Satellites Are Subject To Hacking By Quantum Computers

satellite telemetry Satellites log activities and transmit those data to Earth, but often rely on encryption systems vulnerable to hacking by quantum computers. (credit: L3Harris) Who will believe the space logs in 2040? Space governance and the quantum audit problem by Burak Oktenli Monday, August 10, 2026 Space operations have become exercises in relentless logging. Every thruster firing, every conjunction warning, every proximity operation, and every anti-satellite test generates a stream of telemetry that is meticulously recorded. Today, these logs are operational necessities, but decades from now they will become critical geopolitical evidence. If a state actor claims in 2037 that a commercial satellite intentionally interfered with its military asset, the resolution of that crisis will depend entirely on the historical data recorded today. But how will we know that those logs are authentic and unaltered? The answer is cryptographic signatures. Unfortunately, this is precisely where the foundation of space governance begins to crack. Directing engineers to use new algorithms for new satellites solves only half the problem. The unresolved crisis in space governance is what happens to the historical records. Currently, the cryptographic locks securing space records rely overwhelmingly on classical algorithms like the Elliptic Curve Digital Signature Algorithm (ECDSA). These algorithms are mathematically sound today, but they have a strictly limited shelf life. In August 2024, the National Institute of Standards and Technology (NIST) finalized its first post-quantum cryptographic standards, and its companion transition roadmap, NIST IR 8547, sets a hard deadline. Classical algorithms like ECDSA are slated to be “deprecated” by 2030 and entirely “disallowed” by 2035. Directing engineers to use new algorithms for new satellites solves only half the problem. The unresolved crisis in space governance is what happens to the historical records. “Harvest now, verify/forge later” and the silent vulnerability The defense community is rightly focused on the “harvest now, decrypt later” threat, where adversaries intercept encrypted communications today to decrypt them when quantum computing matures. However, space audit trails face a more insidious variation of this threat. We must confront the reality of “harvest now, forge later.” In this scenario, an adversary collects publicly broadcast or intercepted signed records today. When cryptographically relevant quantum computers arrive, the adversary uses them not to read secrets, but to break the underlying signature scheme. Once the signature is broken, the adversary can retroactively forge historical records. They could alter the telemetry of a 2026 orbital maneuver and present it in 2032 to claim that a specific action never occurred, or that a completely different, hostile maneuver took place. This is not science fiction. According to NIST’s own definitions, a “disallowed” signature scheme must be assumed vulnerable to forgery. Technically speaking, any space record signed with ECDSA will be considered vulnerable to retroactive forgery after 2035. The space domain is uniquely exposed to this vulnerability for three reasons. First, space records have exceptionally long lifespans. A satellite may remain in orbit for 15 to 20 years, and the legal and operational value of its records lasts even longer. Second, these records are the primary mechanism for attribution. Attribution is the cornerstone of space security, and it is exactly what an adversary would want to manipulate. Third, there is no opportunity for “re-measurement” in space. If the telemetry of a specific orbital event is compromised, the only proof of that event is lost forever. The challenge is not merely encrypting the future. It is securing the integrity of the past in a post-quantum world. Why simply changing the algorithm is not enough The seemingly obvious solution is to abandon ECDSA and mandate the immediate use of new, quantum-resistant signatures like ML-DSA. While necessary, this approach is structurally incomplete for several reasons. What space governance needs is not just a new algorithm, but a layered, crypto-agile architecture. Relying exclusively on a single new algorithm introduces a new single point of failure. Algorithms like ML-DSA rely on complex lattice mathematics. If future cryptanalysis breaks that specific mathematical foundation, the space industry will have repeated the exact mistake it is currently trying to fix. Leaning on one pillar is a dangerous architecture for multi-decade records. Furthermore, the transition period itself constitutes a massive vulnerability. Without a clear architecture dictating how and when legacy systems migrate to post-quantum standards, records generated during the transition remain exposed. Finally, historical records are already signed and sealed. You cannot simply go back and “re-sign” decades of telemetry with a new key without fundamentally breaking the chain of custody and the trust inherent in the original record. This is precisely why NIST IR 8547 supports hybrid approaches. During a transition, utilizing both classical and post-quantum algorithms simultaneously ensures that if one fails, the other maintains the integrity of the data. What space governance needs is not just a new algorithm, but a layered, crypto-agile architecture. A layered solution for deep defense in space records Securing the long-term integrity of space records requires a defense-in-depth approach. This aligns with the authority architectures I have previously argued for in these pages regarding on-orbit servicing and the AUTHREX Space Vehicle framework (see “Space autonomy needs an authority architecture before 2027”, The Space Review, June 22, 2026). To protect audit chains against quantum adversaries, we must divide the space record system into three distinct cryptographic layers. Layer 1 is the Entry Commitment. At the moment a record is generated, it must be sealed with two signatures simultaneously. One signature is classical (compatible with today’s infrastructure), and the other is post-quantum. During the transitional decade, both protect the record. An adversary in 2026 cannot break the classical signature because they lack a quantum computer, and by 2032, even if they possess quantum capabilities, the post-quantum signature remains intact. Layer 2 is the Chain Binding. Individual records must be linked together in a continuous hash chain, where every new record carries the digital fingerprint of the previous one. Crucially, this chain must explicitly tag which algorithm was used at every step. This concept, known as “algorithm agility,” ensures that even if algorithms change in the future, older records can still be verified using their original, securely recorded parameters. In the event of a conflict, an orbital collision, or a liability dispute, the deciding factor will be the reliable data logged at the moment of the event. Layer 3 is the Archival Anchor, which provides the long-horizon guarantee. At specific intervals, perhaps annually, a summary digest of the entire record chain is sealed with a stateless, hash-based signature like SLH-DSA. The critical feature of hash-based signatures is that they rely solely on the security of the hash function itself, rather than breakable mathematics like lattices or elliptic curves. The strategic result of this architecture is profound. In the absolute worst-case scenario where all lattice-based cryptography is broken, the assurance that a space record has not been altered reduces to the security of a single hash function. Hash functions are highly resistant to quantum attacks. Quantum computers only halve their effective strength; they do not shatter them entirely. Implementing this layered defense is remarkably inexpensive. My own testing of this prototype architecture demonstrates that sealing a record takes roughly a thousandth of a second, processing over 600 records per second on a single standard processor core. Generating a 30-year archival anchor requires only a few megabytes of storage and seconds of processing time. Long-horizon security is, operationally speaking, practically invisible. The governance gap and the 2030 horizon Current international space governance discussions within the UN Committee on the Peaceful Uses of Outer Space (COPUOS) focus heavily on debris mitigation, anti-satellite testing norms, and the “due regard” principles of Outer Space Treaty Article IX. However, these discussions are ignoring the long-term integrity of the records that make these norms enforceable. Article IX of the Outer Space Treaty requires states to conduct activities with due regard for the corresponding interests of others. This principle is entirely dependent on the existence of reliable, unalterable logs. If we accept that post-quantum adversaries can retroactively forge space records, the entire attribution mechanism of international space law collapses. The 2030 algorithm deprecation date is approaching rapidly. Cryptographic transitions in complex infrastructure typically take five to ten years. An agency or commercial operator that does not begin this migration by 2027 will find its operational margins severely depleted. A comprehensive breakdown of this timeline is available in this analysis of NIST IR 8547. To bridge this governance gap, I propose four immediate actions. First, state and commercial space operators must design their logging systems today to be crypto-agile. Second, the use of hash-based archival anchors must become a mandated standard for any long-lived space record. Third, COPUOS and related international norm-setting bodies must formally address record integrity through a long-horizon perspective. Fourth, the space industry must align its cryptographic migration strategy with a strict milestone plan tied directly to the NIST 2030 and 2035 deadlines. Conclusion In the space domain, power is increasingly tied to who holds the correct records. In the event of a conflict, an orbital collision, or a liability dispute, the deciding factor will be the reliable data logged at the moment of the event. The quantum era threatens to unlock these records, placing not only future operations but our current historical telemetry at risk. The 2030 and 2035 deadlines are not distant theoretical concepts. They are imminent operational realities. If the space community intends for its records to remain credible decades into the future, the foundational architecture must be built now. A record whose lock can be broken loses its value the moment the lock fails. The question is not whether space records will outlive their cryptographic locks. They already are designed to. The question is whether we will give them locks worth keeping. Burak Oktenli holds an MBA and is pursuing a Master of Professional Studies in Applied Intelligence at Georgetown University, where his research focuses on the governance of autonomous and AI-enabled military systems.

