4 ms·

I think (hope?) the Drake equation was never intended as a tool for making realistic estimations, I always thought it was more of a first approximation of how o

by ud_0 5y ago

I think (hope?) the Drake equation was never intended as a tool for making realistic estimations, I always thought it was more of a first approximation of how one might go about thinking about modeling when all you have is a sample of one. Specifically the rate of star formation has always been dubious, because there is such an enormous lag between star formation and the emergence of intelligent life (even if we turn out to be an extremely slow outlier), making it unsuitable for estimating the amount of civilizations active at a specific point in time. My personal drake equation would include only factors that are based on the current state of the galaxy. But that's the Drake equation doing its job: making people think about cosmic odds. It doesn't need an update, because I argue it shouldn't be taken literally in the first place. A few times a year, authors put out "new" estimations of the number of galactic civilizations, and to do that there is really not much alternative to chaining together these kinds of probabilities. While they sometimes may or more often may not use the actual Drake equation to do it, its value lies in the basic structure, not in its specific implementation.

ghaff 5y ago
One of the challenges (I won't say problems) with the Drake Equation is we have some sense for some numbers and they're very very large. Which encourages people to use formulations for some of the other terms like "even if we assume this is one in a billion" and this is "one in a million," you still come up with a significant number of civilizations. When, in fact, some other term may be so vanishingly small that it's only happened once. (Or there may be something else, e.g. a great barrier, that's not being taken into account.) But, as others have written, it's not really intended to be used that way.
derekp7 5y ago
Is it valid to take a data point of one, combined with a time element to get an estimate? For example Earth started gaining life shortly (in geological time) after it was able to sustain it. That would indicate that either life arises easily, or that it pre existed the Earth and got seeded (panspermia). Then at the opposite end of the scale is the long time that it took for life to go multi-cellular, and for a space program to be formed by an instance of that life.
- ghaff 5y ago
  >That would indicate that either life arises easily Not really. While suggestive, all it indicates is that it arose on earth easily.
  - s1artibartfast 5y ago
    not even that. Life could have been incredibly unlikely to arise on earth, but still arise when it did as freak statistical occurrence. We only have one earth to examine, and the anthropic principle severely limits any conclusions we could draw from n=1.
    - ghaff 5y ago
      Yeah, happened to arise on earth quickly would probably be the better phrasing.
- tsimionescu 5y ago
  Well, we know life on earth is at least 3.7 billion years old, but it's not much clear how much older. Since the earth is 4.5 billion years old, that means that life could have appeared after ~17% of its life span so far, which is not THAT little. Furthermore, even if life appeared as early as the earliest estimate - just a few million years after the earth formed - we have no evidence that abiogenesis continued happening after an initial event, or that it happened anywhere except an initial location. Those two factors may mean that abiogenesis could be an extremely unlikely event that randomly happened to occur very early in the Earth's life span.
- pie420 5y ago
  Your line of thinking is exactly what's wrong. The fact that life emerged on earth shortly after being able to sustain life says absolutely nothing about likelihood. It could be a very likely thing. Or it could be a 1 in a 10^35 miracle that happened due to an insanely low probability event involving gamma rays hitting chemical reactions in just the right way. Until we solve abiogenesis and can demonstrably evolve life from abiotic matter, or observe other planets with life on them, this is all speculation.

by ud_0 5y ago

ghaff 5y ago

One of the challenges (I won't say problems) with the Drake Equation is we have some sense for some numbers and they're very very large. Which encourages people to use formulations for some of the other terms like "even if we assume this is one in a billion" and this is "one in a million," you still come up with a significant number of civilizations. When, in fact, some other term may be so vanishingly small that it's only happened once. (Or there may be something else, e.g. a great barrier, that's not being taken into account.) But, as others have written, it's not really intended to be used that way.

derekp7 5y ago

Is it valid to take a data point of one, combined with a time element to get an estimate? For example Earth started gaining life shortly (in geological time) after it was able to sustain it. That would indicate that either life arises easily, or that it pre existed the Earth and got seeded (panspermia). Then at the opposite end of the scale is the long time that it took for life to go multi-cellular, and for a space program to be formed by an instance of that life.

ghaff 5y ago
>That would indicate that either life arises easily Not really. While suggestive, all it indicates is that it arose on earth easily.
- s1artibartfast 5y ago
  not even that. Life could have been incredibly unlikely to arise on earth, but still arise when it did as freak statistical occurrence. We only have one earth to examine, and the anthropic principle severely limits any conclusions we could draw from n=1.
  - ghaff 5y ago
    Yeah, happened to arise on earth quickly would probably be the better phrasing.
tsimionescu 5y ago
Well, we know life on earth is at least 3.7 billion years old, but it's not much clear how much older. Since the earth is 4.5 billion years old, that means that life could have appeared after ~17% of its life span so far, which is not THAT little. Furthermore, even if life appeared as early as the earliest estimate - just a few million years after the earth formed - we have no evidence that abiogenesis continued happening after an initial event, or that it happened anywhere except an initial location. Those two factors may mean that abiogenesis could be an extremely unlikely event that randomly happened to occur very early in the Earth's life span.
pie420 5y ago
Your line of thinking is exactly what's wrong. The fact that life emerged on earth shortly after being able to sustain life says absolutely nothing about likelihood. It could be a very likely thing. Or it could be a 1 in a 10^35 miracle that happened due to an insanely low probability event involving gamma rays hitting chemical reactions in just the right way. Until we solve abiogenesis and can demonstrably evolve life from abiotic matter, or observe other planets with life on them, this is all speculation.