3 ms·
> If you think you know everything, you will miss a lot. However, searching for low probability events in a large search space is not necessarily going to be a
by mikekchar 8y ago
> If you think you know everything, you will miss a lot.
However, searching for low probability events in a large search space is not necessarily going to be a fruitful experience. For example, imagine that there is non-zero probability of a 20 karat diamond being on the beach somewhere, buried in the sand. You can look for it as much as you want, but you will still probably miss it.
Having a small outlet for things that seem crazy because they might turn out to be true is the same kind of logic as buying a lottery ticket because you can't win if you don't play. Yes, you will miss things, but you will almost certainly miss them anyway unless you expend more resources than they are worth -- and then you still aren't guaranteed to make progress.
On the other hand low probability risks that are likely to happen sometime are often worth investigating. For example, if there is a 1 in 1000 year earthquake event, it's highly unlikely to affect you in your life. But it will eventually affect someone so it may be work the effort to study what might happen. To get more esoteric, perhaps we can talk about large meteorites hitting the planet. Do we need a contingency plan for it?
To me, that's the main question you need to ask before you start: are we building up an understanding of something that will almost certainly affect us one day, or are we simply sifting through grains of sand on the beach hoping to find a diamond?
- colechristensen 8y ago>low probability risks that are likely to happen sometime are often worth investigating. >buying a lottery ticket because you can't win if you don't play These are the same situation
- nmca 8y agoWhat? Expectation or risk-adjusted expectation are specified in neither case, they are the defining characteristics and you can't make statements about either of the cases without knowing or estimating them.
- FeepingCreature 8y agoI think the point is you know that beaches don't have diamonds in them in reachable range because people frequent beaches. But it's plausible to me that the CIA is sitting in unique operational territory, and their informational "obviously worthless" periphery is not well frequented. It's not about finding diamonds on the beach, it's about finding beaches with diamonds.
- wslh 8y ago> However, searching for low probability events in a large search space is not necessarily going to be a fruitful experience. The same concern can be applied to areas like basic research and founding startups yet there are a lot of resources, success, and overall patience there.
- marcoperaza 8y agoThere’s a dimension to this that you’re not considering: what you care about most as a government is that no one else gets a lead on you (that is, relative advantage). The difficulty you identify isn’t such an issue if you’re just trying the things in the same realm that other people are suggesting/investigating, to make sure that you don’t fall behind.