent good. Laziness is part of the reason, but not all of it. When you find a workable solution, even if it is only somewhat workable, you begin to experience a mix of relief and satisfaction at another constraint overcome. You know what I’m talking about: It’s like the feeling you get the moment when the lawn is almost raked, the dishes are almost all washed, the second coat of paint looks okay without any final touch-up . . . It takes a lot more effort to get things exactly right. But as I learned working alongside Jack at Sundstrand, things that almost work do not exactly work, which is to say that they do not work at all. I left a few months before the company got sued for declaring nonworking avionics boxes to be real working boxes in its salable inventory—a form of cheating on your taxes. Not everyone there thought like Jack.
Whenever you set out to create something, be it a hook to hang your coat or a string of computer code that lands an airplane, you have to take time with the design. The more time you can set aside to think, Jack reminded me, the better the created thing will be. These are two crucial skills for everything-all-at-once progress: Filter information carefully so you can home in on the best ways to solve your problem, and then develop your ideas fully in the hypothetical before you execute, so that the resulting system really does what you intend it to do.
Easy enough for me to write, but it’s often quite difficult to accomplish.
So there I was, young Bill, laboring away at Sundstrand, creating devices to help direct mining and drilling operations. Meanwhile, I was also absorbing what Jack and my colleagues discussed regarding the general state of domestic engineering in the United States. At the time, particularly in the automotive industry, it seemed as though we were falling further and further behind our international competition. This was a cause for great concern for Jack and the others.
I was not, and am not, an automotive designer. I am a mechanical engineer who loves knowing as much as I can about mechanisms. But for nerdy engineering fun, I routinely dropped the transmission, pulled the engine, or replaced the constant-velocity joints in a friend’s car or truck. Car culture was much more of a thing back then. Even my parents were drawn into it. Because of our French ancestry, we ended up with a few Renaults in our driveway, including the Renault 16 in which my brother and I installed our first THANKS sign. Even as a high school student, I could see that these little French cars were full of clever innovations. The materials might have fallen short, but the ideas were cool. They had front-wheel drive, which frees up more space for passengers because there is no need to make room for a driveshaft connecting the engine to the rear wheels. They had disc brakes for better stopping. Rack-and-pinion steering. MacPherson struts, a compact and efficient suspension system. A rear compartment made bigger by mounting the two rear axles or half-shafts side by side rather than one above the other. From around 1970 on, I began to realize that Detroit was not keeping up with what designers overseas were doing. The Europeans and the Japanese were innovating more than we were.
All through engineering school at Cornell, I felt frustration with what the US automotive industry was doing—or, more accurately, not doing. Right about this time, American automakers created those two famously terrible small cars, the Ford Pinto and the Chevrolet Vega, that continually reaffirmed my low opinion of them. To do routine maintenance on the Vega, you had to loosen the motor mounts so that you could tip the engine to get at one of the spark plugs. The engine block was made out of aluminum; the head, or top, plate of the engine was iron. That was General Motors’ attempt at progressive engineering, but the aluminum block tended to warp at high temperatures, causing it to leak lubricant and coolant. It burned oil, ground itself to a standstill sometimes, and generally made a mess of things. And the fender rust—you could count the days to when it would surely appear.
The Pinto was even worse, notorious for bursting into flame when hit from behind. Managers at Ford knew about the risk but convinced themselves that reconfiguring the gas tank so that the car was less vulnerable to rear impacts was not worth it. The Ford managers were morally wrong, and they turned out to be economically wrong, as well. At least 27 people were killed in fuel-tank fires, at least 117 lawsuits were filed, and Ford’s reputation was greatly diminished for years.
In the decade that followed, imported cars, especially those from Japan, drastically increased their share of the US market. It’s no coincidence. Detroit started with bad designs and then let things get worse from there. My colleagues and I talked a lot about this troubling turn of events. Jack had been my age during the Apollo era. He had developed mechanisms and systems that helped land people on the Moon. And here he was watching the whole country roll downhill, leastways in domestic engineer
ing. He wanted to do his part to push back and encourage good design, and with that in mind, he shared with me a very memorable piece of wisdom.