Cloud Cover And Satellites

optical vs SAR imagery Intel Ohio One construction monitoring. Passing clouds partially obstruct the optical imagery (left), whereas SAR radar (right) penetrates the cover to map exact structural geometry, metallic foundations, and material density. (credit: EOS Data Analytics) Seeing through the storm: SAR imagery’s evolution beyond defense applications by Kateryna Sergieieva Monday, August 10, 2026 Cloud cover shouldn’t get a vote in who survives, but it did for many years. Relief teams sat waiting for a clear satellite pass over a flooded valley or a collapsed building, and the ground kept shifting underneath them the whole time. Synthetic aperture radar (SAR) put an end to that wait. Agencies still filing SAR satellite imagery under “specialty backup” rather than “standard tool” are burning time they don’t have and risking lives they can’t get back. SAR is increasingly crucial for emergency managers, who finally can have eyes on the ground after a hurricane or earthquake. For decades, SAR was largely used by militaries, but that’s over. Commercial operators sell high-resolution SAR images straight to the public now, with growing demand for such imagery. The physics explains the appeal: radar pulses pass straight through smoke, storms, and darkness, conditions that blind an ordinary optical camera. So, for disaster responders, SAR is not optional anymore. Defense and intelligence: the traditional anchor Defense and intelligence remain SAR’s largest market. In December 2025, Germany signed a €1.7 billion deal with Iceye and Rheinmetall to build a sovereign SAR satellite network. Poland, the Netherlands, Portugal, and Finland have all signed similar deals. Japan’s IHI Corporation began receiving its first operational images in 2026 from two ICEYE-built satellites, the opening phase of a contract for four satellites with an option for 20 more. They are part of a sovereign Earth-observation push that Tokyo ties directly to monitoring vessel activity and protecting supply chains across the region. The reason for such popularity is rather simple: radar doesn’t stop working because the sky is overcast or the sun has set. Open-source analysts have taken full advantage of that to track aircraft carrier construction inside Chinese shipyards and watch troop buildups near contested borders, cross-referencing commercial SAR images with shipping data. Disaster response: seeing through the storm SAR is increasingly crucial for emergency managers, who finally can have eyes on the ground after a hurricane or earthquake. Clouds from storms can blind optical satellites but SAR doesn’t have that problem. It sees straight through the weather that grounds everything else. During major flood events, SAR imagery has become the tool of choice for mapping and directing rescue crews to where the water actually is, not where a map from six hours ago said it was. NASA and ISRO built their joint NISAR mission around this exact use case, designing it to hand decision-makers actionable SAR data for disaster response and agricultural monitoring. Greece didn’t wait for a joint mission: it deployed two of its own Iceye SAR satellites specifically to speed up response to floods, wildfires, and landslides. Infrastructure monitoring: protecting critical assets Pipelines, ports, and power plants can’t shut down for a sandstorm. Manual inspections and scattered ground sensors can’t keep pace with assets that run 24/7 in remote terrain, as they’re slow, costly, and often catch problems after they’ve already become expensive to fix. Point a satellite at the same target for a few extra seconds, and the radar starts picking up motion too small to see with the eye. This is where high-resolution SAR imagery benefits. Taskable radar satellites capture detail through darkness, cloud cover, and sandstorms alike. Capella Space’s constellation, for instance, can image the Port of Fujairah—one of the busiest bunkering hubs outside the Strait of Hormuz—at one-meter resolution, picking out terminal layout and vessel traffic at midnight as clearly as at noon. Pipelines get the same benefit. Paired with thermal and hyperspectral sensors, SAR catches the soil discoloration or gas plumes that signal a leak already underway. Detect those leaks early, and you can cut methane emissions from oil and gas operations by up to 70%. Maritime awareness: tracking the invisible fleet Today, some companies run illegal fishing operations, smuggling routes, or sanctions-evasion schemes, hiding their vessels from radar at sea. However, SAR satellite images can highlight “dark vessels” that stop transmitting signals to evade detection. That assists investigations by border enforcement agencies and sanctions monitors. New Zealand is pushing the concept further with low-cost floating radar reflectors designed to help space-based SAR spot people lost at sea: a real answer to the problem of searching 30 million square kilometers of ocean for a single boat or swimmer. In the Pacific Ocean, that can be the difference between a rescue and a recovery. The next frontier: micro-motion detection SAR used to freeze the world in a single frame: one image, one moment, nothing more. That’s changing. Point a satellite at the same target for a few extra seconds, and the radar starts picking up motion too small to see with the eye. That includes phenomena such as an engine idling, a bridge swaying slightly, or a gear turning inside a machine. Each of those tiny movements leaves its own signature in the radar phase data, which is what lets analysts detect them from orbit. Researchers proved it in Glasgow, picking up the idle vibration of a parked van and matching it to an 87-hertz reading from an accelerometer bolted to the vehicle. They’ve done the same with moving ships, extracting not just position but also behavioral clues: what the vessel is actually doing, not just where it sits. Micro-motion detection turns SAR imaging from a mapping tool into a diagnostic one, capable of identifying vehicle types or flagging failing infrastructure before it fails. Conclusion Every example highlights the main point: the ability to see through weather, darkness, and distance saves lives. A decade ago, these technologies were not widely accessible for civilians. But today we finally can track a carrier under construction, guide a flood rescue, and watch a pipeline through a sandstorm, all without approval from military or intelligence services. Iceye alone has launched 62 satellites since 2018. Capella and Umbra have built out fleets of their own. Together, they’ve taken a capability that once belonged to a handful of governments and put it up for anyone serious to order. The debate has shifted from whether SAR is useful to determining who else can’t afford to be without it. Kateryna Sergieieva has a Ph.D. in information technologies and 15 years of experience in remote sensing. She is a scientist responsible for developing technologies for satellite monitoring and surface feature change detection. She is an author of more than 60 scientific publications.

