I’ve been working with a Javascript (+ TypeScript) + Java + SQL stack for the last 10 years.
For 2024 I’d like to learn a new programming language, just for fun. I don’t have any particular goals in mind, I just want to learn something new. If I can use it later professionally that’d be cool, but if not that’s okay too.
Requirements:
Thank you very much!
EDIT: I ended up ordering the paperback version of the Rust book. Maybe one day I’ll contribute to the Lemmy code base or something :P Thank you all for the replies!!!
Welcome to the main community in programming.dev! Feel free to post anything relating to programming here!
Cross posting is strongly encouraged in the instance. If you feel your post or another person’s post makes sense in another community cross post into it.
Hope you enjoy the instance!
Follow the wormhole through a path of communities !webdev@programming.dev
No jokes: pick a language that is in the market, but has a different design philosophy than your background. Your background includes compiled static, and loose scripting, with strong library tooling, so you have diversity there, so a language in which you have to think differently is the right choice.
I recommend:
I’m pretty sure that Lua doesn’t follow functional programming as a philosophy…
I don’t mean to be rude, but I find this baffling; what do you mean by it? One of the primary design goals of Go is to be simple to learn (this is fairly well documented), and it’s one of the few things I really have to give the language credit for. Rob Pike has specifically discussed wanting it to be accessible to recent CS graduates who have mostly used Java. I have never heard anyone before describe learning Go as a “challenge.”
Download GODOT and learn the scripting language and the shader language. Make a game! Or at least neat shader art.
Bonus answer: Csound. Make funky noises with code.
Personally, the language that’s taught me the most to learn has been Haskell. It has a lot of very interesting ideas and a learning curve that plateaus after most other languages. There are several ideas that have trickled down from Haskell to other parts of the programming world and learning about them in the context Haskell is in my opinion better because you’ll learn about them in a context where they fit in with the rest of the language very well instead of being late additions that offer an alternate way of doing things.
Coming from Java and JS, Haskell has a very different approach to a lot of things so you’ll have to re-learn a lot before you get productive in it. This can be frustrating for some but you’ll learn more if you get over that hump on the other hand.
Haskell doesn’t see very much industry use and arguably isn’t very well suited for industrial application (I haven’t used it professionally so I don’t know personally) so it might not directly help you land any new jobs but it is in my opinion it’s a very good way to develop as a programmer.
I recently picked up Rust, still a beginner, but it’s very nice. And the compiler errors and tips are top. Since I used only Python before, Rust is my first compiled language and I’m glad it was Rust. Options and Result’s are also something I appreciate tbh. Other than for AI, simple stuff and maybe prototypes, I would use Rust over Python. Definitely recommend it. The borrow checker is a pain in the ass though. I’ll also be using it later for embedded systems when I become more fluent with it. I am also currently making a big project with Rust that I haven’t been able to do with Python.
DotNet is closest to Java, but hang on to your hat: the state of C# is at least half a decade ahead of Java, if not a full decade. It’s sophistication will make Java use feel like banging rocks together. DotNet Core can now run on all three primary platforms, and with some careful work, you can write a single program that can compile down to each platform and carry along its own required binaries, no pre-install of any framework needed.
My second recommendation would be Rust. Stupidly steep learning curve, but an absolutely game-changing one where safety and security is concerned. It’s my next objective, personally speaking.
Any other language I could recommend starts getting into speciality purposes, which makes general use more difficult or even wholly inappropriate.
For example, if you are dropping into DotNet for business applications, I would also recommend diving into F# for that functional goodness for building complex business rules and data handling. But building an entire app in F# can be jamming a round peg into a square hole under many circumstances, it’s appropriateness envelope does not cover as many cases as C# does. You want to use C# for boilerplate/frameworks, F# for the core bits where it is going to shine.
C# is a great choice.
Incredibly versatile language and should be an easy jump from java.
My two cents: I strongly agree with this. We just deployed an intranet blazor server app running on Linux (don’t know which distro) and apache (we might switch to nginx soon). It works very well and we had to write less than 100 lines of JS (mostly for file download and upload) One of my workmates was hired one year ago and at the time he didn’t know anything about .Net stack. Now he is mostly autonomous and he loves .Net and blazor in particular. Obviously YMMV.
