4 ms·

It's hard to say, though, because the system is totally untested. Even if it works the first time, That's a very small sample size.

bryanlarsen 2y ago
NASA has gotten a lot better at estimating risk, especially after two Shuttle explosions taught them the hard way that they were really bad at it. Applying those methods to Apollo shows us just how lucky we are that we didn't lose any of them. NASA claims that their risk of catastrophe on Artemis 2 is 1 in 400. They'd have to be wrong by a couple orders of magnitude to be comparable to the Apollo missions. I might not trust their 1 in 400 number, but I don't think they're off by 2 orders of magnitude. The endless Starliner delays are another sign of this. They could launch Starliner with the helium leak and everything would probably be fine. But they're not. They've lost astronauts to "go fever" and have instituted a much more risk averse culture.
- metadat 2y ago
  Eh, it's still untested because they don't really do anything involving humans anymore. So how are they better at quantifying risk? Smoke.
  - bryanlarsen 2y ago
    So was Apollo. If I assert that each Apollo mission had a 1 in 4 chance of killing astronauts they didn't do enough flights to prove me wrong. NASA aims for a 1 in 400 chance. You'd need thousands of flight to prove that level of risk. They have to rely on risk modelling.
    - metadat 2y ago
      No? Apollo ran more than a dozen times and failed one time. And even then, what's a human life worth? Not a billion dollars, not even my own life is worth that much if we're being honest. Progress is pretty good. Something has been lost and is missing today with folks more eager to get home before traffic gets too heavy.
      - bryanlarsen 2y ago
        If you roll a 4 sided die twelve times, getting a single "1" is a likely outcome. Getting a "1" 3 times is more likely, but not significantly so.

bryanlarsen 2y ago

NASA has gotten a lot better at estimating risk, especially after two Shuttle explosions taught them the hard way that they were really bad at it. Applying those methods to Apollo shows us just how lucky we are that we didn't lose any of them. NASA claims that their risk of catastrophe on Artemis 2 is 1 in 400. They'd have to be wrong by a couple orders of magnitude to be comparable to the Apollo missions. I might not trust their 1 in 400 number, but I don't think they're off by 2 orders of magnitude. The endless Starliner delays are another sign of this. They could launch Starliner with the helium leak and everything would probably be fine. But they're not. They've lost astronauts to "go fever" and have instituted a much more risk averse culture.

metadat 2y ago
Eh, it's still untested because they don't really do anything involving humans anymore. So how are they better at quantifying risk? Smoke.
- bryanlarsen 2y ago
  So was Apollo. If I assert that each Apollo mission had a 1 in 4 chance of killing astronauts they didn't do enough flights to prove me wrong. NASA aims for a 1 in 400 chance. You'd need thousands of flight to prove that level of risk. They have to rely on risk modelling.
  - metadat 2y ago
    No? Apollo ran more than a dozen times and failed one time. And even then, what's a human life worth? Not a billion dollars, not even my own life is worth that much if we're being honest. Progress is pretty good. Something has been lost and is missing today with folks more eager to get home before traffic gets too heavy.
    - bryanlarsen 2y ago
      If you roll a 4 sided die twelve times, getting a single "1" is a likely outcome. Getting a "1" 3 times is more likely, but not significantly so.