Nim vs Swift

 All Posts Tagged: FLOSS Infrastructure 

This is specifically about the last post here. I forgot to talk about Nim. I think Nim is more directly comparable to Swift, than it is to Zig. Nim might be good to replace Swift in this concept. But I could be wrong about Nim, such that it could actually replace both Zig and Swift to kill two birds with one stone.

Anyway, it's not like Nim clearly falls out of bounds of being a systems language. But I do not think that Nim has a clear enough focus on systems programming to be like C, Rust, and Zig.

I think Nim is in a similar bracket to Swift, where it should be treated much more like a highly performant applications language like C++.

So, the reason to bring this up is to ask the question of whether Nim or Swift should be the premier Linux applications language to dethrone C++ . . . in a tentatively serious hypothetical.

Both Suck Equally at GUI

In a way, Nim is like an applications language that has been misrepresented as a systems language. It is capable of doing a lot, but it has a garbage collector and some other features that seem to make its focus different in practice. But anyway, I think this has made it so that it has a false identity. I think this potentially explains why it hasn't taken off. The focus is confused, and people end up using it for basic scripting. It doesn't get much ecosystem. (Of course, the other reason is simply that that's what happens in the world of programming languages: Very few get the attention they need.)

Swift originally started out very much as an applications language obviously, but it has also ventured into scripting and server-side stuff. But because Swift has been directed to these purposes on Linux rather than applications development, then it has not fully carried the desktop and applications tools and ecosystem with it to Linux. 

So in other words: neither Swift, nor Nim, have very good applications development ecosystem or integration on Linux at this point. In fact, there are NO: modern, "automatic" safety languages that have good applications development integration and ecosystem on Linux. This is terrible.

Both languages seem to be raring to go in such a capacity, but they have essentially just been left alone from any big level of attention.

I do not know how easily Nim could be made to integrate with GTK or Qt, but those are very relevant topics to explore. The idea would ideally be to go to the extent of porting GTK or Qt to this language if possible. The more orgs see the vision, the better.

Why? . . . Linux needs something to stand out to developers that is not another web platform. As well as standing out, it is a practical issue as well.

Could Nim Be Better Than Swift?

Swift just seems to be much further along than Nim. But more than that, it makes sense that Swift has had very many eyes on it, and has been specifically designed for applications development. This counts for a lot in my mind. 

But I am not a developer: I could be wrong, and Nim could be a better fit all around, all things considered, from a feature set perspective, and/or a technical and ergonomics standpoint. etc.

Maybe Nim has:

• a very readable and familiar code style
• multi-thread performance
• garbage collection
• scripting

Nim can turn off garbage collection. Can Swift do that?

Nim is More Grassroots

I don't know exactly what this means, except that maybe Swift just is not the right choice coming from Apple. Maybe Swift is out of touch or too bloated, or too corporate. Or maybe they just don't appreciate being used for Linux applications. Maybe.