I am on the same page with everyone here about the problems with car-centered infrastructure, but I think perhaps we haven’t fully internalized the extent to wh

I am on the same page with everyone here about the problems with car-centered infrastructure, but I think perhaps we haven’t fully internalized the extent to which driverless, electric, shared vehicles will change this equation. It’s an entirely new paradigm - and even more so if cheap tunneling can be achieved. While Musk’s initial ideas about how to reduce the cost may be wrong, at least he’s trying to! His initial ideas about electric cars and rockets weren’t completely right either, because he was not an expert in those areas. He hired the most talented people in those fields and learned and changed over a period of years until he found what worked. It’s a good algorithm, and it helps to be wealthy so that you can survive the failures and dead-ends without running out of money. There are two arguments in the article: that Musk is going about solving the problem in the wrong way, and that he is solving the wrong problem altogether by trying to put cars in tunnels. The author may well be right about the first, but I think he/she is wrong about the second. Let’s think about everything that changes in a fully driverless, electric, shared, and tunnel-augmented vehicle ecosystem. No more greenhouse gas emissions directly from vehicles (and none at all once the grid is fully renewable). No more need to dedicate huge amounts of valuable land in our cities and suburbs to parking (cars can drive themselves to outlying areas to park or, better yet, can be shared). Dramatically higher throughput because the human mistakes that cause congestion can be avoided (I recommend CGP Grey’s video on the causes of traffic: https://youtu.be/iHzzSao6ypE https://youtu.be/iHzzSao6ypE ). Driverless cars could drive faster and closer together, especially if they can communicate with one another. Even more dramatic reduction in deaths and injuries as human error is removed from the equation. Far fewer cars are needed in a shared-vehicle ecosystem. Most privately-owned cars are idle nearly all of the time, but they still take up space. Sharing vehicles allows us to reduce the amount of infrastructure needed for cars. The average size of vehicles in the fleet can also be reduced. Right now people drive unnecessarily large vehicles because they might need the space a few times per year (annual family skiing trip, etc). These vehicles use up more energy and take up more space. In a shared-vehicle ecosystem, most of the vehicles can be something like a Smart ForTwo, with a smaller number of large vehicles that can be used by whoever actually needs them at any moment. A shared-vehicle ecosystem could also include something like a cross between a scaled-up UberPOOL and a scaled-down bus: a Sprinter van (or equivalent) could pick up and drop off a dozen or so commuters who live in area A and work in area B. This improves the density of the system. With affordable tunneling, freeways can go underground instead of destroying communities on the surface. This is one of the greatest evils of our car-centric infrastructure today, but with affordable tunneling we can abolish it. Train-based mass transit will always be desirable for high-traffic corridors because it can be made far more dense. And despite this article’s attempts to paint him otherwise, Musk agrees with this - he is in talks to build a Hyperloop between Washington and Baltimore, as well as a lower-speed mass transit “loop” in Chicago. However, trains will never be all that we need for transit, because they just can’t go everywhere. You will almost always need last-mile transportation at the endpoints, and there will always be routes that are not efficiently served by the train network. That’s where cars (or whatever we call the things they evolve into) come in. A driverless, electric, shared, and tunnel-augmented vehicle ecosystem is probably the perfect mass transit system.