The Producers of Star City Talk About The Brilliant Series

https://www.youtube.com/watch?v=REQmXCsbKwE

My Hero Jareed Isaacman Will Take U.S. Astronauts Back to The Moon First

https://www.youtube.com/watch?v=ZTMZ8z4SS9A

Perserving The Legacy of the I.S.S.

ISS The International Space Station is scheduled for retirement, and a destructive reentry, as soon as 2030. (credit: NASA) Preserving the legacy of the ISS by Jeff Foust Monday, July 27 2026 Last week marked the 15th anniversary of the end of the final Space Shuttle mission, STS-135, and only now are the all the orbiters getting to the final homes. In June, the California Science Center previewed its revised display for the shuttle Endeavour in the new Samuel Oschin Shuttle Gallery. Inside, Endeavour is displayed as though it was on the launch pad, mounted on a external tank with two solid rocket boosters attached and a gantry beside it. “I would love to see this thing in a museum, I really would,” Keaton said. “It’s just not possible in a fiscally responsible, I suppose, or technologically responsible way.” “It is the fulfillment of a decades-long dream and will stand as an enduring source of inspiration for generations of scientists, engineers, and explorers,” Jeffrey Rudolph, president and CEO of the museum, said in a statement about the new center, set to open to the public in November. It also brings to mind another long-running program soon reaching its end: the International Space Station. NASA has maintained plans to retire the ISS at the end of 2030, although there have been some hints that it could be extended at least a couple years. A NASA authorization bill in the Senate, for example, would formally extend the station’s life to 2032. Even if the station continues to operate into the early 2030s, it’s clear it’s time to start thinking about the end of the ISS. That includes not just how to deal with the end of life of the station and how to transition to commercial successors but also what to preserve from the station, and how. Unlike the shuttle orbiters, which could be returned to Earth and put in museums much like aircraft, there is no way to bring the ISS back. NASA has also ruled out trying to preserve the station in orbit, perhaps at a higher altitude than it operates today. “We have a paper on this,” Jacob Keaton, acting ISS director at NASA headquarters, said during a half-day series of panels about preserving the ISS at the AIAA ASCEND conference in May. That 2024 paper looked at several alternatives to NASA’s plans to deorbit the station at the end of its life, including boosting it into a higher, longer-lived orbit. “It’s a lot of delta-V and the debris environment is pretty bad,” he said. That paper estimated that moving the ISS, currently in a orbit a little more than 400 kilometers high, to one between 640 and 680 kilometers, which would keep it in orbit for a century, would require between 18,900 and 22,300 kilograms of propellant. That higher orbit also has more debris, creating the risk of collisions that could cause “complete fragmentation” of the station and jeopardizing the safety of low Earth orbit. The same analysis, he said, also ruled out the partial disassembly of the station to either deorbit pieces of it individually or return them to Earth. “Any disassembly effort to safely disconnect and return individual components (such as modules) would face significant logistical and financial challenges,” the NASA paper noted. “I’ll just speak personally: I would love to see this thing in a museum, I really would,” he said. “It’s just not possible in a fiscally responsible, I suppose, or technologically responsible way, to preserve big chunks of this.” “Though large modules are not feasible for return, NASA has engaged with the Smithsonian National Air and Space Museum and other organizations to develop a preservation plan for some smaller items from the space station,” the NASA paper stated. “The Smithsonian has given us a list of desires,” said Ryan Landon, director of NASA’s ISS research integration office at the Johnson Space Center. She didn’t elaborate on the contents of that list. Cargo Dragon missions won’t be possible once the USDV is installed, Landon said. “So, our last cargo home will be about the middle of 2029.” One challenge will be finding the space to bring those items back. Today, only SpaceX’s cargo Dragon can return significant amounts of cargo from the ISS, with small amounts available on Crew Dragon and Soyuz. Other cargo spacecraft—Cygnus, HTV-X, and Progress—deliver cargo to the station but burn up on reentry. The pace of cargo missions will also slow as the station nears retirement. A key milestone will be the launch of the US Deorbit Vehicle (USDV), a modified Dragon that will dock to the station about 18 months before deorbit. It will use one of the two docking ports available to both commercial crew vehicles and cargo Dragon spacecraft. With the other port used by a Crew Dragon or Starliner spacecraft for the crew, there will be no space for a cargo Dragon. “Once those two vehicles are onboard,” she said of the USDV and commercial crew vehicle, “we will no longer have a vehicle that can return. So, our last cargo home will be about the middle of 2029.” That adds to the urgency for decisions on what to preserve from the ISS. However, historians and curators will be competing for space on those cargo missions with scientists seeking to return experiments as they maximize the station’s research in its final years, along with engineers who need to bring equipment back for maintenance or reuse. “That’ll be an interesting discussion,” she said. ISS A ten-person crew on the International Space Station shares a dinner together earlier this month. (credit: NASA) Researchers have made clear they want to use the ISS as long as they can. “We seek to continue to utilize the International Space Station right up until that last moment when Ryan turns the lights out. So, there will be drivers to maintain as much science capacity on those final return vehicles as we can,” said Michael Roberts, chief scientist of the ISS National Lab. “But, it’s not lost on anyone that there is tremendous scientific knowledge, as well as historical knowledge, that can be retained from the return of some of that instrumentation,” he added. There is, at least, awareness of the need to start planning for preserving items from the ISS. “The ISS is coming on my radar,” said Brian Odom, NASA’s chief historian. “It is the history of the agency’s human spaceflight for a very long time.” However, it’s not clear there is a formal process for determining what can be brought back and how. “There has to be a very concentrated, intentional effort by someone,” said Landon. That individual must be involved in discussions about cargo logistics and can thus advocate for bringing back items when space becomes available on a cargo return mission. That includes “talking to the crews about things to look for and having crews think about heritage while they’re up there,” she said. “The crew is amazing at packing those bags. There is always space somewhere.” “It will be a challenge” determining responsibility for preserving ISS heritage, Keaton said. “The space station program, as currently constituted, is focused on flying the space station.” Odom urged those working on the program “to be their own historian” and think about how to preserve records. “Your work could disappear forever.” “If protecting ISS heritage is something Americans or the US government, or certain portions thereof, care about, that has to be expressed though policy channels. It costs money, it costs time and effort not currently being given to those kinds of things,” noted Gabriel Swiney, director of the Office of Space Commerce’s policy, advocacy, and international division. “If people want to do that, there has to be some organized effort to do so.” Preserving ISS heritage goes beyond hardware on the station today that could be brought back. A big focus will be on preserving physical and electronic records of the station’s development and operations, as well as the oral histories of those involved in the station. That includes how crews actually used the station versus how it was designed to be used. “Archaeology is not necessarily about