▲ 309 points
back
362 comments
I’m kinda confused by the criticism of the industry and NASA not designing for Starship.
The author argues that NASA doesn’t understand Tempe game changer that Starship is, but I don’t think that’s accurate. I think the risk to NASA of assuming the benefits of Starship before it’s proven are just too high.
If NASA (or other industry players), assume Starship will deliver the benefits it promises, they to fundamentally alter their approach to space. But that leaves them with no backup plan in the event that Starship fails to materialize or deliver on its promises.
Whereas this other approach, of treating Starship like a much better vehicle but integrating it into its old processes, gives them an out if Starship fails in some way. Because they’re still using the “old” assumptions, they can fall back to other “old space” style providers.
I think once Starship has proven itself capable of deliver on its promises, you’ll see NASA changing its approach radically in the manner the author describes.
So, fundamentally, I think it’s not a lack of “understanding”, but a different calibration of “risk”. This also applies to many existing space companies, and I also think this gives opportunities for new companies that have a higher risk tolerance to take advantage.
I think the author is arguing for NASA to take a cautiously optimist stance on Starship, instead of assuming it won't work at all until it does.
For example, NASA could start allocating a small-ish budget (e.g. 50M$) to design a parallel mission to their Artemis-IV mission that is currently scheduled for 2026 (but will slip) and that should deliver the Gateway i-HAB module to Moon's orbit.
While 50M$ (excluding launch costs) for a mission like this would be lunacy today given the constraints on weight imposed by current rockets, with Starship that would be a sizeable budget to design a much bigger and more useful module.
If by some cut-off date (say 2024) Starship still hasn't proven itself, then scrap that mission, proceed with the original one and the wasted 50M$ will be a rounding error in Artemis-IV budget.
If Starship proves itself before then, scrap the original mission and go with the redesigned Artemis-IV based on Starship. Billions of dollars will be saved (SLS launch cost + at least 2 years worth of development), and we will have a Lunar Gateway 10 years ahead of what we would have had otherwise in terms of capabilities.
As someone who has worked on NASA contracts pretty much my whole career, you're quite right. Starship is still extremely unproven. If you look at the history of rockets, mayyyybe 1 in 5 that make it to an orbital launch debut (which is like 1 in 50 paper rockets) in turn make it to commercial rollout.
The author's claim that the potential benefits of Starship are not understood by NASA is cringeworthy naive. Everyone at NASA is rooting for SpaceX but they have been burned many times by the promises of the next shiny rocket: DC-X, VentureStar, NASP, Ares, SLS, hell even the shuttle despite its "success" only fulfilled a small fraction of the desired results.
Makes me wonder how old the author is.
NASA just announced that they want to spend $1B per launch for SLS once a year until 2050. For a capability less than what SpaceX might provide for $10M.
That's not sensible risk management, that's head in the sand denialism.
https://arstechnica.com/science/2021/10/nasa-wants-to-buy-sl...
In the days of steam powered factories, a central power plant was used to drive belts that would power various machines. More power intensive machines were sited near the central power plant to maximize efficiency and parts were moved to where the machines were.
Electrification allowed the replacement of belts with motors but machines were expensive so they were upgraded in place and electricity was thought of as just steam, but less messy and cheaper.
It took several decades into electrification before people fully wrapped their heads around: a) many small motors didn't cost more than one big motor and you should redesign your machines so that the motors were where they made sense and b) now that machines could be placed anywhere, you should design your factories so that the machines were in the path of the parts.
Sometimes, quantitative changes become qualitative changes and you need to start from a clean sheet because all of the existing received wisdom stemmed from obsolete assumptions. It takes a while for this process to happen because we don't fully know how all of our assumptions stack on top of each other and we resist the change because our power/identities rest on elements of the status quo.
Disruption theory is premised on upstarts recognizing opportunities that incumbents are structurally unable to exploit because they derive too much benefit from the old status quo.
They're happy enough to put all their eggs in the even more untried SLS basket.
I've read that the fundamental problem for NASA is that they are a political football. They have at most 8 years of one administration before the next comes in, and Congress can change hands several times even in that span. As such, the only way to create a program that won't be announced to great fanfare one year and cancelled the next, is to make it out of solid, Congressional grade pork. Which would be SLS. Since nobody benefits by cancelling it, it won't be cancelled. Even if it becomes an obsolete and ridiculous white elephant.
