Almost 20 years ago. Was Python even a teen?
Python does not attract smart young iconoclasts anymore. It's the scripting language everyone writes because everyone writes Python.
What languages do attract smart young iconoclasts? Nim? Elm? Whatever you're doing that is not yet old enough to drink by the time you comment?
Interesting how it started to attract people with the same mindset that likes Java: great care for trivial correctness guarantees s.a. those ensured through type hints or producing programs through ticking boxes in existing templates. And disregard to large-scale design issues, something that was initially thought to be helped by more concise syntax and greater freedom to explore / less fear of novelty.
However, the thought that dedication to a (marginal) language will help you hire better programmers isn't new, and has some supporting evidence. I remember similar argument coming from Linus Torvalds, when he didn't want C++ programmers working on Git, for example.
I also happen to think that a lot of new(-ish) languages today are trying to ride the hype wave, and by doing so are trying to become instantly more popular rather than appealing to the tastes of a small but dedicated community. Eg. Rust is trying to be a "better C++", but in a way C++ programmers would most likely not be repulsed by it. Both superficial decision (s.a. "familiar" curly bracket syntax) and more fundamental ones are there to ensure that large audience of C++ programmers would have an easy migration path.
I think that, today, if you wanted to match the effect the author saw created by hiring Python programmers in 2004, you'd have to go after hiring people dedicated to undeservedly forgotten languages s.a. Ada, Scheme or Erlang. Rust is posed to become just a slightly different flavor of C++ with the large volume of low-skill programmers pumping out low-quality software from big corporate shops. For now, it may attract enthusiastic people, but that's not going to last.
If anyone cares, here's some comments from one of those authors: https://github.com/StefanSalewski/NimProgrammingBook/issues/... . And from the other: https://twitter.com/d0m96/status/1592827547582332929?t=IqKBz... .
These quarrels with Araq differ obviously, but both are pretty discouraging as far as Nim's prospects are concerned.
https://github.com/nim-lang/Nim/issues/9026
Nothing as strong as the above but it definitely rubbed me the wrong way. So much advertising about Nim being efficient/fast and the default way to read a file is incredibly slow and inefficient .... and they don't care.
Almost 10 years ago, I translated this article to Telugu.