Here's a great small point from Ryan Avent:
Consider the automobile, for instance. Back in the late 19th century, it would have been easy to dismiss the potential importance of the car. Cars weren't very good, for one thing. But more importantly, it wasn't at all clear how they might be of much advantage. Cities were dense places, with tangles of streets crowded by pedestrians and horse-drawn wagons and omnibuses. Cars were expensive and offered little in the way of travel-time advantages. Petrol was very expensive in real terms, and there was no infrastructure available to deliver it to would-be drivers. It was hard to see a market for cars outside of the realm of playthings for the rich.
The striking thing about this line of thinking is that it underestimates the automobile for the exact reason that the automobile turned out to be so important. A transformative new invention, by definition, doesn't fit very well into the world as it exists. Automobiles were transformative—so useful that vast new public and private infrastructure projects were undertaken over a period of decades to better-and-better accommodate their presence. But that's an iterative process. At the outset nobody wants to build the infrastructure for something that nobody uses, and nobody wants to use something that there's no infrastructure for. Eventually it tips.
The flipside of this is that as a matter of personality, inventors and innovators are bound to be prone to over-enthusiasm for their own gizmos. So folks who adopt the curmudgeon role and dismiss the hyper-men are going to be right nine times out of ten. But the successes that really matter are exactly the ones that have lots of barriers standing between them and real utility.