Estimated reading time: 7–8 minutes
“My dissertation: A Comparative Analysis of the Languages ALGOL, FORTRAN, COBOL and PL/I.”
When I saw the list of dissertation titles, I did not hesitate for a moment.
One subject leapt out at me, as though it had been written for me:
“A Comparative Analysis of the Languages ALGOL, FORTRAN, COBOL and PL/I.”
The supervisor was Conf. Dr. Dragoș Vaida, a well-known name with international visibility, who was known to lecture at the University of Bucharest.
He also held classes at the Cybernetics department at ASE, however.
Much later I learnt that he had been involved in the reorganisation of economic computer science education at ASE.
For me, to work for six months under the supervision of Dragoș Vaida was not merely an honour, but an opportunity I had no intention of missing.
I was already in love with programming.
This project gave a clear sense of purpose to my passion.
We met roughly once a month.
I would go to him, at the Computing Centre of the University of Bucharest, which operated in a discreet building on Strada Ștefan Furtună.
He had been appointed its director, and one could sense immediately that this space was far more than a mere place of work for him — it was a miniature factory of ideas, a zone in which theoretical plans took on flesh and bone in real code, run on real machines.
I can no longer recall the exact contents of the dissertation — too many years have passed and I kept no copy.
But I know that I worked for nearly six months on the IBM 360 at ASE, where I ran hundreds of lines of code in FORTRAN and COBOL, analysed the syntax of ALGOL and tried to make sense of the ponderous but promising logic of PL/I.
I know that I learnt an enormous amount during that time.
Professor Vaida possessed a rare quality: he did not try to impress through erudition (though he could have done so at any moment).
He preferred to push things towards the concrete.
He encouraged me to leave the theories neatly filed away in a drawer and to concentrate on what I could verify: how one writes a real programme, how one compares two compilers, how one measures efficiency.
It was evident that he excelled in theory, but he had the wisdom not to lose me in the labyrinth of definitions.
He challenged me to see the real differences between these languages, not only in their syntax, but in the philosophy behind them.
Looking back, I believe I would have written a much better dissertation had I had three to five years of experience behind me.
At the time, I was learning as I wrote. I was making comparisons as I discovered the concepts.
But perhaps that is precisely why I learnt so much about programming in those six months.
The dissertation was not a crowning achievement.
It was, in truth, a training laboratory, and those months set me definitively on the course of my profession.
And just when I thought everything was heading towards an orderly conclusion, the bombshell arrived: in addition to the dissertation, there was to be an oral synthesis examination in economics, covering every subject with an economic profile from all the years of study.
Incredible!
In other words, everything: from general accounting to socialist planning and the modelling of financial flows.
ASE would not let us leave its broad embrace so easily.
Even though we had specialised in automatic computation and algorithms, we had to prove that we could still speak the language of macroeconomics.
But we got through that as well.
And to conclude, that dissertation taught me to understand why some ideas take hold, whilst others remain merely beautiful dreams written in textbooks.
Indeed, FORTRAN and COBOL each had its own backing — FORTRAN from IBM, and COBOL from the American government and the major corporations.
But that was not the sole reason these languages dominated the computing world well into the 1990s.
The reason was, I believe, that they were languages in which the code you wrote produced correct and repeatable numerical results, and in a world without modern debugging, without IDEs, without Stack Overflow, predictability was gold.
They were not necessarily “beautiful” languages. They were clear, and above all, auditable.
ALGOL left its mark in silence, in the structures of modern languages.
PL/I aspired to be everything and ended up being nothing.