I regularly see news stories saying that some new technology is coming that will change the world "within the next five years". Sometimes these stories are optimistic: this new technology will make your life better. Sometimes they are dire warnings: this new technology will take away your job or otherwise ruin your life. But they are almost always wildly exagerrated.

Read the details, and it usually says something like, "Engineers say they are close to making a machine that does X". Somehow the writer assumes that this means they will have a working model within a few months, and then they'll immediately begin mass production and within a couple of years this machine will be in every home or office or car.

That's not how real inventions work. In real life, most new invention goes through a number of stages.

1. Get an idea how to make a machine to do whatever, or to do it better.
2. Build a proof-of-concept prototype. Usually, the proof-of-concept only works under ideal conditions: the machine is in perfect condition, it is being operated by an engineer who knows exactly how to operate it correctly, environmental conditions are perfect -- the weather or whatever, etc. Some inventions fail at this step, when it turns out that something that sounded like a great idea doesn't actually work. A POC is not in any way a practical machine that would work in the "real world". And that's fine, because that's not the point. The point is to test whether the basic idea works.
3. Build a real-world prototype. Build a version that will actually work in the real world, when the machine is no longer brand new, when the user is only vaguely aware of how it's supposed to work, when conditions are not ideal, etc. Some inventions fail at this stage. It turns out that they just won't work in real world conditions. Self-driving cars worked great on a test track, with clearly painted road markings, other cars all behaving predictably, etc. But there were many problems when they got to the real world.
4. Often the prototype is ridiculously expensive. It has to be hand-built by a team of experts. Or it uses materials or components that are very expensive. So the next step is to make a practical working model that can be mass produced for a reasonable cost.
5. Build a factory to mass produce them.
6. Market them and convince people to buy them. This is often tricky. You have to convince people that the new product is not only better than what they were doing before, but enough better to be worth the time, trouble, and cost of switching. Like, if you made a car that gets 1% better gas mileage, few people would immediately junk their old cars to buy the new one. The savings wouldn't be worth the cost. Even if the new product is demonstrably better, some people won't want to take the trouble to learn how to use it. Etc.
7. Build any supporting infrastructure required. For example, for a new type of engine, you need places to refuel it. For cell phones, they needed to build a network of cell phone towers. There is often a catch-22 here: No one wants to invest in infrastructure until the product is widely used, but no one wants to buy the product when there's no infrastructure available. For example, there has been some research on hydrogen-powered cars. But who would want to buy a hydrogen car until there is a network of hydrogen fueling stations? But who would want to invest in a network of hydrogen fueling stations until there are a lot of hydrogen cars on the road?

Even if people are rushing out to buy the new product, it could be many years until it dominates the market. If there are lots of the old product around, people may not be willing to immediately throw them away and replace them with the new one. Like, suppose someone came out with a new kind of coffee maker that was clearly superior in every way to existing coffee makers. Would you immediately throw away your old coffee maker to buy a new one? Some would, but many would not. They'd wait until their current coffee maker breaks and then buy the new kind.

Getting from step 1 to dominating the market can take years, sometimes decades.

Often when you read these "golly-wow" stories, the engineers are still on step 1. Someone said, "Hey, I bet we could build a robot to do X." But he doesn't even have a working proof-of-concept prototype yet. All he has his his personal belief that it's possible. Very often this turns out to be a naive hope. Even if he really does get the new idea to work, going from "here's a new type of machine that almost mostly sort of works" to convincing people to throw out their old whatevers and use the new one can be a very long haul.

I saw a political debate once where once candidate had a proposal to increase energy production by encouragnig more oil drilling. His opponent said this was impractical: from the time an oil company started searching for new oil supplies until they got the oil to market would take over five years! Then he described his better plan: invent a totally new type of engine that doesn't use oil! Now there may be advanctages to developing such an engine, but he presented this as the fast-track alternative to drilling for more oil. He expected the audience to believe that the country could begin research to develop this new engine, invent such an engine, work out all the practical problems, mass produce it, build a chain of fueling stations across the country to dispense the new fuel, train mechanics to maintain it, and get people to buy enough of them to make a serious dent in oil consumption ... all in less time than it would take to drill more oil!

Many, probably most, new inventions or products fail. They turn out to have practical flaws, or to be too expensive to produce, too specialized, etc.

Of course there are inventions that have changed the world. The automobile, microwave oven, computer, and cell phone come to mind. But none of these went from first concept to widespread use in five years.

Getting from step 1 to dominating the market can take years. Sometimes decades.

© 2024 by Jay Johansen

Comments

No comments yet.

Add Comment

Name
E-mail
Comment