Latest post is more of a continuation of this one: https://caseyhandmer.wordpress.com/2021/03/04/sls-what-now/
I'm going to presume what author thought and quote him:
------
What does NASA do with teams that cost a few billion dollars a year to feed and whose main expertise is building rockets and space probes whose entire architectural philosophy is threatened by current launch costs, let alone the order of magnitude improvement that’s in the pipeline?
The Perseverance Mars rover cost $2.4 billion, which works out to a few thousand salaries for just under a decade. Thousands of people are needed to build this rover because landing stuff on Mars is so hard that subsystem masses must be tracked to a tenth of a gram, on a system that weighs a tonne. The whole thing is meticulously handcrafted from custom silicon, PCBs, titanium tubes, motors, cameras, and other awe-inspiring instruments. Starship will be able to land 100 of them per flight. Now what? How can NASA feed a team that makes one feather light rover per decade for a billion dollars if the demand just jumped by a factor of a thousand and the unit cost fell by the same amount? Set up a production line? Work out how to make them with a team of ten? Build one every two weeks?
In short, in a world where SLS’s ongoing failure is justified and/or ignored while Starship races towards transformational capabilities, NASA needs to think very deeply about its place in a human spaceflight program that appears poised to proceed without NASA at its center.
-----
So the point is it seems to be a question - what is NASA planning to do AT ALL in the world where Starship is flying?
> I think the risk to NASA of assuming the benefits of Starship before it’s proven are just too high.
I think the risk of not doing so is to high.
And NASA is already relaying on it. They have had all the data and they selected it for Human Landing System. They knew what it required and they were comfortable with it.
> If NASA (or other industry players), assume Starship will deliver the benefits it promises
It doesn't have to deliver on everything to be worth designing for it.
> But that leaves them with no backup plan
Going back what they always did before is the backup plan.
How come wasting 20 billion on SLS is totally reasonable, but risking a few billion on starting with designing for a rocket that doesn't exist isn't.
> I think once Starship has proven itself capable of deliver on its promises, you’ll see NASA changing its approach radically in the manner the author describes.
But that will leave a years to decade long gap of actually profiting from it.
> but a different calibration of “risk”
Yes an incredibly conservative one that would have meant the US could never go to the moon.
I'm not sure that NASA designs much. JPL designs and builds most of the probes launched by NASA. JPL is an anomaly in the Federal Government where it gets good results. It would be interesting to look into how JPL does so well relative to other agencies and contractors such as Boeing (which is a national embarrassment).
The author trots out the "$10,000 a pound" launch cost which is really indicative of the overall quality of the article.
For non-Space Wonks, the "$10,000/lb" to orbit number was the rough cost of launching some payload on the Space Shuttle to orbit. Costs for the Space Shuttle were basically NASA's entire manned spaceflight budget divided by the number of Shuttle launches performed in a year. It's not just a naive number it's wholly misleading.
No/few commercial space launch approach $10k/lb. It's a meaningless number for any discussion of launch costs since the Space Shuttle is retired and was not a commercial launch vehicle anyways. Even the most expensive ULA rockets aren't even close to that price.
If you're going to talk about space launch capabilities and bring up stupid numbers it impacts your credibility.
I have to agree with you here. Given NASA's position it has to wait and see how successful Starship will be. NASA cannot afford to lose ground financially, politically, or publicly if they bet on Starship only for there to be issues with it down the line. They're not the owners of this tech which gives them time and flexibility to wait for the right moment to capitalize on it.
I can only assume they are internally very excited at the prospect of what Starship can be to them in the future but that can't be fully expressed if something out of their control has not been fully proven. They will jump on this as soon as the risk is outweighed by the benefit which, as the author argues, is in the process of happening.
I would buy your argument if they weren’t assuming the benefits of SLS before it’s proven. Why give SLS, but not Starship, the benefit of the doubt?
In other words, NASA doesn't want to move fast and break things.
>I think once Starship has proven itself capable of deliver on its promises, you’ll see NASA changing its approach radically in the manner the author describes.
Like how SLS has proven itself?
I think part of the point that the author misses is that while designing a spacecraft for a Starship launch does remove a lot of the constraints there is not the infrastructure to take advantage of that (yet).
There have been some preliminary studies on spacecraft that could take advantage of a Starship launch capability, but we currently can't transport a spacecraft that large over the road. It would most likely have to be transported by ship from the manufacturing facility to the launch site. Additionally, new test facilities like thermal vacuum chambers, acoustic cells and anechoic chambers would likely have to be created to handle the large spacecraft that a Starship could enable. And NASA is not going to make that investment until there is a higher probability of the Starship succeeding.
