I mean, nuclear might be expensive, time consuming to build, and difficult but at least we've been doing it for 70 years. It's not clear that projections of long term costs on things we haven't even done at scale are going to be realistic when we try them in the real world.

Even if it takes 20 years to build a nuke, that might still be sooner than "power to x" becomes a viable, affordable option for most of the world.

This is the weird thing about this argument: it so often rests simultaneously on an argument for practicality and/or cost effectiveness, but literally any criticism is met with a gish gallop of unproven technology. You can't have this both ways.

In the end, my central thesis here is just that if you think the be all and end all of this issue is cost effectiveness you are optimizing for the wrong goals and it leaves open a giant window for carbon emitting sources to justify their use when renewables don't work for various local reasons.

The thing to be black and white about is carbon emissions. "What energy mix works for place X right now if we try to eliminate carbon emissions asap" is not a decision that's gonna be made on an internet forum.