Python wins the triathlon by being second best in all the disciplines. Matlab is better at linear algebra, R is better at statistics, Mathematica is better at

Python wins the triathlon by being second best in all the disciplines. Matlab is better at linear algebra, R is better at statistics, Mathematica is better at notebooks, C# is better at desktop applications, C is better at low level, JavaScript is better at web, Perl is better at job security. And so forth and so forth. But if you ever spent time with python in any one of those fields and though “ok, I don’t hate it but still prefer the leading player” and then switch to doing something else and you have the option of using the second best tool, or spend a ton of time learning the best, your bound to just go “I’ll just use python” > discover the rough edges and debates - a few that have come up lately for me at work - typing or no typing, oop or functional? pytest or unittest? poetry vs pipenv None of that has anything to do with python, it’s just the typical team bike shedding debate about what tools and features to use. There’s always going to be the typed vs untyped debate. But normally that’s settled when you decide to use node, or when you decide to use rust. Using one unit test framework or another is always potential discussion independent of python, and you’ll always need the team to deal with versioning and dependencies. If your “rough edges” in python is literally just that your team can’t agree on things, your not going to have it any easier by switching to Rust or Perl, Julia or C#.