Hacker News Reader

Comment by altacc | Hacker News Reader

It is wrong to compare these two very different eras. The Apollo program was an ideological race, so there was a mush higher tolerance of risk and financial and human costs. Remember, people died on the US & Soviet space programs as they were figuring things out. There might not have been many launch scrubs but there were plenty of failures.

Today, the tolerance for risk is much lower, the costs higher and the budget much more controlled. There are many politicians who want to greatly reduce NASA's budget in favour of private ventures. A failed launch is ammo for their cause. So whereas Apollo could proceed (with some delays) after people burnt to death in a capsule, if that happened today the whole program would be halted indefinitely and NASA heavily questioned as to why it needed to take such risks. 50 years on from Apollo, NASA has learnt that an abundance of caution is needed.

View on HN

I agree that the tolerance for risk is different today. But the argument would seem to imply, that NASA was then somehow more willing to endanger the lives of the crew. As mentioned, people died, so the increased caution was already a part of the program.

Let's not diminish the brilliant, flawless execution of the Apollo program, with the argument that they cared less for the lives of Astronauts.

Sorry, they were more willing to endanger the lives of the crew, and they even admitted it. I remember watching a documentary where they were estimating the percentage chance of the loss of the crew, and comparing that to the chance the Russians would beat them to the moon.

The comment you replied to is correct. Every launch has a non-zero chance of death, and the Apollo program was simply willing to tolerate a higher probability of loss than we are today. I think if you want to call that "they cared less for the lives of Astronauts" that that's your own interpretation.

Brilliant, but certainly not flawless.

"Flawless execution"? Apollo 1 burned three astronauts to death during a rehearsal test. They had astronauts back on top of a rocket in under twenty months.

Acknowledging that their risk tolerances were higher than politics necessitates today doesn't "diminish" anything.

> They had astronauts back on top of a rocket in under twenty months.

In a redesigned spacecraft (based on the Block II that was already in development) that replaced the pure oxygen environment at launch with a Nitrogen/Oxygen atmosphere that was slowly bled off to pure O2 at the desired partial pressure once it was in space.

Some of the critical changes were previously vetoed until Apollo 1 confirmed the worth of doors that opened to outside rather than inside, despite increased weight of the sealing mechanism.

And yet the inward-opening door was itself a response to a near-fatal failure of the hatch on Liberty Bell 7. That almost killed the second American in space, Virgil I. "Gus" Grissom.

Grissom was one of the three astronauts that died in the Apollo 1 fire.

There's a movie about one of the flawed launches.

"Houston, we have a problem"

This is dangerous. It's going to kill people. We need to make a decision on whether or not it's acceptable and proceed or not. But this seems to be politically not possible. Really the only people's opinions that I care about at all about this are the actual astronauts that are putting their lives at risk, but somehow it always seems to come down to some bureaucrats whose definition of risky is leaving blanks on a form unfilled.

I don’t think this explanation adds up very well.

NASA had a manned spaceflight program after Apollo that was specifically designed for safety and efficiency—the Space Shuttle—which turned out to be expensive and unsafe. After the Columbia disaster the Shuttle was cancelled, but the subcontractors for the Shuttle cried foul and Congress pretty much forced NASA to build a new rocket out of Shuttle parts—first Ares, then SLS.

If it were up to NASA, they would either build the rocket out of something other than old Shuttle parts or just outsource rocket development and focus on operations. The politicians are the ones making them build SLS.

It's not nice what I'm about to say, but maybe it would be better in the long run if the SLS blew on the launchpad. It's not manned, so it won't lead to loss of life. It would lead to program termination, which is a good outcome.

It wouldn't just terminate SLS though, it would probably kill Artemis, which would be a bad thing.

Maybe. But then again, Musk is hellbent to go to Mars. If SpaceX indeed builds thousands of Starships, then a private-funded Artemis-like program will be possible for what will amount to spare change.

We are pretty close to (if not already at) the point where we don't need Government funding for space exploration.

> Musk is hellbent to go to Mars.

He's hellbent on building Starship I think, but Mars? I don't see it. He's invested a lot of time and money into building a booster that will excel at putting satellites into orbit for Starlink and DoD contracts. But how much has he invested in Mars colony habitat technology? Virtually nothing.

Starship is a satellite launcher, not a Mars rocket. The Mars talk is a recruiting tactic.

You have a point, but in the end it doesn't matter. If you can bring 100 tons of cargo in Low Earth Orbit and refuel there, then a Moon mission is easy enough for a VC-funded startup to put together. Musk will be more than happy to sell ride tickets on his Starship.

Given his behavior over the past few years, i'd rather have two baskets with two sets of eggs.

