Drivin' Online

There's been a lot of talk about self-driving cars. This week a study prophesied that in twenty years, most cars will come without a strong wheel or pedals.

I find that hard to believe. Self-driving had become a promising technology, advancing a lot in just the last decade.  So I'm sure it will be possible.  And a car that can take over for you would be very useful. But the study isn't claiming it will be possible, or even common, but that most people will be content with it as the only option.

It's easy to get carried away with predictions of technological progress. For instance, digital cameras have gone from unheard of to dominant in the past twenty years. But there are also examples of technology being slower to catch on; over the same twenty years, online shopping has gone from unheard of to common but not standard.

People tend to overestimate change involving technology, because of those situations where the change is so fast. A that any change involving technology will be as fast. But fast changes happen when:

• The advantages of the new way are complete and undeniable.
• There's no emotional connection to the old way.

I doubt self driving cars will be the only way because:

Lives are involved

It sounds nice to talk about self driving cars, but we still have to see how the public reacts to a fatal accident involving one of the cars. Even if we get to the point where self-driving cars are safer overall, they won't be perfect, and the thought of dying because of a bad decision by a machine you have no control over will be a powerful one.

Regulations are involved

There are a huge number of regulations involving cars. It's because - as stated above - lives are involved. The regulations-are-evil-job-killers crowd is surprisingly accepting of auto industry rules, since they have made cars dramatically safer over the years. So a major alteration to how cars work and are used will not be a matter of just making a new car and putting it on the road.

Companies will fight it

Often, fast technological change comes when new companies build around new technology overtake old companies with old technology. But that’s not likely here. To start a new car company, you need huge manufacturing facilities all over the world, and develop relationships with an army of parts suppliers.
And (relating to the previous two points) there’s a much higher bar when it comes to quality. Tech startups benefit from our low expectations of new developments. Or, to put it another way, look down at your Android phone and ask, would you buy a car from these people?

And the existing car companies aren’t going to want to build self-drive-only cars; taking away the driving aspect takes away a big differentiator between makes and models. A no-steering-wheel car is just an appliance, and is reduced to a commodity.