5 ms·

What an oddly pedantic comment. 10^21, 10^22 and 10^23... Whichever number, the point that the parent poster was making doesn't change at all. >You are implic

What an oddly pedantic comment. 10^21, 10^22 and 10^23... Whichever number, the point that the parent poster was making doesn't change at all. >You are implicitly assuming a probability that life can arise. What are you suggesting? We're all actually not alive? Come on.

pfdietz 4y ago
No, it's not oddly pedantic. It's more an expression of anger, because this sort of argument is common among SETI people. It's a very common error: there's lots of opportunities for life to arise, therefore life must be common in the universe. And this is utter bullshit. Probability does not work that way!
- haswell 4y ago
  > there's lots of opportunities for life to arise, therefore life must be common in the universe. And this is utter bullshit. Why anger? And will you expand on why this is utter bullshit? In what way does probability not work this way? You are making strong assertions without expanding on why you believe those assertions. Your assertion also goes against some pretty big names in the field; not that big names are automatically trustworthy, but they do seem to carry more weight than a random HN comment (no offense).
  - pfdietz 4y ago
    The argument that gets my goat is basically this: "if N is large enough, then for any probability p > 0, N p > 1." And that's clearly false. The SETI people making this busted argument are implicitly assuming p, the chance that life arises on a suitable world, is not "too small". And there is no basis for making that assumption.
    - johnday 4y ago
      On all of the Goldilocks planets we have so far been able to image with the resolution needed to determine the presence of life, we have determined the presence of life. That is not mathematically rigorous, but it is, from the Bayesian perspective, enough to move the needle.
      - pfdietz 4y ago
        Observer selection bias blows that argument out of the water. Ask yourself: if life were extremely rare, what would be different here on Earth? Nothing, because life has to have started on Earth for us to be here to be thinking about it.
      - thaumasiotes 4y ago
        > That is not mathematically rigorous, but it is, from the Bayesian perspective, enough to move the needle. That is obviously false; zero observations will never cause a Bayesian adjustment.
    - haswell 4y ago
      I don't think the common form of argument is that because N is large enough, life must exist. The argument is generally that scale and probability favor the existence of intelligent life, assuming Earth is not somehow exceptional in the universe. Assuming Earth is exceptional seems just as problematic. Welcome to the Fermi paradox.
      - pfdietz 4y ago
        And for any N, if p is small enough, other life is very unlikely to exist. Why are you holding p fixed and allowing N to increase? You are implicitly assuming N is in that regime where N > 1/p. That's just assuming p > 1/N, and there's no good reason to make that assumption.
        gcanyon 4y ago
        I sympathize -- I've argued your side of this many times. I won't jump in today, but I just wanted you to know your words aren't falling on entirely deaf ears.
        pfdietz 4y ago
        I think observer selection falls into a cognitive blind spot many people have.
        nyokodo 4y ago
        This fallacy has always bothered me so thank you for stating it so well!
      - gcanyon 4y ago
        > I don't think the common form of argument is that because N is large enough, life must exist. The comment that started this thread: "That's 10^22 opportunities to get it right. We can't possibly be alone."
- richardfey 4y ago
  As far as N is smaller than the available combinations, the hypothesis is probable.

pfdietz 4y ago

No, it's not oddly pedantic. It's more an expression of anger, because this sort of argument is common among SETI people. It's a very common error: there's lots of opportunities for life to arise, therefore life must be common in the universe. And this is utter bullshit. Probability does not work that way!

haswell 4y ago
> there's lots of opportunities for life to arise, therefore life must be common in the universe. And this is utter bullshit. Why anger? And will you expand on why this is utter bullshit? In what way does probability not work this way? You are making strong assertions without expanding on why you believe those assertions. Your assertion also goes against some pretty big names in the field; not that big names are automatically trustworthy, but they do seem to carry more weight than a random HN comment (no offense).
- pfdietz 4y ago
  The argument that gets my goat is basically this: "if N is large enough, then for any probability p > 0, N p > 1." And that's clearly false. The SETI people making this busted argument are implicitly assuming p, the chance that life arises on a suitable world, is not "too small". And there is no basis for making that assumption.
  - johnday 4y ago
    On all of the Goldilocks planets we have so far been able to image with the resolution needed to determine the presence of life, we have determined the presence of life. That is not mathematically rigorous, but it is, from the Bayesian perspective, enough to move the needle.
    - pfdietz 4y ago
      Observer selection bias blows that argument out of the water. Ask yourself: if life were extremely rare, what would be different here on Earth? Nothing, because life has to have started on Earth for us to be here to be thinking about it.
    - thaumasiotes 4y ago
      > That is not mathematically rigorous, but it is, from the Bayesian perspective, enough to move the needle. That is obviously false; zero observations will never cause a Bayesian adjustment.
  - haswell 4y ago
    I don't think the common form of argument is that because N is large enough, life must exist. The argument is generally that scale and probability favor the existence of intelligent life, assuming Earth is not somehow exceptional in the universe. Assuming Earth is exceptional seems just as problematic. Welcome to the Fermi paradox.
    - pfdietz 4y ago
      And for any N, if p is small enough, other life is very unlikely to exist. Why are you holding p fixed and allowing N to increase? You are implicitly assuming N is in that regime where N > 1/p. That's just assuming p > 1/N, and there's no good reason to make that assumption.
      - gcanyon 4y ago
        I sympathize -- I've argued your side of this many times. I won't jump in today, but I just wanted you to know your words aren't falling on entirely deaf ears.
        pfdietz 4y ago
        I think observer selection falls into a cognitive blind spot many people have.
      - nyokodo 4y ago
        This fallacy has always bothered me so thank you for stating it so well!
    - gcanyon 4y ago
      > I don't think the common form of argument is that because N is large enough, life must exist. The comment that started this thread: "That's 10^22 opportunities to get it right. We can't possibly be alone."
richardfey 4y ago
As far as N is smaller than the available combinations, the hypothesis is probable.