Going to Mars will cost between $100B and $10T (the latter includes building a city). There's no way that Musk can do this on his own without government assistance.

> Today, the tolerance for risk is much lower,

yet SpaceX can launch a lot more often and a lot more reliably, and cheaper too. Strange, huh?

I think you might not be taking a significant numbers of factors into account as these are not comparable programs. Artemis & the SLS have very different requirements to SpaceX’s falcon. The closer comparison is to Starship & Super Heavy, which has also not yet launched. The Artemis program is ahead of SpaceX.

Compare SLS and SpaceX, not Artemis and SpaceX. SpaceX became a part of Artemis when it was selected to take Astronauts to the moon. SLS will only take them into orbit near the moon. SpaceX will take them to the actual surface of the moon. That’s a kinda important part of Artemis.

Yeah. It’s maddening to imagine that the spaceship to bring them to the moon is smaller than the lander that will put them on the moon.

That landing ever actually occurring seems about as likely as pigs flying (or pork barrels flying?) NASA selecting the absurd Starship lander proposal seems like a tacit admission of this.

The lander contract is a 'safe' contract for SpaceX to win, despite their absurd proposal, since the whole program is very unlikely to ever get to the stage where the lander is actually put to use and sent to the moon.

Frequency and cost yes, but reliability is more complicated.

ULA has 148 launches with NO accidents. SpaceX has put up more, but had an accident in flight with the Falcon 9 in June 2015, and another on the pad in September 2016.

Mind you, ULA started with a known rocket design. While SpaceX has been iterating as it goes.

But still, we don't have significant statistical evidence about which is actually more reliable.

As you said, ULA when it started already had mature rockets. You need to compare rocket families not operators.

And ULA had a number of issues in that time. For example a failure in an Atlas first stage and the only just managed to deploy the sat because the upper stage had just enough margin.

And ULA had major delays, way longer then what SpaceX could afford for things like their Delta 4 Heavy launches.

ULA of course also got 1 billion $ in subsidies per year to ensure launch safety, a luxury SpaceX never had.

> But still, we don't have significant statistical evidence about which is actually more reliable.

Actually we do. It you look at the launch rates and how often Falcon 9 has launched already the statistical significant of the results gets bigger and bigger.

Unfortunately I can't find the list that shows this data and explains the methods used right now.

But Falcon 9 is very clearly on top of the list.

Every time they do a run-of-the-mill Starlink launch that goes without a hitch, Falcon 9 becomes safer and more valuable. Every time Falcon 9 flies, the necessity for Starship to succeed soon decreases.

I doubt Starlink is commercially viable in the long run if they don't get Starship operational. The Starlink constellation requires continuous maintenance, replacement of old satellites that have expended their fuel and fall out of LEO a few months later. To do this with Falcon 9 instead of Starship, they need something on the order of 10x more launches, and each time they throw away the second stage.

Furthermore the Starlink constellation 'only' has 3,000 satellites so far, but SpaceX has sought approval for 30,000 more than the 12,000 already approved. To launch and maintain a constellation that huge with Falcon 9? It might be commercially viable, but I doubt it, they already have a very rapid launch cadence with Falcon 9 starlink launches and they evidently think they need much more.

No, we really don't have significant statistical evidence. Even for the rocket family. And there is no need to look for the reference you can't find since this is all verifiable from public data.

The ULA workhorse was the Atlas 5. Over its whole lifetime, both under Lockheed and ULA, it had 95 launches of which one was a partial failure (declared success). Wikipedia is good enough for that.

The SpaceX workhorse is the Falcon 9. https://www.spacex.com/launches/ puts it at 176 launches with 1 failure and 1 partial failure. This is not counting one that blew up on the pad with the satellite aboard.

The measured failure rate of the Atlas 5 is 0/95. Of Falcon 9 is 1/176. Note that 0/95 < 1/176.

The measured "not perfect success" rate of the Atlas 5 is 1/95. Of Falcon 9 is 2/176 = 1/88. Note that 1/95 < 1/88.

Also anyone with basic statistical knowledge will immediately see that the sample size is sufficiently low that we cannot draw any conclusion at 95% confidence.

Now what Falcon 9 does have the crown on is the longest string of uninterrupted successes of any rocket family. They are now at 148. (Not counting, of course, the one that blew up on the launch pad with the satellite on board.) But not enough to be statistically strong evidence that they are best. Also ULA has over 150 launches across their rocket platforms without any being declared a failure - which exceeds SpaceX's current winning streak.

That said, I agree with you about ULA's advantages, and I agree that SpaceX is the future. But I'm not going to lie about the math. And the math says that there isn't significant statistical evidence that SpaceX is better on safety.

The statistical measure I was talking about does some more complex calculation and takes into account the different iteration of the rocket as well.

