Am I the only one who fears that we’re entering a new era fraught with peril for lunar astronauts? It’s bad enough that NASA has essentially thrown over the whole manned lunar program to Musk, but now it appears that Musk has to do it on a shoestring budget. IT folks know very well what suffers when budgets are decimated: disciplined development methodologies are replaced by shoddy thrown-together crap, and rigorous testing methodologies are replaced by cursory ones, or no meaningful ones at all.
Manned space travel is an enterprise that, more than any other, is ruled by the Murphy principle that anything that can go wrong, absolutely will go wrong, and it will go wrong in spades, in more ways than you can possibly imagine. So NASA is apparently thinking, let’s give it all to an egotistical eccentric who may not be completely sane, and who is going to run this mega-project as a for-profit business; what can possibly go wrong?
Fortunately, we actually have a good point of comparison here: the Commercial Crew Program, i.e. the program that funded development of crewed vehicles from SpaceX and Boeing. SpaceX received $2.6B and Boeing $4.2B for essentially the same services (6 crewed flights).
And what happened? SpaceX succeeded, with multiple seamless flights under their belt, and Boeing fucked up. Boeing’s demonstration mission failed and they may not even fly before SpaceX has finished all of their missions. They’ve yet to fly even a second demonstration mission to prove they’ve fixed the issues.
In particular relevance to your (and my) industry, it was Boeing’s poor software development program that failed them. And more specifically, they failed to use a full hardware-in-the-loop development process. It would have uncovered some stupid bugs on their part. SpaceX has always had an end-to-end, hardware-in-the-loop process and as such their software seems to be top notch.
So it’s the experienced contractor, with the higher bid, that used shoddy development and testing methodologies here, and it cost them. Their process was far more important than what they charged NASA.
Minor correction here now that I’m reading the source selection release from NASA:
The CO thus opened price negotiations with SpaceX on April 2, 2021. As contemplated by the solicitation, the Government instructed SpaceX that it was permitted to change certain price and milestone-related aspects of its proposal (e.g., the Government requested a best and final price, as well as updated milestone payment phasing to align with NASA’s budget constraints), but was prohibited from changing content within its technical and management proposals or otherwise de-scoping its proposal in any capacity. SpaceX submitted a compliant and timely revised proposal by the due date of April 7, 2021. Although SpaceX’s revised proposal contained updated milestone payment phasing that fits within NASA’s current budget, SpaceX did not propose an overall price reduction. After I reviewed this revised proposal and consulted with the SEP Chairperson and CO, it was evident to me that it would not be in the Agency’s best interests to select one or more of the remaining offerors for the purpose of engaging with them in price negotiations. Following a final review of the offerors’ SEP reports and 4 SpaceX’s revised pricing proposal, I made final Option A selection and award determinations, as documented herein.
So they didn’t reduce the price; they only spread out the milestone payments so that NASA could afford to pay them on time. That’s more reasonable than simply cutting the price to fit (because you wonder where the slack came from). SpaceX has other funding sources and so has some wiggle room with regards to when the checks arrive.
By the way, I recommend reading the PDF I linked to above, as it talks a lot about risk. So for example, about the refueling risk:
Indeed, despite SpaceX’s concept of operations relying on a high number of launches, there is some flexibility in the timing of its required propellant tanker launches prior to the time-critical HLS Starship. This flexibility will allow NASA to time its crewed mission only after SpaceX has successfully achieved its complex propellant transfer activities and is ready to commence launch of its lunar lander. It is this flexibility that allays my concerns with regard to the admittedly riskier aspects of the first phase of SpaceX’s concept of operations. And, I further acknowledge that bounding more of the risk associated with these activities within the first phase of SpaceX’s mission actually enables the use of a single-element lander for the crewed portion of its mission.
So while they acknowledge that refueling is risky, it is in a phase of the mission that doesn’t endanger anything else. And it enables a single-element lander, which is less risky than the alternatives.
I downloaded the full PDF and will read it, but this paragraph jumped out at me. It sounds like they are saying that fhe SpaceX proposal either came in much lower than the others, or the others couldn’t delay payment like SpaceX could, and the difference was large enough that there was no point in further negotiations.
But this is turning out to be a strange program. What do you need SLS and Gateway for, if you can just refuel a Lunar Starship in LEO and fly directly to the moon and land? And if Lunar Starship works, it means SpaceX will have worked out in-space refueling and high launch cadences with Starship. In that case, SLS really looks like a white elephant. It can’t even put Orion into low Lunar orbit, which is what necessitated the Gateway in the first place.
