He's missing parts of the problem. Yes, nukes are too slow and expensive--let's look at why in a bit more detail: Slow: There are two big factors here. Firs

He's missing parts of the problem. Yes, nukes are too slow and expensive--let's look at why in a bit more detail: Slow: There are two big factors here. First, they aren't using stock designs. Most of the design time goes away if you're using an off-the-shelf design. Second, litigation. The greens figured out how to tie things up in court to the point that power companies won't touch nuke. Especially with things like regulators telling them they can't operate their newly-built plant (nothing wrong with it) and can't pass it along to customers in their rates. Costly: As he says, ALARA guarantees nuke is too expensive because it automatically precludes any actual improvements. There's no reason to come up with a better design because in the end it will cost just as much because of the extra safety systems they will have to pile on it. There's also the delays caused by litigation that can really drive up the bill. My fixes: 1) One bite at the apple. Once a plan has been approved that's it, the government doesn't get to say no unless what's actually built has a meaningful deviation from what was approved or there have been major changes in understanding of the risks. 2) Scrap ALARA. Instead, require nuke plants to be 10x safer than the next safest *competitive* source of power. This is natural gas which is currently approximately 50x as dangerous. (Solar and wind are intermittent--to be competitive you need to count their storage systems also. The storage systems don't exist, thus they can't be counted. Hydro is limited by available sites--and pretty much every worthwhile place already has a dam. I haven't seen safety numbers on geothermal but it has the same problem--it can't be scaled up.) (And I would apply this principle to industry in general--the government can't impose safety requirements on something that's too far above the alternative.)