the distant past,” said Justin Walsh, a professor of art history, archaeology, and space studies at Chapman University. “It’s really about understanding human activity, understanding human culture, understanding adaptation to environments, and specifically doing that through the lens of material culture.” Walsh is the co-principal investigator on an ISS archaeology project that included documenting six areas on the station daily for two months to see how those areas were used by crews, including in “unexpected and unanticipated” ways, such as one astronaut who stored their toiletry kit on a wall in a high-traffic area of the station, illustrating the challenges of storing items on the ISS. “Those kinds of things tell a story that is really important for us to understand about what it means to live in space at our current level of technology.” That heritage can be documented in many ways. He pointed to the “ISS in Real Time” website that brings together photos, videos, transcripts of space-to-ground calls, and other documentation from nearly every day of ISS operations. He added he’s looking for other ways to document the station, including working with NASA to sample smells in different parts of the ISS to understand the chemicals that create it. “We would actually be able to know what the chemical composition of the atmosphere is to a degree that we could recreate smells for exhibits in science museums,” he said. (“Unfortunately, the first time I crossed over from STS-126, it even smelled like a submarine, which is not a nice thing to say about any place,” Stephen Bowen, a submarine officer who became a NASA astronaut, said of the station during another panel.) Despite that rich record, though, historians like Odom are worried they will lose valuable insights into the ISS because of ephemeral electronic records and fading memories. “We might know about Apollo at any minute than we’ll know about ISS in five years,” he said. “The way the bureaucracy created a record during Apollo was pretty substantial.” He urged those working on the program “to be their own historian” and think about how to preserve records of the decisions they made. “Don’t think that somebody else is going to do it for you,” he said. “Your work could disappear forever.” “I’m not big person on holding things,” Bowen said. “The biggest legacy, and what we should be preserving, is to continue to fly similar missions.” Regarding physical preservation of the station, some panelists got creative and suggested that private astronaut missions might be able to help, “taking down that they can,” said Todd Mosher, scholar in residence at the University of Colorado Engineering Management Program. Roberts called on historians to buy a SpaceX Dragon mission “and load it up with whatever you want.” (Perhaps overestimating the financial resources of historians.) Asked what one object from the ISS they would like to preserve, panelists offered answers ranging from the station’s cupola to the table that served as a communal dining space on the ISS to the station’s microbiome. Robert Pearlman, editor of collectSPACE, suggested something large enough that could be cut into small pieces and then mounted on cards or melted into medals “just so that everyone who remembers the space station, everyone who had a good memory of it over the last 25+ years, can have a piece of it if they want.” And what about Bowen, who spent half a year on the ISS in 2023 after three short-duration visits during the final years of the shuttle program? “I’m not big person on holding things,” he said. “The biggest legacy, and what we should be preserving, is to continue to fly similar missions.” Jeff Foust (jeff@thespacereview.com) is the editor and publisher of The Space Review, and a senior staff writer with SpaceNews. He also operates the Spacetoday.net web site. Views and opinions expressed in this article are those of the author alone.

The Nancy Grace Roman Space Telescope Flies On August 30, 2026

Roman and OSO Nancy Grace Roman with a scale model of OSO-1 in 1962 (credit: NASA) Orbiting Solar Observatories: Nancy Grace Roman and solar science from spinning spacecraft by Trevor Williams Monday, July 27, 2026 When NASA was created on October 1, 1958, a major focus of its work was to be on space science. The new agency absorbed the facilities of the National Advisory Committee for Aeronautics (NACA); incorporated two Army establishments, the Army Ballistic Missile Agency at Redstone Arsenal and the Jet Propulsion Laboratory; and took over the Vanguard rocket program from the Naval Research Laboratory. These covered the areas of aeronautics and launch vehicle design, but did not address spacecraft design and operations. A new center was therefore established to cover this area: this was initially called the Beltsville Space Center [1, p. 28], but was renamed Goddard Space Flight Center (GSFC) on May 1, 1959, in honor of the American rocketry pioneer Robert H. Goddard, who launched the first liquid fueled rocket on March 16, 1926. There were significant obstacles to women having careers in science during the post-war period. Despite this, a key member of the early NASA science staff was Nancy Grace Roman. GSFC was initially intended to carry out all aspects of spacecraft design, even human spaceflight. Later, when the wide scope of space work became apparent, Goddard became focused on robotic missions, in particular those involving science: this has remained a key thrust of the center ever since. Surprisingly, given the difficulties that faced women when trying to enter scientific and technical fields in the 1950s and 1960s, much of the groundwork for this work was put in place by a woman. Nancy Grace Roman There were significant obstacles to women having careers in science during the post-war period. Despite this, a key member of the early NASA science staff was Nancy Grace Roman. Roman, an astronomer who obtained a Ph.D. from the University of Chicago in 1949, later described her early experiences in science in a 2016 essay as follows [2]: When I was a girl, women were not supposed to be scientists. When I asked my high school guidance counselor for permission to take a second year of algebra, she sneered, “What lady would take mathematics instead of Latin?” The college environment was similar. If the dean of women could not dissuade a girl from majoring in science or engineering, she had nothing more to do with her. My first bit of encouragement came in my third year of college, when the physics department chairman said to me, “I usually try to talk women out of majoring in physics, but I think maybe you might make it.” In graduate school, it was clear that the faculty did not like educating women. This environment affected her employment choices: for instance: I realized that, as a woman, I had little chance of getting tenure in an astronomy research department. So, to stay in astronomical research, I changed my specialization and accepted a position in the Naval Research Laboratory (NRL). After working in radio astronomy at NRL, her move to NASA came about as follows, as she described in a 1980 interview [3, p. 41]: One day Harold Urey was giving a colloquium at NASA, which was of course very young in those days. And as you know, a good percentage of the science staff in NASA in the early days did come from NRL. The old Vanguard group and the NRL rocket group were taken over en masse by NASA. I came down to hear him, just because it sounded like an interesting lecture, and Jack Clark, who had been at NRL, but was now with NASA, came up to talk to me afterward, and said, “By the way, do you know anyone who would like to come and work for NASA and set up a program in space astronomy?” Well, as I say, I'd been staying very far away from anything having to do with the rocket group or the Vanguard group at NRL, but the idea of coming in with an absolutely clean slate to set up a program that I thought was likely to influence astronomy for 50 years was just a challenge that I couldn't turn down. Roman interpreted Clark’s question as an invitation to apply, and joined NASA in late February 1959 as Head of Observational Astronomy, becoming Chief of Astronomy in 1960. She was the first woman to hold an executive position at NASA, and was often referred to as the “Mother of Hubble” after advocating for the Hubble Space Telescope (HST) at NASA Headquarters for two decades. One of her key observations was that it was important for Hubble to be designed to support planetary science. To enable this, she had the vidicon imaging tubes that were originally baselined for Hubble, a 1950s technology, replaced by more modern charge coupled devices (CCDs) to gain improved performance [4, p. 248]. This led to the Wide Field/Planetary Camera that has been a key HST instrument. Summarizing, as Dr. Roman pointed out: “I am glad I ignored the many people who told me that I could not be an astronomer. I have had a wonderful career in a field that I love.”