How many SpaceX rockets and failures have there been? (And that's not a knock on SpaceX, either - this stuff is hard. Combining precision and technology with 'controlled explosion' is going to be a challenge).
But as a NASA person said - NASA-funded contracts "can't" have failures. They obviously do, but he was more talking to the acceptability, political and otherwise. One or two launchpad explosions of a taxpayer funded vehicle and you're fighting Congressional demands to shut down the entire program. SpaceX provides a layer of abstraction and indirection to that, so they can move faster - "Who cares if we blow up 10 in the next couple of years to get to one that works".
But they didn't, they went for the first time right approach, but that failed too. If you are going to have failures, maybe just accept that first time right doesn't exist, or just takes much much longer.
Either you make fast iteration acceptable, regardless of politics, or you fail.
It's not just political process either, it's the technical process. You need to be able to debug, fix, and manufacture the iterations quickly.
Looking at https://www.space.com/every-spacex-starship-explosion-lesson...
there have been many many prototype and other losses. And incidents, some catastrophic, some less so.
SN1, 3, 4, 7, 8, 9, 10, 11, 15 and the orbital Starship launch attempt all had failures losing hardware. If NASA/publicly funded work had that many failures (or a fraction of them) there'd be Congressional enquiries and calls to shut down the program and stop burning tax dollars.
It's true that SpaceX enjoys more latitude to destroy test hardware in its private development programs that aren't funded with somebody else's money (public or private), but why is that relevant here? Commercial Crew was funded by NASA with public money, and SpaceX developed Dragon 2 in a relatively conservative and conventional program with NASA looking over their shoulder the whole time. There is no double standard.
The overall perspective is that SpaceX developed their crewed capsule much much faster and cheaper than Boeing. The data also indicates that flying with SpaceX is safer.
Congress doesn't care about buring tax dollars as long as it is spent in their districts. Otherwise Artemis and SLS wouldn't exist.
The issue isn't "burning tax dollars". Congress is too busy selling out the country's future to give a shit about that. The issue is that they'd already rather not be giving any money to NASA in the first place. They'd just give the defense contractors tax payer funded 'donations' directly if they could.
SpaceX has an assembly line in Hawthorne and test site in Texas. (Both send kit to the space coast for launch.)
The reliability of what comes out of the former exceed’s Boeing’s. The innovativeness of what comes out of the latter exceeds them once again. Muddling statistics between the two would be like considering Boeing’s experimental drones when measuring its commercial airliners’ reliability. They’re totally different departments.
So SpaceX is the only company truly innovating on all fields, has the highest launch success rate, highest launch cadence, the most capable rockets, and launches for far cheaper than any other company (or country).
I think it's worth pointing out that the failures were early in F9's life, and that the current configuration "F9 v1.2FT Block 5" has had no failures to launch and remarkable success when it comes to landing the booster.
> How many SpaceX rockets and failures have there been? (And that's not a knock on SpaceX, either - this stuff is hard. Combining precision and technology with 'controlled explosion' is going to be a challenge).
IMO, this really misunderstands the two kinds of "tests".
SpaceX is engaged in a development program. And as a part of that development program, they're doing test flights to discover how to properly build Starship. Those flights are expected to fail in various ways. The exact way they fail gives SpaceX vital information that's used to improve the rocket.
A big part of the reason SpaceX is doing this is because simulation and modeling have a limited ability to give good answers to questions about novel behaviors when it comes to rockets - the speeds and just too high. And the only way to find the true unknown unknowns is to interact with reality.
In contrast, Starliner's tests are supposed to be demonstrations that the system is complete, functional, and ready for service. They are not supposed to have anything wrong with them at all.
It's worth pointing out that Boeing chose to do less testing and more paperwork as part of Starliner's certification. If Boeing had done an in-flight abort test instead of a pad abort test like SpaceX did, they probably would have caught the OFT-1 problems then.