Starship III: Risk

43 min readNov 25, 2024

We must be bold.

*Starship’s sixth test flight launches from Starbase, TX. Credit: SpaceX*

This is a continuation of a series I started last year. If you haven’t read Starship I and Starship II: A Reprise, I would recommend reading those for a sense of perspective.

Welcome back. It’s been a minute. Let’s recap?

We left off at May of last year, just after Starship’s first test flight. In the time since — Starship has completed five more largely successful flights, not the least of which marked by a successful “catch” recovery in October. The program has come from crawling — to walking — and now running to a flight 7 in the coming weeks. This is, by all means, a good thing.

The world around Starship has also changed. The Biden-Harris administration is coming to a close, as America chose to return Donald J. Trump to the helm. With him comes Elon — far more politically vocal than ever before, and now seemingly involved in Trump’s every step.

Starship has by all means become a political symbol. Trump speaks of Starship extensively in his rallies and speeches. Right-wing influencers use it as a symbol of American progress and dominance. To much of the left, it represents waste and excess — a grandiose vanity project in the face of inequality.

There is a lot to unpack here. With it comes the future of America’s space program.

Let’s circle back for a second. I left you last time with a few calls to action and predictions in Starship II. They boiled down to:

Blue Origin’s team must be selected as a second option for the Artemis HLS under Sustaining Lunar Development to allow for dissimilar redundancy.
Relying on Starship to be capable of carrying crew to the Moon is a risk due to extreme program complexity and should be reconsidered as the sole option.
The timelines for Artemis III and IV must be reviewed and re-baselined as soon as possible to set realistic expectations for the Artemis program.
Starship would need to conduct around 18 launches per lunar sortie — a number that is untenable for two crewed missions and a demo by 2028.

So how did we do?

Well, let’s start with Blue, which was selected for SLD within two weeks of that post, and are now developing their own lander to be ready for Artemis V, notionally targeted for 2029. Again, this is a good thing. But I do want to address the amount of weight I assigned to this.

Blue’s approach is not a perfect solution to Starship’s woes. It too is dependent on orbital refueling — a technology that remains untested. Blue has significantly less flight experience than SpaceX — which today is the de-facto launch provider of the United States. Just because we have both Blue and SpaceX competing does not mean we are guaranteed a functioning vehicle. It is, however, still better than nothing.

But I want to zoom out here. The thing about most of this — the refuelings, flight cadence, etc. — is that this is all technical risk — quantifiably in the physical realm.

The kicker for much of this lies in programmatic risk — an order up from technical. Starship’s successes over the last year or so have proven that from a technical perspective — yes, this will be challenging, but it’s not inherently impossible. 18 launches (a number that was months later confirmed by the NASA Advisory Council as “high teens” — you’re welcome) is quite difficult and does leave a lot of room for things to go wrong. But it isn’t existential.

Do you remember the quote I started that post with?

As time has gone on, I’ve reflected a lot on how Starship’s first flight has been recieved and what actually happened. To put it bluntly, it’s not good. Compared to Artemis I? Well, the Artemis hype train has been going ever since that launch. Yet it feels like most of the people I know want to forget Starship’s test flight ever happened. Or if they don’t, they want to remember the cool launch, the awesome photos we got, but not the implications of what actually happened.

Returning to this two years after Artemis I — it’s hard to feel the same way. SLS will likely not fly for at least another year — a number that is frankly untenable. Moreover, the excitement around Artemis, or really spaceflight at all — is slowing tremendously.

The golden age of American spaceflight may just be a gilded age.

gild·ed — adjective
/ˈɡildəd/
covered thinly with gold leaf
(alt.) given a specious or false brilliance
gilded age
a period in American history between the Reconstruction and Progressive eras marked by materialistic excess and corruption

That’s a big claim.

It is important to say that yes, the space program of the 2020s is still one of progress. You can’t look at SpaceX catching a Super Heavy booster out of thin air and not believe that at least a little. You can buy an honest-to-god ticket to space from multiple companies. NASA is still intent on its return to the Moon under Artemis. Every other factory in El Segundo is a startup making spaceships. If you’d talk to them, they’d tell you “hard tech” is cool now.

It is still hard, though. And the track record of the median American space startup over the last few years shows it far more than what CEOs would like to tell their investors.

This is not a problem by itself. A free market will see entrepreneurs try their hand at the impossible all the time and convince investors to pay for it. Sometimes this even works. However, American space policy since the fall of Constellation and the Shuttle program in 2011 has largely pivoted to relying on these commercial enterprises as much as possible instead of in-house state capacity. These moves — which encompass practically everything NASA does with the exception of SLS and Orion— were lauded for their purported cost savings and early successes with SpaceX.

And… so far, just SpaceX.

“For some time, [NASA] has been hoping that other new space companies would step up and similarly thrive like SpaceX in an environment of purely fixed-price contracts. To succeed over the coming decade in low-Earth orbit and on the Moon, the agency is counting on a new generation of companies, such as Axiom Space and Intuitive Machines, to take this next step.
But what happens if they don’t?” — Eric Berger, Ars Technica

The problem with many of the commercialized enterprises that the space policy of the Obama and first Trump administrations gave us is that beyond SpaceX, performance has been poor, yet beyond the occasional Inspector General report, these programs have come to dominate the American space sector without question.

To make matters worse — the little state capacity we do have left hasn’t been doing much better either. SLS and Orion are years delayed and billions over budget. On the uncrewed side, Mars Sample Return is stuck in a death spiral, and budget cuts have forced delays and cancellations for many, many missions.

What gives? How did we get here? And how do we get out?

Artemis — despite being the most successful attempt so far — is not NASA’s first try at rekindling its lunar exploration program.

Constellation was a program that roughly came to be with the 2005 NASA Authorization Act (2005 S. 1281), which directed NASA to “develop a sustained human presence on the Moon, including a robust precursor program to promote exploration, science, commerce and US preeminence in space, and as a stepping stone to future exploration of Mars and other destinations.” Supporting it was NASA administrator Michael Griffin, who set out to make the program “Apollo on steroids.”

The program called for an orderly end to the Space Shuttle program — already in its final days due to waning support post-Columbia. Replacing the Space Shuttle were two launch vehicles derived from the Shuttle — Ares I and V. Ares I had the task of lifting the Orion capsule to LEO, regardless of whether the mission was to the ISS or beyond. Ares V, an uncrewed super-heavy lift vehicle, would meet Orion in LEO and ferry it and a payload to the Moon or beyond.