[2] Early NASA space science and the Orbiting Solar Observatory Program There was early interest in performing stellar astronomy using spacecraft. However, the technical challenges were formidable, which led to solar astronomy being focused on first and carried out by the Orbiting Solar Observatory (OSO), a program which Roman oversaw at NASA Headquarters from 1961 to 1963. As described by Roman [3, pp. 46-47]: The solar work went a little faster than the stellar in both rockets and satellites because pointing at the sun was an easier job. You could do it with two axis stabilization instead of three, and you had a bright object to lock onto. And as a result the pointing systems worked and worked satisfactorily much earlier than they did for the stars. I think that's why solar physics really became a valid observational technique earlier than stellar… I don’t think it was possible to do much in nonsolar astronomy, until you had the three axis pointing controls… Goddard was working very hard particularly to get pointing controls. Ames was also involved in this, at the time… Actually Ames probably had more pointing control and gyroscope experience at that time than Goddard did. There was a group there that was fairly active. In fact, they developed the first solar pointing system for rockets and were responsible for the solar pointers for quite a while, after Goddard was the main group for the other rockets. Pointing in fact dictated the overall OSO configuration, which was quite novel. Specifically, OSO-1 was the first “dual-spin” spacecraft, where one section of the satellite spins and another does not, being despun by a motor at the bearing between the sections. In the case of OSO, a nine-sided “wheel” section spun to stabilize the spacecraft and allow a subset of the scientific instruments to scan the sky, while the despun “sail” section kept a solar panel and other instruments pointing towards the Sun to within 1 arcminute. Following launch, the sail was spun up (to 30 RPM for the case of OSO-1) by cold gas thrusters that were mounted, along with their associated tanks, on arms that were deployed from the spacecraft by its rotation. OSO-1 was the first “dual-spin” spacecraft, where one section of the satellite spins and another does not, being despun by a motor at the bearing between the sections. A key property of dual-spin spacecraft is that any damping that is present in the despun section will stabilize the spin, regardless of the geometry of the vehicle: departure from a pure spin would create nutation, producing a nodding motion of the despun section, which can be dealt with by a nutation damper. The situation is quite different from that for a simple spinner satellite, where internal damping stabilizes the spin if the body is oblate (“tuna can” shape), but destabilizes it if the body is prolate (“soup can” shape). This lesson was learned after the prolate Explorer 1, the first American satellite, went into a flat spin shortly after launch.[5] Following that experience spinning spacecraft were designed to be oblate, despite the fact that packaging these into a launch vehicle shroud became more challenging. The early OSOs had an oblate configuration, but later ones used a more efficient prolate packaging, once the lessons of dual-spin stability had been fully absorbed. Roman and OSO OSO-3 under construction. (credit: BBRC) As shown in the following diagram for OSO-5,[6] the pointing control hardware for these spacecraft included not only a nutation damper but also thrusters to control the spin rate of the wheel section, as well as “pitch control” hardware (consisting of sensors, thrusters, and a magnetic coil) to keep the spin axis correctly perpendicular to the Sun line and the sail pointed at the Sun throughout the mission. Roman and OSO OSO-5 attitude control hardware. (credit: NASA) Key Orbiting Solar Observatory Program points The first spacecraft, OSO-1,[7] was launched on March 7, 1962; launch mass 458 pounds (208 kilograms), instrument mass 75 pounds (34 kilograms) on sail, 100 pounds (45 kilograms) on wheel. Spin rate 30 RPM. The final spacecraft, OSO-8, was launched on June 21, 1975; launch mass 2,350 pounds (1,066 kilograms). Spin rate 6 RPM. OSO-1 to 7 were built by Ball Brothers Research Corporation (BBRC); OSO-8 was developed by Hughes Space and Communications Company. The spacecraft were launched on Thor-Deltas into orbits of about 575 kilometers altitude and inclination 32.8 degrees. All launches were successful with the exception of OSO-C, which experienced a second stage failure and did not achieve orbit. Instruments included X-ray and gamma ray telescopes, coronagraphs and ultraviolet sensors. The study of solar flares was a focus for OSO. In addition, OSO-7 was equipped with a spinning occulting disk to produce an artificial eclipse and so enable studies of the corona.[8] Spacecraft design lifetimes were typically six months; these were usually greatly exceeded in practice. Specified pointing accuracy was 1 arcminute. The OSO program covered an entire 11-year solar cycle. The OSO spacecraft carried many international experiments: for example, solar X-ray instruments jointly developed by University College, London (UCL) and the University of Leicester were flown on OSO-4 and 5, and a Lyman alpha instrument from UCL flown on OSO-6. These built upon the experience that these investigators had gained by flying experiments on the Ariel 1 satellite.[9, p. 348; 10; 11; 12] Two significant unplanned incidents occurred during the OSO program: OSO-7 deployment problems resulting from second stage failure:[13] In the nominal OSO separation sequence, the sail section of the spacecraft is spun up to 40 RPM prior to the second stage burn in order to provide spin stabilization against tip-off torques. Then, after separation, the sail section is despun and the wheel spun up to its mission rate of 30 RPM. However, a vehicle malfunction during the OSO-7 second stage burn caused the stage/spacecraft stack to tumble, reaching a rate of 55 RPM. The resulting post-separation spacecraft attitude motion was quite complicated and required a great deal of non-standard commanding to correct. The whole recovery sequence took several orbits longer than nominal: the battery reached a depth of discharge of 75%, but recovery fortunately occurred in time to avoid power problems. The only hardware impacts that were observed were additional attitude control gas usage, although there remained more than enough to cover the mission, and the failure of one of the two spacecraft tape recorders, thought to have been caused by the high accelerations (up to about 8 g) experienced during the launch sequence. In addition, the failure of the second stage led to a final orbit with lower than planned perigee: reentry therefore came three years after launch, somewhat earlier than originally planned. Roman and OSO Intended vs actual launch sequences for OSO-7. (credit: BBRC) OSO-2 inadvertent ignition of solid rocket third stage during ground testing (an event referred to in a NASA report[14] as “The Disaster”): On April 14, 1964, the solid rocket third stage of the OSO-2 launch vehicle was in the Spin Test Facility at Cape Kennedy with the spacecraft mounted to it. It was covered with a polyethylene shroud as a dust protector: when this was adjusted a crackle was heard and the rocket ignited. As a result, three men were burned fatally, and a further 11 had non-life-threatening burns. The only positive to come from this tragedy was that an extensive study was performed of the various mechanisms that could have contributed to it: it was found that a polyethylene cover could in some circumstances charge the spacecraft to 15,000 volts, and that the accident had been caused by an electrostatic discharge through the rocket igniter squib. This study led to measures being identified that were taken on future missions to prevent any similar occurrences. The OSO-2 spacecraft was extensively damaged in this incident, but was rebuilt using flight parts, prototype parts, flight spares and new parts. This reconstituted spacecraft was launched on February 3, 1965 and operated successfully. Advanced Orbiting Solar Observatories and Skylab Soon after the launch of the first OSO spacecraft, planning began for a follow-on series of Advanced OSOs (AOSOs), also known as Helios.