There's a second factor, which is that certain specific fusion designs are heavily over-funded, in a way that doesn't necessarily reflect their practicality. A

There's a second factor, which is that certain specific fusion designs are heavily over-funded, in a way that doesn't necessarily reflect their practicality. A case in point: NIF isn't even supposed to be the basis for a reactor design. ITER is, but it's still a fair way off and even assuming it lights it'll be 2 more generations before possible industrial application. The problem is that these approaches can't work without enormous budgets, and while they produce lots and lots of very interesting science, they hoover up talent and funding resources that might be better spent exploring other avenues.