Those in the industry are pretty cautious about Elon's promises. Back in 2011, SpaceX advertised that the first Falcon heavy flight was going to be in 2013 [0], but the demo flight didn't occur until 2018, with the first launch of an actual payload in 2019.
[0] https://www.reuters.com/article/space-business-rocket-idUSN0...
>I think the risk to NASA of assuming the benefits of Starship before it’s proven are just too high.
This seems to be a theme of many of Elon Musk's ideas: assume they will work and talk about what a difference it will make. Just look at the media coverage surrounding the Tesla Robot.
So, this stuff is cool and I'm glad it's happening, but there's a ton of capability to prove before any sort of external organizational budget should be allocated to the Starship project. It's still a rough prototype.
Yeah if starship gets to orbit and lands both stages, NASA will start talking about it more. At the moment brief static fire tests still knock off multiple heat tiles. Still a long way from knowing it will be viable for humans to fly on.
Spacex is building the transcontinental railroads of the solar system. There is mind boggling upside for them. I'm kinda disappointed Blue Origin seems to have fallen in a rut. I'm sure Musk will be happy to provide transportation for Bezos space colony construction projects!
https://en.wikipedia.org/wiki/Railroad_Tycoon
Edit My point isn't to disparage Bezos, but to highlight the fact that anyone with grand aspirations in orbit/solar system will have to go through Musk and Spacex until a competitor does appear. RE space colonies, I believe the natural sequence of events for space tourism is: sub orbital, oribital, long term orbital (a small space "hotel") culminating in orbital habitat ala O'Neil Cylinders. Cheap kg to orbit enables all of that.
Transcontinental railroads provided routes between already populated areas, with viable trade between them.
Solar system beyond Earth is more like a desert, possibly rich with resources, but nearly no resource is precious enough to be worth carrying by a starship.
The incentives for space expansion past LEO and GEO will not be economical.
Yeah but watching Bezos turn into a household joke is worth it.
Do we know the Chinese part of the story? I feel like the western media hardly report anything on them while e.g. Wikipedia suggests that there are new developments happening all the time:
https://en.wikipedia.org/wiki/China_Manned_Space_Program
How significant are those in practice?
Can we just build private space, lunar, and martian colonies? They could be financed by Tesla shareholders.
The colonisation of the world by Europeans started with government investment, but eventually became self-sustaining.
If these offworld colonies are free of political correctness, crime, taxes, welfare systems, and disease, and contain only beautiful, healthy, and intelligent people (via careful selection, including genetic screening and IQ tests, of permitted colonists), they could draw the best of humanity to them.
>Starship is intended to be able to transport a million tonnes of cargo to the surface of Mars in just ten launch windows.
So at 100tons per starship sent to Mars that's 10,000 trips or 1000 per launch window. But the current plans call for something like 16 starship launches just for refuelling the one starship for the lunar mission. Suppose we can do the Mars trip with the same number of flights that's 170,000 launches to transport that 100million tons? And this guy is mad that nasa is not assuming that this is achievable?
Blue Origin said it would be 16, Musk has claimed it won't be more than 8.
https://wccftech.com/musk-rejects-blue-origins-claim-of-16-s...
It ends up being ~100,000 launches over 10 launch windows. Elon has talked about building 1000 starships (presumably you would need less boosters)
https://www.tesmanian.com/blogs/tesmanian-blog/1000-starship...
Anyway, it's absolutely the case that transferring a million tons of cargo will require a new way of thinking about the scale of launches.I would also not assume that the refueling launches have to all take place within the window, that's a bad assumption to start with. You would start your refueling launches well ahead of the window.
added point - I do think the article overlooks that while Starship will change things, the timing is likely to be slow enough that NASA will be able to adapt. They do have ridiculous timelines on some of their projects, but just like everything else, they will just rewrite those timelines if new solutions present themselves.
By the time you're sending a thousand Starships in one go, the fuel's probably coming from an orbital refinery processing chunks of captured comets, not Earth launches.
Or they've been replaced with some big Aldrin cyclers built using Starships.
"Mars trip with the same number of flights that's 170,000 launches to transport that 100million tons? And this guy is mad"
I think the idea that Starship will get us cities on mars is pure fantasyland. You could have a decent base on the moon, and maybe a visit to Mars, but if you want million tons to Mars, you have to go beyong chemical rockets, build a mars transit vehicle, etc.
It makes no sence to forgo those technologies if you are moving billions of tons.
