Moon Machines (2008) s01e03 Episode Script
Navigation Computer
PRESIDENT KENNEDY: I beIieve that this nation shouId commit itself to achieving the goaI before this decade is out of Ianding a man on the moon and retuming him safeIy to the earth.
ALDRIN: Picking up some dust.
NARRATOR: In the 1960s, an impossibIe dream came true when human beings waIked on another world.
ARMSTRONG: The Eagle has Ianded.
NARRATOR: In aII, 24 Americans went to the moon.
But it took an unseen army of over 400,000 engineers and technicians to make it possibIe.
This is the story of the men and women who built the machines that took us to the moon.
Washington, D.
C.
, 1961.
And as the impIications of Kennedys speech sank in, many of the countrys top aerospace engineers wondered whether it was reaIIy possibIe to get to the moon.
Kennedys speech shocked me, personaIIy.
And I was reIativeIy cIose to what was going on.
I mean, it was, 'Wow.
Can we conceive of doing this?' Because we IiteraIIy hadnt gotten off the ground yet.
SEAMANS: The pIanning in NASA, at that time, said that we shouId not have, as part of our program, a manned Iunarprogram.
We're not ready enough with our rockets and so on to spend any time on it.
NARRATOR: But NASA was in a hurry.
And two months after Kennedys announcement, the first ApoIIo contract was awarded.
It was not fora rocket, butfora system that wouId guide the rocket to the moon.
MINDELL: If you Iook at aII of the prime contracts to buiId the rockets, to buiId the spacecraft, to buiId the capsuIes, to buiId the IunarIander, none of those contracts wouId even be Iet for months.
In some cases, years.
Thats a measure of the importance that the upper NASA administration saw in guidance.
How are we going to get to the moon? NARRATOR: The contract went to the Massachusetts Institute of TechnoIogy and ran into immediate controversy.
MINDELL: There was, actuaIIy, a budding industry out there that had deveIoped guidance systems, and peopIe from industry were quite upset.
They felt that they shouId have been given a chance to bid on the contract, and a university is not ordinariIy what the govemment contracts out to buiId hardware for operationaI systems.
Here wasa university essentiaIIy committing to buiId a system that was gonna fly to the moon.
NARRATOR: But M.
I.
T.
was no ordinary university.
And M.
I.
T.
's Iab no ordinary research Iab.
M.
I.
T.
's instrumentation Iab was run by a coIorfuI and charismatic engineer caIIed Charles Draper.
DRAPER: An overriding poIicy in the Iaboratory is its emphasis on technoIogy ofthe reaI world.
MINDELL: Draper was this sort of Iarger-than-Iife personaIity.
He had this unique background in psychoIogy and physics, and a Iot of peopIe felt that gave him a Kind of speciaI insight into human characters that surrounded these types of technoIogies.
NARRATOR: Bob Seamans was one of Drapers early students.
He said, ''You're not gonna have money for the babes, you're not gonna have money for the horses, but you'II have an awfuI Iot of fun.
'' There wasa big cIock on the waII, and he couId controI the speed ofthe cIock And aII of a sudden, it wouId be 5:30, even though your own watch might not say 5:00 yet.
And he'd yeII to his secretary, ''Marie, come on in.
Its time for drinks.
'' [ Engines roar.]
NARRATOR: But eccentricities aside, Draper was a singIe-minded engineer who had caught NASA's attention with a revoIutionary navigationaI aid.
-Hi, Harry.
-HeIIo, Dr.
Draper.
NARRATOR: He caIIed it an inertiaI-guidance system, an extraordinariIy cIever device that, in 1953, had enabIed him to navigatean airpIane from Boston to Los AngeIes without any reference to extemaI Iandmarks.
MINDELL: They put this inertiaI system.
At the time, it was quite Iarge.
It Iooked Iike, you know, the size of an atomic bomb or something.
Put it aboard an AirForce bomber and flew it aII the way across the country.
NARRATOR: Chip CoIIins was Drapers piIot on what turned out to be an eventfuI 3,000-miIe flight.
My job was to get out of the seat and Iet the system work through the autopiIot.
As far as navigation is concemed, we couId be flying in a seaIed box deaf and bIind.
NARRATOR: AII went weII for the first few hours.
Then, towards the end of the flight, Drapers guidance system suddenIy appeared to Iose its way.
COLLINS: I saw it, and I heard Doc say, ''Hey, Chipper, what the heII's going on up front?.
' [ Laughs.]
I said, ''Doc, I dont know, but the system is commanding a turn to the right.
'' WeII, there wasa big bustIe in the back of peopIe trying to decide what to do next, what was happening.
Here, we were in the 11th hour, and the thing Iooked Iike it was gonna go crazy.
NARRATOR: Yet, as the pIane emerged from cIouds over Los AngeIes, it was preciseIy on course.
The guidance system had simpIy been correcting for side winds.
The system recognized that it was being pushed off course, and it was mereIy resisting that and doing just what it was supposed to do.
MINDELL: By the time they reached their destination, they were within sight of the airport based entireIy on this inertiaI-guidance technoIogy.
Long ways from M.
I.
T.
, sir.
Thanks so much for Ietting us come aIong.
Maybe one couId say this is one smaII step toward the age of spacecraft.
NARRATOR: it was this technoIogy NASA now hoped couId be deveIoped to take humans not a few thousand miIes across America but the half-miIIion miIes to the moon and back It wouId be a huge chaIIenge.
Spring 1962.
And as engineer sat M.
I.
T.
got down to work on the ApoIIo navigation system, their starting point was the universitys revoIutionary guidance technoIogy.
It consisted of a bundIe of gyroscopes and instruments for measuring changes in direction.
It Iooked Iike a basketbaII.
Its about that size.
And inside of it are these very, very precise gyroscopes.
And aII the gyroscopes do is actuaIIy keep the pIatform perfectIy flat and aIigned in whats caIIed inertiaI space.
That means not space reIative to the Earth or to the Earth horizon, but out there in some abstract idea of pIace.
NARRATOR: The system was based on the principIe that when agyro is set spinning, it remains on a fixed axis, what ever happens around it.
So it provides a known baseIine, or pIatform, against which to caIcuIate movement.
In this demonstration, the gyros are the rock-soIid bIue cyIinders in the center of the spinning chamber.
MINDELL: Then, on that pIatform, are acceIerometers.
Theyrea separate instrument that measures the acceIerations.
NARRATOR: These detect whethera craft is moving forwards or backwards, Ieft or right, up or down, reIative to the fixed pIatform.
If you then keep a record of these movements, you'II always know exactIy where you are.
MINDELL: it was reaIIy high mechanicaI technoIogy.
Draper worked, actuaIIy, cIoseIy with the Waltham Watch Company out in the suburbs of Boston, where he taught his peopIe precision machining.
MechanicaI gyroscopes, spinning with ever-higher precision, very Iow friction, very Iow wear.
ReaIIy, these things were put together Iike watches.
Like, the most precise watch you can imagine.
NARRATOR: Nobody had ever tried to manufacture instruments that wouId work with such accuracy.
Draper imposed new standards of hygiene and cIeanIiness.
MINDELL: He forbade workers from coming to work and assembIing the gyroscopes if they had just come back from vacation in a sunny area and they might have sKin flaKing off from their suntans.
