It was the 70's and I continued working as a programmer for the Neuroscience Department using the laboratory computer, which in other articles I have explained that it only had 32 Kilo Words of 12 bits each, a typewriter that could punch 8-bit paper tapes. channels (bits), and an 8-channel paper tape reader.

I have also commented that I programmed coding "by hand", and as soon as it was available in the machine language assembler, and also in the LEAL language, which was "postfix".

Then think about my joy and amazement when I was able to "get hold of" a language like FORTRAN!

FORTRAN comes from "FORmula TRANslating" and was in fact the first high-level language in history, developed by the genius of John Backus and his great team, back in the years 1953 to 1957. Highlighted mention to the genius of Grace Hopper who developed the first compiler in history, the A-0.

I could now program all kinds of scientific calculations! But remember how it worked:
After thinking and scribbling papers, I started to program directly in FORTRAN, also on paper, and when I was "almost certain" of "almost everything" then I wrote by hand character by character on specially designed papers to later be able to "punch" paper tapes or punch cards. The paper tapes were continuous, so any errors had to be corrected by hand, covering or opening little holes, or cutting and tying pieces of tape, where each little hole represents one bit. Each punched card corresponded to a line of code or data, so to correct errors only a new card had to be punched. Read "How we developed the ANFACT" so that you can see what can happen with these technologies ...

But hey, the use of FORTRAN accelerated our research results and allowed us to make giant leaps in the application of mathematics and statistics in the Neurosciences, all despite the 32 Kilo Words, or rather, because of that, that is, the technological limitations were so many and so varied that the inventiveness in programming had to, and in fact did, overcome everything and achieve programs that ran on several computers at different times, to then join results and run again on several computers, and so on until the final results are obtained. It was frequent that a program, or part of it, was "running" 24, 48, and 72+ hours ...

FORTRAN really was, and is, a wonderful language where, by the way, it was programmed with GOTO and that is why sometimes it was very difficult to finish without errors and giving the correct results! But the programmers were so free to think and do what that we wanted to! Until Dijkstra and his "Goto considered harmful" ...

Octavio Báez Hidalgo.

No thoughts on “FORTRAN. The best for calculus.”

Leave your comment

In reply to Some User