I think he's handwaving a bt on how easy "space tractors" are going to be. Caterpillar etc are as fully invested int heir corporate structures and current practices as Kodak and all the other dead companies he mentions were; just because they could adapt things with "vacuum rated bearings" etc doesn't mean they will.
I fear we'll need some legal reforms before we can have "the remainder of human industry" keep up with SpaceX here. They've got Musk's "laws don't apply" card shielding them, or something, to explain how they've managed to innovate this far and this fast. The company that produces a cheap vacuum capable drone tractor won't have that, and will be rendered a wet stain by big competitors before they can get one of their products on a rocket. I expect.
NASA is hamstrung by Congress here. Congress mandated the SLS program. It's not meant to be competitive with SpaceX. SLS is a jobs program for key districts, no more, no less. It's welfare for Boeing.
I don't know if NASA personnel cannot plan for or even discuss SpaceX. It's probably not that overt. The powers there probably just realize it'd be career suicide to do so.
I completely agree that reducing LEO payload costs (in $/kg) by 2 or even 3 orders of magnitude will be game-changing.
Personally I'd like to see a viable competitor to SpaceX here because competition is good and it drives innovation. But it isn't Boeing, ULA or even Blue Origin. I actually think Blue Origin is so culturally broken that you'd almost be better off starting from scratch.
>> A space-adapted tractor needs better paint, a vacuum compatible hydraulic power source, vacuum-rated bearings, lubricants, wire insulation, and a redundant remote control sensor kit. I can see NASA partnering with industry to produce and test these parts, but that is no way to service the institutional overhead embodied by a team of hundreds of people toiling on a single mission for a decade.
If one function of NASA is to redistribute pork spending around the country, that seems like an even better way to do it. Rather than have contractors that strictly focus on NASA activities, use existing companies and fund projects to make space-capable versions of existing equipment. That also means the money can be shuffled around more as needed.
> We need a team of economists to rederive the relative elasticities of various design choices and boil them down to a new set of design heuristics for space system production oriented towards maximizing volume of production. Or, more generally, maximizing some robust utility function assuming saturation of Starship launch capacity.
This is not really what economists do. Elasticity is not something an economist can “derive” or predict in advance of market developments with any reliability. It can only be reliably observed.
That is the whole point of a free market. What is possible—changes in elasticity between factors—emerges from complex behavior. As it emerges, new efforts can compete to take advantage of it, thereby themselves creating new possibilities.
This is why people are not re-inventing everything around the Starship vision yet; because it hasn’t actually happened yet.
There are many projects and companies sitting on the garbage heap of history because of, basically, a branch prediction error. They guessed wrong on direction or timing. Even if Elon is right about Starship costs, but it takes significantly longer than expected to get there, that’s a huge risk to projects that set it today as a dependency.
A big part of the cost of space missions is not making things as light as possible, but developing the scientific instruments. More upmass doesn't always make designing sensitive scientific instruments easier.
If said scientific instruments need to go on a rover, mass still matters. More mass can limit rover mobility and at some point the rover requires more power than can be provided with RTGs or solar cells.
Progress at Starbase, SpaceX's base in Boca Chica, TX is done at an astonishing pace and very impressive.
Some of the stuff seen in the video series from August by Everyday Astronaut has changed already.
When I worked at a national laboratory, it was always evident that research priorities lagged funding priorities. No scientific program director would allocate significant resources into an area that wasn't either already funded or practically guaranteed to be funded. I don't how NASA and JPL operate organizationally, how their budgetary decisions are made, but I would guess that no movement toward re-orienting around Starship (or any new commercial technology) will happen until funding agencies dictate that it happen.
Ironically Related: https://news.ycombinator.com/item?id=29026781
The article is really interesting, but the stress the author expresses about the potential downfall of NASA seems misplaced. This is just creative destruction at work. If NASA can adapt to capitalize on the new paradigm, then it will, and there will be NASA logos everywhere in future space stuff. If not, then NASA will fade away, and some of us will wax nostalgic about how cool NASA seemed when we were kids but nobody will really care because humanity will have moved far past the limited space exploration that was possible to-date.
Like most old US government programs, NASA is fully of wasteful political constraints - for instance, facilities scattered around the country and to satisfy pork-barrel politics and get random senators to support the overall program. I think NASA is still a net good, and I'd love to see it adapt, but if it fails to adapt, that probably just means the institutional dysfunction between Congress and NASA administration was too great, and even though it will be a little sad, it will still be a net win for humanity to replace that dysfunction with a new wave of highly functional new players.
