2 ms·

The problem is that humans, you and me included, are very bad at extremely large and extremely small numbers. We don't have infinite planets. We have 10^21 sta

by mabbo 8y ago

The problem is that humans, you and me included, are very bad at extremely large and extremely small numbers. We don't have infinite planets. We have 10^21 stars, roughly, and some constant P average planets per star. If the odds of life are 10^-14 per planet, and the odds of intelligent life are 10^-7 per life hosting planet, and we presume every star has one planet that might support life, then in the 10^21 stars in the universe, we'd expect only one intelligent life. That is one possibility! But we simply don't know what the odds of life and the odds of intelligent life are in the slightest. We don't even know what range of possible values are reasonable. And humans are so bad at contemplating numbers like 10^21 and 10^-14. They're too extreme for one to think of them rationally. And saying "well, it happened here, be it must be common" is selection bias- we can only contemplate the problem because it happened here. The right answer is to get more information to improve our estimates of the probabilities. If we can find life at Alpha Centauri, that would either be a hell of a coincidence, or a very good indicator that the probability of life given a planet is high.

Razengan 8y ago
> If we can find life at Alpha Centauri, that would either be a hell of a coincidence, or a very good indicator that the probability of life given a planet is high. Why would finding life on just one more planet – out of the 195 or so "worlds" that we know about, including moons – suddenly make "the probability of life given a planet" high? By your own reasoning, whether we find life at Alpha Centauri, or do not find any life there, will not really change the probability of discovering future life or intelligence among the rest of all the 10^21 * P planets. We could always just keep saying that "There are only 2 planets in the universe with life. No more." "There are only 5 planets in the universe with life." and so on and on.
- mabbo 8y ago
  > Why would finding life on just one more planet – out of the 195 or so "worlds" that we know about, including moons – suddenly make "the probability of life given a planet" high? Because sampling. It's like election polling. If you ask one person who they're going to vote for and they say "Candidate A", you really have no idea how likely it is that candidate A has massive support. Perhaps you just happened to find the only person voting for that candidate! But if you ask the next random person and they say the same thing, well, the odds that you found the only two people voting for candidate A in your first two random asks is very low. On the other hand, if the second person is not supporting candidate A, that also helps you get a better guess at support levels. You can never know the true level without asking every single person, but with every additional piece of data you can better estimate just what the truth is. Just the same, as soon as we find another star with life, the number of stars we had to search to find that second one gives us a fantastic start to coming up with an estimate on how common life is. We presume that our corner of the universe is a good sampling of the rest, and extrapolate based on what we know here. The key point is that extrapolation from a single data point is wildly innacurate.
  - Razengan 8y ago
    > As soon as we find another star with life, the number of stars we had to search to find that second one gives us a fantastic start to coming up with an estimate on how common life is. We presume that our corner of the universe is a good sampling of the rest, and extrapolate based on what we know here. We currently know of at least 8 or 9 planets in our system + at least 185 moons (new ones were discovered around Jupiter last month.) So, finding life on 2 out of 194~ places will take us from "There is only 1 planet in the entire universe with life!" to "how common life is"? > The key point is that extrapolation from a single data point is wildly innacurate. But extrapolating from 2 points out of (100 billion x 200 billion) is not? – Why wouldn't people continue to use the same arguments they do now to say Earth is the only planet with life, to say there are only 2 planets in the universe with life?
  - dragonwriter 8y ago
    > Because sampling. Sampling requires a random selection from the universe of interest.

by mabbo 8y ago

Razengan 8y ago

> If we can find life at Alpha Centauri, that would either be a hell of a coincidence, or a very good indicator that the probability of life given a planet is high. Why would finding life on just one more planet – out of the 195 or so "worlds" that we know about, including moons – suddenly make "the probability of life given a planet" high? By your own reasoning, whether we find life at Alpha Centauri, or do not find any life there, will not really change the probability of discovering future life or intelligence among the rest of all the 10^21 * P planets. We could always just keep saying that "There are only 2 planets in the universe with life. No more." "There are only 5 planets in the universe with life." and so on and on.

mabbo 8y ago
> Why would finding life on just one more planet – out of the 195 or so "worlds" that we know about, including moons – suddenly make "the probability of life given a planet" high? Because sampling. It's like election polling. If you ask one person who they're going to vote for and they say "Candidate A", you really have no idea how likely it is that candidate A has massive support. Perhaps you just happened to find the only person voting for that candidate! But if you ask the next random person and they say the same thing, well, the odds that you found the only two people voting for candidate A in your first two random asks is very low. On the other hand, if the second person is not supporting candidate A, that also helps you get a better guess at support levels. You can never know the true level without asking every single person, but with every additional piece of data you can better estimate just what the truth is. Just the same, as soon as we find another star with life, the number of stars we had to search to find that second one gives us a fantastic start to coming up with an estimate on how common life is. We presume that our corner of the universe is a good sampling of the rest, and extrapolate based on what we know here. The key point is that extrapolation from a single data point is wildly innacurate.
- Razengan 8y ago
  > As soon as we find another star with life, the number of stars we had to search to find that second one gives us a fantastic start to coming up with an estimate on how common life is. We presume that our corner of the universe is a good sampling of the rest, and extrapolate based on what we know here. We currently know of at least 8 or 9 planets in our system + at least 185 moons (new ones were discovered around Jupiter last month.) So, finding life on 2 out of 194~ places will take us from "There is only 1 planet in the entire universe with life!" to "how common life is"? > The key point is that extrapolation from a single data point is wildly innacurate. But extrapolating from 2 points out of (100 billion x 200 billion) is not? – Why wouldn't people continue to use the same arguments they do now to say Earth is the only planet with life, to say there are only 2 planets in the universe with life?
- dragonwriter 8y ago
  > Because sampling. Sampling requires a random selection from the universe of interest.