I had that problem too (from the commentary here, this lack of specific examples is the post’s biggest issue) -- whatever examples I could come up with seemed distinctly unspectacular.
However, I think avoiding common failure modes—being less wrong—is a decent way to increase the expected value of your power.
Unfortunately, it seems much easier to list particularly inefficient uses of time than particularly efficient uses of time :P I guess it all depends on your zero point.
I had that problem too (from the commentary here, this lack of specific examples is the post’s biggest issue) -- whatever examples I could come up with seemed distinctly unspectacular.
However, I think avoiding common failure modes—being less wrong—is a decent way to increase the expected value of your power.
Unfortunately, it seems much easier to list particularly inefficient uses of time than particularly efficient uses of time :P I guess it all depends on your zero point.