Losing Science Marketshare to Non-Python Languages

I can’t comment on 1), since I am not an expert in this use, but there are many areas in which Julia is a tool that provides a huge amount of benefit and utility, it just depends on the problem.

On 2), I think the most common and recommended IDE for Julia is now VS Code? The language server does allow autocomplete which has improved a lot over the years and runs in the background without you having to have an open REPL. Unfortunately the design of multiple dispatch doesn’t have a convenient way to have great autocomplete like in single dispatch languages (Python with typing, C# etc) as discussed in this thread (Why there is no OOP (object oriented programming) in Julia? - #27 by LaurentPlagne).

I think a lot of people come from languages like Python, MATLAB or R which have varying levels of tooling. While Python has PyCharm, which is very good, the tooling that a lot of people use is very limited (Jupyter notebooks, Spyder etc) and coming to Julia with VS code, you get similar or better support in VS code. I still think there’s a long way to go for sure, but Julia is still quite young, and most other languages have at least a decade of lead time developing good tooling. The tooling has improved a lot over time, and I hope this area gets more attention in the future, but it takes a lot of effort and dev time to achieve.

2 Likes