3 ms·
> We've just seen the one, but we've only had the capability to detect something like that for ~5 years. So in the absence of further data, it's reasonable to
by earenndil 9y ago
> We've just seen the one, but we've only had the capability to detect something like that for ~5 years. So in the absence of further data, it's reasonable to conclude that `Oumuamua-like objects pass through the solar system once every 10 years
No it's not. It's impossible to deduce anything from a sample size of 1. (And yes I realise that this invalidates my point as well.)
- nkoren 9y agoToss a coin (known to be non-evenly-weighted) five times. It comes up heads once. Can you now estimate the probability of the coin coming up heads? Of course you can. What you can't do is put error-bars on that estimate: you might have just seen the only "heads" that you'll see in the next ten thousand tosses. So the error bars are essentially infinite, but you can make a first-order estimate. Analogously, we've had the capability to to detect `Oumuamua-like objects for ~5 years, and after searching for 5 years, have found one such object. That's good enough to start making first-order estimates of probabilities. Such estimates will obviously change, possibly significantly, as more observation time goes by. Absent error bars, you can't be certain of that first-order estimate is even approximately correct. But if it's not even vaguely correct, then that means your initial "heads" observation was very, very, unlikely. Hence my conclusion is valid: either we got extraordinarily lucky in detecting `Oumuamua when we did, or Carl Sagan is wrong.