In low sample size cases such measures are a better predictor of the future.

Insurance rates also reflect this same reality, as SpaceX has industry leading safety rating.

If I can find this document again I will send it, but somehow I can't find the damn thing.

Falcon 9 flight 1 was delayed more than 10 times, for about 6 months, from it's initial NET launch date of Nov 29 2009 until the actual launch on June 4 2010.

https://en.wikipedia.org/wiki/Dragon_Spacecraft_Qualificatio...

That’s a great point, because it bellies the core strategic problems with SLS. When your rocket has already flown a lot, and getting it human-rated does not slow your rate of launches, you can afford to take your time not to kill anyone. The same plan is being put in place with Starship, which will fly a lot before they put people in it.

SLS is one and done. It’s the old Apollo and Shuttle strategy, with the post-Columbia and Challenger risk management. It’s a bad strategy.

The first 19 Falcon 9s were "one and done". But even after the first, those next 18 still had a fair few launch delays between them too.

Getting your launches right is always a challenge for any new rocket family, and delaying your early test flights when (not if) you see unexpected telemetry data during the countdown should be seen as normal while you're learning how your rocket works in the real world.

Even with Starship, where cratering the first few test article hops was considered pretty likely, they had a bunch of holds and scrubs during the countdowns to those launch attempts. I expect the next test flight not to have zero delays either.

And SLS is being cautious. Delays should be expected, and not seen as some terrible bit of mismanagement. "Go fever" is real, and is a bad thing.

https://en.wikipedia.org/wiki/Go_fever

When I said one-and-done, I meant that they're flying an empty one and the second one already has people in it. That's the losing strategy.

> There are many politicians who want to greatly reduce NASA's budget in favour of private ventures.

No actually they want to give NASA more money so NASA can buy what it needs from private ventures.

Never launching is also a risk and a mode of failure.

How many versions of Starship did Space X explode or crash?

They seem to have produced better results on less money — so if NASA can’t follow the strategy that works, why should we throw more into the money pit?

SpaceX want to produce the least rocket they need, NASA want 0 failures, so they'll always have too much rocket

Also SpaceX plan to make 1000 Starships, and NASA plans to make about 5 of theirs

Unfortunately NASA doesn't have the ability to even appear to fail explosively while SpaceX can afford to laugh about it. Any significant explosion would have the tech illiterate Congress breathing down their necks.

This. This right here. I absolutely hate seeing the two compared. They're doing two very different things with two very different goals.

What do you think is different about their goals?

SpaceX is a profit motivated company with large degrees of flexibility. They move fast and break ships. NASA is a research institution focused on scientific advancement and space research. They also are very rigid due to the way they receive funding. NASA cannot afford to move to a starship style system because the SLS contract was locked in.

All that scientific advancement to end up 50 years later with a Moon vehicle not much more different than Apollo.

Their goals are similar, NASA is just hamstrung by Congress and the general public. If NASA had the freedom to choose its direction they too would be working on a Starship style rocket by now, not attempting to pass off a dinosaur as modern.

Or they might have ran their original design of STS instead of one hamstrung and made hilariously expensive due to contract with DoD

> NASA is just hamstrung by Congress and the general public

I don't think you can assert that NASA hasn't or wouldn't fall to Pournelle's Iron Law of Bureaucracy. What makes NASA special to be immune to such dysfunction?

> Unfortunately NASA doesn't have the ability to even appear to fail explosively while SpaceX can afford to laugh about it.

It's the curse of waterfall budgeting. A democratic government with typical constraints wants to minimize spending and to have all spending fully accounted for in advance.

"I dunno, maybe it will blow up this time?" is not an acceptable line-item under either standard.

Therefore, budgets must plan for success and design failure out of the process. Minimizing risk maximizes process and cost, ironically sacrificing the very low-cost public policy goal at the feet of risk intolerance. Instead of fifty SLS launches, of which the first dozen or so might fail as learning experiences, we get five that all must go flawlessly.

True "risky" development is reserved for programs of true national urgency (e.g. the original defense focus of the space race) or for programs that can operate well outside the public's watchful eye.

>> why should we throw more into the money pit?

You don’t get to decide where the money goes at all. Senators do so they can boast about bringing jobs to their state in reelection campaigns.

blah, blah, blah, and after the first mishap with the new system… No amount of money will make the system perfect.

We’ve heard the same story for 50 years.

Cancel the project now. There should be a price for taking the wrong path.

There was an article from The Economist that was posted on Hacker News earlier today, arguing the SLS wasn't a good use of resources[1]. It got flagged and was removed, for some reason.

[1] https://news.ycombinator.com/item?id=32636643

Reply on news.ycombinator.com