OT, but not much, the Apollo 11 Guidance Computer code:
that stuff actually made the man land on the moon (and safely back).
Anyway, and just for the record, not only it was common at university, up to the mid/late '70's, to manually punch cards, but you only had the computer available for a very limited time slot, so each bug would usually mean that you had next chance to run your program, hopefully corrected, "next tuesday":
Besides the claims (mostly for fun) of being old, I never had the occasion to work with those, though.
AFAICR the CP/M was still in use in the first years of DOS on "professional" computers, typically accounting ones, as the PC (and hence DOS) was seen initially as a PC (Personal Computer) not something used in actual business, I wouldn't be surprised that a few of those survived through the '80's.
Year 1992 (or maybe 1993, not too sure), on the 6th or 7th of august (it was a friday) the (DOS, on PC) computer program that was used - experimentally - to keep the official accounting of a project site, reached and surpassed the amount of Lit. 9,999,999,999 (10 billions lire) and bombed out.
The programmers/software house that sold the program was immediately contacted (that was the last working friday before they closed three weeks) they said "no problem, as soon as the programmer comes back from holidays on the 31th of August he will be able to solve the issue".
Every technician available (in the construction firm I was working with) was called to manually write the registries, some 10-15 people that worked practically non-stop for the whole weekend frantically writing down the data in order to be able to send/cash the progress report on next monday as expected.