John Backus, the Turing Award-winning inventor of FORTRAN — arguably the world’s first high-level programming language — has died at age 82.
A lot of programs of a certain age started their careers with Fortran. I am just a young’un compared to those guys, who had to write programs on punched cards and walk uphill to the mainframe . . . both ways . . . in the snow . . . with no shoes.
But I did give technical support to MATLAB users who needed to integrate their “legacy” Fortran code with MATLAB. I didn’t know Fortran, but that usually wasn’t a problem because no one was really programming much of anything new in it in 1998. So it was mostly a matter of coaxing the users’ compilers to do the right thing.
So, I’m not the right person to eulogize Backus or lionize his contribution to computing, but I’ll give a couple reminiscences about Fortran.
For a long time Fortran had funny syntax rules where the indentation of your code mattered for program compilation. That cracked me up when I first learned about it.
Years ago I was talking to Cleve Moler and asked for some pointers about learning Fortran. (That’s the kind of place I work at: You can just casually talk to the company founders.) Cleve, ever the straightforward kind of guy, asked, “What do you want to do that for?” Cleve first wrote MATLAB in order to shield his engineering students from needing to learn Fortran to do numerical analysis, so that response makes a bit of sense and shouldn’t be a reflection on Fortran, per se.
I wish I had more complementary things to say about Fortran, but I give Backus great kudos for it.
However, I can say that I owe him a great debt for co-developing BNF, that especially pedantic way of expressing a grammar which is so often used by languages and file formats of a certain age.
“You need the willingness to fail all the time. You have to generate many ideas and then you have to work very hard only to discover that they don’t work. And you keep doing that over and over until you find one that does work.”