Constellation saw just one major flight — Ares I-X, a demonstration of the single Space Shuttle booster-derived launch vehicle’s first stage. When Obama took office in 2009, the Office of Science and Technology Policy under his administration requested a review of the nation’s human spaceflight policy in the name of ensuring a “sustainable path to achieving its boldest aspirations in space.” The Review of United States Human Space Flight Plans Committee — better known as the Augustine Committee of 2009— would release its report on October 22 of that year, finding Constellation far too expensive, delayed, and underfunded.

When the Presidential Budget Request was released on February 1, 2010, Constellation’s budget — and the lofty goals it carried — were entirely struck out. Presidential Policy Directive-4, released June 29, 2010, simply stated NASA’s exploration goals as:

Set far-reaching exploration milestones. By 2025, begin crewed missions beyond the moon, including sending humans to an asteroid. By the mid-2030s, send humans to orbit Mars and return them safely to Earth.

NASA was already looking at a near-inevitable gap in American access to space, unavoidable without an extension of the Shuttle program beyond 2011. The agency instead now found itself on the ledge of a cliff it would have to jump.

What does a space program do when the infrastructure, teams, vehicles, jobs, and plans it has spent years building become useless overnight? What do the people — the thousands upon thousands of highly skilled, critical workers a space program needs — do when their job no longer has a reason to exist?

For the people — that question is easy. When Space Shuttle Atlantis’ commander Chris Ferguson radioed over the words “wheel stop” on July 21, 2011 — the words that echoed the end of the Space Shuttle — three thousand workers across the Shuttle program packed their bags. They would go home July 22, taking with them everything they learned about flying the world’s only operational reusable spacecraft.

But for the program — it’s a different story. If you are a program manager or an executive, you are left with the pieces of billions of dollars and decades of work, though with none of the people that made it happen. Sometimes, it may even be your fault. Nonetheless, the only option left beyond failure is to try to put the pieces into something new and make a phoenix rise from the ashes. It starts with asking what pieces you have left to begin with.

The Augustine Commission did not fully abandon NASA in its reccomendation to cancel Constellation. It laid out what it called the “Flexible Path” — a proposed option that built upon work done with the Commercial Resupply Services program started in 2008 and called for commercially developed crewed vehicles to conduct flights to the ISS starting in the late 2010s. It then advocated for a new super-heavy launch vehicle — either related to Ares V, the Space Shuttle, or a derivative of a commercial heavy vehicle — to begin crewed deep-space flights in the 2020s.

This is, for the most part, the timeline we ended up on.

To a certain group of policy advocates, the commercial spaceflight work begun under the CRS program was the answer not just for NASA’s immediate woes — but the future. The program had already narrowly saved nascent SpaceX from bankruptcy by providing it a financial lifeline. Two years after contract award, SpaceX launched its first demonstration — Falcon 9 Flight 2, better known as COTS Demo-1, on December 8, 2010 — flying the capsule that would come to be known as Dragon twice around Earth before a successful splashdown.

Dragon’s first flight offered a glimmer of hope. Could commercial enterprises — with a bit of help — take on the brunt of the risk and do it better, faster, and cheaper? To NASA’s new Associate Administrator, Lori Garver —previously Obama’s top space policy advisor and a key leader in the 2008 presidential transition team for NASA — this was a moment to go all in, but not without consequences.

The death of Constellation and the Shuttle program simultaneously was not simply catastrophic due to a loss of capability or direction. Much of the biggest backlash to the programs’ cancellations stemmed from the second-order effects of their existence. NASA had to grapple with its existence not just as a program of raw spaceflight progress — but a high-tech jobs program across the United States.

To win support from Congress, NASA’s programs had to be architected to benefit all 50 states. To some extent it was just plain politicking — but the jobs program was (and remains to be) wildly effective. While not the most cost or time-efficient method, NASA returned far more to the economies of any locality that contributed than was ever put in — a net positive. Commercial enterprises did not have nearly as strong as an incentive to decentralize their operations.

This pitted the 111th and 112th Congress — which favored Constellation heavily due to its traditional approach — right against Obama’s executive branch and its heavy advocacy for a commercialized space program.

2010 S.3729, the 2010 NASA Authorization Act passed by the 111th Congress, therefore became the genesis of the Space Launch System program and extended a lifeline to Constellation’s Orion. At the same time, it reiterated support for NASA’s nascent commercial programs:

(Sec. 302) Declares that it is U.S. policy that NASA develop a Space Launch System as a follow-on launch vehicle to the space shuttle that can access cis-lunar space and the regions of space beyond low-Earth orbit…
…for such System to be designed as a fully-integrated vehicle capable of carrying a total payload of 130 tons or more into low-Earth orbit in preparation for missions beyond low-Earth orbit.
(Sec. 303) Directs the Administrator to continue the development of a multi-purpose crew vehicle… for use with the Space Launch System. Requires the vehicle to continue advancing the development of the human safety features, designs, and systems in the Orion project.
(Sec. 401) Continues NASA’s support of the existing Commercial Orbital Transportation Services program which is aimed at enabling the commercial space industry in the development of a reliable means of launching cargo and supplies to the ISS.
(Sec. 402) Continues, and authorizes the expansion of the number of, participants and activities of the Commercial Crew Development (CCDEV) program in FY2011.
— NASA Authorization Act, 2010 (abbreviated excerpt for brevity)

The Act, which also included clauses relating to the preservation of Constellation and the Shuttle’s workforce and contracts, was intended to try to smooth the transition to the incoming SLS program. However, the executive branch sought to obstruct the program from coming online in time to retain any of these workforces. As STS-135 flew out and much of NASA’s critical Shuttle workforces stared down layoffs, a critical decision on the final configuration for SLS was publicly stalled, forcing nearly the entire team that had built the Space Shuttle out.

“The continued delays to the announcement are now causing numerous managers and workers — at least those remaining after the massive jobs losses shortly after Atlantis’ return — to question if the delay is based on politically-aligned tactics to kill the SLS…
In effect, those opposed to SLS — such as the architects of the FY2011 plan — only need to delay another month before they can cite the “difficulties and costs” of having to rehire workers to build a vehicle which could have been announced when the workforce was still in place.”
— Chris Bergin, NASASpaceFlight, July 27, 2011

SLS as a program persisted — but barely. The destruction of NASA’s workforce and infrastructure was so severe that the Michoud Assembly Facility — which had built everything from the first stages of the Saturn rockets to every Space Shuttle external tank — suffered such a total workforce loss that the empty facility was temporarily repurposed as a movie studio.

