> However, working with Clojure and Scheme, I understand the power of repl driven development that is difficult to explain outside the experience of it.

The only lisp I've dabbled with was Racket and I found the repl driven development to be a frustrating necessity. I would spend a lot of time trying to figure out the actual type of the thing that I a given function (even a stdlib function) needed, and futzing around in the repl seemed to be the fastest way to do it, but it was still quite slow.

It reminded me a lot of my extensive experience with Python where things that are easy in, say, Go, are quite hard. People rave about their repl, but it feels like they're comparing it to a script iteration loop rather than static analysis tooling.

That said, I completely buy the argument that Python's REPL is particularly bad and that there are other use cases where REPL-driven development shines. My mind is open, but these are the sorts of experiences the pro-REPL folk should be prepared to engage with in their evangelism. :)