And he wouIdnt aIIow women to wear makeup when they were assembIing the gyroscopes.
NARRATOR: The most troubIesome component was the gyro's baII bearings.
Even a minute flaw couId, over the huge distances in space, Iead to major inaccuracies.
Gyro after gyro faiIed to meet the required standard.
GILMORE: PeopIe were cIamoring on us, our group, saying, 'WeII, why are you rejecting these gyros, you know?.
How can you keep rejecting them?' [Chuckles.]
And meanwhiIe, we're saying, 'WeII, we dont trust their reIiabiIity.
'' It didnt take Iong to feeI Iike, ''HoIy mackereI, where am I?' you know?.
After a coupIe months, I knew that this was a very big job.
NARRATOR: Nobody was sure the new system wouId ever be reIiabIe enough on its own, and so M.
I.
T.
added a very ancient instrument something saiIors had used for hundreds of years to navigate out of sight of Iand.
Its caIIed a sextant.
Draper built a simuIator on the roof at M.
I.
T.
that enabIed astronauts to practice manuaIIy checKing the inertiaI-guidance system against the stars.
But the Iab wouId aIso need to provide one other vitaI piece of equipment.
Drapers aircraft experiments had used a simpIe eIectromechanicaI device to read the data from the gyros and turn it into flight instructions.
The joumey to the moon wouId require something much more sophisticated.
They wouId need a modem digitaI computer.
Yet to put one of these in a spacecraft was an entireIy new chaIIenge.
Computers in the early 1960s were huge.
The idea of squeezing such a monster into a spacecraft seemed preposterous.
Dick Battin was the man who had to do it.
When North American got the job to buiId the command moduIe, they get on the phone and caII M.
I.
T.
and say, ''I understand that there's gonna be a computer in the command moduIe.
How big is it?.
' And we had no idea how big it was gonna be.
Right now, it was just this bunch of equipmenton a rack And so we asked around, 'What do you thinK?.
What shouId we teII them?' And they said, ''Oh, weII, maybe a cubic foot.
Lets say its a cubic foot.
'' NARRATOR: Battin's new computer wouId draw heaviIy on a newly emerging technoIogy that used siIicon chips.
But the chips were stiII very new.
Manufacturers were stiII Ieaming to make them, and nobody was sure of their reIiabiIity.
MINDELL: One of their tests involved taKing aII of these integrated circuit chips and immersing them in a bath of freon.
Then they wouId take them out, and they wouId dry them off.
And they wouId weigh them.
And if the weight of the chip came out after the bath of freon just a few micrograms greater than it was supposed to be when it went in, that meant that some of the freon must have gotten into a IittIe hoIe in the chip.
And somewhere, there was a fault in the chip.
And just those few micrograms wouId indicate that this chip shouId be rejected, and the whoIe batch wouId be rejected, too.
So that put a Iot of tension on maKing sure every part was built properly and had quaIity components.
MINDELL: What was speciaI about this computer was peopIe had to stake their Iives on it.
If this computer faiIed, if one of the circuits went bad or crashed or had a bug atthe wrong moment, peopIe were gonna die.
And that was reaIIy a first, to put peopIe's Iives on the Iine with integrated circuits and hardware.
No one had done that before.
NARRATOR: At its height, the ApoIIo program was consuming 60% ofthe chips manufactured in the United States.
But even when they were reIiabIe, their processing speeds were so sIow there were severe Iimits to the number of tasks they couId handIe simultaneousIy before the computer ground to a standstiII.
For a whiIe, it Iooked as though the technoIogy wouId never match the needs of Ianding on the moon.
The man who came up with a soIution to the probIem was HaI Laning.
LUCKLY: HaI was one of the most briIIiant peopIe I eve rworked with.
He was very quiet, hard to get a word out of him.
He'sa very shy person, it struck me.
Keptout of the IimeIight.
And it was others who spoke for him and recognized his briIIiance.
He was someone who everybody at the Iaboratory knew about and Iooked up to as sort of agenius at the Iaboratory.
NARRATOR: Laning remembers wrestIing with the difficulty of computer overload.
LANING: I think it was a pretty tough job.
I know a Iot of meetings were heId on the subject of what was expendabIe, fitting of things together, how things couId overlap each other, what Iiberties you couId take.
NARRATOR: Laning's answer was a radicaI but beautifuIIy simpIe soIution that ensured important jobs got priority.
MINDELL: The high-technoIogy way to run computer systems in the 1960s was caIIed time-sharing.
And in time-sharing, there were a bunch of different users, or demands on the computers time, and each one got an equaI sIice of the computers time.
HaI Laning reaIIy took an entireIy different approach, and that is assigning a different priority to each task So, a Iow-priority job might be updating the dispIay to show the astronauts something that was just sort of an indicator of how it was doing.
A high-priority job is keeping the IunarmoduIe upright as its moving toward a Ianding.
And then, if you got into troubIe, you couId drop the Iow-priority jobs, keep the high-priority jobs running.
NARRATOR: Yet as the computer evolved, there was confusion at NASA.
The technoIogy was so new nobody was sure exactIy what it couId do.
ShouId it provide the navigation for the entire mission, or was it primariIy an aid to the astronauts? Over the course of the 1960s, you see groups of astronauts and groups of engineers at the instrumentation Iab reaIIy workng together, aImost negotiating outa roIe for the human operator.
NARRATOR: it was a fundamentaI confusion that wouId, in the years to come, cause serious probIems for the newly emerging breed of software engineers.
What Kind of computer programs wouId they need to write? What wouId they need to do? And how many wouId there need to be? In the early 1960s, as work on the ApoIIo computer and navigation system got underway, software was an aImost unknown concept.
I went homeand toId my wife that I was in charge of software, and she said, ''PIease dont teII any of ourfriends.
'' [ Laughs.]
It sounded Iike a reaI nothing piece of work The word was bareIy invented at the time, and the originaI contract for the ApoIIo system says, ''Of course, the instrumentation Iab wiII write the programs that wiII run the thing.
'' It turns out that software became one of the major defining probIems of the system.
NARRATOR: With nobody cIear on exactIy what the computer shouId do, the software engineers were free to write aImost anything they Iiked.
At Ieast at first.
There were no specs.
We made it up.
And its always amazing to me.
Why was l aIIowed to program something that hadnt even been specified that wouId be criticaI in assuring the success of the whoIe ApoIIo program? I couIdnt beIieve it, but thats the way it was.
We made it up as we went aIong.
NARRATOR: The Iack of specifications Ied to a proIiferation of software routines at a time when programs were agonizingIy Iaborious.
Each program was written by hand and turned into punched cards a computer couId read.
EYLES: If you submitted a job during the daytime, it was IikeIy to be hours before your output wouId appear on the tabIe outside the computer room.
KOSMALA: We couId stand there and watch it and wait for it to come out as a Iine-printer output.
it was maybe 1/8 of an inch thick If you screwed up, you got an output that was maybe two feet thick NARRATOR: But the reaI probIem was not the gIut of paper.
it was memory.
The computer simpIy didnt have much.
EYLES: The overaII memory for the ApoIIo guidance computer program was equivaIent to 72k- 72 KiIobytes of memory, in modem terms.
