News for hackers. Stuff that matters.

Ask HN: Does a strong editor improve language adoption?

Does having an IDE for a new language matter? Does it help with language adoption? Do you bother investing in a language that doesn't have a Language Server Protocol ( LSP ) or mode for your favorite editor? Does a tightly coupled set of tools help elevate a new language above others? Should a language "win" even though it doesn't have these tools but just because it is "good"?
( 6 comments | 7 points | on HN )
Re:Ask HN: Does a strong editor improve language adoption? by smt88 on Wednesday December 04, 2019 @09:46PM (#21709547)
You’re basically asking if usability matters for a language, and of course the answer is yes. Lots of languages have useful and verbose type systems that would be painful without IDEs. With them, they make refactoring and autocompletion look close to magical.
Re:Ask HN: Does a strong editor improve language adoption? by PaulHoule on Wednesday December 04, 2019 @07:42PM (#21708838)
I started coding Java in a text editor while Java was still in beta. There were good IDEs for C, Pascal, assembly, etc.

It was clear that a great Java IDE was possible, but it took a decade to get there.

Java IDEs compensate for the pain of Java being wordy. The debuggers work, even when you are debugging a process with both Jdb and gdb at the same time.

Re:Ask HN: Does a strong editor improve language adoption? by auslegung on Wednesday December 04, 2019 @08:01PM (#21708972)
I did Haskell for a while in nvim with basically no features (there are several, I just didn't set them up). I simply had ghcid running in one tab, nvim in the other, and did what the compiler told me to do.

So that didn't matter to me because the compiler is great and ghcid is very fast and sufficient.

Re:Ask HN: Does a strong editor improve language adoption? by whb07 on Wednesday December 04, 2019 @11:28PM (#21709988)
Haskells tooling is abysmal, which saying out loud it doesn’t make sense. Such a great compiler with the most über of type systems, one would think that the tools used for it would be a shining beacon of hope.

Yet setting up the most basic system from scratch can easily take hours.

I definitely appreciate the sense of pride and joy one has with surviving the challenges, but jeez, stop for a second and take a quick look at Rust’s package manager or F# and Ocaml’s ide experience.

It might, just might, help bring in new blood and users to grow the community and with it bring better tooling and other useful libraries.

Re:Ask HN: Does a strong editor improve language adoption? by auslegung on Saturday December 14, 2019 @09:48PM (#21794124)
In general I agree but have you used ghcid? I’m not really sure what people want in ide tooling, I’m quite fine with compiler output and everything else I want from an ide should be language agnostic: goto, search, replace, file navigation, etc.
Re:Ask HN: Does a strong editor improve language adoption? by proxybop on Friday December 06, 2019 @03:34PM (#21725718)
I sure hope so. As a developer for a language, we’re planning on creating a language server and debugger to make it easier for people to use our language.