Jack made a sketch, which I’ve done my best to live by over the years. He called it an “upside-down pyramid,” a triangle with one vertex pointing to the bottom of the page. I embellished his sobriquet into the “upside-down pyramid of design,” for obvious reasons, and refined it into a master plan for setting yourself up for design success. We divided this pyramid into horizontal layers, each of which represented a step in the design or production of a thing. The upside-down pyramid of design is not (yet) a nerd icon like the slide rule, but it should be. It is a visual shorthand that illustrates the best way to turn ideas into action.
Since I’ve been going on and on about cars just now, we’ll start with one of those as our example. Down at the bottom vertex is where the design takes place. Automotive designers take into account a great many things: drivetrain, number of seats, safety features, and overall look. How will the handling and interior space change if the engine drives the rear wheels instead of the front ones, for instance? How many fasteners, how much paint, and how many tires or hinges will you need? It’s all got to be figured out. In general, design is where the fewest number of people are involved. This is the cheapest step in the process of bringing a product to market. It’s just people sitting and thinking thoughts about requirements, shapes, materials, looks, and feels. But once that is done, then your car company (or construction firm, or movie studio, or whatever entity is involved) starts spending real money, which is what makes this phase the most important one, as well.
Above the design on the upside-down pyramid comes procurement: getting the stuff you need to make something. For a car, there are sheets of steel, plastic, glass, rubber, wires, and so on. In an architectural undertaking, it would be cement, wallboard, insulation, and glass. For a seamstress or seamster, it’s when you buy the fabric, cut it from the bolt, and start shearing shapes that you really commit to a pattern or design. That’s when there’s no turning back, from a financial standpoint or a sewing-table stool. This is the level where you start to really shell out the cash. Someone has to get all those things. We’re talking about the purchasing department and manufacturing engineers, people who figure out where to get the materials, which have to be of a certain quality, in a certain quantity, at a manageable price, available in the right place in a manageable amount of time.
In the case of the design, you’re paying salaries to the designers. But on the procurement level, you’re paying salaries and you’re buying stuff. If it’s a car that you plan to produce millions of, there’s millions and millions of dollars’ worth of raw car stuff. There are important decisions to be made at the procurement stage, as well. The materials you choose have a huge influence on your final design. Cheap cloth can ruin the cut of a well-tailored suit. Beautiful new buildings can be marred by ugly mineral runoff from cheap prefab brick cladding. Hard, shiny plastic can make the interior of a luxury car feel like the low-rent option from the airport parking garage.
Then you go up to the next layer of the pyramid. You start using the raw materials to make parts: wheels, fenders, bumpers, transmissions, and gas tanks. Go up to the next layer, and you’ll find welders and painters. Move up again, and you start fitting all the purchased parts together. Now we’re really spending money, because all these steps require skilled laborers paying attention to what they’re doing. Millions and millions of dollars, euros, or yen are flying around. Finally, up at the top of the inverted pyramid is the mythic layer broadly called “marketing.” Finished cars have to be shipped all over the continent or all over the world, and there is nothing but competition. A great many other manufacturers are trying to climb up the same pyramid. Marketing can make or break your vehicle’s success.
Naturally, the same principles hold true for a million other products, and not just physical manufactured ones. The pyramid applies equally to software packages and cable TV providers. If the design is not good, neither shall the product be any good. Apple got the design of the iPhone right and thrived, even though plenty of people were already selling mobile phones. Google figured out the right way to do Web searches and thrived, even though there were already a lot of search engines. If I may toot my own horn (sorry), The Science Guy show is still used in classrooms and homes. I claim that is because we thought it through and were disciplined in the curriculum and format.
In each case, there is the amazing thing about that pyramid: The whole wide tottering top—the place where a creation finally sees the light of day and captures the imagination of the public, or not—depends on the little triangle of design at the bottom. You could have your car painters come to work and do calisthenics together to be an ace painting team. The welders could come in to work and sing the “What a Great Day to Weld a Quarter Panel” song, crooning like Sinatra, taking tremendous pride in their jobs. The upholsterers could lovingly install the finest reasonably priced seats money can buy. The wiring team could lay those wire bundles in with graceful curves and cute little harness tie straps. But if the initial design sucks, the best thing that all those people will come together to create, even on your very best day, is a crappy Ford Pinto. You’re never going to get anything better from this assembly line than what was designed on the white board, graph paper, or design tables.