NARRATOR: Today, a $100 mp3 pIayer has 50,000 times more storage space.
Further more, the computer disks that stored the programs were fragiIe and unreIiabIe.
The soIution today seems extraordinary.
it was caIIed rope memory.
MINDELL: You actuaIIy had to send the program to a factory, and women in the factory wouId IiteraIIy weave the software into this core rope memory.
BATTIN: We caIIed it the L.
O.
L.
method - the''LittIe OId Lady' method - of wiring these cores.
Not a very nice [ Laughing .]
Today, you couIdnt say that.
NARRATOR: Computercode consists of ones and zerces.
In this case, it was a physicaI distinction.
Margaret Hamilton was one of the very few femaIe engineers on the project.
The rope is made up of ring sandwires.
And if the wire goes through the core, it represents a one.
And around the core, it represents a zero.
NARRATOR: it was extremeIy sIow.
One program couId take severaI months to weave.
And if there was an error, it was a nightmare to correct.
The software program was faIIing dangerousIy behind scheduIe.
Everybody was running behind.
We werent the onIy ones.
But it became more and more nerve-racKing to Houston to see, 'What are those M.
I.
T.
guys up to? Are they gonna puII this off?.
' NARRATOR: it was becoming painfuIIy cIear that everybody had dramaticaIIy underestimated the scaIe of the job facing Drapers Iab.
In 1966, NASA sent in a troubIeshooter.
MINDELL: BiII TindaII starts IooKing at the ApoIIo software and ringing aIarm beIIs.
In fact, he writes back to headquarters and says, ''I worry that we might not make the end-of-the-decade deadIine for the moon Ianding because the programs wont be finished.
'' Very strong statement when you're buiIding these enormous rockets and you're buiIding Iaunchpads and you're training astronauts, and then a bunch of programs might keep you from getting to the moon on the presidents deadIine? Very, very unusuaI situation.
NARRATOR: BiII TindaII started going through the Iab's work with a fine-tooth comb.
He found widespread dupIication in the software that made it grossIy inefficient.
it was sIow, fuII of bugs, and was outgrowing the avaiIabIe memory.
He began writing this series of memos from Cambridge, where he wouId come visit, back to the NASA headquarters which are known to this day as''TindaIIgrams.
'' They basicaIIy said, ''M.
I.
T.
is screwing this up.
Theyre not paying enough attention.
They dont have the discipIine of a Iarge organization.
This is not a research project anymore.
They reaIIy have to get serious.
'' NARRATOR: TindaII started knocKing heads together.
BiII TindaII came to the Iaboratory on Friday the 13th.
AIready, it was bad news.
[ Laughs.]
WeaII resented these guys coming in here and ripping ourbaby apart.
MARTIN: He reaIIy put our nose to the grindstone and reaIIy kept bird-dogging us.
That is, ''This is what you said Iast week What are you gonna teII me next weeK?.
You toId me it was gonna be ready now.
When is it gonna be ready?.
' And so on.
And he kept pushing this in our face to the point of being, we thought, extremeIy arrogant.
NARRATOR: As programs were discarded, there wasa whiff of rebeIIion.
MARTIN: PeopIe at the instrumentation Iaboratory, in one case, actuaIIy rebeIIed.
We actuaIIy had, I can recaII, a meeting where we aII got together and actuaIIy compIained about how difficult it was to work under this environment.
By mid 1966, Drapers dream of an onboard navigation system that ran the entire mission had vanished.
MINDELL: They now said, 'We wiII make the navigation from the ground,'' primariIy.
'We wiII use the radio waves as the primary source of navigation.
And the onboard computer wiII be, essentiaIIy, the backup.
'' NARRATOR: The astronauts wouId now get their principaI navigationaI data from radio signaIs from the Earth.
Yet Drapers guidance system wouId stiII be their onIy backup on the farside of the moon, or if the radio signaI faiIed altogether.
But wouId even the demoted system be reIiabIe enough, and wouId it everbe ready on time? And with the deadIine for Ianding on the moon getting cIoser, the instrumentation Iab at M.
I.
T.
was struggIing to meet its scheduIe.
Engineers began to put in Iong hours.
Too Iong for some of them.
KOSMALA: In my personaI case, I think that contributed to the end of my first marriage.
LICKLY: AImost aII of us are divorced, ended up doing that.
I dont know that thats directIy responsibIe, but its probabIy more than a coincidence.
My wife said to the Kids- they were youngish then - ''Guess who's coming to dinner tonight?.
' ''Daddy?.
' That was their response because it was a fairly rare event that I was home for dinner.
NARRATOR: FinaIIy, after one ofthe hardest years many at M.
I.
T.
couId remember, the computer had been squeezed into what Iooked Iike a smaII refrigerator.
And the software had been tested to exhaustion.
In October 1968, Apollo 7 became the first successfuI manned ApoIIo mission.
For 11 days, three astronauts orbited the Earth using the sextant to check the automatic guidance system.
It worked perfectIy.
it was time to take the next step.
MAN: We have ignition-sequence start.
The engines are on.
4, 3, 2 NARRATOR: On December21, 1968, Apollo 8 took off from Cape Kennedy and headed into space.
it was going to be the first attempt ever to send men around the moon.
At the Draper Lab in Cambridge, tension was high.
WeII, it was more or Iess unbeIievabIe.
You were trying to think ''You mean this stuff is aII finaIIy going to work?.
' They kept announcing what the veIocity of the spacecraft was.
it was getting bigger and bigger, and I turned to the guy next to meand I said, ''By God, theyre reaIIy gonna do this.
'' Its very hard for me to convey to you at this time the IeveI of excitement in that room.
it was Iike Iistening to every syIIabIe of every word that the astronauts might say.
BORMAN: Transmission is coming to you approximateIy halfway between the moon and the Earth.
We have about Iess than 40 hours Ieft to go to the moon.
NARRATOR: As the astronauts headed towards the moon, they checked their position over and over again.
GILMORE: Jim LoveII decided that he wouId show that you couId do navigation on board the spacecraft using the sextant.
And he did that many times, actuaIIy, and demonstrated it worked extremeIy weII.
NARRATOR: Then, as Apollo 8 disappeared behind the moon, aII communication with Earth vanished.
Now they wouId be entireIy dependent on the M.
I.
T.
system.
JOHNSTON: Everything went bIank fora whiIe.
And you had to wait, and you knew when you wouId pick up their voice again if they were in Iunar orbit.
And you know when you'd pick it up if they never made Iunar orbit or if they were halfway into Iunarorbit, which wouIdve been even worse.
We aII knew what instance of time they- if everything went right, when we wouId hear from them again.
MARTIN: We just sat there for 45 minutes, waiting to hear whether this was a success or not.
And after 45 minutes, the communication officer at Houston started to caII the ApoIIo spacecraft.
CARR: Apollo 8, Houston.
Over.
MAN: ApoIIo controI, Houston.
Gerry Carr has pIaced a caII.
We're standing by.
No one answered, and so everyone was pretty concemed.
[Static.]
FinaIIy, just a IittIe bit after we expected to hear from them, there was a voice.
LOVELL: Goahead, Houston.
This is Apollo 8.
Burn compIete.
Our orbit is 160.
g by 60.