It seems to me that choosing to use a Starship variant for the Lunar landing requires a rethink of the entire program architecture. The logic that’s been used to justify SLS over Starship is that we don’t know if Starship will actually work. But now they’ve tied the program to Starship anyway. If Starship works as promised, it’s a far more capable system than SLS at maybe 1/5 the price.
The answer is simple: SLS must exist and have a reason for existing. Everything else is secondary to that fact.
That said–and I’d have run some numbers to be sure–I think that Starship will only have enough delta-V to go from LEO to LLO and then to the surface and back on one full propellant load. It won’t be able to rendezvous with the Lunar Gateway or be able to bring astronauts back from there. Orion or some other shuttle-like craft is needed for that.
Some members of Congress already hate the selection; for instance:
However, as long as SLS has some reason for existence, NASA will probably not get too much blowback.
However, the SEP did identify two instances of proposed advance payments within Blue Origin’s proposal. Pursuant to section 5.2.5 of the BAA, proposals containing any advance payments are ineligible for a contract award. The solicitation’s advance payment prohibition applies to proposed CLIN payment amounts and, separately, to proposed milestone payment amounts within those CLINs. Blue Origin’s proposal is not
compliant with the latter of those two requirements.
It later notes that there would be opportunity later to fix the proposal, but it still sounds like a big screwup, especially as it was already more expensive.
You raise some good points here. My only counterpoints are to say that, first, Boeing isn’t particularly a high standard of comparison considering their incredible screw-ups recently, to the point that aviation media are running with articles like “What the hell is wrong with Boeing?”. Second, SpaceX is being funded to the extent of $2.9 billion. Now, I don’t fully understand what the scope of this contract is – I’m sure you know better than I – but it might be worth noting the the Apollo program had an estimated total cost of about $25.4 billion in 1973 dollars, roughly equivalent to $194 billion today. So Musk’s $3 billion is chump change in comparison, though I gather that this is mainly for the design of the spacecraft and not the mission costs. Still, it does smell like a seriously constrained budget for the magnitude of the task.
Yeah, but that’s why this pick is strange - it’s only going to work if the Starship architecture works, and if it does there is no need for Gateway or SLS.
Yes, that all seems accurate. But does it matter? Just send two Starships. One stays in LLO, the lunar version goes down to the Moon and back, then stays in lunar orbit and the astronauts transfer to the other Starship and fly home. If they need to refuel in LEO before landing, that’s a piece of cake becausevthe whole system depends on easy, cheap in-space refueling.
I’m pretty sure they can send tankers to refuel the Lunar Starship as well, so that it can be reused. They just need more of them. But then going to the surface and back only requires about 5 km/s delta-V, sonyoubwouldn’t need to fully refuel the thing.
Once this system is up and running and SpaceX is flying an SLS-sized rocket almost every day, SLS (and every other rocket system) will be antiquated. So this choice seems contradictory to the political goal of keeping SLS and its giant workforce in play.
I wonder what the new NASA administrator will have to say? Bill Nelson is an old-school porkmeister who has always been opposed to private space and was a huge booster of SLS. He only came around to COTS and CCS reluctantly. I wouldn’t be surprised to see him reverse this, assuming he is confirmed. I hope he’s not, as he’s a terrible choice.
Yeah, but government costs are whacked. In 2011, NASA did an estimate of the cost to build Falcon 9:
Key takeaway: They estimated that development would cost 3.977B, based on ‘NASA culture and environment’. They further estimated that the private sector, without NASA’s expensive culture and environment, could do it for 1.7B.
Actual cost to develop Falcon 9: $300 million dollars. Less than a tenth of what NASA said it would take for them to do it, and 1/5 of what they thought private industry could do.
Given that, $2.9B for Lunar Starship, which will use much of the hardware of regular Starship, sounds feasible to me.
I wonder if the decision was made now precisely because they wanted to get the commitment in place before Nelson comes in and tells them to use only NASA hardware or some stupid thing.
This contract is “just” for the lander. While that’s no small component, it’s only a part of the whole system, which includes SLS, Orion, and the Lunar Gateway. These latter components cost in the ballpark of $35B. I could talk your ear off about why SLS is a gigantic waste of money, but regardless, it’s going to be a component, and has used a more Apollo-style contracting process.
It’s only for two landers, the first an uncrewed demo. Apollo paid for seven landings, plus several more test flights.
Part of the Starship development costs are shared with SpaceX’s other projects. Starship was being developed anyway, so the contract here is to produce a lunar-capable version and two vehicles. That’s much less than paying for the entirety of development, as NASA had to do for the Saturn V.
Space is easier now. Not surprising since NASA and others blazed the trail. SpaceX hasn’t been shy that they built on the work of others.
