Programming in 1969 # Programming in 1969
An interview with a pioneer (my mum)

July 8, 2019

![Punch cards [(source)](](2019_07_08_programming_1969/punch_card_box.jpg width="480") !!! Note I conducted this interview with my mother, Marianne Ernerfeldt, in December 2018. This is a slightly abridged English translation. The full interview can be [read in Swedish here](2019_07_08_programmering_1969.html). ## Why did you want to become a programmer? I decided to become a programmer 1965-'66. I had read an article with a picture of a flowchart and I thought "this would suit me". When I graduated high school in 1967 there were no universities teaching programming, but there was a 6-month course in Solna that would become a 12-month course, and it was eligible for student loans. So I applied for that. At the same time SJ (the Swedish state railway company, then a monopoly) advertised for a trainee program with a one year paid training program where you got to learn all the different parts of the SJ operations. SJ had a computer division, so I applied for the SJ program too with the hope of maybe ending up there. However, SJ ended up with 700 applicants for 50 positions, so there was a tough selection process with various tests. And I got in! So because I needed the salary to get my own apartment, I accepted. Incidentally, the SJ management was quite unhappy when they found out that the trainee program brochures had been sent out to both male *and female* students! There weren't many women who got accepted into the program, but we were a few. During the training we went around to all divisions of SJ and learned about everything from the trains and rails to how the communications worked (SJ had its own electricity and phone lines!). After a year of that I got a job, but it was a boring job. Finally in 1969 I heard that SJ was starting an internal training program for programming, so I applied for that. After more tests I and three others started the programming training in 1969. We were two girls and two guys. ![My reference for IBM System/370 from 1976. To the right I have written in ÅÄÖ (Dec. 5B, 7B and 7C), because the Swedish letters were not part of EBCDIC.](2019_07_08_programming_1969/ibm_370_reference.jpg) ## What was the training like? First we visited the SJ computer division and got a rundown of what computers are etc. Then we took classes at IBM who had an "training machine" in a huge building in Stockholm. We were maybe 50-100 people in the same class, but we were split up so that we were eight people in each room. There we watched two TV screens at the front of the classroom. The teacher and his blackboard were broadcast on the screens from another room. Each teacher had maybe ten rooms with students, and each room could ask questions using a microphone and call attention using a button. It was super modern! First we learned a bit about IBM OS, and then we learned PL/I which was IBM's own programming language. It was a more modern version of Cobol with features that Cobol did not yet have (but would get later), like making tables and queries. So PL/I was a much better language back then and much simpler: you could write the code using English words, `DO WHILE` etc. A really nice programming language! It was the government agencies that were at the forefront of data processing in Sweden. Banks and others came far behind. So at the IBM courses it was mostly people from other agencies, but SJ was at the forefront. After the first IBM course I went back to SJ to do my first practice programs. The four of us made a dating program where you would input men and women and their traits, and then generate a good matching between them with an algorithm of our own invention. And after that we started writing production code! Later I took more courses, for instance I learned assembler the same way. So my education was really a week here and a class there, and then we had a supervisor at work that helped us. ![My stencil for drawing flowcharts. Used to visualize how the data would flow and in which order things were to happen.](2019_07_08_programming_1969/flowchart_2.jpg width="480") ## What was your job like? First we would draw flowcharts and then we would write the code using a pencil. We then handed the code to the puncher unit where the code would be punched onto punch cards. The punch cards had 80 columns – 72 for the program and 8 for sequencing – so each line of code could be at most 72 characters wide. You had to write the code clearly so that the women operating the punchers could read it. After a few years as SJ we got someone dedicated to reading our code, and that helped a lot. Otherwise they would mostly punch data cards: time reports from SJ, how far each train carriage had traveled (so they could get called in for service), etc. The punch machine looked like a regular typewriter that would punch holes in the cards. Above each column it would also type the letter in clear text. We also used to serve cake on punch cards, so they were quite versatile. When I first started, the programs were small, but later they could be several meter-long boxes of cards. Each line of code became one punch card. So one instruction per line and card. So the puncher unit would return the program (thousands of cards) to us. We would also have to create "control cards" which encoded whether the punch cards were to be compiled or executed, and what language it was etc. The control cards had a separate color. The first card was a work card with my name on it, so they knew whom to return everything to. We put the punch card boxes on a special table. The operators came wearing their white coats and took the boxes and ran the programs. Sometimes we only got one run per day, because we programmers had the lowest priority in the machine hall. So we had to work on several programs simultaneously to keep busy. Finally the cards were returned together with "pajama paper" containing lists with error codes and line numbers. ![Pajama paper](2019_07_08_programming_1969/pyjamaspapper.jpg width="480") We had access to a couple of hole punches so we could make minor corrections ourselves. Then we had to create test files and see if the program produced the expected result. If not you would sit and "desktop test" (think with pen and paper) and try to figure "what the hell went wrong?". So it could take quite a while to get a program right. We had several machines. We had IBM 360 from the beginning as well as some even older machines. Later on we got IBM 370. Towards the end of the 70's we got terminals. We never had our own terminals, but shared a terminal room. We had to fight over terminal time when we wanted to make changes to a program. The program would flash up on the screen, and we could modify it. We had Alfaskop terminals in yellow and brown. I never got my own terminal before quitting SJ in 1979. ![Alfaskop 3700 [(source)](](2019_07_08_programming_1969/alfaskop_3700.jpg) ## Tell me about your colleagues The computer division at SJ had around 40 programmers and system engineers. All my colleagues were, like me, trained in-house, with a few exceptions: there were a few people the same age as me who had gone the 6 month course in Solna I mentioned earlier. But otherwise there was no other way to learn this – it was a brand new profession! Some colleagues used to be engine drivers! Most did not even have a high school diploma. The system engineers worked mostly with specifications, what the inputs and outputs of the programs were to be etc. As programmers, we were the problem solvers, making flowcharts and figuring out how to solve the task. There were around 10 machine operators in the machine hall. They wore white coats and handled the tape storage, disks and fed the punch cards. We were rarely allowed to enter the inner sanctum of the machine hall. The entrance had a "Closed Shop" sign. The machine hall was big. The earlier machines (IBM 1400) took up 10-20 m², but later machines were small as refrigerators. The punch unit was made up of 50 young women. If there were any visits to the computer section from other parts of SJ it was always the punch unit that was the most interesting, as most people at SJ were men! Of the 40 programmers we were 5 or 6 women. I was part of the young gang, but most programmers were around 10 years older than me (I was born in 1947). Towards the end of the 1970's we got a few new recruits that had done 12-month training courses. Before I started I met Solveig who worked at the SJ computer division. She told me all doors had signs saying "Miss X" or "Mrs. X", but the guys' doors had no corresponding information about marital status. The women had been so pissed off by this that they had unscrewed the signs. So when I started they just looked at me and said "I guess you don't want a Miss/Mrs either?". "No thanks!", I said. ## What kind of things did you do? In the spring of 1969 – before I started – the SJ online booking was launched, with 24/7 uptime. It was extremely advanced for its time, and written entirely in assembler. This was one thing where SJ were really outstanding – there was no other company in Sweden that even came close to us. I was very lucky to get my education at a company that was such a trail blazer! We created the programs, and once they were completed and tested we handed them off. Others were responsible for maintaining them, we just wrote new ones! Mostly it was programs that gathered statistics about the operations (e.g. payroll), and then those programs would be run at regular intervals (e.g. every month). In 1979, after ten years at SJ, I quit to start working in the bank sector instead. ![Marianne at her desk at SJ](2019_07_08_programming_1969/marianne_sj.jpg width="480")