I was reading a blog about an innovative method for teaching a robot to anticipate its human partner’s needs, for example, knowing that an assembler would need a screwdriver for their next task. Something the engineering professor on the project said triggered a series of memories for me — and I made some important connections:
“I would encourage engineers to be fearless and target the hardest problems, but also understand that this is a trial-and-error process. We had to run more than 10 studies, and each study lasted about one or two months until we got this right. So, it’s not going to work the very first time, but I think that’s what makes the problem interesting; if it worked the first time then the problem would not be interesting, nor would it be challenging.”
This reminded me of an interview I did more than 10 years ago with Joel Spira, the man who developed the solid-state home lighting dimmer and in 1961, started the Lutron Electronics company to bring them to market. He described reading a book published by GE, the Controlled Rectifier Manual, which presented a newly invented semiconductor component called the silicon controlled rectifier (SCR). Introduced by General Electric in 1958, the SCR was a “high-current, high-voltage, high-efficiency switch.” It was different than the existing controllable solid-state devices, transistors, which were amplifiers. Spira, who had already been working on a concept for a light dimmer, told me that he realized that these new components would be ideal for compact, efficient, and inexpensive home lighting dimmers.
Try, Try, Try…
So, on a ping pong table in his New York City apartment, he set to work building a prototype. According to Spira, he worked on it for about a year before arriving at a practical prototype. During the course of that year, although he burned out transistors almost daily, he persevered, and that was the start of lighting dimmers becoming a common household feature.
But even more than that, since Spira’s solid-state dimmers could be controlled by a low-voltage, low-power input, they could easily be controlled remotely. And that led to systems where you could have a menu of different lighting setups; lighting that could be automatically controlled and coordinated with other building systems; lighting that could be programmed to make buildings significantly more energy efficient. All of that was built on Spira’s not being discouraged by large numbers of early failures.
Hearing that, made me think back on my engineering career. Especially in my early years, if I burned out a bunch of transistors, I would decide I was on the wrong track — not yet understanding that failing repeatedly was the normal course of affairs for designing something new. When I finally learned that lesson, I became a much better engineer.
Edison and Tesla
Then of course, there was Thomas Edison. Largely self-taught, he is viewed as the epitome of trial-and-error inventing — for example he tried 6,000 different light bulb filaments before settling on one he was happy with. Edison was quoted as saying that 99 percent of genius “was knowing the things that could not work.”
Nikola Tesla, a competitor of Edison’s, was just the opposite. Tesla had serious training in electrical theory and practice. He approached problems theory first.
For example, Edison was convinced that the electric power to light his lamps should be DC and built a generating and transmission system to prove it. On the other hand, Tesla, was convinced that AC, based on theory, had many advantages over DC. With AC you could transform the output of a generator to high voltage at low current. That would make it possible to transmit the power over much longer distances than Edison’s DC, which was limited to about a mile.
However, despite the radical differences in their approaches — Edison’s was bottom-up and Tesla’s was top-down — they both experienced repeated failures before they could see their ideas become practical. Edison started with some basic scientific principles for his trial-and-error approach. When he finally discovered something that would work to his satisfaction, he would go on from there, with the help of expert employees, to develop practical applications.
Tesla, on the other hand developed elegant theoretical plans and then had to find financial backers to enable him to implement them. So, when they did not pan out at first, he faced the problem of convincing his backers to keep financing his successive attempts at perfecting his inventions. He was convinced, because of the beauty and clarity of his ideas, that they would just work, but he did not reckon with the need for learning from successive failures. So, while his ideas proved extremely fruitful for generations of scientists and engineers, he himself had minimal real-world success.
A Simple Lesson
The simple lesson I take from these examples is that failure is an inevitable component of innovation — how you deal with it is what determines whether you are ultimately successful.