There wasn’t much of anything for SLS or Orion to do anyway. In Obama’s speech to the NASA workforce soon after Constellation’s cancellation in April of 2010, he’d openly dismiss the idea of a return to the Moon:

“Now, I understand that some believe that we should attempt a return to the surface of the Moon first, as previously planned. But I just have to say pretty bluntly here: We’ve been there before.”

Instead, Obama would opt to establish what would become the Asteroid Redirect Mission — a vaguely defined goal to send a crewed mission to an asteroid and return a small chunk in the late 2020s, in lieu of using the Moon as a stepping stone to Mars. To demonstrate the mission’s importance, NASA would also disband its Exploration Systems Mission Directorate — the top-level office at NASA responsible for beyond-Earth human spaceflight, and roll it into its Space Operations directorate to create the Human Exploration and Operations Mission Directorate. The move made sense — with the amount of actual planned beyond-Earth human exploration NASA aimed to do, there’d be no need for a separate top-level directorate.

With the complete lack of urgency now established, progress on Orion, SLS, and their related ground systems would move at a snail’s pace. One mission — Exploration Flight Test-1, would launch in late 2014 as a test of just the crew module of Orion.

It would be eight more years until Orion would take flight again.

The story from here — SpaceX’s rise to power — is what we all know.

CRS-1, the first operational Dragon mission, would launch just two years after COTS Demo-1 and re-establish American cargo capability to the ISS. Just two months later, in December of 2012, SpaceX would win two launch contracts from the U.S. Air Force intended to certify Falcon 9 to compete for missions under the Department of Defense’s stringent and prestigious Evolved Expendable Launch Vehicle launch acquisition contract.

By 2014, NASA would award industry giant Boeing — as well as up-start SpaceX — the long awaited Commercial Crew contract intended to close the gap in American human spaceflight capability. And by 2015 — Falcon 9 Flight 20’s first stage would make history by successfully returning to its launch site at Cape Canaveral — marking the beginning of the end of expendable launch. SpaceX quickly became unstoppable — and perhaps showed commercialization was the winning formula after all.

This was not without a lot of help, though. Many of SpaceX’s contracts were set up as public-private partnerships, with NASA offering heavy assistance both directly and indirectly via research, testing, and advisory capabilities. So too did the many highly experienced and skilled engineers that had left NASA, often involuntarily, after 2011 and the end of Shuttle. SpaceX hardware had even flown on the Shuttle — with Endeavour testing critical sensors for Dragon on STS-127 in 2009.

But while NASA had an unusually close-knit relationship with SpaceX — these were all still capabilities the agency once had on its own, especially some 30 years earlier in the early days of the Shuttle program, which saw the Space Shuttle launching as often as every other week, bringing crew and comsats up and down until it became almost normal.

Seven astronauts — the crew of STS-51L — would give their lives on Space Shuttle flight 25. NASA had rushed the mission, overlooking critical design flaws and launching on a day far too cold for the Shuttle, despite warnings. Ronald Reagan’s extremely conservative administration saw it as yet another failing of the state to do what private industry should do better. Reagan himself would sign National Security Decision Directive-254 into law in December of 1986 — barring the Space Shuttle from ever launching another commercial payload.

The aftershocks of that decision and the Challenger disaster as a whole would permanently stunt the Shuttle program and lead to the first Augustine commission — in 1990 — nearly entirely cancelling human spaceflight. A reusable launch vehicle would not routinely carry another commercial payload until Falcon 9’s first landing 30 years after Challenger.

Looking back — it’s clear NASA saw something valuable in outsourcing much of its capabilities to SpaceX. Should the Shuttle fail, it held extreme consequences — not the least of which due to the lives of the astronauts’ at stake with each flight. But failure would also generate tremendous negative sentiment towards the agency. When NASA failed, America failed. There was no room for iterative development — systems would have to be perfect on day one, a technical impossibility.

SpaceX showed a way out. Falcon 9’s many failures in its quest for routine reusability were hardly a sore spot — often, they were openly celebrated by the company. The entirely automated nature of Falcon meant lives were hardly ever on the line when things went wrong. And SpaceX would learn how to fail fast and fail forward — returning to the pad in record time with another attempt. By decoupling operational capabilities from technological risk — in other words, ensuring Falcon 9 was still a perfectly good launch vehicle if the landings failed — SpaceX found a way to make failures pay for themselves.

SpaceX also simply would not be what it is without the leadership of Elon Musk — who assigned SpaceX the existential goal of making life multiplanetary. Critically, doing so was not just a catchphrase for SpaceX — but a simple goal every individual working at SpaceX — and many of its supporters — wholeheartedly believed. The spirit of the NASA of old was well and truly alive within SpaceX.

The incredible momentum behind SpaceX’s ascension would come to a crest at the International Astronautical Congress of 2016 — where Musk took the stage to unveil SpaceX’s next big thing — the Interplanetary Transport System. ITS was a vision for a massive evolution from Falcon 9 — a giant, rapidly reusable first stage that would return directly to the launch pad coupled with a multi-purpose reusable spacecraft doubling as a second stage. While the spacecraft would have exhausted nearly all of its fuel to get to orbit — it could then later be refuelled by a series of tankers, enabling deep space transport on a scale orders of magnitude beyond anything done since.

To an America that hadn’t even returned crew to orbit from its own soil since the end of the Space Shuttle — the scale of ITS was entirely incomprehensible. But it — and SpaceX’s willingness to make the impossible possible — was a welcome change to a country whose government still found the most worthwhile goal for the world’s most advanced space program was visiting an asteroid in the ambitious timeframe of “eventually.”

As Obama’s second term drew to a close — the administration that had overseen the end of Constellation and the Space Shuttle, and with it the systematic dismantling of NASA’s exploration goals — would have to go. But so too would the administration that brought SpaceX to power. Would an incoming Democratic administration preserve the status quo? On the other hand, would a Republican administration — whose voter base in the South benefitted disproportionately from the work done at critical NASA centers in that region — boost SLS and Orion instead?

America voted — and on November 9, 2016, Hillary Clinton conceded the race to be 45th President to Donald J. Trump.