[15] Negotiations began in October 1963 [4, p. 249] between NASA Goddard and Republic Aviation Corp. for a series of four such spacecraft, with launches planned to start in 1966.[15, p. 27] This timing was selected in order to have the AOSOs operational during the next solar maximum in 1969, with operations continuing into 1971. Not only did the AOSO instruments need a new means for reaching space, but the Apollo Applications Program (AAP)—the Skylab space station—needed a high-quality science payload. The AOSOs were to have been larger than the OSOs, with better resolution and ten times the data storage: this would have allowed them to observe transients such as solar flares,[16, p. 69] although returning that much data would have been challenging. The pointing accuracy specified for AOSO was a considerable improvement over that of OSO: 5 arcseconds versus 60. Attitude control would have used a set of three orthogonal reaction wheels, with magnetic torque coils for desaturation; initial detumbling would have been performed using cold gas thrusters.[15, p. 36] Launch into Sun-synchronous orbit would have been performed by Thor-Agena, although a minor increase in launch vehicle performance would have been required. Given the increase in performance of AOSO relative to OSO, its development would have been expensive. Furthermore, the NASA science budget experienced cuts, leading to AOSO being cancelled in 1965.[17, p. 70] To keep open the option of perhaps being able to fly the AOSO instruments on some future mission, their development was kept alive after cancellation,[17, p.74] and they were made general-purpose.[17, p. 167] The main modification to their design that was considered was the use of film for imagery return, given the large amounts of data generated. Studies were carried out of the feasibility of mounting these solar telescopes in either an Apollo Service Module (SM) or Lunar Module (LM) flying in Earth orbit, with the crew returning the film to Earth. A drawback of these concepts, however, was that the mission durations could not exceed about two weeks. It was then recognized that not only did the AOSO instruments need a new means for reaching space, but the Apollo Applications Program (AAP)—the Skylab space station—needed a high-quality science payload. Flying the AOSO telescopes on Skylab solved both these problems: they were mounted in the Apollo Telescope Mount (ATM), which began as a derivative of the LM. Data was collected on film, with these canisters retrieved by astronauts during spacewalks and then returned to Earth in the Apollo Command Module. Roman and OSO Advanced Orbiting Solar Observatory. (credit: NASA) In addition, collaborative research was carried out between Skylab (launched in 1973) and the final OSO, OSO-8 (launched in 1975). Specifically, OSO-8 studied energy transfer between different solar layers,[18] with Skylab data on chromospheric structure used to target these OSO measurements. So, OSO and AOSO data were eventually used in tandem to improve the study of the Sun. Roman and OSO Skylab Apollo Telescope Mount. (credit: NASA) P78-1 Solwind: eventual anti-satellite test target Following the end of the OSO program, the DoD Space Test Program P78-1 Solwind spacecraft was launched on February 24, 1979, from Vandenberg Air Force Base. An Atlas F launch vehicle placed the 2,934-pound (1,331-kilogram) Solwind into a 500 -kilometer altitude Sun-synchronous orbit with an inclination of 97 degrees. The spacecraft was designed to observe space weather by examining the solar corona and the upper atmosphere of the Earth. It was built by Ball Brothers Research Corporation (BBRC) from the modified OSO-7 flight spare. Using the same approach that was taken for the OSO spacecraft, Solwind was a dual-spin spacecraft equipped with a passive nutation damper for spin stability. P78-1 carried seven scientific instruments, including the flight spare white light coronograph from OSO-7. Solwind was the first satellite to observe comets, notably nine Sun-grazing comets of the Kreutz group, which were imaged between August 1979 and August 1984. Roman and OSO OSO-7 and P78-1 Solwind spacecraft. (credits: NASA and USAF, resp.) By early 1985, Solwind had experienced significant degradation of its batteries, and the last of its three onboard recorders had failed. These factors conspired to render further operations challenging, so discussions started on terminating the mission. Meanwhile, there was a desire to test the ASM-135A anti-satellite weapon, then under development, against an orbiting spacecraft: such a test had not been carried out to date. The ASM-135A, developed by LTV Aerospace, involved a three-stage missile launched at 38,100 feet (11,600 meters) altitude from an F-15A aircraft in a zoom climb. The 30-pound (13.6-kilogram) spinning Miniature Homing Vehicle (MHV) third stage used line-of-sight guidance based on an infra-red seeker to impact the target at a closing rate of around 15,000 mph (6.70 km/s), performing a kinetic kill. Given that Nancy Grace Roman proposed an early exoplanet study concept in 1959 based on using a space telescope, associating her name with the new telescope is certainly very fitting. The target satellite for a test of this system would have to be under DoD control, and there was too little time to develop one from scratch, as a congressional ASAT ban was expected in the near future (it actually took effect in October 1985.) Since Solwind was operating in a degraded state, it became a leading candidate for use as the ASM-135A test target. This test was authorized by President Reagan on August 20, 1985, and carried out on September 13.[19] It generated 285 trackable pieces of debris; only eight remained in orbit as of January 1998, with the last piece reentering on May 9, 2004. Despite this successful demonstration, the ASM-135A program was terminated in 1988 because of technical problems and significant cost overruns. Roman and OSO ASM-135A anti-satellite weapon test launch. (credit: USAF) The continuing space science legacy of Nancy Grace Roman NASA is shortly to launch a major space observatory named in honor of Nancy Grace Roman. The Roman Space Telescope is an observatory sensitive to visible and near-infrared radiation, with resolution comparable to that of the Hubble Space Telescope (HST) but with a field of view roughly 100 times larger. The main instrument on Roman is the Wide Field Instrument (WFI), which will be used to study dark energy, the mysterious quantity that is believed to cause the acceleration of the expansion of the universe, as well as to observe exoplanets using the technique of gravitational microlensing. Roman and OSO Roman Space Telescope undergoing final testing at NASA Goddard. (credit: Seth Shulman) Roman makes use of a spare spacecraft primary mirror donated to NASA by the National Reconnaissance Office (NRO). It has a diameter of 7.9 feet (2.4 meters), the same as that of HST. Development of the spacecraft, originally called the Wide Field Infrared Survey Telescope (WFIRST) and renamed Roman in 2020, began in 2016, led by a team at NASA Goddard. Launch to a Sun-Earth L2 libration orbit, ahead of schedule and under budget, is planned for August. 