Bundle size is my only complaint with blazor, has to send the .net runtime in webassembly to the client.
Aside from this, C# on the browser is an absolute joy to use. I’d use for everything if I could.
You are talking about Blazor webassembly, I am talking about Blazor server side, which loads as fast as a “normal” website.
Server side Blazor has other caveats, that’s why I specified it is an intranet project, where server side Blazor fits very well. Anyway, at the moment, Microsoft is still putting effort in polishing both type of Blazor hosting model.
This is not our first Blazor intranet web app and some of them are running in production for one year more or less.
It is really a joy to program using Blazor, especially if you need cross tab/browsers/device/user real time communication, which comes almost free thanks to underlying SignalR channel.
Ah you mean Razor then. Blazor lets you run C# in the browser, but Razor is the one that needs a server and streams changes to the client using signalR.
I know Microsoft didn’t get this right with naming and you got caught in the trap but there are 2 (actually 3) ways of hosting Blazor.
I also see that this confusion won’t help OP choose Blazor over some more coherent dev environment hehehehe
I can recommend Rust - I’m training a few people on it.
And a pure functional programming language like Haskell or Scheme, if you don’t know what functional programming is, or are not comfortable with it. Functional programming needs a different mental approach to traditional (imperative) programming paradigm. Some of the more modern languages like Rust, JS and Python incorporate a lot of functional programming constructs. So it makes sense to learn them.
And a lisp - Common Lisp’s popularity is a public secret. Scheme is also fine. This family is homoiconic (program and data are treated more less the same). The syntax is actually very close to its AST. This gives Lisp unparalleled metaprogramming capabilities - mostly through macros. Macros in traditional languages are nowhere near Lisp Macros.
If it interests you, study a stack based language like Forth or Factor. Though they feel very different from Lisps, they have similar underlying properties. And you get more or less the same advantages.
+1 for Rust, the learning curve can be pretty daunting, but once you’re over the hump you’ll never want to go back
Have you ever wanted to do more with regular expressions? Then give Perl a try.
Do you want to awaken the Elder Gods, and consequently suffer a slow and inevitable descent into horror and madness? Then give Perl a try.
Truly efficient perl code is write-only.
Ada particularly the SPARK subset. It’s approach is quite different than most languages, focusing on minimising errors and correctness. It’s fairly difficult but I like to use it to teach people to actually understand the problem and how to solve it before they ever write the code.
I advise you to learn something different and hard for you. Only this case will help you to grow and realize a lot of new.
All these languages are efficient and forget about hype and popularity. Language does not matter if you have what to write with it.
I’ll suggest Elixir. It’s a language that runs on the same virtual machine as Erlang, which has proven to be great for ultra-reliable and excellent at managing many, MANY concurrent processes.
Elixir itself builds upon this great foundation with a syntax similar to Ruby, but entirely functional. It’s a delightful language to read and write.
C or C++, specifically with the use of compiler explorer so you can get a feel for how code actually runs.
Common Lisp or Haskell to get a taste of something really different.
Go. It’ll be just different enough from what you have experience with to make you think about things differently (in a good way!) from now on. And it’s also a fantastically well-designed language that’s great for getting real work done. And it’s lightning fast as languages go, and compiles to an actual executable. Really a pleasure to work with. It’s my (no pun intended) go-to language for every new project I start. (Excluding what I write specifically for a paycheck. I don’t have a choice there.)
Don’t really want to invest in a language designed by Google.
Rust, haskell, python, c++ are all interesting choices. I would argue that c# is too close to what you already know to be interesting.
If I were you though, I’d pick a project first, then decide what language makes sense for it.
This is my favourite list in here, but I’d throw a Lisp in too.
Lisp, Haskell, and Rust should all teach you something new and profound about programming.
I’m a huge fan of Haskell and (for pragmatic purposes) Purescript. Purescript is hard to find much in the way of documentation but it is so similar to Haskell in that the steep learning curve is worth it, IMO. I rarely find a project that I couldn’t accomplish with one or the other or both.