Presidential Policy Directive-4 of June 28, 2010 (National Space Policy), is amended as follows:
The paragraph beginning “Set far-reaching exploration milestones” is deleted and replaced with the following:
“Lead an innovative and sustainable program of exploration with commercial and international partners to enable human expansion across the solar system and to bring back to Earth new knowledge and opportunities. Beginning with missions beyond low-Earth orbit, the United States will lead the return of humans to the Moon for long-term exploration and utilization, followed by human missions to Mars and other destinations;”.
Space Policy Directive-1, Donald J. Trump, December 11, 2017

Space Policy Directive-1 marked the decisive end of the Obama era of space policy. It sent a clear message — that abandoning the goals of Constellation was a mistake, and that America still believed in those goals. But it also came with an acknowledgement of NASA’s current position — an agency that had increasingly become a client of SpaceX and the rest of the commercial sector to continue many of its spaceflight objectives — nevermind still dependent on Russia for crewed access to space for the coming years.

While NASA may have had a deep-space launch and crew capability in development, SLS and Orion were still half-baked and moving forward on a minimum viable yearly budget. They also were hardly being developed with any real urgency yet — nevermind by an agency that had for the most part institutionally forgot how to implement and operate a large-scale launch capability. However, they were still pieces to the puzzle that NASA had at their disposal.

SPD-1 came alongside other sweeping changes to the nation’s space program. Earlier in 2017, Trump had re-established the National Space Council, a cross-agency board that briefly existed from 1989 to 1993, with the goal of cohesively steering the nation’s space policy. But the first signals of Trump’s intent for space came with the Presidential Budget Request for FY2018.

The budget continued increased funding for both SLS and Orion, as well as NASA’s commercial enterprises. It ended the nebulous Asteroid Redirect Mission in favor of a Moon to Mars path that would later be enshrined under SPD-1. However, it also called for defunding, delays, and cancellations across the board for NASA’s Earth science missions, and the full deletion of NASA’s Office of Education.

NASA got extremely lucky. During the FY2018 budget cycle, the 115th Congress — which had previously come to a near standstill on the budget including a minor government shutdown — decided to lift nation-wide spending caps. NASA’s proposed funding increases for exploration passed — without the cancellations and delays the Trump administration had proposed to incorporate to even out the budget.

With funding secured, a slate of commercial exploration programs began development under NASA as the agency set its eyes on a 2028 lunar landing. A request for proposals for privately developed small cargo landers — the Commercial Lunar Payload Services program — went out in September 2018. Later that year, studies began for a crew-capable lunar lander program, also privately developed, under NextSTEP-2 Appendix E.

With this, it became clear SpaceX had changed NASA’s space strategy for good. Instead of taking end-to-end responsibility on its exploration programs, NASA could offload risk almost wholly to its contractors. It began using a new standard of contracting that embraced this— firm-fixed-price.

Firm, fixed price contracts are exactly what the name says on the tin. When a contractor submits a proposal against a request from NASA, they name their bid price. If the contractor ends up under-budget, they get to pocket the extra money as profit. If the project overruns — it’s the contractor’s to pay.

Furthermore, NASA issued many of these contracts as services contracts — similar to how launch services had been procured for many years. NASA was no longer owning and operating these spacecraft — but instead allowing private industry to. In this way, NASA saw itself as an “anchor customer” for each service it procured — services that, while directly relevant to NASA’s goals, would also be open to other companies and private entities’ use. In theory — should the private sector be interested enough, NASA could save a lot of money on development costs that would instead be paid for by investors and shared across several customers.

While this approach had already been highly effective with launch services and had seen early successes with SpaceX and Orbital Sciences’ performance on the Commercial Resupply Services contract, firm-fixed-price would see its first outstanding success in 2019. SpaceX’s Crew Dragon, a massively upgraded descendant of the capsule first flown in 2010 — completed Demo-1, its first uncrewed mission to the ISS. The end of America’s nearly decade-long human spaceflight gap was in sight.

Just over a year later, in the midst of a new global pandemic, NASA astronauts Bob Behnken and Doug Hurley climbed onto what would become Crew Dragon Endeavour. Demo-2 would launch on May 30, 2020, and capture the American flag the crew of STS-135 had left on the International Space Station 8 years and 10 months before. Critically — SpaceX would beat long-time industry giant Boeing to the punch, to which NASA had awarded billions more to develop the Starliner space capsule. Starliner would end up failing its first test flight due to several critical software errors.

Maybe it was time to bet on the underdogs.

Just one month before Demo-2, NASA would award the Human Landing System development contract to SpaceX, Blue Origin, and Dynetics — directing the three companies to spend the next 10 months developing the next American lunar lander. After the 10 month “base period,” NASA would select anywhere from zero to two proposals to be eligible for selection for Artemis missions.

One small catch.

When Space Policy Directive-1 was signed, NASA began planning for a 2028 deadline to return to the Moon. This was seen as the soonest politically and technologically viable date given the current funding and state of the space program, including private development. It also was the timeline established with the expectation of a sustainable return to the Moon — explicitly called out in SPD-1 to avoid another “flags and footprints” campaign like Apollo. NASA’s new program was a return to the Moon — to stay.

The 2028 deadline did not last long. On March 26, 2019, Vice President Mike Pence would call for it to be accelerated by four years to 2024 under the program NASA would quickly soon brand as Artemis. Named for Apollo’s sister — it carried the goal of the next man — and the first woman — on the Moon by the end of 2024.

This sudden imposed schedule crunch forced large-scale change across NASA’s programs and their requirements. It would also have to be passed onto the HLS competitors — who now faced a near-impossible deadline that would need major funding to be feasible.

The Trump administration — namely NASA administrator Jim Bridenstine — once again returned to Congress asking for a $3 billion increase in funding to HLS for the FY2021 budget. He proposed achieving this by cutting funding to missions such as the Nancy Grace Roman Space Telescope and PACE — missions NASA was actively committed to — seemingly with the expectation that the 116th Congress would be as kind as the previous and fund both.

Appropriators flatly rejected the request — returning with a fraction of NASA’s proposed budget for the program. They also weren’t happy with Bridenstine’s attempt to cheat funding by putting otherwise safe programs on the chopping block. “Going forward,” a report accompanying the Senate’s markup of NASA’s funding bill stated, “NASA should refrain from requesting only part of the funding it requires to accomplish all of its missions.”

Bridenstine would not have the time to fight back. The first Trump administration would come to a turbulent end as 2020 closed out — seeing Bridenstine replaced in the interim by Associate Administrator Steve Jurczyk — standard during presidential transfers.