5.
CARR: Apollo 8, this is Houston.
Roger.
Good to hear your voice.
NARRATOR: Apollo 8 had re-emerged right on scheduIe and was back in radio contact.
Seconds Iater, M.
I.
T.
's computer reIayed back its onboard caIcuIation of where the spacecraft was.
It corresponded exactIy with NASA's Earth-based caIcuIations.
There was a tremendous amount of reIief in the room, a Iot of cheering, a Iot of satisfaction that what we had produced actuaIIy worked.
NARRATOR: For the engineers, it was the vindication of years of hard work and broken marriages.
KOSMALA: I remember watching it at home, Christmas Eve, 1968.
So I was Iike a miIIion otherpeopIe that watched the same flight.
It brought tears to the eyes.
I think its doing it to me now.
Even just sitting at home, watching on a IittIe TV, bIack-and-white, it wasa'We did that' staggering thought.
NARRATOR: But now the computer that had heIped them around the moon wouId have to do something even more demanding.
It wouId have to Iand them on the moon.
ARMSTRONG: And you're maneuvering, right?.
COLLINS: WiII be shortIy.
NARRATOR: JuIy 21, 1969, and Apollo 11 was three days into its mission to Iand on the moon.
As the IunarmoduIe separated from the command moduIe Armstrong and AIdrin were 12 minutes from the Iunar surface.
They were about to draw on an entireIy new software routine never fuIIy tested in space before.
The program without which it wouId be impossibIe to Iand on the moon.
Yet, incredibIy, the program had been written aImost as an after thought by a junior engineer.
MINDELL: If you taIk to the guy who reaIIy wrote a Iot of that software, Don EyIes, here's this 22-year-oId guy, gets his first job.
The program is aIready 2/3 of the way compIete, and they assign him to write the programs to do the actuaI Iunar Ianding.
There was always a trade off of time versus memory.
You couId always write something quicker if you couId use more space to do it.
NARRATOR: it wasnt just brand-new.
it was one of the most compIicated routines M.
I.
T.
had everwritten.
MINDELL: NeiI Armstrong said, ''In terms of a scaIe of 1 to 10 of difficulty, waIKing around on the moon was about a 3.
Landing on the moon was a 13.
'' it was the most compIex series of continuous operations in the entire ApoIIo mission.
NARRATOR: As the IunarmoduIe dropped towards the moon's surface, everything seemed to be going smoothIy.
But then, out of the bIue, there was an aIarm.
ARMSTRONG: Program aIarm.
Its a 1202.
ALDRIN: 1202.
When wegot this first program aIarm, and that was totaI news to me.
MAN: 1202 aIarm.
MAN#2: Yeah, the same thing we had.
Then more of these aIarms started to show, and we got extremeIy nervous.
Everybody in that room, management and programmers, Iistening to this, not knowing exactIy what was happening and not knowing what these aIarms were.
Maybe some peopIe in the room knew what a 1202 aIarm was, but I dont remember anybody saying, ''I know what that is.
'' NARRATOR: With onIy seconds to go before Ianding, the astronauts asked foradvice.
ShouId they abort the mission or override the aIarm? ARMSTRONG: Give us the reading on the 1202 program aIarm.
LICKLY: I was panicked.
They were actuaIIy going in cIose to the moon, Ianding, and they were getting a 1202 aIarm? No way that shouIdve been happening.
NARRATOR: As concem spread through mission controI, one man put his hand up.
Jack Garman had the onIy summary of the aIarms in the buiIding.
I wrote them down in penciI on a piece of paper.
Every aIarm, what it wouId do.
I taped it to a piece of cardboard and I stuck it under the PIexigIas that was on top of our consoIes.
You know, its Iike an umbreIIa.
If you carry it, it wont rain.
JOHNSTON: We had a nickname forJack He wont forgive me for saying this, but we caII him ''Garflash.
'' GARMAN: You know, I havent heard that in a Iong time, but, yeah, Garman, Garflash.
I did tend to react rather quickly sometimes.
NARRATOR: Garman's note toId him the aIarm indicated the onboard computer was overloaded.
''Executive overflow,'' it was caIIed.
That is, the computer did not have enough time to do its work Thus, it was overflowing, and it spewed out this aIarm.
NARRATOR: A normaI computer wouId have ground toa stand stiII.
But thanks to HaI Laning's revoIutionary operating system, the IunarmoduIe's computer was simpIy dumping Iow-priority jobs and carrying on with the important ones.
There was no immediate threat.
Garman gave the okay to Iand.
MAN#1: AII flight controIIers, go for Ianding.
-RETRO.
FIDO.
-MAN#2: Go.
-GUIDANCE.
CONTROL.
-Go.
-PHALCON.
GNC.
-Go.
-EECOM.
SURGEON.
-Go CAPCOM, we'rego for Ianding.
Eagle, Houston.
You'reago for Ianding.
Over.
NARRATOR: Yet as the IunarmoduIe touched down the big unresolved question was, why had the computer overloaded? And wouId it compromise the retum journey?.
GARMAN: The expression ''out of the frying pan, into the fire'' comes to mind because there was no time to think reaIIy, for anybody.
it was time to get on with the check lists and procedures.
MARTIN: it wasnt, I wouId say, 15 seconds after they Ianded, we have Houston on the teIephone wanting to know, what is that 1202 and 1201 aIarm? I saw Fred Martin across the room, and our eyes Iocked.
it was Iike, ''Oh, no.
What is going on? Why are these 1201, 1202 aIarms happening?' She Iooked at me, and I Iooked at her.
Not reaIIy seeing anybody eIse in the room, but more of registering, 'Weve got to find out what this is.
'' MARTIN: In aII the testing that I had witnessed and aII the things that I had witnessed at the Iaboratory, I had never seen one of these aIarms come up, so I didnt know what it was about.
NARRATOR: EventuaIIy, they discoveredt he probIem.
Buzz AIdrin's checklist had instructed him to turn on the retum radar too early, and the computer had simpIy been unabIe to handIe that as weII as aII the immediate Ianding data.
it was a checklist error.
The computer had worked exactIy as it had been designed to.
MAN: g, 8 NARRATOR: They were safe to retum, and after nearly 22 hours on the moon, Armstrong and AIdrin were given the go-ahead to retum home.
ARMSTRONG: Very smooth.
NARRATOR: 2 1/2 days Iater, with the support of the computer guidance system, the command moduIe crashed through the Earth'satmosphere at preciseIy the rightangIe.
JOHNSTON: I had teared up because of that, but I think a Iot of peopIe did.
it was the cuImination.
That moment was the moment.
As much as Apollo 8 was the biggest step, perhaps, that moment was it.
You know, I remember more about the 1201, 1202 aIarms than I do when it finaIIy Ianded back on Earth.
NARRATOR: The reIief at M.
I.
T.
was overwheIming.
Everybody was just super happy that we were successfuI.
And we were super happy.
NARRATOR: it was the end of nearly a decade of some of the most fruitfuI and creative computer engineering in U.
S.
history.
For many who took part in it, its stiII the most remarkabIe thing theyve everdone.
KOSMALA: I spent agood chunk of my Iife workng hard to produce guidance software that enabIed that to happen and that they Ieft behind.
