Or you can vibe code 350k LOC https://github.com/space-bacon/my_rust_cms/
Production ready!
That’s just the LLM rambling. Not battle tested yet. I will get to that after I ship the Semiotic-Reflexive Transformer.
High-Level Rust: Getting 80% of the Benefits with 20% of the Pain
111–120 of 122 posts
Re: High-Level Rust: Getting 80% of the Benefits with 20% of the Pain
#112Re: High-Level Rust: Getting 80% of the Benefits with 20% of the Pain
#113Java is a fine language and has sufficiently expressive types. It's the most consistently overlooked language and frankly it's completely annoying. Java is a powerhouse. If you can live with a VM, it's an amazing language. The disdain for Java is honestly just weird. The boiler plate is only marginally annoying to write and makes it considerably easier to read. The 'enterprise ecosystem' is definitely bloat, but that…
Overlooked? It's one of the most used languages of all time maybe second only to C.
The distain likely comes from a time period of its history where "Enterprise" patterns had been taken to an absurdist end before the community final regained it sanity. Modern java is quite expressive and ergonomic.
Those in industry in those enterprise days simply got sick of those silly FactoryFactory classes and their like.
Today there's almost no "boilerplate" as they've even made it nice even for scripting use cases and allow floating main methods, var, etc.
However, to actually compete with rust they'd need first class native support and an ARC GC replacement. Iirc they already have "pluggable" gc implementations so it's not out of the question. Java tends to be fairly conservative in features though, so may take a couple decades to get there if they even care to.
Re: High-Level Rust: Getting 80% of the Benefits with 20% of the Pain
#114Re: High-Level Rust: Getting 80% of the Benefits with 20% of the Pain
#115Earlier quoted context omitted.
1. Java is mentioned in their comparison table. They just don't use it much. 2. There is really no reason to include Java in the search for your preferred language, since Kotlin is strictly better along every relevant axis.
I strongly disagree with Kotlin being strictly better than Java. It pretends to fix things that haven't been problems in Java for more than 10 years while introducing a layer of syntactic complexity that's completely unwarranted. It just thrives on hype and Android development.
I've "ported" several legacy server side applications to kotlin over the years. I have no regrets.
That said I feel kotlin is almost a testbed language for java to steal features from at this point. Modern java is "good enough" now to warrant sticking with java these days. But back before some of the more recent java editions, kotlin was a boon to productivity, at least for me.
Luckily with "big" (feature and keyword wise) languages, you get to pick and choose what features you actually use. Obviously there are pros and cons, but in most cases you can control the complexity. The issue that remains is when a library or framework you use evolves to use more bells and whistles than you're comfortable with, but I'm general the kotlin community is finally large enough that there's always alternative libraries etc.
I generally just consumed the java libs directly in kotlin, sometimes with my own tiny shim layers for ergonomics. That way nothing crazy gets foisted on me w.r.t. orms etc.
Re: High-Level Rust: Getting 80% of the Benefits with 20% of the Pain
#116I'm surprised no one in the comments is mentioning Kotlin. Out of all the languages I've worked with it has been the most enjoyable by far. I agree with the article that there isn't much community but I feel like that's arguably the least important category there. You should definitely give it a shot if you've never tried it before.
For a while using kotlin as a simply better java "front end" syntax shim increased productivity.
These days java has been "catching up", aka stealing the good ideas from kotlin--so it's less appealing to me. I think where it still shines is in cross platform targeting like ios, android, desktop and server as few other languages really enable that. C# did, but only because of one vendor (Ximian or whatever it was called, some offshoot of Mono)
Re: High-Level Rust: Getting 80% of the Benefits with 20% of the Pain
#117Isn’t this “high level rust” idea similar to swift? Everything is ARC. Clones are cheap. But you still have a reasonable sound type system and no garbage collection. I get it. Tooling on swift is meh the further you are from Apple, so I’m not suggesting it’s better. But from a language point of view; is it not essentially a high level rust? Ps. I don’t really know swift. Just asking where/why I’m wrong really.
Swift is a great language, sadly hampered by poor stewardship with terrible compile times, small community outside of apple ecosystems, etc.
I do hope it improves. Having a second compiler vendor that actually cares about making it usable without reaching for 3p compiler caching hacks would go a very, very long way.
Re: High-Level Rust: Getting 80% of the Benefits with 20% of the Pain
#118Isn’t this “high level rust” idea similar to swift? Everything is ARC. Clones are cheap. But you still have a reasonable sound type system and no garbage collection. I get it. Tooling on swift is meh the further you are from Apple, so I’m not suggesting it’s better. But from a language point of view; is it not essentially a high level rust? Ps. I don’t really know swift. Just asking where/why I’m wrong really.
It's not ARC though. They use fancy threading mechanisms to avoid having to check on every access. Much faster.
It is still primarily ARC. There's simply some optimizations in place, but still no real GC. So you'll never get those dreaded pauses at runtime. It COULD be a real rust competitor if a couple issues were addressed, mostly devx (in the standard tooling sense not the misuse of term this article has, meaning expressiveness, correctness, etc. which Swift already has in spades)
Re: High-Level Rust: Getting 80% of the Benefits with 20% of the Pain
#119Crystal lang deserves a look given the article writer’s preferred language attributes. It has OO paradigms, elegant composability, efficient codegen etc.
Crystal's issues list 785 open issues tagged as bugs, as of today, 16. april 2026: https://github.com/crystal-lang/crystal/issues?q=state%3Aope...
Is that normal? Or does someone just really need to close some stale bugs?
Re: High-Level Rust: Getting 80% of the Benefits with 20% of the Pain
#120It is absolutely bizarre to not check DevX for Rust, it was literally the most loved programming language for years according to SO surveys.
I think it's because majority of people using Rust come from C++ where the devx is way worse. But I think what pulls down the devx is the perceived steep learning curve, plus lifetimes.