The Biden administration’s pick for Administrator would be Sen. Bill Nelson — one of NASA’s most vocal supporters in Congress. Nelson had even flown on STS-61-B, the last successful Shuttle mission before the Challenger disaster, to drum up support for NASA.

Nelson would not be confirmed in time. Jurczyk would oversee the sole selection of Starship for the HLS contract — and still commit to the 2024 deadline. He would hand the post to Nelson two weeks later.

This, for the most part, brings us to today. Sen. Nelson — and the Biden administration as a whole — would largely embrace the NASA programs that were left to them by the Trump administration. Privatization had become the norm, manifesting in many different forms.

Most visible was the slew of startups that would attempt to come to market — born from a period of near-zero interest rates in the United States meant to stimulate the economy, especially throughout the pandemic. Several different launch companies — all hailed to be SpaceX 2.0 — tried their hand. NASA would openly encourage these companies under the new VADR (Venture-class Acquisition of Dedicated and Rideshare) launch contract — assigning them payloads considered risk-tolerant enough to fly on unproven vehicles.

Success was few and far between. Rocket Lab, an early entrant, would manage to begin operationally flying its Electron vehicle it first tested in 2018 — likely the largest success story of the smallsat boom.

Relativity — a company praised for constructing its rockets via a novel method of 3D printing 85% of the rocket by mass — would test its Terran 1 vehicle from Cape Canaveral once, setting extremely low expectations of hoping the rocket survived peak aerodynamic pressure. The rocket would make it to space — and then fail to ignite its second stage.

Terran 1 was cancelled quickly after in lieu of a much larger vehicle. Rumor is that the 3D printed construction Relativity was hailed for was so weak that tank assemblies were regularly cracking along layer lines. It was also so inefficient that the rocket — despite being advertised as capable of over a metric ton to LEO — could likely only carry a couple kilograms of useful payload in reality. Terran R — the company’s supposed pivot — adopted a much more traditional construction.

Virgin Orbit — under Richard Branson’s multinational brand — would also try its hand. It would see some early success, going public via special-purpose acquisition company (SPAC), a move that seemed to pull nearly infinite funding from eager public investment. Virgin Orbit would have the most abrupt fall of any competitor — failing a critical and widely publicized launch billed as the first from the United Kingdom via its 747-launched LauncherOne rocket. Just three months after the January 2023 launch, Virgin would file for bankruptcy and become nearly immediately defunct.

Next up — Firefly. The company has had five launches of its Alpha rocket so far in a 3-year span — most of which stretched the definition of success. Its first flight dramatically exploded over the Californian coast in late 2021. Flights 2 and 4 made it to an orbit — but a functionally useless one. Rumor states that Flight 3, a “tactical responsive” mission for the U.S. government, also reached a bad orbit. The company’s first decisive success was only in July of this year.

In that same three-year span, SpaceX launched north of 200 Falcon 9 rockets — and flew Starship, the single most powerful rocket ever built, six times in half that timeframe.

There is a truth in all of this that anyone serious about spaceflight must contend with and take to heart: SpaceX is decisively America’s space company.

The numbers are not kind to anyone else trying to claim second place. In just 2024 so far alone, of 137 launches carrying the American flag — 119 were SpaceX.

And of ~20,450 payloads launched to space throughout history, SpaceX’s Starlink satellites — built nearly end to end by the company — represent 7,337 of them — 36% of all payloads ever launched since the dawn of spaceflight in 1957. All of which in the last four years.

As for Commercial Crew — Dragon has flown 56 crews so far — taking over America’s human spaceflight capability entirely. Boeing’s Starliner — the spacecraft NASA paid more for to fill that same role — has so far only flown two astronauts.

Those two — Barry “Butch” Wilmore, and Sunita Williams — will be coming home on Dragon after a string of on-orbit faults led NASA leadership to lose confidence in the capsule’s safety.

But there is a second truth in this. The rise of SpaceX only fueled the large-scale privatization of NASA’s space programs which has been ongoing from the end of the Space Shuttle to today. SpaceX’s successes in every contract they’ve competed on have only fed into this feedback loop.

However, it really is just SpaceX.

Take for example the Commercial Lunar Payload Services contract — one of a slew of privatizing measures taken by the Trump administration throughout the initial Artemis program rush. The emphasis in this contract is on services.

NASA is essentially hoping to procure a “service” to deliver a payload from Earth to the Moon — no different than shipping a package from A to B on Earth. The contract’s terms follow this too — bids are under firm-fixed-price terms, and structured as indefinite delivery — indefinite quantity (IDIQ), meaning NASA may issue any amount of “task orders” to competitors for transporting lunar payloads.

This isn’t a new approach to spaceflight — the Commercial Space Launch Act, which enabled the full privatization of launch services away from the Space Shuttle, led to launch acquisition under several contracts on identical terms. However, by the time the Act was in effect, there were several launch vehicle families produced by private companies (even if owned and operated by the Government) that had reached a mature operational capability. Critically, the Government was not paying for development of a new vehicle — but individual flights with well established cost and schedule baselines.

This approach breaks down heavily when applied to new or bespoke programs. With firm fixed-price, the contractor bears the costs for any unpredicted risks that result in overruns — risks that are incredibly difficult to predict at the concept stages of a development program. At the same time, a contractor that hedges its bets and proposes the most realistic timeline or cost it can provide is likely to lose a competitive procurement — as established baselines to filter out unrealistically optimistic bids from companies looking to undercut simply don’t exist.

“In [Boeing Defense & Space], our performance on fixed-price development programs is simply not where it needs to be. We expect substantial new losses in BDS this quarter, driven by the work stoppage on commercial derivatives, continued program challenges and our decision to complete production on the 767 freighter. I will be providing additional oversight of this business and these programs.” — Kelly Ortberg, Boeing CEO

Starliner may be the canary in the coal mine of firm-fixed-price programs. Boeing has lost an estimated $1.85 billion to date on the program — one which was arguably set up for failure from the start, and played even worse by Boeing. SpaceX was given a significant head start due to Dragon’s substantial flight heritage under Commercial Resupply Services — one the company leveraged heavily towards Crew Dragon development to save cost and timeline.

In several instances, lessons learned from the Dragon program also failed to transfer to Starliner — one major incident including a type of wire harnessing tape (P213) used extensively on both capsules but removed from Dragon after DM-2 due to a discovered flammability risk. It took nearly 3 years for NASA to inform the Starliner team of the same.