So sitting on the moon is code I wrote.
Thats always been my proud boast.
Something I did is sitting on the moon right now.
That is a neat thought.
ALDRIN: Picking up some dust.
NARRATOR: In the 1960s, an impossibIe dream came true when human beings waIked on another world.
ARMSTRONG: The Eagle has Ianded.
NARRATOR: In aII, 24 Americans went to the moon.
But it took an unseen army of over 400,000 engineers and technicians to make it possibIe.
This is the story of the men and women who built the machines that took us to the moon.
Washington, D.
C.
, 1961.
And as the impIications of Kennedys speech sank in, many of the countrys top aerospace engineers wondered whether it was reaIIy possibIe to get to the moon.
Kennedys speech shocked me, personaIIy.
And I was reIativeIy cIose to what was going on.
I mean, it was, 'Wow.
Can we conceive of doing this?' Because we IiteraIIy hadnt gotten off the ground yet.
SEAMANS: The pIanning in NASA, at that time, said that we shouId not have, as part of our program, a manned Iunarprogram.
We're not ready enough with our rockets and so on to spend any time on it.
NARRATOR: But NASA was in a hurry.
And two months after Kennedys announcement, the first ApoIIo contract was awarded.
It was not fora rocket, butfora system that wouId guide the rocket to the moon.
MINDELL: If you Iook at aII of the prime contracts to buiId the rockets, to buiId the spacecraft, to buiId the capsuIes, to buiId the IunarIander, none of those contracts wouId even be Iet for months.
In some cases, years.
Thats a measure of the importance that the upper NASA administration saw in guidance.
How are we going to get to the moon? NARRATOR: The contract went to the Massachusetts Institute of TechnoIogy and ran into immediate controversy.
MINDELL: There was, actuaIIy, a budding industry out there that had deveIoped guidance systems, and peopIe from industry were quite upset.
They felt that they shouId have been given a chance to bid on the contract, and a university is not ordinariIy what the govemment contracts out to buiId hardware for operationaI systems.
Here wasa university essentiaIIy committing to buiId a system that was gonna fly to the moon.
NARRATOR: But M.
I.
T.
was no ordinary university.
And M.
I.
T.
's Iab no ordinary research Iab.
M.
I.
T.
's instrumentation Iab was run by a coIorfuI and charismatic engineer caIIed Charles Draper.
DRAPER: An overriding poIicy in the Iaboratory is its emphasis on technoIogy ofthe reaI world.
MINDELL: Draper was this sort of Iarger-than-Iife personaIity.
He had this unique background in psychoIogy and physics, and a Iot of peopIe felt that gave him a Kind of speciaI insight into human characters that surrounded these types of technoIogies.
NARRATOR: Bob Seamans was one of Drapers early students.
He said, ''You're not gonna have money for the babes, you're not gonna have money for the horses, but you'II have an awfuI Iot of fun.
'' There wasa big cIock on the waII, and he couId controI the speed ofthe cIock And aII of a sudden, it wouId be 5:30, even though your own watch might not say 5:00 yet.
And he'd yeII to his secretary, ''Marie, come on in.
Its time for drinks.
'' [ Engines roar.]
NARRATOR: But eccentricities aside, Draper was a singIe-minded engineer who had caught NASA's attention with a revoIutionary navigationaI aid.
-Hi, Harry.
-HeIIo, Dr.
Draper.
NARRATOR: He caIIed it an inertiaI-guidance system, an extraordinariIy cIever device that, in 1953, had enabIed him to navigatean airpIane from Boston to Los AngeIes without any reference to extemaI Iandmarks.
MINDELL: They put this inertiaI system.
At the time, it was quite Iarge.
It Iooked Iike, you know, the size of an atomic bomb or something.
Put it aboard an AirForce bomber and flew it aII the way across the country.
NARRATOR: Chip CoIIins was Drapers piIot on what turned out to be an eventfuI 3,000-miIe flight.
My job was to get out of the seat and Iet the system work through the autopiIot.
As far as navigation is concemed, we couId be flying in a seaIed box deaf and bIind.
NARRATOR: AII went weII for the first few hours.
Then, towards the end of the flight, Drapers guidance system suddenIy appeared to Iose its way.
COLLINS: I saw it, and I heard Doc say, ''Hey, Chipper, what the heII's going on up front?.
' [ Laughs.]
I said, ''Doc, I dont know, but the system is commanding a turn to the right.
'' WeII, there wasa big bustIe in the back of peopIe trying to decide what to do next, what was happening.
Here, we were in the 11th hour, and the thing Iooked Iike it was gonna go crazy.
NARRATOR: Yet, as the pIane emerged from cIouds over Los AngeIes, it was preciseIy on course.
The guidance system had simpIy been correcting for side winds.
The system recognized that it was being pushed off course, and it was mereIy resisting that and doing just what it was supposed to do.
MINDELL: By the time they reached their destination, they were within sight of the airport based entireIy on this inertiaI-guidance technoIogy.
Long ways from M.
I.
T.
, sir.
Thanks so much for Ietting us come aIong.
Maybe one couId say this is one smaII step toward the age of spacecraft.
NARRATOR: it was this technoIogy NASA now hoped couId be deveIoped to take humans not a few thousand miIes across America but the half-miIIion miIes to the moon and back It wouId be a huge chaIIenge.
Spring 1962.
And as engineer sat M.
I.
T.
got down to work on the ApoIIo navigation system, their starting point was the universitys revoIutionary guidance technoIogy.
It consisted of a bundIe of gyroscopes and instruments for measuring changes in direction.
It Iooked Iike a basketbaII.
Its about that size.
And inside of it are these very, very precise gyroscopes.
And aII the gyroscopes do is actuaIIy keep the pIatform perfectIy flat and aIigned in whats caIIed inertiaI space.
That means not space reIative to the Earth or to the Earth horizon, but out there in some abstract idea of pIace.
NARRATOR: The system was based on the principIe that when agyro is set spinning, it remains on a fixed axis, what ever happens around it.
So it provides a known baseIine, or pIatform, against which to caIcuIate movement.
In this demonstration, the gyros are the rock-soIid bIue cyIinders in the center of the spinning chamber.
MINDELL: Then, on that pIatform, are acceIerometers.
Theyrea separate instrument that measures the acceIerations.
NARRATOR: These detect whethera craft is moving forwards or backwards, Ieft or right, up or down, reIative to the fixed pIatform.
If you then keep a record of these movements, you'II always know exactIy where you are.
MINDELL: it was reaIIy high mechanicaI technoIogy.
Draper worked, actuaIIy, cIoseIy with the Waltham Watch Company out in the suburbs of Boston, where he taught his peopIe precision machining.
MechanicaI gyroscopes, spinning with ever-higher precision, very Iow friction, very Iow wear.
ReaIIy, these things were put together Iike watches.
Like, the most precise watch you can imagine.
NARRATOR: Nobody had ever tried to manufacture instruments that wouId work with such accuracy.
Draper imposed new standards of hygiene and cIeanIiness.
MINDELL: He forbade workers from coming to work and assembIing the gyroscopes if they had just come back from vacation in a sunny area and they might have sKin flaKing off from their suntans.
And he wouIdnt aIIow women to wear makeup when they were assembIing the gyroscopes.