Starship will be reusable. Plus it’s just a simpler architecture than the Saturn V. Starship is just two stages, both methane-based and using the same engines. The Saturn V was three stages, used both kerosene and hydrogen fuel, and different engines on all stages. It was just a very expensive vehicle overall.
Well, you certainly know that I agree with this. Just… don’t tell Congress or the Senate, because if they figure out that a fully-developed Starship makes SLS and the Lunar Gateway redundant, they might cancel the program.
If you send just a single Starship with LEO refueling, then you need an extra crew transport. That’s the “need” that SLS serves, and allows NASA to fund SpaceX without creating the apparent risk of SLS being made useless.
That said, our legislators aren’t completely dumb. For instance, Richard Shelby told NASA “No more fucking depots.” (i.e. orbital propellant depots) back in around 2011. He knew that the existence of refueling depots made the case for SLS weaker and increased the chance of it being canceled (thereby reducing the amount of pork flowing to Alabama). So he fought against them.
I hope this decision isn’t somehow reversed. But it does seem that the dominant risk is that it makes SLS, Orion, and the Lunar Gateway seem dumb. I mean, we’re going to see four astronauts living in a cramped mini-space-station transfer to Orion, and from there into Starship. Starship will be a cavernous luxury hotel in comparison to the others. Talking with the astronauts or even just watching the videos will raise the question of “why aren’t we using these for everything instead of that tiny shit?” Knowing Musk, he’ll install a trampoline or some other low-gee toy and the astronauts won’t be able to hide their glee.
It’s cool to have a harmless fantasy and I didn’t mean to crap on yours. But I have a feeling that in the very very very unlikely event this actually happens when the first people step out and realize they will never breathe or feel fresh air again, that they will never move freely again, that they will succumb to suffocation or radiation poisoning before long; their one thought will be “I have made a terrible mistake.”
There have always been explorers and people who veered far from the path most people take. All the arguments here could have been applied to many explorers: "Why do you want to travel across the ocean, a voyage that kills most people, to find new land when it would be easier to just find some land here?
In complexity theory, this tendency for some elements of a complex system to veer away from the norm is known as Lévy flight. In social complexity, it’s the Lévy flight foraging hypothesis:
When complex animal systems are in equilibrium and there is no stress on them, the animals tend to explore and move in a sort of Brownian motion. But when resources are scarce, you see a change in behavior - some of the animals engage in very long-range random walks away from the group in search of resources.
It seems that optimized search behaviour in complex systems requires the majority of searchers to engage in brownian-motion tyoe random walks, while a small percentage engage in longer, more speculative searches.
In the human case, the searching is the exploration of the ‘adjacent possible’ in Daniel Kahneman’s words - the area of potential resource discovery that has opened up due to other resources being discovered. Invention is the realm of the adjacent possible.
For example, SpaceX’s self-landing rockets could not exist until we had computers capable of controlling the gimballing of the engines. Once computers got small and powerful enough, rockets that could land on their tails became part of the adjacent possible, waiting for someone to discover them. Invention is all about exploring the adjacent possible.
Explorers and risk-takers may be the human equivalent of sharks that break off from a group and swim a thousand miles and return. Most people stay within the possible, looking for small improvements or simply harvesting resources they know exist. But some people take on flights of fancy or engage in high-risk behaviours, because complex human systems benefit from having a certain amount of Lévy flight. High risk, high reward.
It seems to me that early expeditions, including individuals leaving islands in small boats, were Lévy flights, and they probably increased and decreased in proportion to the dissatisfaction some people had with life where they were - either because resources were low, or because they became wealthy and bored…
So don’t criticize people willing to attempt to colonize Mars. It may sound stupid to you, but that would be because you didn’t become one of the elements of the system motivated to engage in Lévy flight.
I think the type of people who would actually go on such a mission have such different goals and ways of thinking that you simply don’t understand them, just like I will never understand someone willing to free-climb a mountain without safety lines. So I try to avoid judgeing others who are not like me. I just assume I don’t understand what makes them tick, and that’s okay, Diversity is good.
And as a social good, this type of way out there thinking and exploration tends to accrue benefits we didn’t see coming. The first Mars colony might fail, but the lessons learned will help other colonies succeed in the future, or lead to discoveries we didn’t even know to look for. Serendipity needs people who don’t think and act like everyone else.
I don’t think anyone is telling you that your personal tastes are bad. What I was saying earlier is that the concept of “hey, I’m on another world, millions of miles from Earth” gets old really really fast when there’s actually nothing to see.
I would be quite happy, though, to put on a VR headset – one of the good ones that has laser motion tracking – and be able to walk and look around in a detailed virtual environment reconstructed from lander imagery. That would be cool for about fifteen minutes. Then I could take off the headset and still have a life.
