> Unix became popular not because it was Good, but because it was Just Good Enough to entertain the cowboys who built and used it ... and so made cowboy coders

> Unix became popular not because it was Good, but because it was Just Good Enough to entertain the cowboys who built and used it ... and so made cowboy coders the status quo for much of this profession and culture I disagree very strongly with these insults directed at programmers from 50 years ago because now, in retrospect, half a century later what they did doesn't live up to some flawless system written in Rust that exists only in one's imagination. Doesn't is seem a little bit like calling Thomas Edison a cowboy that made the terrible mistake of giving us electric lighting through filament bulbs when LED lights would have been so much better. In these early days of computer science I read virtually every important published article on programming languages and operating systems, the field was still that small. MIT didn't even think it warranted a separate department, it was just a subsidiary branch of EE like say communications. Researchers like Edsger Dijkstra, Tony Hoare, Niklaus Wirth, Per Brinch Hansen, Leslie Lamport, David Gries, Donald Knuth, Barbara Liskov, and David Parnas were all trying to figure out how to structure an operating system, how to verify that a program was correct, how to solve basic problems of distributed systems, and how to design better programming languages. Practitioners working on operating systems would have been familiar with almost everything written by these giants. It's easy to insult C, I myself wouldn't choose it for work today. But in 1989, 20 years after the birth of Unix, I did choose it for my company's development of system software--it still made sense. And back in the 1960's what alternatives were there? Fortran? PL/1? Pascal? Lisp? We were still programming on keypunch machines and relational databases hadn't been invented. The real competition back then for system programming was assembly language.