It's somewhat of a chicken and egg problem: if you don't give programmers the ability to choose their tools, you can't hire top talent; if you can't hire top ta

It's somewhat of a chicken and egg problem: if you don't give programmers the ability to choose their tools, you can't hire top talent; if you can't hire top talent, you can't give programmers the ability to choose their tools. Recruiting is also a zero-sum game: if you don't hire the best, your competitors will. That means if technology is an enabling factor in what you do (it doesn't have to be your core competence), empowering programmers is a competitive advantage. I was at CUFP (Commercial Users of Functional Programming) recently where I had a conversation with Yaron Minsky of Jane Street. He mentioned that due to their use of OCaml (and being one of the few places in their industry where software engineers are not treated as second class citizens) they have a much higher quality of not only hires but applicants. He mentioned a job listing on a caml mailing list yielding a fairly small number of resumes, but a much higher percentage of these resumes translated into phone screens, on-sites and finally hires. I am not sure I'm allowed to disclose the percentage, but it's a fairly incredible number compared to what I've seen elsewhere; even the people they passed on, were of high quality. They are effectively choosing "the best of the best". If someone is willing to learn spend the time to learn a non-mainstream language (this used to be Ruby and Python in 2004, but now they're Haskell/OCaml/Scala, CL/Scheme/Clojure, Erlang; this may change again, I see at least one of Scala, Clojure or Erlang becoming mainstream) -- and I am talking about actually learn, beyond "fizz buzz" -- chances are they have the passion and the work ethic to be a good programmer. Of course this isn't a bijective map: there are good programmers who aren't interested in non-mainstream languages (in many cases they have highly specific knowledge e.g., machine learning where the language used doesn't matter as much), but I've yet to see someone take the time to learn Haskell on their own (without being forced to do so for a university course) who wasn't a good programmer: they may not be best of the best or a good fit for every position, but they are certainly not the kind of person you're going to regret bringing on-site to interview.