Previous: Satellite Death Rattle
Next: A Strangely Cool Supermassive Black Hole!



View count:229,543
Last sync:2023-05-20 22:00
The software running Apollo’s guidance computers was literally woven by hand by "little old ladies."

Hosted by: Hank Green
Support SciShow by becoming a patron on Patreon:
Dooblydoo thanks go to the following Patreon supporters -- we couldn't make SciShow without them! Shout out to Kathy & Tim Philip, Kevin Bealer, Andreas Heydeck, Thomas J., Accalia Elementia, Will and Sonja Marple. James Harshaw, Justin Lentz, Chris Peters, Bader AlGhamdi, Benny, Tim Curwick, Philippe von Bergen, Patrick Merrithew, Fatima Iqbal, Mark Terrio-Cameron, Patrick D. Ashmore, and charles george.
Like SciShow? Want to help support us, and also get things to put on your walls, cover your torso and hold your liquids? Check out our awesome products over at DFTBA Records:
Looking for SciShow elsewhere on the internet?

Memory size discussion:
[SciShow intro plays]

Hank: Back in the 1960s and 70s, the Apollo missions blasted their way from Earth to the Moon. And they carried two of the smallest, most sophisticated guidance computers ever invented … which were running on software knitted by little old ladies. No, really. The software running Apollo’s guidance computers was literally woven -- by hand -- out of wires and magnetic rings that looked like tiny donuts. It was called Core Rope Memory.

The Apollo missions were a huge hurdle for both navigation and portable computing. The orbital mechanics were complicated, and they needed guidance, especially while they were on the far side of the Moon, unable to communicate with Earth.

Navigating there and back was a serious problem, and NASA needed computers to solve it. A team at MIT invented the navigation software to run on these computers. Programmers wrote it from scratch and tested it on huge mainframe computers, using paper punch cards to input the programs.

Running any given program could take an entire night. And, of course, the software had to be bug free, because once the programs were loaded onto the hardware of Apollo computers, they couldn’t be changed. So they had to be perfect. Why couldn’t they be changed? Because the program was hardware, essentially.

There were a few different forms of storage that existed in the 1960s that could hold a computer program. One involved paper punch cards with holes in them, read in a giant reader. There were also disk drives that were so big they had to be pushed on wheeled steel carts, and magnetic tape on reels. But these options were all way too heavy to fly into space. Or, in engineer-speak: they weren’t flight-weight.

Even if they were light enough to fly, they’d still need to be able to withstand the shock, vibration and G-forces of launch, temperature changes, and cosmic radiation. And if they couldn’t withstand all that, the astronauts could die. So, the memory storage had to be small, lightweight, safe, strong and robust enough that even if you lost power, you didn’t lose the program.

The only technology at the time that met these specs was core rope memory, which coded ones and zeros -- the fundamental language of programming -- into wires and magnets. It was woven on a type of loom, by threading individual wires through various holes with large needles -- kinda like knitting needles. Engineers at the time called it LOL memory, a not-very-nice acronym for “little old lady” memory -- because it took highly skilled garment workers, often older women, to weave it.

To represent a one, a seamstress wove a wire through a little magnetic donut called a core. The donut acted like a transformer -- a device that changes the voltage of an electrical current running through it. If the computer saw a voltage change at the other end of the wire, it assigned it the number one. To get a zero, they weaved the wire outside of the core. Electrical current through it wouldn’t change. The computer would interpret that lack of voltage change as a zero.

They'd weave the entire program out of wires going through or around cores. There were lots of wires and donuts, which meant that Core Rope Memory was incredibly hard to manufacture. It came out looking a lot like a rope, but it was really a program made out of woven electrical pathways.

It also provided the most storage per cubic centimeter at the time -- the Apollo Guidance Computer came with a whopping 36 kilobytes of memory. This tiny microSD card has almost a million times that. But core rope memory is Read Only Memory.

You can’t write to it, which is really good if you don’t want to accidentally record the 1960s equivalent of a podcast over what would be steering you to the Moon. But it also meant the programs had to be perfect the first time around. When each core rope was finished, the program was run and compared with the program stored on magnetic tape from MIT -- they actually had a defense contractor build a machine to do this automatically.

If they found a mistake, the program could be rewired before it left the factory -- though fixing it was an enormous pain. So there’s a lot more to knitting than scarf patterns: it can also take you to the moon and back.

Thank you for watching this episode of SciShow Space, and thank you especially to our patrons on Patreon who help make this show possible. If you want to help us keep making episodes like this, just go to to learn more. And don’t forget to go to please and subscribe!