I just finished reading a DevX article on programming in the "Old Days". The article is entitled "Adjust or Go Extinct: 50 Years of Conventional Wisdom and Eye-raising Anecdotes from Programming Veterans".

Inspired by the exploits of the old geezer in the article (Gerald Weinberg) I thought I would relate some of my earliest programming experiences.

I learned the rudimentary of programming on a time-share system in high school in about 1973-74. The computer we connected to was an IBM 360 at the College of William and Mary. We wrote and ran programs in APL (A Programming Language). One of our pastimes was creating tight obsfucated code. I wrote a prime number generator that was only 4 lines of code long. We worked on a remote terminal that had a paper feed - no display. Whatever you typed into the computer ended up on paper, and so did the computer's response.

When I started college, the Engineering School had its own computer. I believe it was an HP 2000. These computers were time shared and ran a rudimentary BASIC programming language. I actually found some information on the models here. Users sat at teletype systems that used paper for the display. In 1978-79, or so, the University swapped out the paper teletypes for DEC CRT terminals. I'm sure this was a cost-cutting measure, because we sure were using a lot of paper.

The byproduct of these DEC CRTs was that they could be programmed to move the cursor arond the screen and place text characters at those positions. Of course, this led a couple of smart individuals to write game programs, because now you had the beginnings of a graphics capability. One very smart guy wrote some very sophisticated D&D and Star Trek games. The Star Trek game gave players the opportunity to play against each other, while the D&D (Dungeons and Dragons) game used a rudimentary artificial intelligence - players played individually against the computer and the opposition characters got more sophisticated and evil as you descended levels.

Besides games, the HP 2000 was very good (at the time) for creating programs to solve engineering and math problems and they could be "saved" to the system. Of course, you had a space limitation. There was even a pretty decent writing program, which saved me a bunch of money. We had to write a senior thesis, and in those days you took your hand-written notes over to a typist that you paid money to for typing your notes into a paper - because few people could afford a decent typewriter. I wrote mine on the computer, and still have a few copies lying around the house.

Of course, to the University, the computers in the Engineering School were just there to support students, and therefore were unimportant. The really important computers were over at the business school, down in the basement. These were called MAINFRAMES. I don't recall the model that was there, but these were not time-sharing systems. They ran one job at a time.

I took a couple of courses on programming languages, and specifically in FORTRAN, that required me to learn to program these beasts using decks, or IBM paper cards that had holes punched in them. One card represented either one line of code, or one record of data. These were fed into the MAINFRAME computer.

Of course, we weren't allowed to touch the computer. It had it's own accolytes in attendance that would feed in our boxes of cards (for large programs you actually kept your deck of cards in a cardboard box) into a metal hopper that would read them into the mainframe for processing. The mainframe would run the program and create a printout of the results. The computer operators would then place the printouts in boxes labeled alphabetically. You got your printout from the box that was labeled with the first initial of your last name.

You dropped off your card deck to be processed (run your program) and sometime, usually hours later, you would get back a printout back - and your card deck. No IO, no storage, no graphics, no individual workstations - at least we had that at the Engineering school.

Barbaric by today's standards, and a step backward from what we had at the Engineering School. But the MAINFRAME was still king. And this was a decade after watching Spock, and other characters, program computers on Star Trek where you could store results or information on a shiny plastic card. There are still companies and organizations using mainframe computers today. But, thankfully, they are few and far between.

Today, you can go to lots of stores and buy a computer, and learn to program on your own computer. But in the early days, when computers (i.e. mainframes) were so expensive, the chance that you would touch a computer was slim to none. There was a whole "cast" system built around the computer, especially the mainframes. Of course, the mainframe computers were hugely expensive in relationship to the cost of personnel to support them, so no one was willing to just let anyone touch them. 

Now days, it's the other way around. As I recall, the breakout came when some manufactures started selling kits for computers that could be owned by individuals back in the late 70s. It was a mind-boggling proposition at the time - to be able to own your own "personal" computer. Couple that concept with the rebellious nature of the times and next thing you know you have Apple computers. IBM did themselves in with their own invention - the IBM PC. And, as they say, the rest is history.