Starliner also took months of delays to allow Dragon time at the ISS to complete now operational missions —as the capsule frequently used both IDA docking ports on the ISS after the earlier Dragon 1 (which berthed to separate CBM-type ports) stopped flying.

This type of glacial inertia kills programs. Two years between flight tests raises much of the same institutional faults that doing nothing at all causes. Engineers that are truly passionate about their work flock to programs that operate and execute swiftly — not those that are mired in failure a decade later.

This is a lesson Boeing has had to learn across all of its programs. After a high-profile string of failures with the 737 MAX airliner, as well as massive cost overruns and delays across its aircraft, spacecraft, and other defense systems, incoming CEO Kelly Ortberg has promised large change to the company.

Ortberg’s first move was undoing the large-scale outsourcing of its programs by buying Spirit Aerosystems. Spirit was once just Boeing’s Wichita plant — but was spun off at the dawn of the 787 program in the name of a restructuring that would offload financial risk from Boeing to its subcontractors — of which Spirit became one of many, building airframes for the 787 and 737.

346 dead and a serious near miss on Alaska 1282 later, Boeing agreed to buy back Spirit — at a $4.7 billion dollar loss.

Ortberg’s second move? He doesn’t seem very keen on Boeing’s space systems, making clear in his first earnings call that “Boeing is an airplane company.”

“We’re better off doing less and doing it better, than doing more and not doing it well,” added Ortberg. “Clearly, our core of commercial airplanes and defense systems are going to stay with the Boeing company for the long run, but there’s probably some things on the fringe there that we can be more efficient with, or that just distract us from our main goal here.”

If Boeing chose to exit space systems, Starliner may very well be the first to go — an up to $5 billion failure for NASA.

Boeing’s billions in losses on fixed-price programs have also largely dissuaded other “prime contractors” — large military industrial complex corporations like Lockheed, Northrop, and other institutions — from competing on novel space systems.

When looking at any of NASA’s large commercialized programs — whether it be CLPS, or the ISS replacement program under Commercial LEO Destinations, or any of the other auxiliary Artemis programs — very few if any bids come purely from established contractors not named SpaceX. At most, primes have chosen to team up with startups to allow those companies to bear the risk instead.

These startups also make aggressive proposals towards many of these programs — as a contract won under tough and likely loss-inducing terms can still be a massive boon for an upstart company attempting to get their foot in the door, further dissuading established companies from competing.

These contracts, once won, provide every incentive for a startup to take as many shortcuts on cost and schedule as possible to attempt to deliver in time. No example more fitting exists than Astrobotic’s Peregrine lander — the first of the CLPS missions. The mission failed under an hour from separating from its carrier rocket when a type of valve that was discovered to be unfit for use was knowingly kept on Peregrine due to concerns that work to replace it would stretch the timeline too far.

The valve in question controlled a tank of helium pressurant at several thousand psi to pressurize the lander’s fuel tanks. Upon commissioning, the valve failed open and subsequently ruptured a tank immediately.

NASA, however, had already issued a task order to Astrobotic for its next mission — to carry the VIPER rover to the surface of the Moon on Astrobotic’s Griffin lander. The work that would now be required to investigate and fully remediate the failure mode seen on Peregrine delayed VIPER to point where NASA could no longer contend with the cost overruns incurred from the delays — exacerbated by an extremely constrained budget forced by Republican austerity measures in the 118th Congress.

NASA notified the 118th Congress of its intent to cancel VIPER on July 17, 2024. However, the agency was still on the hook for the $320 million task order it had awarded Astrobotic — for which it would now only fly a mass simulator (fancy space word for paperweight) to the lunar surface. NASA also refused to subsidize other science payloads for the mission — instead directing prospective teams to buy space from Astrobotic. It then decided to essentially auction off the $505 million nearly flight-ready rover to the best proposal.

Added up — between the rover, Peregrine, and Griffin, NASA lost nearly a billion dollars between a lander that exploded on startup, a rover it’d never get to fly, and the paperweight it’d try to put on the Moon — all in the name of what was proposed as a services contract no different than asking FedEx to ship a package.

CLPS’ next mission, Intuitive Machines’ IM-1, also suffered major failures. The lander suffered several major navigation issues on its way to the lunar surface, leading to an off-target crash-landing that left the lander irrecoverably — and uselessly — tipped over.

When the program was audited by NASA’s Office of Inspector General, it found several faults in CLPS’ contracting method:

If there were to be another SpaceX, it wasn’t here.

The same tone rang true with yet another program — the Exploration Extravehicular Activties Services (xEVAs) contract. xEVAs was a direct privatization of the xEMU spacesuit project required for Artemis moon missions. xEMU fell into hot water around 2021 in large part due to the unrealistic 2024 deadline imposed previously by the Trump administration, forcing the agency to search for a way to cut costs.

xEVAs was born — best described as “spacesuits as a service.” Under a similar FFP IDIQ based contract to CLPS — task orders would be issued for use of spacesuits privately developed and owned by contractors. Two companies won contracts: Collins Aerospace — which led a team of contractors largely identical to those assisting NASA’s previous xEMU effort, and Axiom Space — a startup already competing under the Commercial LEO Destinations program that had never done any spacesuit work. Instead of a common path to flight, Collins was awarded a task order for ISS suits, while Axiom was entrusted with the lunar suits.

“With these awards, NASA and our partners will develop advanced, reliable spacesuits that allow humans to explore the cosmos unlike ever before. By partnering with industry, we are efficiently advancing the necessary technology to keep Americans on a path of successful discovery on the International Space Station and as we set our sights on exploring the lunar surface.”
— Vanessa Wyche, Director of NASA’s Johnson Space Center

It took just two years for the Collins team to bow out. According to SpaceNews, “industry sources said they believed that Collins had suffered delays and cost overruns and concluded that it was no longer feasible for the company to continue work on [xEVAs], particularly given the fixed-price nature of the contract.”

As for Axiom — the company was struggling to make payroll recently, strapped for cash after operating net-negative missions to the ISS alongside turbulent development of its replacement for the ISS. A real possibility emerges that Axiom, too, is forced to bow out — leaving NASA’s moon program without a spacesuit.

There is a lot more that could be covered here — but I hope the point is clear. Already, we are left with some clashing truths and harsh realities to navigate as we architect the space program of the 2020s and beyond.

