Hacker News Reader

Comment by gilrain | Hacker News Reader

As someone who frequently posts about my personally-excellent experiences with Rust, what makes you suspect it's astroturf rather than just turf? You think Mozilla is paying people to post about Rust using puppet accounts? C'mon.

View on HN

Think about it. Mozilla Rust -> Godzilla (C)Rust -> "Godlike" giant lizard crust -> God is radiant -> light -> illumination -> Illuminati lizard people from beneath the Earth's flat crust are funding paid protestors to shill Rust, it's really the only explanation that makes sense.

(The person you're replying to is using the transparent internet argument tactic of trying to cast doubt upon genuine enthusiasm by implying some vague sinister motive, conveniently without bothering to articulate what that motive might even possibly be. Let's all recognize bad-faith arguments, downvote, and move along.)

It does seem a little odd that every comment about Rust gushes about its progress and stability without discussing any downsides. One would be similarly suspicious if, e.g., Perl6 were discussed this way.

Try it out and write the critique, then. The posts are positive because Rust is genuinely achieving rapid progress and stability... but certainly not without flaws.

I've experienced plenty of frustration, albeit outweighed by the massive benefits for my use case. And there are whole problem domains to which it's just not suited. I see these mentioned pretty frequently.

For once there is no consensus how to do parallelism an concurrency in Rust since none of the library are matures.

> none of the library are matures.

Rayon, the Rust library getting most of the discussion in this thread, is now 1.0: https://github.com/rayon-rs/rayon/blob/master/RELEASES.md .

Don't most people opt for something called Tokyo? I don't do Rust much, but even I have heard of it by now...

Tokio for IO concurrency, Rayon for data parallelism.

Exactly! Rust doesn't solve any parallelism problems. It's no better than established languages like C# or F#, and probably worse.

I'd consider handling data races gracefully a pretty large step forward in doing parallel data computation.

>It does seem a little odd that every comment about Rust gushes about its progress and stability without discussing any downsides.

Doesn't seem odd at all.

First, a lot of people commenting about Rust are enthusiastic recent adopters that haven't seen much of the language, including any real ugly sides yet.

Second, it's not entirely true, almost all Rust threads mention the steep learning curve, the slow compiler, and other issues such as the variadic generics (or lack thereof to be precise).

Third, we have seen the same "early adopter enthusiasts seeing it all rosy" circle for RoR, Go, Node, and Mongo, nothing out of the ordinary here.

That's what happens when you have a new hip language/framework that has a lot of hype.

You see the same with Rust, Elixir, VSCode, Elm, Purescript, ReasonML, etc.

It's not astroturfing, it's just a lot of beginners that are overly excited to proselytize their new discovery.

I don't think it's fair to claim these people are all beginners. More likely it's a phenomenon where mature languages have been around long enough that you already know their strong and weak points, but when you look at a language in development you only see its potential.

I mean, it's not just true for new languages. For Haskell, you see beginners claim that "it's a perfect language and can do no wrong", while experts in the language will acknowledge its faults, including: long compile times, the downsides of laziness, etc.

Reply on news.ycombinator.com