You’re staring at a screen or a massive piece of steel and something goes wrong. Not just "oops" wrong, but "the world is watching" wrong. We’ve all seen those moments. A bridge wobbles, a telescope sees blurry images, or a nuclear reactor starts doing things it definitely shouldn't. The immediate thought is always the same: how did they fix that? It's never as simple as a software update. Honestly, the solutions are usually a mix of terrifying risks, duct-tape-style ingenuity, and math that would make your head spin. Sometimes, the fix is actually more impressive than the original invention.
The Hubble Space Telescope: The Ultimate Vision Correction
When Hubble launched in 1990, it was supposed to be the pinnacle of human achievement. Instead, it was a $1.5 billion embarrassment. The first images came back blurry. A tiny flaw—literally a fraction of the width of a human hair—in the primary mirror meant the light wasn't focusing right.
So, how did they fix that when the thing was orbiting 340 miles above the Earth?
NASA couldn't just bring it back down. They had to perform surgery in space. They developed COSTAR (Corrective Optics Space Telescope Axial Replacement). Think of it as a pair of very expensive contact lenses. Astronauts on the Endeavour mission had to physically swap out instruments to install these corrective mirrors. It was a high-stakes gamble. If they bumped the wrong thing, the whole mission was toast. Story Musgrave, one of the lead astronauts, famously spent years practicing the movements in a giant pool just to make sure he didn't drop a bolt. It worked. The blurry blobs became the crystal-clear nebulae we see on every second-grader's bedroom wall today.
The Leaning Tower of Pisa: Pulling it Back from the Brink
People forget the Leaning Tower of Pisa was actually closed for over a decade. By the 1990s, the tilt was so bad it was mathematically certain to fall over. Gravity is a jerk like that. The solution wasn't to just push it back; you can't push a 14,000-ton stone cylinder without it snapping.
The engineers got weird with it. First, they tried lead weights. They piled 600 tons of lead on the north side to act as a counterweight. It helped, but it was ugly. Then they tried "ground freezing" with liquid nitrogen. Terrible idea. The soil expanded, then contracted, and the tower actually lurched.
The real "eureka" moment? Soil extraction.
John Burland, the soil mechanics expert on the project, realized they didn't need to touch the building. They needed to touch the dirt. They drilled 41 diagonal holes under the high side (the north side) and slowly sucked out small amounts of silt. Basically, they invited the tower to sink into a more upright position. It took years. They moved it back by about 17 inches at the top, which sounds like nothing but was enough to stabilize it for another 200 years.
The Citicorp Center: The Secret Near-Disaster in NYC
This is the craziest "how did they fix that" story in architecture because for a long time, nobody knew it happened. In 1977, the Citicorp Center was finished in Manhattan. It sits on four massive stilts, but because of a church on the corner, the stilts are in the middle of each side, not the corners.
A student named Diane Hartley was looking at the plans and realized something was off. The wind calculations only accounted for winds hitting the faces, not the corners. When an engineer named William LeMessurier double-checked her math, he turned pale. In a "quartering wind" (hitting the corners), the joints would fail. A hurricane could literally knock the building over into Midtown.
The fix was a secret race against time. Every night, after office workers went home, welders snuck into the building. They welded two-inch-thick steel plates over the bolted joints. They did this for months while keeping the public in the dark to avoid a mass panic. They even had a weather plan: if a hurricane hit before they finished, they were going to evacuate the entire surrounding blocks of Manhattan. Luckily, they finished just before Hurricane Ella approached.
Apollo 13: The Carbon Dioxide Crisis
We’ve all heard "Houston, we have a problem," but the technical fix for the CO2 buildup is a masterclass in improvising under pressure. The astronauts were breathing out carbon dioxide, and the scrubbers on the Lunar Module (the "lifeboat") were square. The spare scrubbers from the Command Module were round.
You can't fit a square peg in a round hole. Literally.
📖 Related: How to Change G Shock Battery: What Most People Get Wrong and How to Fix It Yourself
Ground control had to invent a fix using only what was on the ship. They used plastic bags, cardboard from a flight plan cover, and a lot of grey duct tape. They basically built a funnel that forced the air through the wrong-shaped canisters. It’s the ultimate "how did they fix that" because there was zero room for error. If the tape didn't seal, they’d all be dead within hours.
When Software Goes Rogue: The Knight Capital Glitch
Not every fix involves steel and hammers. In 2012, Knight Capital Group lost $440 million in 45 minutes. A piece of old, "dead" code was accidentally activated during a software deployment. The system started buying high and selling low at a rate of thousands of trades per second.
The fix here was a nightmare. They couldn't just "turn it off" because they didn't know which server was doing it. They had to manually go through 28 different servers and uninstall the new software, then try to figure out how to stop the hemorrhaging of cash. They eventually had to get a massive line of credit just to stay in business. The "fix" was a total restructuring of how DevOps and deployment work in high-frequency trading—moving away from manual pushes to automated, fail-safe pipelines.
Lessons from the Great Fixes
When you look at these cases, a few patterns emerge about how massive problems actually get solved:
- Don't fight the symptoms; find the root. For the Leaning Tower, the building wasn't the problem—the dirt was.
- Use what you have. The Apollo 13 crew didn't have a 3D printer; they had tape and cardboard. Creativity thrives under constraints.
- The "fix" often becomes the new standard. The Citicorp disaster led to completely new building codes regarding wind loads that we still use today.
- Admit the mistake early. If LeMessurier hadn't listened to that student, a building would have collapsed in the middle of New York.
Actionable Insights for Your Own "How Do I Fix This?" Moments
You might not be fixing a space telescope, but the logic applies to your business or home life:
- Isolate the variable. Stop changing five things at once. Change one thing, observe, and move to the next.
- Look at the "soil." If a project is failing, is it the project itself or the environment (the culture, the tools, the timing) it's sitting on?
- Build a "COSTAR." Sometimes you can't fix the core flaw. Instead, build a "lens" or an interface that corrects the output.
- Document the "Welds." If you perform a quick fix or a "hotfix" under pressure, go back later and make it permanent. Don't leave the duct tape there forever.
If you're dealing with a system failure right now, stop looking for a magic button. Start looking for the small, incremental adjustments that pull the tower back toward center. Real engineering isn't about perfection; it's about the relentless pursuit of "stable enough to work."