Jeff writes Second they never covered the what happens after we use the punch card in Computer Science classes.
I used 'em for bookmarks. Oh ... remember books? Anyway, lots and lots and lots and lots and lots and lots and lots and lots and lots and lots and lots and lots and lots and lots and lots and lots and lots and lots of punchcard bookmarks ... until I was getting smothered in bookmarks. Well, I'm exaggerating. But anyway, by that time, I was sick of punchcard bookmarks, and fortunately, the data-processing world was sick of punchcards. Now I'm besieged by old floppy disks...
Well, punch cards were harder to use in all respects, including finding mistakes. But you would have to be more careful from the outset for many reasons. So I believe there were truely less mistakes to be found.
One thing punch cards did do for you. You had to get it right the first time. They were hard to correct. Like the typewriter to the MS Word times 10. If your program failed, sometimes, you had to wait days to re-run, this would force you to be more careful.
They were re-using, alright -- but think of all the trees they killed in the first place. Whenever I hear "punch card" an immediate vision of all the CS and engineering students scurrying around campus with their big stacks of punch cards -- and I attending a big engineering school, so there were tons and tons of 'em. I was always vaguely intriqued and massively relieved that I didn't have to add a punch card stack to eveything I had to schlep across campus.
Databases in general and relational databases in particular are intended to be most useful for knowledge representation in the context of validation -- when a theory has been formulated -- and the logical model representing it in the database can be used to validate it and further analyzed to derive additional implications of a theory.