If the design is no good, no matter how hard everyone works, the product will be no good, either. On a television show, you can have the charmingest host ever, with the greatest director ever, with the quickest, sharpest camera operators in the business, but if the idea for the show sucks, the show is going to suck. The videotape storerooms and hard-drive servers of TV networks are fully stocked with shows that never made it past a few episodes because the initial design or conception didn’t get done well.
On the other hand, if the design is great, then you have a solid shot at an equally great product that people will take pride in creating and that consumers will want to use or watch. That was the indelible message I took away from my experiences at Sundstrand. A good design doesn’t guarantee a great product, because there are plenty of places to go wrong in execution; but you will never, ever have a great product without a very good design.
UPSIDE-DOWN PYRAMID OF DESIGN
There’s another saying, one I picked up while I sat on the Seattle Bicycle Advisory Board: “Good engineering invites right use.” For many years, I’ve commuted in big cities by bicycle. It’s so much more pleasant than sitting in a car or on a train, so long as everyone understands how the bike lane is supposed to be used. But if a stranger is driving along a street with a bike lane, will he or she be able to instantly figure out where the bikes should ride and where the cars should drive? A lot of dangerous or deadly bicycle–car interactions happen because of avoidable confusion about lane markings; drivers should always know exactly where they need to keep a lane clear for cyclists, and there should always be unambiguous signs or lights indicating who has the right of way. The examples of inviting right use are infinite. The instruments in an airplane cockpit should be self-explanatory. I’ve often mused that when I come upon a door handle or bar, I should never have to guess whether to push or pull. Embrace the idea, and you have a shot at creating something of real value.
As hopeful as each of us might be about the beginning of a project, remember that things generally don’t come out right the first time through. Even with this upside-down pyramid design idea, you are almost certainly going to run into glitches with your project, be it a dress pattern, a new spacecraft atmospheric braking system, or a software module. But when you’re in the design phase, that’s okay! It’s all too easy to get swept away by someone else’s sales pitch or to go along with the will of a large group because it’s the path of least resistance. To that I say: Don’t do it. Accepting other people’s values without question is exactly what gave the world the Pinto. Before you start spending money, apply the core nerd values instead. Ask yourself: Do
you honestly believe in the course of action? Does it address a meaningful problem? Does it invite right use? Would it make the world a better place? If it succeeds, will you be proud of it? If it fails, will you have learned something in the process and be glad that you tried?
It’s why I strongly encourage everyone to be willing to take a second shot, to have the expectation that even if the first thing you build is good, it’s probably not going to be good enough to sell to a stranger. In my opinion (once again correct, obviously), you almost always have to build the second prototype.
Spend time with the initial design, but budget and plan to spend more time with your second bite at the apple. Trust that you will make mistakes, and plan on learning from them. It’s a hard lesson for most of us to learn because it costs the one thing so few of us have—time—but it will save you much more time (plus energy, money, and reputation) if you commit to making every conceivable improvement before you move to the next phase of development.
Taking that time to evaluate your work and find ways to strengthen it makes all the difference. A small vehicle that was created around the same time as the Vega and Pinto was the Datsun B210, produced by the company now known as Nissan. It was made of the same raw stuff as its American counterparts, but it worked better and ran well over 100,000 miles. The designers started with a similar concept (a small, economical car), but they just did more with what they had. Because of their quirky looks and extraordinary reliability, those cars won a devoted following. Then, more than a decade after the B210, Mazda, another Japanese company, came out with the Miata. Like the Datsun, the Miata resembled its competitors in design and materials but trounced them in execution. It matched the looks and handling of the British roadsters that inspired it. The big difference was that it ran flawlessly all the time. Now in its fourth generation, the Miata is the bestselling sports car in history.