image

I’m interested in the latter.

You might think, “Well, they’re both time savers, so why not just go faster?” But the latter would be counterproductive with respect to productivity. See Candle Experiment

As for the latter, since “whether it is a goal” is unknown, performing many acts to confirm “whether it is a goal” is the royal road to shortening the time to reach the goal. However, if the number of trial-and-error attempts itself is used as a KPI, even useless trial-and-error attempts will be done without hesitation. Instead of the dichotomy of “is it a goal or not?” we need to consider “what is the probability that it is a goal” and start with the highest probability. In the absence of information, everything is equally probable, so it is better to start immediately without worrying about which one to start with, and the difference in probability will gradually become visible according to the information obtained.

  • Eratosthenes’ sieve
  • Algorithm for finding prime numbers
  • It is inefficient to try to see if they can be divided in turn.
  • Addition is less costly than division.
  • At the stage of finding a prime number, that knowledge will tell you “there is no prime number here”.
  • By recording this information, we can efficiently discover prime numbers.

This page is auto-translated from /nishio/äșŒçšźéĄžăźç”Ÿç”Łæ€§ using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.