When I joined the COBOL development department in 1970, I saw how the nightly builds of the compiler ran. The COBOL compiler comprised many assembly language modules. The developers would keypunch their corrections and additions to the code on IBM cards and put their cards into a big deck of cards in a metal tray that would be sent by a courier truck to our main building about a mile away. This went out at the end of each day and returned the next morning. We hoped the build would succeed and the test programs would compile and run.
The latest code for the compiler was stored on an 800 Bits Per Inch (BPI) tape that had ten cards of 80 characters per block, a Master History Tape (MHT). The blocks were separated by an inch of blank tape. The control card they used for multiple years had a parameter on it “CIT” for Card Image Tape. It should have been MHT. Instead of storing the current code merged with the new code on an MHT, they were writing out the temporary update tape as a ten times longer and slower blocked-at-one CIT. This was not only slower, but when many updates were made, the temporary 2400-foot CIT could overflow and ruin the nightly run.
I saw this error in the control card and pointed out the problem. The other developers were reluctant to follow my advice because I just got there, and they had been doing the updates that way for years. They tried my advice and it worked!
This is an example of how we did software development in the old days of batch processing before interactive interfaces, with tapes before disks, and this was the computer technology used to get us to the moon.
Six years earlier, I asked Neil Armstrong if NASA planned to have computer games to entertain the astronauts on long missions. He laughed, probably thinking of how primitive their computers were. My high school science project was writing a program that taught a computer how to play the game of Mancala. Part of my awards for that project was to attend the National Youth Science Camp in West Virginia, where I met our special guest, Neil Armstrong, five years before he went to the moon.