NASA’s next step in the Artemis program is its Artemis II mission — currently forecast to happen about three years after Artemis I (late 2025). Stacking for Artemis II has just begun — while SLS has been ready, Orion faced significant delays due to time needed to understand excessive erosion seen on Artemis I’s heat shield.

A three year cadence between missions — even if they are between a first test flight and a first crewed flight — is untenable. Neither is the $4 billion life-cycle cost of each SLS mission. Yet it is also the price of flying crew on flight number two, unless you aim to make the same mistakes as the Space Shuttle.

Should Artemis II fly in late 2025, I expect some 10–15 Starships to have already flown by then — a number that puts the existence of the Space Launch System justifiably in question. These are questions an incoming Trump administration — much more influenced by Musk this time — can and will ask. But striking down SLS — and in turn destroying any state capacity NASA has left — is not the answer.

The way the space program stands today — there are two major players: NASA, and SpaceX, in a relationship that blurs the line between symbiotic and parasitic. NASA and the Department of Defense are either largely or wholly dependent on SpaceX — which controls a vast majority of all American mass-to-orbit. SpaceX, in turn, is dependent on the billions given to it by the American government to fuel its massive momentum in the industry.

With this comes the first harsh truth. There will not be a second SpaceX — at least not in the ways American space policy has attempted to court one over the last decade. These attempts have cost the American taxpayer billions of dollars — and force often program-ending budget cuts on the missions that surround them. To sell off our space program for pennies on the dollar in the hopes of cultivating a space economy full of SpaceX-like juggernauts is delusional. A majority of these programs that have gotten far enough to show results have been large-scale failures. We must first stem the bleeding.

It does mean keeping SLS. And it does mean finding as many uses for SLS as possible, for SLS is just one example of America’s complete lack of real state capacity across infrastructure projects. Similar pains are felt in other high-tech industries in America — whether it’s nuclear power, or modern high-speed railways, or any other technologically or sociologically complex project in which it seems timelines and costs have only gone up higher and higher with less to show for it.

Are we truly incapable of building a Moon rocket efficiently even though we did it 50 years ago? Or have we systematically deconstructed our own capability as a nation to do so?

There was a time when the Space Shuttle flew every other week. That time was 40 years ago now. SLS — a vehicle that is overall far simpler, and builds on the knowledge of 135 Space Shuttle flights — has no excuse to not match that, especially when the risk to personnel is far, far lower (or practically zero on uncrewed missions).

That said — we do have SpaceX. Falcon 9 flies on a practically daily basis. I do fully believe that if realized, Starship will do so as well. That should not be a shock to anyone — not when you plot the trajectory America’s space program could have followed had the Space Shuttle lived up to its initial cadence. The Shuttle’s pitfall was the crew whose lives it invariably put in jepoardy with each flight. Starship, as an uncrewed system, solves this.

The solution to the problem of cost — and risk — is through flight rate. This is expensive at first — a reality difficult to square with the reality of yearly Congressional funding cycles in any industry. It may be the one saving grace of private investment — which can conjure large sums of money in one go to fund intensive development cycles. Perhaps we could learn a thing or two from long-term planning. This is a field China excels in — and one in which we should watch their behavior.

China is the only power that comes close to paralleling SpaceX’s might. The Long March rockets’ cadence is not quite SpaceX’s near-daily rate — but has still scaled aggressively in response. China, too, is planning a Starlink-like constellation. It is also rapidly adapting its future plans in response to Starship — with modern renditions of the super-heavy-lift Long March 9 looking increasingly like a Starship clone.

This lends to an interesting point. China does have its own semi-private launch sector forming. But these startups act more as ways of offloading risk for research and development in the launch sector without being given massive responsibility, as the country still conducts all of its key launches under its state-controlled space programs. What would be considered controlled proprietary information is also culturally irrelevant — with information freely shared across China’s aerospace sector.

This is a key role NASA had within the American aerospace sector — and one that is quickly diminishing due to privatization. Private companies working with NASA will mark anything they touch as confidential and proprietary — massively reducing the potential for information exchange across the American aerospace industry. During contract negotiations, companies will also aim to prevent the government from having any rights to their work that may be in question — as each technology handed to the government is something the company — or a competitor — could use for profit.

Yet, most of these companies would not be capable of performing without the vast sums of research freely available from NASA centers’ work — much of which dates back to the 1960’s. A long list of NASA spinoffs — many of which have become as fundamental to day to day life as modern microprocessors — would not exist without NASA research and development either.

In our top level discussions of space programs — those concerned with cost and schedule — this is often missed. But it is as critical as ever. How many key technologies did we miss out on from cancelling Constellation? How much was developed with NASA dollars, but plausibly under “private expense,” that will be forgotten to time once those programs or companies fold?

The Artemis program is not currently a serious space program. It is “sustainable” — in that SLS, Orion, Gateway, and the rest of the NASA-operated programs under that umbrella — can operate on a practically insignificant budget. Yet in terms of launch capacity, well, Starship runs laps around NASA.

It is true that SLS went to the Moon and back before Starship made it to step two of its contractual obligations for HLS. But Starship shows a real and true promise for a multiplanetary future. Will there be Starships on Mars? I couldn’t tell you. But reducing the cost of mass to orbit — even if just LEO — is step one. Starship may not be capable of what it claims it can do now — but it has ample room for improvements that will be made.

These vehicles do not have to be pitted against each other. They also may be critical to getting NASA out of the pinch it is in — with the looming end of the ISS and no real replacement lined up. Will we cede this too — to a private entity that will have to haplessly search for profit in running what is one of America’s most critical national labs — if it can even build a replacement? China, for its part, is rapidly deploying its Tiangong space station.

America can turn its gilded age of spaceflight into a golden age. It can also avoid ceding its 50 years of hard-earned leadership in spaceflight to China — or anyone else. But we cannot afford to cancel any more projects. We cannot afford to keep selling off our space program in the name of chasing short-term cost benefits or cargo-culting our way into getting a second SpaceX.

And a second thought: Outsourcing and competing everything does not work. If we can’t learn from SpaceX’s own incredible success with vertical integration — or Boeing’s fatal failures with Spirit Aerosystems and the like — maybe we can learn from our own achievements under the highly centralized Apollo program.

