Science6 distinct publishers3 min readPublished
Katalyst's Link spacecraft never got close enough to grab the sinking telescope. Swift now reenters later this year, and on-orbit rescue remains a demonstration, not a planning assumption.
The Scientist · Science desk

Compiled by The ScientistSomething wrong?How this is made
NASA and Katalyst Space Technologies said Wednesday that the company's Link spacecraft will not attempt to capture and boost the Neil Gehrels Swift Observatory, because of continuing problems controlling the vehicle [1][2]. NASA had paid $30 million for the boost, and what it got instead is a proximity-operations experiment next to a telescope that will reenter the atmosphere later this year, no earlier than October [3][4][5].
The mechanics of the failure are ordinary and therefore instructive. Link launched in early July and went into an uncontrollable spin a few weeks later [6]. On July 28 NASA said hardware malfunctions had caused the tumble and that communications with the spacecraft were sporadic [7]. Controllers slowed the spin using the same xenon-gas thrusters that were meant to lift Swift, but pointing and positioning problems persisted [8][9].
The schedule was the other variable. NASA found Swift's orbit decaying faster than expected because of intense solar activity, which set a hard deadline, and Katalyst had nine months to design, build and launch Link, a spacecraft intended to grab the telescope with robotic arms [10][11][12]. Chief executive Ghonhee Lee described it as an ambitious mission on an aggressive timeline put together in under a year, and said the company took on a high-risk, high-reward challenge and had already learned a great deal [13][14]. NASA Administrator Jared Isaacman said the outcome was not what the team worked toward but does not change why the mission was worth attempting [15].
The cost to science is concrete. Swift launched in 2004 and its instruments were switched off earlier this year to slow the descent, so the observations stopped before the rescue did [16][17]. As of Wednesday the telescope was at 214 miles, against the 373-mile orbit Katalyst hoped to restore, a shortfall of 159 miles [18][19][20]. Scientific American reports that Swift's loss leaves a gap in high-energy observations that is unlikely to be filled soon [21].
The planning consequence is larger than one telescope. Hubble, now 36 years old, is also losing altitude because of solar activity and had been next in line for a servicing attempt if this one had gone well; whether that is still on the table was not immediately known [22][23][24]. Programs writing mission-extension into their risk plans should note that the money bought a launch, a spin and a set of lessons, not the orbit.
Katalyst will keep flying Link near Swift to gather data for future rescue work, and Lee says the aim is a repeatable playbook for rendezvous, proximity operations and servicing [25][26]. Three things to watch: whether the proximity campaign produces enough engineering data to make the $30 million defensible [3][25], the actual reentry date once atmospheric drag is measured against the October estimate [5], and what NASA now says about Hubble [24].
Ranked by verification strength, evidence, and original report placement.
The mission to save NASA's sinking Swift Observatory was called off Wednesday; NASA and Katalyst Space Technologies announced that Link will not attempt to capture Swift and boost it to a higher orbit because of problems with controlling the spacecraft.
In a LinkedIn post, Katalyst said the company and NASA had agreed that, due to ongoing spacecraft control issues, "the mission will not include the capture and boost of NASA's Neil Gehrels Swift Observatory."
NASA paid $30 million to Katalyst in an attempt to raise the telescope to a safer, longer-lasting orbit.
NASA said the decision means the telescope will plunge through the atmosphere after more than 20 years of tracking events such as gamma ray bursts and exploding stars.
Swift's demise is not expected before October.
The rescue spacecraft went into an uncontrollable spin a few weeks after its liftoff in early July.
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Two publishers, on-record agency and vendor statements
Both sources independently report the cancellation, the control failure, the $30 million contract and the reentry outlook, quoting NASA's administrator and Katalyst's CEO directly and citing specific dates and altitudes. What is missing is any technical root-cause analysis or contract documentation, so the record is strong on outcome and thin on causation.
One rescue attempt, zero captures, residual demo only
Adoption of on-orbit rescue as a working capability remains near-nil: a single $30 million mission failed before rendezvous, the target telescope will reenter, and the only continuing activity is a proximity-operations demonstration. The one queued follow-on customer, Hubble, has no confirmed plan.
Servicing playbook language ahead of demonstrated capability
The subjects' framing — a 'repeatable playbook' for rendezvous and servicing, and capabilities 'America will need' — runs ahead of a mission that lost attitude control before approach and spent its boost propellant on despin. The publishers themselves are sober, labelling the mission a failure and quantifying the shortfall, which keeps the gap moderate rather than large.
Vendor LinkedIn post and agency statement drive the framing
Nearly all interpretive language in the cluster comes from two interested parties: Katalyst, which needs future servicing business and issued its disclosure via a LinkedIn post, and NASA leadership, which must justify a $30 million expenditure that produced no boost. Both sources reproduce those quotes prominently; neither adds independent technical or oversight commentary.
Outcome well established, causes and next steps unresolved
Confidence is high that the capture was cancelled, that $30 million was spent, and that Swift will reenter later this year, because two publishers converge on dated, attributable facts. It is lower on why the spacecraft failed, whether any servicing capability was validated, and what happens to Hubble, all of which the cluster leaves open.
product
Swift will burn up, and LINK's consolation prize is the honest state of satellite servicing5 distinct publishers
science
Katalyst's servicer spent its fuel on its own tumble instead of raising NASA's Swift telescope1 distinct publisher
product
NASA stops trying to save Swift, and gamma-ray burst alerts get a hard deadline1 distinct publisher
science
Two robots in orbit will decide whether satellites stop being disposable1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 20, 2026
1 article · August 20, 2026
1 article · August 20, 2026
1 article · August 20, 2026
1 article · August 20, 2026
1 article · August 19, 2026