Common Lisp is a great choice for many purposes. (And, if you're doing a startup, a fringe language like CL is a good way to find and attract some of the best

Common Lisp is a great choice for many purposes. (And, if you're doing a startup, a fringe language like CL is a good way to find and attract some of the best hackers, and avoid all the Leetcode grunts.) Just comments on the Scheme/Racket parts... > I looked at Scheme, but that community seemed to still be fractured over the r6rs/r7rs mess. Things were fractured before R6RS, with very little portable ecosystem code, and then R6RS didn't solve that, but AFAIK, people got back up on that horse, started over, and have been embracing R7RS. > Also, there wasn't a big enough ecosystem to suit my liking. There's a reasonably-sized ecosystem, but three things: 1. Unless things have changed recently, much of the good stuff is still in the package system for a particular implementation (like Racket, "https://pkgs.racket-lang.org/ https://pkgs.racket-lang.org/", or Chicken, "http://eggs.call-cc.org/5/ http://eggs.call-cc.org/5/"). 2. It's nothing the size of Python or JavaScript. 3. Don't believe any claims of "batteries included"; you often have to roll your own basic packages for real-world work. (But this can actually be a blessing, even for startups that have to move fast, depending on what you have to do, and how capable you're willing to rise to be.) > I'd already tried Racket in school and didn't like it. The runtime was a bit slow and bloated for my tastes. Although CS professors are some the greatest friends of Scheme (having designed and made much of it), CS professors can also be the worst enemies for real world use of Scheme. Before Racket, most people who had heard of Scheme knew it only from school, in problem set homework for the dense SICP course, or from whatever pet intro CS textbook their professor wrote. That's a good way to never want to try Scheme again. (And it got worse as many CS departments became optimized Leetcode->FAANG hiring funnels.) Then people never saw Scheme used, nor even described by, real-world programmers, for real-world things. So Racket (PLT Scheme) comes along, and half of the handful of people thinking of Scheme as a language for real things gravitate to Racket, because they are doing some real-world things. And overall it's one of the best programming languages out there. But still, although the Racket CS professors include some great programmers, Racket is determined by CS professors, who do PL and CSE research, and write and teach textbooks. So, Racket's use in intro CS classes seems to perpetuate the tradition of CS professors ensuring that students will never again want to touch Scheme after they get a passing grade for the class.