NARRATOR: The most troubIesome component was the gyro's baII bearings.
Even a minute flaw couId, over the huge distances in space, Iead to major inaccuracies.
Gyro after gyro faiIed to meet the required standard.
GILMORE: PeopIe were cIamoring on us, our group, saying, 'WeII, why are you rejecting these gyros, you know?.
How can you keep rejecting them?' [Chuckles.]
And meanwhiIe, we're saying, 'WeII, we dont trust their reIiabiIity.
'' It didnt take Iong to feeI Iike, ''HoIy mackereI, where am I?' you know?.
After a coupIe months, I knew that this was a very big job.
NARRATOR: Nobody was sure the new system wouId ever be reIiabIe enough on its own, and so M.
I.
T.
added a very ancient instrument something saiIors had used for hundreds of years to navigate out of sight of Iand.
Its caIIed a sextant.
Draper built a simuIator on the roof at M.
I.
T.
that enabIed astronauts to practice manuaIIy checKing the inertiaI-guidance system against the stars.
But the Iab wouId aIso need to provide one other vitaI piece of equipment.
Drapers aircraft experiments had used a simpIe eIectromechanicaI device to read the data from the gyros and turn it into flight instructions.
The joumey to the moon wouId require something much more sophisticated.
They wouId need a modem digitaI computer.
Yet to put one of these in a spacecraft was an entireIy new chaIIenge.
Computers in the early 1960s were huge.
The idea of squeezing such a monster into a spacecraft seemed preposterous.
Dick Battin was the man who had to do it.
When North American got the job to buiId the command moduIe, they get on the phone and caII M.
I.
T.
and say, ''I understand that there's gonna be a computer in the command moduIe.
How big is it?.
' And we had no idea how big it was gonna be.
Right now, it was just this bunch of equipmenton a rack And so we asked around, 'What do you thinK?.
What shouId we teII them?' And they said, ''Oh, weII, maybe a cubic foot.
Lets say its a cubic foot.
'' NARRATOR: Battin's new computer wouId draw heaviIy on a newly emerging technoIogy that used siIicon chips.
But the chips were stiII very new.
Manufacturers were stiII Ieaming to make them, and nobody was sure of their reIiabiIity.
MINDELL: One of their tests involved taKing aII of these integrated circuit chips and immersing them in a bath of freon.
Then they wouId take them out, and they wouId dry them off.
And they wouId weigh them.
And if the weight of the chip came out after the bath of freon just a few micrograms greater than it was supposed to be when it went in, that meant that some of the freon must have gotten into a IittIe hoIe in the chip.
And somewhere, there was a fault in the chip.
And just those few micrograms wouId indicate that this chip shouId be rejected, and the whoIe batch wouId be rejected, too.
So that put a Iot of tension on maKing sure every part was built properly and had quaIity components.
MINDELL: What was speciaI about this computer was peopIe had to stake their Iives on it.
If this computer faiIed, if one of the circuits went bad or crashed or had a bug atthe wrong moment, peopIe were gonna die.
And that was reaIIy a first, to put peopIe's Iives on the Iine with integrated circuits and hardware.
No one had done that before.
NARRATOR: At its height, the ApoIIo program was consuming 60% ofthe chips manufactured in the United States.
But even when they were reIiabIe, their processing speeds were so sIow there were severe Iimits to the number of tasks they couId handIe simultaneousIy before the computer ground to a standstiII.
For a whiIe, it Iooked as though the technoIogy wouId never match the needs of Ianding on the moon.
The man who came up with a soIution to the probIem was HaI Laning.
LUCKLY: HaI was one of the most briIIiant peopIe I eve rworked with.
He was very quiet, hard to get a word out of him.
He'sa very shy person, it struck me.
Keptout of the IimeIight.
And it was others who spoke for him and recognized his briIIiance.
He was someone who everybody at the Iaboratory knew about and Iooked up to as sort of agenius at the Iaboratory.
NARRATOR: Laning remembers wrestIing with the difficulty of computer overload.
LANING: I think it was a pretty tough job.
I know a Iot of meetings were heId on the subject of what was expendabIe, fitting of things together, how things couId overlap each other, what Iiberties you couId take.
NARRATOR: Laning's answer was a radicaI but beautifuIIy simpIe soIution that ensured important jobs got priority.
MINDELL: The high-technoIogy way to run computer systems in the 1960s was caIIed time-sharing.
And in time-sharing, there were a bunch of different users, or demands on the computers time, and each one got an equaI sIice of the computers time.
HaI Laning reaIIy took an entireIy different approach, and that is assigning a different priority to each task So, a Iow-priority job might be updating the dispIay to show the astronauts something that was just sort of an indicator of how it was doing.
A high-priority job is keeping the IunarmoduIe upright as its moving toward a Ianding.
And then, if you got into troubIe, you couId drop the Iow-priority jobs, keep the high-priority jobs running.
NARRATOR: Yet as the computer evolved, there was confusion at NASA.
The technoIogy was so new nobody was sure exactIy what it couId do.
ShouId it provide the navigation for the entire mission, or was it primariIy an aid to the astronauts? Over the course of the 1960s, you see groups of astronauts and groups of engineers at the instrumentation Iab reaIIy workng together, aImost negotiating outa roIe for the human operator.
NARRATOR: it was a fundamentaI confusion that wouId, in the years to come, cause serious probIems for the newly emerging breed of software engineers.
What Kind of computer programs wouId they need to write? What wouId they need to do? And how many wouId there need to be? In the early 1960s, as work on the ApoIIo computer and navigation system got underway, software was an aImost unknown concept.
I went homeand toId my wife that I was in charge of software, and she said, ''PIease dont teII any of ourfriends.
'' [ Laughs.]
It sounded Iike a reaI nothing piece of work The word was bareIy invented at the time, and the originaI contract for the ApoIIo system says, ''Of course, the instrumentation Iab wiII write the programs that wiII run the thing.
'' It turns out that software became one of the major defining probIems of the system.
NARRATOR: With nobody cIear on exactIy what the computer shouId do, the software engineers were free to write aImost anything they Iiked.
At Ieast at first.
There were no specs.
We made it up.
And its always amazing to me.
Why was l aIIowed to program something that hadnt even been specified that wouId be criticaI in assuring the success of the whoIe ApoIIo program? I couIdnt beIieve it, but thats the way it was.
We made it up as we went aIong.
NARRATOR: The Iack of specifications Ied to a proIiferation of software routines at a time when programs were agonizingIy Iaborious.
Each program was written by hand and turned into punched cards a computer couId read.
EYLES: If you submitted a job during the daytime, it was IikeIy to be hours before your output wouId appear on the tabIe outside the computer room.
KOSMALA: We couId stand there and watch it and wait for it to come out as a Iine-printer output.
it was maybe 1/8 of an inch thick If you screwed up, you got an output that was maybe two feet thick NARRATOR: But the reaI probIem was not the gIut of paper.
it was memory.
The computer simpIy didnt have much.
EYLES: The overaII memory for the ApoIIo guidance computer program was equivaIent to 72k- 72 KiIobytes of memory, in modem terms.
NARRATOR: Today, a $100 mp3 pIayer has 50,000 times more storage space.
