Actual Space Systems Engineer here (and not for SpaceX): Yes. One of the more recent ones had a communications failure, and self-destructed to make sure it had very little chance at causing the damage the above people have their arses in a knot over. It’s rapid prototyping. Why?
NASA projects run overbudget and over deadline because they’re trying to get it perfect in the first few launches. That’s only part of the problem, but it’s a significant part. Look at Artemis: 1) launch, 2) launch to the moon, 3) launch to the moon for a long duration stay with humans. That takes so much time and money and simulation and testing of everything that even a government has trouble. So what do they do? They adapt, extend deadlines, increase funding, etc.
Private industries don’t have that luxury. If SpaceX decided to run Starship 500x overbudget to get it right in the first few attempts, they’d be bankrupt. How do you remedy this?
Give it your best guess, strap a bunch of sensors to it, watch it (probably) explode (which really is any failure, as it’s required to explode for safety if it can’t land), use that data to improve the design, and then try again a few hundred times until it doesn’t explode anymore.
And in the end, it’s cheaper than spending years predicting every mode of failure and preventing them like NASA does. It’s a different mode of operation, because industry and government have different resources and norms down to the way the project is structured from a leadership point-of-view.
And that’s why commercial rockets are supposed to explode.
If private industries can’t afford to do it right, they shouldn’t do it. You need only look at the quality of software to see why. Once the bar is set low, they will never raise it.
Give it your best guess, strap a bunch of sensors to it, watch it (probably) explode (which really is any failure, as it’s required to explode for safety if it can’t land), use that data to improve the design, and then try again a few hundred times until it doesn’t explode anymore.
You just described the V1 and V2, with the only difference being their intent WAS to explode. Otherwise same stratagy. Which isn’t exactly surprising that musk is running plays from the nazi playbook.
I’m honestly expecting him to try to create a car called “The Everyman” or some shit. But of coarse that would require the cars to not catch fire or explode.
So they were supposed to explode?
Actual Space Systems Engineer here (and not for SpaceX): Yes. One of the more recent ones had a communications failure, and self-destructed to make sure it had very little chance at causing the damage the above people have their arses in a knot over. It’s rapid prototyping. Why?
NASA projects run overbudget and over deadline because they’re trying to get it perfect in the first few launches. That’s only part of the problem, but it’s a significant part. Look at Artemis: 1) launch, 2) launch to the moon, 3) launch to the moon for a long duration stay with humans. That takes so much time and money and simulation and testing of everything that even a government has trouble. So what do they do? They adapt, extend deadlines, increase funding, etc.
Private industries don’t have that luxury. If SpaceX decided to run Starship 500x overbudget to get it right in the first few attempts, they’d be bankrupt. How do you remedy this?
Give it your best guess, strap a bunch of sensors to it, watch it (probably) explode (which really is any failure, as it’s required to explode for safety if it can’t land), use that data to improve the design, and then try again a few hundred times until it doesn’t explode anymore.
And in the end, it’s cheaper than spending years predicting every mode of failure and preventing them like NASA does. It’s a different mode of operation, because industry and government have different resources and norms down to the way the project is structured from a leadership point-of-view.
And that’s why commercial rockets are supposed to explode.
(All this said, Fuck Muskrat)
I think they skip this step more often than they should
If this is a good way to operate, why doesn’t NASA just do that themselves?
They did, in the 50s and 60s.
If private industries can’t afford to do it right, they shouldn’t do it. You need only look at the quality of software to see why. Once the bar is set low, they will never raise it.
If we’re putting people in them at some point shouldn’t we be trying to predict every mode of failure?
You just described the V1 and V2, with the only difference being their intent WAS to explode. Otherwise same stratagy. Which isn’t exactly surprising that musk is running plays from the nazi playbook.
I’m honestly expecting him to try to create a car called “The Everyman” or some shit. But of coarse that would require the cars to not catch fire or explode.