I always call this a "local maximum" problem. Once you've optimized the crap out of your tech, any change makes it worse (e.g., replacing crossbows with primitive guns). But if you do switch, then optimizing that technology takes you to an even higher maximum.
The problem is that you have to go backwards to go forwards, and you can't always predict (or convince the powers-that-be) that the end result will be better.
Extremely relevant to electric cars. Looks like we are close to electric > ICE, (or past that point, whatever), but it was a long painful time of hyping subpar cars by those who believed in the potential of the technology.
Agreed! I also believe that once we flip (EV > ICE) the momentum goes the other way.
For example, let's say that 50% of cars on the road are EVs. Now gas stations have a problem. You can't survive with half your customers gone, so maybe half the gas stations go out of business. But that means your nearest gas station is much further away, so now the incentive for EV goes up.
In California (and the Bay Area, particularly), I bet we'll see this relatively soon.
The problem is that you have to go backwards to go forwards, and you can't always predict (or convince the powers-that-be) that the end result will be better.