30. The mission will continue the strong tradition of international involvement in Goddard science missions: organizations participating in RST are the European Space Agency, CNES in France, JAXA in Japan and the Max Planck Institute for Astronomy in Germany. Roman is also equipped with a coronograph instrument for direct imaging of exoplanets. This will be used as a technology demonstration for the Habitable Worlds Observatory. Given that Dr. Roman proposed an early exoplanet study concept in 1959[20] based on using a space telescope—in her design positioned on the Moon—associating the name Nancy Grace Roman with the new telescope is certainly very fitting. References Venture into Space: Early Years of Goddard Space Flight Center, A. Rosenthal, NASA Center History Series, 1968. “Following my Lucky Star”, N.G. Roman, Science, Dec. 9, 2016, Vol. 354, Issue 6317, p. 1346. N.G. Roman, interview with David DeVorkin, Niels Bohr Library & Archives, American Institute of Physics, Aug. 19, 1980 (transcription version July 16, 2025). The Space Telescope: A Study of NASA Science, Technology, and Politics, R.W. Smith, Cambridge University Press, Cambridge, 1989. “Rotation of Artificial Earth Satellites”, R.N. Bracewell and O.K. Garriott, Nature, Vol. 182, pp. 760-762, Sept. 20, 1958. OSO-5 Press Kit, Release 68-13, NASA, Jan. 20, 1969. “OSO to Study Sun in Quiet Period”, Robert R. Ropelewski, Aviation Week & Space Technology, July 26, 1971, pp. 34-36. “OSO Will Provide Own Eclipse”, Aviation Week & Space Technology, Sept. 20, 1971, p. 17. History of British Space Science, H.S.W. Massey and M.O. Robins, Cambridge University Press, Cambridge, 1986. Ariel 1: The First International Satellite – Experimental Results, NASA SP-119, 1966. “NASA Goddard and the Dawn of International Cooperation in Space”, T. Williams, The Space Review, Dec. 8, 2025. “British X-Ray Astronomy”, K.A. Pounds, Quarterly Journal of the Royal Astronomical Society, Vol. 27, 1986, pp. 435-444. OSO-7 Orbiting Solar Observatory One-Year Performance Summary, Report F72-10 (NASA CR-130184), Ball Brothers Research Corporation, Dec. 31, 1972. History of Orbiting Solar Observatory OSO-2, NASA TM X-55590, NASA Goddard Space Flight Center, Apr. 1966. The Observatory Generation of Satellites, Session II of Special Astronautics Symposium held at the Franklin Institute Dec. 27, 1962, NASA SP-30, Mar. 1963. Living and Working in Space: A History of Skylab, W.D. Compton and C.D. Benson, The NASA History Series, NASA SP-4208, 1983. “Helios to Relay More Accurate Solar Data”, Warren C. Wetmore, Aviation Week & Space Technology, Mar. 4, 1963, pp. 48-53. “OSO-8 Program Keyed to Skylab Data”, Aviation Week & Space Technology, June 30, 1975, pp. 45-47. “The First Space Ace – F-15 vs Satellite”, P. Glenshaw, Air & Space Magazine, Apr. 2018. “Planets of Other Suns”, N.G. Roman, The Astronomical Journal, Vol. 64, No. 1273, pp. 344-345, Oct. 1959. Trevor Williams is a space dynamicist who grew up avidly following the Apollo missions, and has long been fascinated by space history. He recently retired from NASA Goddard.

Starship May Become Useful Before It Becomes Reuseable

CLPS landers Starship lifts off on its 13th suborbital test flight July 24. (credit: SpaceX) Starship may become useful before it becomes reusable by Paul Almond Monday, July 27, 2026 SpaceX’s Starship-Super Heavy system is usually discussed in terms of its most ambitious promised capabilities: full reusability, rapid turnaround, very large payloads to orbit, and eventual operations beyond low Earth orbit. Those aims matter, but they can make the program look more binary than it really is. Starship is often treated as either an experimental vehicle still working through flight-test failures, or as a future fully reusable transport system that will radically reduce the cost of reaching orbit. That framing misses an important intermediate possibility: Starship may become commercially useful before it becomes fully reusable. The first commercially meaningful Starship may not be one that lands, survives inspection, and returns quickly to flight. The most plausible early use case is not crewed flight, lunar operations, or Mars, but Starlink deployment. Through Starlink, SpaceX has a large internal demand for mass to orbit and does not have to persuade an external satellite operator to accept the same level of risk from an immature launch system. The nearer-term question is therefore not only when Starship becomes fully and rapidly reusable, but when it can become a payload-first vehicle: one that can deploy Starlink satellites to useful orbits and then dispose of the ship safely, even if the ship is not recovered. The first commercially meaningful Starship may not be one that lands, survives inspection, and returns quickly to flight. It may be one that delivers Starlink satellites to useful orbits and then comes down predictably. The missing middle The transformative Starship case still depends on full and rapid reuse. Recovering both stages, inspecting them quickly, refueling them, and flying again at high cadence would be a different kind of launch system from anything currently in routine service. Starship’s significance lies not just in its scale, but in its ambition to move orbital launch closer to transportation infrastructure than to bespoke expeditionary rocketry. But a system does not have to achieve its final form before it has operational value. There is a possible middle ground between “test vehicle” and “fully reusable transport system.” In that middle ground, Starship’s upper stage might be expended, or recovery might remain an optional development objective rather than a condition of commercial mission success. The Super Heavy booster may reach operational reuse before the ship does. The ship might deploy payloads and then be deliberately disposed of over a remote ocean area. Reentry data could still be gathered, but the payload mission would not depend on the ship surviving. This distinction matters because “expendable Starship” is not, by itself, a very interesting claim. Any launch vehicle can be expended. Expendability buys performance by removing the need to bring hardware back. The more interesting point is that Starship’s first useful commercial phase may be neither pure test flight nor full reuse. It may be Starlink deployment with secondary reuse testing attached. That is different from simply “putting payloads on test flights.” A test-led mission chooses its trajectory, timing, and objectives mainly to gather vehicle data and manage public risk. If a payload is carried, it is opportunistic. A payload-led mission reverses the priority. It must reach a useful orbit or deployment condition, protect the deployment objective, and have a credible end-of-mission disposal plan. Reentry and recovery tests may still be attempted, but only within that mission envelope. The fact that the ship must reenter does not by itself justify turning the end of the flight into a landing experiment if doing so changes the trajectory, increases public risk, complicates licensing, or undermines the payload objective. That transition, from test-led to payload-led flights, may be one of the most important thresholds in the Starship program. Why Starlink is the natural first customer Starlink is what makes this intermediate case commercially plausible. SpaceX is developing Starship alongside a satellite network that can absorb large amounts of launch capacity, rather than relying only on external customers for early demand. Starship’s first useful commercial phase may be neither pure test flight nor full reuse. It may be Starlink deployment with secondary reuse testing attached. The current Falcon 9-based Starlink deployment system is already formidable. Falcon 9 has become the workhorse of modern orbital launch, and SpaceX has used it to build one of the largest satellite constellations ever deployed. But next-generation Starlink satellites are larger and more capable. Reuters reported that SpaceX was aiming to begin launching V3 Starlink satellites in the second half of 2026, likely on Starship, and that Starship’s payload bay was tailored for these upgraded satellites, with capacity for up to 60 in a single flight compared with roughly two dozen smaller Starlink satellites on Falcon 9. SpaceX has also begun testing this deployment path directly. On Flight 12, Starship successfully deployed 20 Starlink simulators similar in size to next-generation Starlink satellites, while also pursuing in-space and reentry test objectives. On last week’s suborbital