The Singularity Is Nearer

Kurzweil grading his own predictions two decades on, which is more interesting than the new ones. He was directionally right about compute and wrong about what it would feel like. I read it for the scorekeeping, not the forecast.