fwiw i think the interesting part about the original study wasn't so much the slowdowm part, but the discrepancy between perceived and measured speedup/slowdown

fwiw i think the interesting part about the original study wasn't so much the slowdowm part, but the discrepancy between perceived and measured speedup/slowdown (which is the part i used to bring up frequently when talking to other devs)