With this, I believe the American space program should:

Immediately implement a super-heavy launch applications program under NASA. Subsidize both private and public development of payloads that take advantage of the tremendous capability afforded by SLS and Starship. Make both of these vehicles fly as often as possible.
Implement a plan to sunset the CLPS and xEVAs programs. Transition xEVAs back to a fully in-house project. Develop an in-house, low-cost lunar lander under JPL to avoid further staffing cuts and the loss of critical institutional knowledge within the Lab. Take advantage of high flight rates to rapidly develop, test, and iterate such a vehicle.
Begin planning and development of a true government-operated ISS successor as soon as possible. Take advantage of large-scale capabilities offered by access to super-heavy lift. Transition CLD from a sole ISS replacement to the option of procurement of modules for a new station, or lesser contracts to free-flying stations.
Issue a Commercial Cargo to Crew follow-on contract to CRS and Commercial Crew. Invite proposals to begin as cargo vehicles similar to Dragon that recieve the option for a conversion program if proven reliable. This may be a viable exit strategy for Boeing’s Starliner. It may also allow for new entrants to gain the ability to develop complex space systems without the overload of human-rating.
Push for further development and iteration for Starship to the extent that is feasible. SpaceX should not recieve a “get out of jail free” card for blatant offenses commited over development of the program — but the Government should continue to work closely with SpaceX to ensure compliance is achieved without excessive delay.
Strengthen the National Space Council and implement a joint applications board between the Space Force and NASA. Create more opportunities for national security and exploration missions to work hand in hand.

These are not an exhaustive list of reccomendations. Nor do I promise you that they will all work on their own — or that any of them would be considered. But I do earnestly believe them to be key to a multiplanetary future that works for all.

SpaceX has shown incredible power. Musk remains a critical character — key to the creation of SpaceX, but potentially key to its destruction. That is a risk we run. These types of enterprises always crumble from the top down.

I leave you with a story of a launch company I didn’t mention — the one I worked for. I did a lot of concept design for entirely new space systems — proposals under contracts very similar to the ones discussed for what could have become entirely new space programs. My job was not just to tell the story — but to write it.

Practically everything we did depended on one rocket working. We struggled to test fast — taking a year and a half between a failed first flight due to a design misstep — to a second flight that never flew at all after a critical fire on the launch pad. Faces came, and then they went.

I wrote a large part of the final public statement for that program — an acknowledgement of our failure. I spent my last couple weeks there pouring over everything we had left — trying to come up with ways to pull out of the dive at first, but then just knee-deep in the piles of rubble left from both flights, all held in the decrepit wastes of a previous launch company that went under.

I held a melted, charred chunk of aluminum — the remains of a rocket we’d all poured our hearts into. I thought about the work some of the smartest people I’d met in my life put into it — friends that worked at early SpaceX before coming here would say we must’ve had more talent concentrated under our roof than SpaceX did when they were under that very same roof. We just didn’t know how to use it.

I looked back to the statement I wrote — and to one of its final lines that I was surprised had made it in. “This is not the outcome we wanted,” it said, “and it eclipsed the incredible endurance and execution of our team that got Flight 2 just short of the finish line.”

What would come of all the painful work everyone had put in — knocked down over decisions they had no part in? What would come of all the concepts for the future my team had created — of programs we knew were critical for a rapidly accessible future in space?

On August 29, 2024, myself and many of my friends got the same call those many others did in this industry. We went home. Many of our friends followed us soon after. I do still think about the statement that marked that company’s official end — at least of the launch program.

“However, over the past few years, we’ve seen our ability to make a meaningful impact in the launch industry diminish. Take a look around. U.S. rockets fly every couple of days, with perfect success. It’s revolutionary.”

It’s easy to say “U.S. rockets.”

It’s a lot harder to say that it really is just SpaceX.

P.S.

Thanks for reading. This is the culmination of years of work to tell a story that’s still continuing to be told. I couldn’t tell you how many initial takes I made at this — only that 10,000 words later, it still doesn’t feel like enough. I’d loved to have covered uncrewed spaceflight, especially Mars Sample Return more — or taken some time to look into the future. Companies like Blue still show a chance to disrupt — but it’ll be an uphill climb.

On the topic of personal stories — some thoughts I couldn’t find space to fit into this: Space systems are nothing without the people behind them. This incoming administration does present serious challenges to a lot of those folks, myself included. We should strive for a multiplanetary future, for all. American leadership in space is only worth anything if it’s a promise to do so under justice and liberty for all.

The industry is going to see some turmoil in the next few months. I don’t think we’ll be the last to fall. Should your friends get the call — take care of them. It’s tough to see something you were deeply passionate for collapse.

But — should you get the call. The most important thing you can do for a multiplanetary future is to persist and keep trying. There is no dollar value equal to the hard-earned lessons from actually trying something. Failure hurts. It often feels personal.

There was a poster that I had helped steer the design of in our office. It was a FOD awareness poster — one that showed a giant bolt hanging over a rocket — and I found it weirdly emotional. It read: “Don’t let the small stuff keep us down.”

P.P.S.

For Nik, Sydney, Eva, Alex, Davide, Ashley, and everyone else I’m certain I’ll feel bad about missing here. Thank you.

And no, I don’t have a new role yet. You may see some more writing from me — or some other projects. You’re welcome to get in touch at hi (at) lavieohana.com if you’d like to work on things together. Until then — watch this space. The story’s just begun.

— Appendix —

There’s a lot that went into this article — and a lot of great sources that didn’t make the final cut. I’ve attempted to hyperlink as many of them as possible, but here’s the full list of those that did:

NASA Selects Blue Origin as Second Artemis Lunar Lander Provider — NASA

Marcia Smith, @SpcPlcyOnline — X

SpaceX just stomped the competition for a new contract — that’s not great — Ars Technica

S.1281–109th Congress (2005–2006): National Aeronautics and Space Administration Authorization Act of 2005 | Congress.gov | Library of Congress

NASA’s New Moon Plans: ‘Apollo on Steroids’ | Space

Review of U.S. Human Spaceflight Plans Committee — Final Report

National Space Policy of the United States of America — June 2010

S.3729–111th Congress (2009–2010): National Aeronautics and Space Administration Authorization Act of 2010 | Congress.gov | Library of Congress

Preliminary NASA plan shows Evolved SLS vehicle is 21 years away — NASASpaceFlight.com

MAF goes Hollywood as movie teams utilize the facility’s once-busy expanses — NASASpaceFlight.com

Remarks by the President on Space Exploration in the 21st Century | whitehouse.gov

NASA Creates Human Exploration And Operations Directorate — SpaceNews