Flight 13, it deployed 20 functioning Starlink V3 satellites for brief tests before the satellites reentered. That does not prove that Starship will be ready on that schedule. It does show why Starlink is the obvious bridge case. A Starship flight that deploys a large batch of V3 Starlink satellites could be commercially valuable even if the ship is not recovered. It could add network capacity, test deployment systems, and provide operational experience with real payloads, while SpaceX continues working toward the more difficult goal of full ship reuse. The regulatory filings around Starlink also show that the deployment problem is not abstract. Reuters reported that the FCC had approved another 7,500 second-generation Starlink satellites, bringing the authorized total to 15,000, subject to deployment milestones. The underlying FCC order also links certain orbital shells and inclinations to the possibility of Starship launches from Starbase, with alternative inclinations if the FAA does not allow particular Starship launch profiles. That is a revealing detail: a payload-led Starship is not just a test vehicle with satellites bolted on. It must be able to reach useful Starlink orbits and inclinations under a flight profile regulators are willing to license. The real threshold: safe disposal If Starship is used in an early payload-first mode, it does not necessarily need to land. It may not even need to survive reentry intact. But it must not become a large object making an uncontrolled reentry. Test objectives and commercial deployment objectives can conflict. This is where the second stage’s size matters. A Falcon 9 upper stage can be disposed of or allowed to reenter in ways regulators and operators understand well. Starship is in a different class: it is much larger, carries more energy, and, if stranded in orbit, would present a more serious uncontrolled reentry problem. Under the FAA’s commercial launch rules, operators must normally satisfy public-risk criteria for launch, reentry, and disposal, including limits on collective public risk. To fly its near-orbital Starship trajectory, SpaceX sought and received an FAA waiver, published in the Federal Register, from 14 CFR 450.101(a)(1)(i). In that waiver, the FAA stated that, at Starship’s then-current stage of development, the public risk associated with a random Starship reentry would be 40 to 50 times higher than the normal collective-risk threshold. The near-orbital trajectory, by contrast, allowed SpaceX to avoid random reentry risk and predict debris impact locations with high certainty. That is central to understanding any intermediate commercial mode. Starship does not need to be reusable to be useful, but it does need a licensed end-of-mission state. For a payload-first Starlink mission, the minimum success condition would include not only launch and satellite deployment, but controlled disposal of the ship. That disposal could in principle be achieved in several ways. The most straightforward would be to use Starship’s own propulsion and guidance systems for a controlled deorbit or targeted reentry over a broad ocean area. A transitional vehicle might carry additional disposal-assurance hardware, though that would add mass, complexity, and new safety questions. At this stage, the controlling requirement is not intact recovery but predictable disposal. A stripped-down cargo Starship could discard systems needed for intact recovery, but not those needed for safe disposal. It would still require attitude control, power, command logic, tracking, telemetry, deorbit capability, and a credible failure analysis. Expendability would reduce the recovery burden; it would not remove the obligation to control where the vehicle goes. Payload-led flights may conflict with test-led flights This is why the intermediate regime is more subtle than simply adding Starlink payloads to Starship tests: test objectives and commercial deployment objectives can conflict. A test flight may prefer a trajectory that maximizes engineering data while keeping debris in a known remote corridor. A Starlink deployment mission may require a different orbit, altitude, inclination, or phasing. A recovery test may consume propellant margin or impose reentry constraints that are not needed for the payload mission. A deliberately conservative disposal profile may be more attractive for an early commercial flight than an aggressive recovery experiment. The FAA will not ignore optional post-deployment tests simply because the payload has already been released. If SpaceX deploys Starlink satellites and then attempts a demanding reentry, landing, or catch profile, that entire sequence is part of the licensed risk picture. A payload-led Starship mission may therefore need to treat recovery experiments as secondary. They may be useful when they fit within the same acceptable risk envelope. Once they begin to determine the trajectory, timing, disposal plan, or licensing case, the mission has become test-led again. The analogy is not Falcon 9’s upper stage so much as Falcon 9’s early booster recovery attempts. Customers bought a launch service, and SpaceX then attempted booster recovery after the primary mission. Early Starship Starlink flights might follow a similar pattern: deployment is the mission; recovery or reentry testing is a SpaceX development objective layered on top. What early usefulness would and would not prove A payload-first Starship used for Starlink would be important, but it should not be confused with proof of the full Starship architecture. Such a mission would not prove rapid turnaround, cheap heat-shield refurbishment, ship catch, long-duration orbital operations, orbital refueling, crew safety, lunar landing, or Mars transport; those would remain separate milestones. The public debate around Starship often jumps from spectacular test flights to sweeping claims about Mars, lunar bases, or airline-like rocket operations. They are not necessarily the right benchmark for judging the first possible commercial threshold. It would prove something narrower but still significant: that Starship can move from experimental trajectories toward operational payload service. It would show that Starship can deliver useful mass to useful orbits, deploy payloads, and end the mission safely. For SpaceX, that could matter even before full reuse because Starlink creates a large internal demand for precisely that capability. This is also why the intermediate mode should not be oversold. If an expendable or semi-expendable Starship launches V3 Starlink satellites, the economics may still be far from the final promise of rapid full reuse. Expending the ship would consume hardware and engines. It could reduce payload margin if additional disposal-assurance systems are needed. It could complicate the test program if a payload-optimized ship diverges from the reusable configuration SpaceX ultimately wants. Yet the existence of those trade-offs does not erase the middle ground. A vehicle can be commercially useful before it is economically revolutionary. Useful before reusable The public debate around Starship often jumps from spectacular test flights to sweeping claims about Mars, lunar bases, or airline-like rocket operations. Those ambitions matter to SpaceX’s long-term story, but they are not necessarily the right benchmark for judging the first possible commercial threshold. For Starlink, the relevant near-term question is simpler and harder: can Starship fly a payload-led mission to a useful orbit and dispose of the ship safely enough for regular use? If it can, then Starship may begin to matter commercially before it becomes fully reusable. That would not be the final Starship vision, and it would not settle the question of rapid reuse or the economics of the complete system. It would, however, mark a transition from vehicle testing toward operational infrastructure. Starship may become useful before it becomes reusable. The missing middle is not glamorous, but it may be where the program first becomes commercially real. Paul Almond is an independent researcher based in London whose research interests include launch systems, spacecraft engineering, and space policy. He has previously published in the Journal of the British Interplanetary Society.