Well, I guess that’s where we disagree. But I’m obviously biased and it’s understandable that to someone looking in from the outside it appears like “just another language”. I’m loathe to link to my https://www.youtube.com/watch?v=kc9HwsxE1OY video yet again, but it’s still the best explanation around for why something new and different is going on here. The evidence is an ecosystem where program composition and code reuse actually works and happens on a regular basis. That doesn’t have anything specifically to do with building a CAS, but it’s why so much powerful software can be built with relatively little time and effort: ubiquitous, fast multiple dispatch allows totally independent software components to be composed into programs that do more than the sum of their parts. Yes, Common Lisp has opt-in multiple dispatch. And no, that’s not good enough. It’s not transformative unless it’s ubiquitous and in Common Lisp it’s relatively slow and opt-in, both of which stop it from being ubiquitous. It also doesn’t work with CLOS parametric types, which means that you can’t use it where you especially need it for numerical work.
In any case, I want to thank you for your thoughts on this. I think the only way we’re going to see how this plays out is to try it and see what happens.