[tzs]

I bet a large part of that is that there is a good chance that someone writing FORTRAN in the'80s would work out their algorithms, how they would code those algorithms, and how they would structure their code on paper, and have pretty much the whole thing worked out before they ever sat down at a terminal to actually enter code.

When actually entering code, their focus can then be on the fairly straightforward translation of their notes specifically into FORTRAN, and on adding good comments for those who deal with the code later. Most of their creative energy at this stage can go into those comments.

Part of this is that it was still common in the '80s to either not have interactive access to the computer the code was for, or for such access to only be via shared terminals in a computing center away from your office. You needed to arrange things so that when you actually got to a terminal, you were efficient.

Since then, we've almost always got access to our own private computers that are powerful enough to run at least test versions of whatever we are working on even if it is ultimately meant for some big server somewhere else. Now we can sit down and start coding while still designing the program in our head. And so we do, even if sometimes it probably be better to separate design and coding.

I think this also might have something to do with why BASIC was good as a teaching language, as was noted recently in some other HN discussions. BASIC, especially with some of the limits put on it to fit it in some smaller computers, was constraining and painful enough to deal with that people quickly learned that trying to do that while also trying at the same time to figure out the design and algorithms they needed was way too hard. They naturally learned to separate figuring out how to do something from coding it.

[ams6110]

In the 1980s, for FORTRAN, coding your program on punch cards was still not that uncommon.

[james_s_tayler]

I wonder if they were made to do leetcode algorithms on a chalkboard in FORTRAN 77 back in the day during interviews.

[hermitdev]

Back then, you likely didnt have anything resembling a PC or a white board (probably available but less common than today), so it was more likely on a blackboard with chalk and pen or pencil on paper before working on your punch cards. I started my internship at a Fortune 500 in 2001, and they had just retired their last card reader for the mainframe. Still had plenty of blank/partially used cards that made for great notecards.