Re Python: It’s fine to be slow considering the huge infrastructure Python has. NumPy & SciPy are fabulous. And don’t forget that, like Bash, Python is designed to wrap around C/C++ easily so Python itself being slow does not interfere with your mathematics being fast.
As a JS programmer, I often have Python Envy! : ) But other than the math libraries which I covet, JS with it’s great “window system” (the browser) and webgl access, async Promises, and web workers etc makes up for the lack of the great Python libraries. And it’s everywhere! It’s in your phone, it runs servers (node), the Electron desktop version builds great apps, .. heck, it’s even in your watch!
And to tell the truth, with all languages having good networking, if I need math, I just build a web-app that uses Python on the back end as a micro-service.
So it’s the old “best tool for the job” idea, and with great networking, we can mash up all the languages, no problem.
.. but I still, sob, lust for numpy in JS. Sigh.