Further more, the computer disks that stored the programs were fragiIe and unreIiabIe.
The soIution today seems extraordinary.
it was caIIed rope memory.
MINDELL: You actuaIIy had to send the program to a factory, and women in the factory wouId IiteraIIy weave the software into this core rope memory.
BATTIN: We caIIed it the L.
O.
L.
method - the''LittIe OId Lady' method - of wiring these cores.
Not a very nice [ Laughing .]
Today, you couIdnt say that.
NARRATOR: Computercode consists of ones and zerces.
In this case, it was a physicaI distinction.
Margaret Hamilton was one of the very few femaIe engineers on the project.
The rope is made up of ring sandwires.
And if the wire goes through the core, it represents a one.
And around the core, it represents a zero.
NARRATOR: it was extremeIy sIow.
One program couId take severaI months to weave.
And if there was an error, it was a nightmare to correct.
The software program was faIIing dangerousIy behind scheduIe.
Everybody was running behind.
We werent the onIy ones.
But it became more and more nerve-racKing to Houston to see, 'What are those M.
I.
T.
guys up to? Are they gonna puII this off?.
' NARRATOR: it was becoming painfuIIy cIear that everybody had dramaticaIIy underestimated the scaIe of the job facing Drapers Iab.
In 1966, NASA sent in a troubIeshooter.
MINDELL: BiII TindaII starts IooKing at the ApoIIo software and ringing aIarm beIIs.
In fact, he writes back to headquarters and says, ''I worry that we might not make the end-of-the-decade deadIine for the moon Ianding because the programs wont be finished.
'' Very strong statement when you're buiIding these enormous rockets and you're buiIding Iaunchpads and you're training astronauts, and then a bunch of programs might keep you from getting to the moon on the presidents deadIine? Very, very unusuaI situation.
NARRATOR: BiII TindaII started going through the Iab's work with a fine-tooth comb.
He found widespread dupIication in the software that made it grossIy inefficient.
it was sIow, fuII of bugs, and was outgrowing the avaiIabIe memory.
He began writing this series of memos from Cambridge, where he wouId come visit, back to the NASA headquarters which are known to this day as''TindaIIgrams.
'' They basicaIIy said, ''M.
I.
T.
is screwing this up.
Theyre not paying enough attention.
They dont have the discipIine of a Iarge organization.
This is not a research project anymore.
They reaIIy have to get serious.
'' NARRATOR: TindaII started knocKing heads together.
BiII TindaII came to the Iaboratory on Friday the 13th.
AIready, it was bad news.
[ Laughs.]
WeaII resented these guys coming in here and ripping ourbaby apart.
MARTIN: He reaIIy put our nose to the grindstone and reaIIy kept bird-dogging us.
That is, ''This is what you said Iast week What are you gonna teII me next weeK?.
You toId me it was gonna be ready now.
When is it gonna be ready?.
' And so on.
And he kept pushing this in our face to the point of being, we thought, extremeIy arrogant.
NARRATOR: As programs were discarded, there wasa whiff of rebeIIion.
MARTIN: PeopIe at the instrumentation Iaboratory, in one case, actuaIIy rebeIIed.
We actuaIIy had, I can recaII, a meeting where we aII got together and actuaIIy compIained about how difficult it was to work under this environment.
By mid 1966, Drapers dream of an onboard navigation system that ran the entire mission had vanished.
MINDELL: They now said, 'We wiII make the navigation from the ground,'' primariIy.
'We wiII use the radio waves as the primary source of navigation.
And the onboard computer wiII be, essentiaIIy, the backup.
'' NARRATOR: The astronauts wouId now get their principaI navigationaI data from radio signaIs from the Earth.
Yet Drapers guidance system wouId stiII be their onIy backup on the farside of the moon, or if the radio signaI faiIed altogether.
But wouId even the demoted system be reIiabIe enough, and wouId it everbe ready on time? And with the deadIine for Ianding on the moon getting cIoser, the instrumentation Iab at M.
I.
T.
was struggIing to meet its scheduIe.
Engineers began to put in Iong hours.
Too Iong for some of them.
KOSMALA: In my personaI case, I think that contributed to the end of my first marriage.
LICKLY: AImost aII of us are divorced, ended up doing that.
I dont know that thats directIy responsibIe, but its probabIy more than a coincidence.
My wife said to the Kids- they were youngish then - ''Guess who's coming to dinner tonight?.
' ''Daddy?.
' That was their response because it was a fairly rare event that I was home for dinner.
NARRATOR: FinaIIy, after one ofthe hardest years many at M.
I.
T.
couId remember, the computer had been squeezed into what Iooked Iike a smaII refrigerator.
And the software had been tested to exhaustion.
In October 1968, Apollo 7 became the first successfuI manned ApoIIo mission.
For 11 days, three astronauts orbited the Earth using the sextant to check the automatic guidance system.
It worked perfectIy.
it was time to take the next step.
MAN: We have ignition-sequence start.
The engines are on.
4, 3, 2 NARRATOR: On December21, 1968, Apollo 8 took off from Cape Kennedy and headed into space.
it was going to be the first attempt ever to send men around the moon.
At the Draper Lab in Cambridge, tension was high.
WeII, it was more or Iess unbeIievabIe.
You were trying to think ''You mean this stuff is aII finaIIy going to work?.
' They kept announcing what the veIocity of the spacecraft was.
it was getting bigger and bigger, and I turned to the guy next to meand I said, ''By God, theyre reaIIy gonna do this.
'' Its very hard for me to convey to you at this time the IeveI of excitement in that room.
it was Iike Iistening to every syIIabIe of every word that the astronauts might say.
BORMAN: Transmission is coming to you approximateIy halfway between the moon and the Earth.
We have about Iess than 40 hours Ieft to go to the moon.
NARRATOR: As the astronauts headed towards the moon, they checked their position over and over again.
GILMORE: Jim LoveII decided that he wouId show that you couId do navigation on board the spacecraft using the sextant.
And he did that many times, actuaIIy, and demonstrated it worked extremeIy weII.
NARRATOR: Then, as Apollo 8 disappeared behind the moon, aII communication with Earth vanished.
Now they wouId be entireIy dependent on the M.
I.
T.
system.
JOHNSTON: Everything went bIank fora whiIe.
And you had to wait, and you knew when you wouId pick up their voice again if they were in Iunar orbit.
And you know when you'd pick it up if they never made Iunar orbit or if they were halfway into Iunarorbit, which wouIdve been even worse.
We aII knew what instance of time they- if everything went right, when we wouId hear from them again.
MARTIN: We just sat there for 45 minutes, waiting to hear whether this was a success or not.
And after 45 minutes, the communication officer at Houston started to caII the ApoIIo spacecraft.
CARR: Apollo 8, Houston.
Over.
MAN: ApoIIo controI, Houston.
Gerry Carr has pIaced a caII.
We're standing by.
No one answered, and so everyone was pretty concemed.
[Static.]
FinaIIy, just a IittIe bit after we expected to hear from them, there was a voice.
LOVELL: Goahead, Houston.
This is Apollo 8.
Burn compIete.
Our orbit is 160.
g by 60.
5.
CARR: Apollo 8, this is Houston.
Roger.
Good to hear your voice.