One way to summarize this article and relate it to software engineering: Before Starship, orbital launches and their payloads were like "snowflake" servers. Meticulously crafted one-offs. After Starship, launches and their payloads should become like containerized servers. (Starlink is already going this direction. No wonder, since Falcon is just a stop-gap and that system was designed for Starship)
I'm more worried about NASA's goals than I am their costing strategy.
SLS and Orion are taking up much of NASA's budget, and I don't see the payoff. What science are we going to add with another manned moon mission? Going to the moon and returning with stuff like the Genesis Rock was important, but that's been done. Likewise I'm really not sure what science we've gotten from the Space Shuttle and the ISS, outside of biology on long-term human space flight.
Whereas stuff like Voyager, Galileo, Cassini, Juno, Hubble have had _clear_ science gains, and cost much less than manned missions. Now we the Webb Telescope and Lucy, we should be doing much more stuff like that.
We tend to overestimate science and technology progress in short term, while underestimating it in long term. Of course, starship will fail on it's promises and artemis program will lag it's schedule and cost of kg to orbit will be 2000$ and not 100$ For next two decades that is.
But in 2045 it would be unimaginable to design new rocket not taking in account starship. Starship itself will be on it's 10-th iteration. Everyone: EU, China, Japan will have starship-inspired space launch vehicles. In-orbit refuelling will be finally figured out.
That is when slow progress will change thing forever. Not Artemis program, not today.
What am I missing here? Why is spacex valuation 100bn if this is true? I would think it would be 1tn easily in this kind of capital abundant environment. Is the risk that high? Is the market size being overestimated?
The very second that Starship gets to orbit it'll be politically easy for NASA to make a sudden shift
I’d love to know if any Universities or Research organizations are planning big observatories based around launching on Starship. What was once a rare NASA budget item could now be in the reach of many more researchers.
I would like to know what the Ozone impact these rockets will have is. It's my understanding that they burn quite a bit of the ozone up as they pass through; but my information could be wrong.
Damn right, it's not. It's the best shell prompt generator and few use it!
I think it was SN11 that stuck the landing. I was there.
Space is becoming a religion for atheists, with SpaceX serving the role of the church and Elon Musk in charge of it as the techno-utopian Pope.
No wonder really smart people like Bill Gates, Warren Buffett and George Soros are staying out of it.
The amount of mass that we have to bring up there to live and thrive and be happy is so enormous that Nobel astrophysicist Kip Thorne and Christopher Nolan had to essentially cheat their way into the epilogue of Interstellar. The hero of the movie finds the answer to the problem of how to lift humanity off the Earth into the Gargantua Black Hole, which by definition is the place where all the rules are off, and there is no knowledge of what really happens in there, so that Kip Thorne could, in good conscience, allow it in the script.
People who love space should work on mind uploading and laser transmission of the aforementioned upload. That's the only way we can manage to lift ourselves.
The entire space reporting has the problem that simultaneously there is an established genre of fanboy fawning about space x, which is kinda unconnected to space x actually starting to disrupt space.
Right now, it looks like one could save a few billions if Artemis is not green lit a decade ago. However, private launches were still rare at that time, so in absence of a, likely expensive, time machine that's just not an option.
So looking at the planned timelines, Artemis I will circle the moon next year, and a crewed landing is planned for 2024, while star ship does not have a planned launch to orbit yet.
Also on a personal note, I am utterly astonished that there will be people on the moon before the subway station down the street is operational.
> That is, Starship is a powerful logistical system that puts launch below the API.
Lol this sounds like pure BS
> The Artemis program to the Moon requires a Gateway and separate Human Landing System (HLS) because even the SLS doesn’t have enough lift capacity to be execute the mission on its own.
The Artemis programme requires a Gateway because senators decided it would be cool to have a gateway. AFAIK, anyone with actual technical expertise thinks it's stupid.
There's a bunch of logic in this piece i just can't follow. Like:
> How can the space industry saturate this increased launch supply?
It doesn't need to?
> What “lunar exploration objectives” can be “fulfilled” with such an architecture?
The only objective which has been pursued since the end of the Cold War, that of pumping billions of dollars into politically important states.
Who is this guy? Is he new?
I know that going into space is sexy and nerdy and cool and that it’s a difficult problem. What I don’t get is why so many people think about going to space and spend so much money on it when there are people in the US who drink water that catches on fire..and people elsewhere who don’t have access to water at all. Seems like we should work on the truly hard problems rather than the ones that are basically “just go up, really fast”