I have to say I have never had to deal with punch cards. At school, in the fifth form, when there were options to learn Fortran and Cobol, I chose car maintenance instead.
The engine block was removed, the valves ground in, the engine block replaced, and everything else was fitted in reverse order to removal. The car was started and ran sweetly.
Meanwhile, the punch cards were sent off to the computing centre, and returned two weeks later with green-lined manuscript paper printoffs which mostly consisted of "syntax error" or "fatal stack overflow in..." or some such thing...The luckier ones got things like printf " hello world " or 2 + 4 = 6...driving an old (resurrected, even) Ford Prefect in tight circles around the quad certainly felt good to me.
I felt that programming a Texas Instrument calculator to do certain physics calculations was rather cool, except we could not take electronic calculators into exams - we had to rely on log tables and slide rules.
At University, computing remained a small department, I think that all the computing hardware was probably off-campus, despite Leeds being one of the largest UK universities outside Oxbridge and London.
Twelve years after graduating, I first used a PC, a Compaq Deskpro with two floppy 5 and a bit inch disks. It was dedicated to a special flatbed laser scanning densitometer that was used to determine protein concentration in two-dimensional separation gels. We programmed in basica, calculating the yields of cDNA and mRNA from our experimental procedures, and plotting standard curves and regression lines using very primitive graphics on screens with limited resolution.
Then someone had a 386 with a colour EGA (or was it CGA) screen, and a GEM graphical environment let me use their system. Something went wrong, and I thought the friendly sounding DOS v3 RESTORE command might just do the trick...
It did: File.001 File.002... it was amazing how many files could be fitted onto a floppy disk, at least 1000, anyhow...
I stuck to molecular Biology for a while longer, but in 1989 I started a Master's course in Information Science, and started to use Borland turbo pascal and Prolog, and some dos-based office called Smart suite, I was writing up my dissertation on Microsoft Word, and using Windows 2 Pbrush to illustrate my work, moonlighting in the computer labs in the evening on Apple Macintosh versions of Word.
The work had to get done somehow...