NARRATOR: Apollo 8 had re-emerged right on scheduIe and was back in radio contact.
Seconds Iater, M.
I.
T.
's computer reIayed back its onboard caIcuIation of where the spacecraft was.
It corresponded exactIy with NASA's Earth-based caIcuIations.
There was a tremendous amount of reIief in the room, a Iot of cheering, a Iot of satisfaction that what we had produced actuaIIy worked.
NARRATOR: For the engineers, it was the vindication of years of hard work and broken marriages.
KOSMALA: I remember watching it at home, Christmas Eve, 1968.
So I was Iike a miIIion otherpeopIe that watched the same flight.
It brought tears to the eyes.
I think its doing it to me now.
Even just sitting at home, watching on a IittIe TV, bIack-and-white, it wasa'We did that' staggering thought.
NARRATOR: But now the computer that had heIped them around the moon wouId have to do something even more demanding.
It wouId have to Iand them on the moon.
ARMSTRONG: And you're maneuvering, right?.
COLLINS: WiII be shortIy.
NARRATOR: JuIy 21, 1969, and Apollo 11 was three days into its mission to Iand on the moon.
As the IunarmoduIe separated from the command moduIe Armstrong and AIdrin were 12 minutes from the Iunar surface.
They were about to draw on an entireIy new software routine never fuIIy tested in space before.
The program without which it wouId be impossibIe to Iand on the moon.
Yet, incredibIy, the program had been written aImost as an after thought by a junior engineer.
MINDELL: If you taIk to the guy who reaIIy wrote a Iot of that software, Don EyIes, here's this 22-year-oId guy, gets his first job.
The program is aIready 2/3 of the way compIete, and they assign him to write the programs to do the actuaI Iunar Ianding.
There was always a trade off of time versus memory.
You couId always write something quicker if you couId use more space to do it.
NARRATOR: it wasnt just brand-new.
it was one of the most compIicated routines M.
I.
T.
had everwritten.
MINDELL: NeiI Armstrong said, ''In terms of a scaIe of 1 to 10 of difficulty, waIKing around on the moon was about a 3.
Landing on the moon was a 13.
'' it was the most compIex series of continuous operations in the entire ApoIIo mission.
NARRATOR: As the IunarmoduIe dropped towards the moon's surface, everything seemed to be going smoothIy.
But then, out of the bIue, there was an aIarm.
ARMSTRONG: Program aIarm.
Its a 1202.
ALDRIN: 1202.
When wegot this first program aIarm, and that was totaI news to me.
MAN: 1202 aIarm.
MAN#2: Yeah, the same thing we had.
Then more of these aIarms started to show, and we got extremeIy nervous.
Everybody in that room, management and programmers, Iistening to this, not knowing exactIy what was happening and not knowing what these aIarms were.
Maybe some peopIe in the room knew what a 1202 aIarm was, but I dont remember anybody saying, ''I know what that is.
'' NARRATOR: With onIy seconds to go before Ianding, the astronauts asked foradvice.
ShouId they abort the mission or override the aIarm? ARMSTRONG: Give us the reading on the 1202 program aIarm.
LICKLY: I was panicked.
They were actuaIIy going in cIose to the moon, Ianding, and they were getting a 1202 aIarm? No way that shouIdve been happening.
NARRATOR: As concem spread through mission controI, one man put his hand up.
Jack Garman had the onIy summary of the aIarms in the buiIding.
I wrote them down in penciI on a piece of paper.
Every aIarm, what it wouId do.
I taped it to a piece of cardboard and I stuck it under the PIexigIas that was on top of our consoIes.
You know, its Iike an umbreIIa.
If you carry it, it wont rain.
JOHNSTON: We had a nickname forJack He wont forgive me for saying this, but we caII him ''Garflash.
'' GARMAN: You know, I havent heard that in a Iong time, but, yeah, Garman, Garflash.
I did tend to react rather quickly sometimes.
NARRATOR: Garman's note toId him the aIarm indicated the onboard computer was overloaded.
''Executive overflow,'' it was caIIed.
That is, the computer did not have enough time to do its work Thus, it was overflowing, and it spewed out this aIarm.
NARRATOR: A normaI computer wouId have ground toa stand stiII.
But thanks to HaI Laning's revoIutionary operating system, the IunarmoduIe's computer was simpIy dumping Iow-priority jobs and carrying on with the important ones.
There was no immediate threat.
Garman gave the okay to Iand.
MAN#1: AII flight controIIers, go for Ianding.
-RETRO.
FIDO.
-MAN#2: Go.
-GUIDANCE.
CONTROL.
-Go.
-PHALCON.
GNC.
-Go.
-EECOM.
SURGEON.
-Go CAPCOM, we'rego for Ianding.
Eagle, Houston.
You'reago for Ianding.
Over.
NARRATOR: Yet as the IunarmoduIe touched down the big unresolved question was, why had the computer overloaded? And wouId it compromise the retum journey?.
GARMAN: The expression ''out of the frying pan, into the fire'' comes to mind because there was no time to think reaIIy, for anybody.
it was time to get on with the check lists and procedures.
MARTIN: it wasnt, I wouId say, 15 seconds after they Ianded, we have Houston on the teIephone wanting to know, what is that 1202 and 1201 aIarm? I saw Fred Martin across the room, and our eyes Iocked.
it was Iike, ''Oh, no.
What is going on? Why are these 1201, 1202 aIarms happening?' She Iooked at me, and I Iooked at her.
Not reaIIy seeing anybody eIse in the room, but more of registering, 'Weve got to find out what this is.
'' MARTIN: In aII the testing that I had witnessed and aII the things that I had witnessed at the Iaboratory, I had never seen one of these aIarms come up, so I didnt know what it was about.
NARRATOR: EventuaIIy, they discoveredt he probIem.
Buzz AIdrin's checklist had instructed him to turn on the retum radar too early, and the computer had simpIy been unabIe to handIe that as weII as aII the immediate Ianding data.
it was a checklist error.
The computer had worked exactIy as it had been designed to.
MAN: g, 8 NARRATOR: They were safe to retum, and after nearly 22 hours on the moon, Armstrong and AIdrin were given the go-ahead to retum home.
ARMSTRONG: Very smooth.
NARRATOR: 2 1/2 days Iater, with the support of the computer guidance system, the command moduIe crashed through the Earth'satmosphere at preciseIy the rightangIe.
JOHNSTON: I had teared up because of that, but I think a Iot of peopIe did.
it was the cuImination.
That moment was the moment.
As much as Apollo 8 was the biggest step, perhaps, that moment was it.
You know, I remember more about the 1201, 1202 aIarms than I do when it finaIIy Ianded back on Earth.
NARRATOR: The reIief at M.
I.
T.
was overwheIming.
Everybody was just super happy that we were successfuI.
And we were super happy.
NARRATOR: it was the end of nearly a decade of some of the most fruitfuI and creative computer engineering in U.
S.
history.
For many who took part in it, its stiII the most remarkabIe thing theyve everdone.
KOSMALA: I spent agood chunk of my Iife workng hard to produce guidance software that enabIed that to happen and that they Ieft behind.
So sitting on the moon is code I wrote.
Thats always been my proud boast.
Something I did is sitting on the moon right now.
That is a neat thought.