>For example Zig is probably the most impressive new language, and it’s got a ton of awesome stuff but the chance that I’m going to adopt it over a language with often comparable performance that is ALSO much safer? Rounds to zero.
1) Rust, in practice, is "safer" than Zig but doesn't seem to be "much safer".
> Our Zig codebase had one leak and one undefined memory access. That was really surprising to me (in a good way). Our Zig codebase is large, complex, and uses tons of memory tricks for performance that could easily lead to unsafe behaviors. I thought we'd have a lot more issues, honestly. Also, the one leak found was during calling a 3rd party C API (so Zig couldn't detect it). So this is a huge success.
Take that as you will. And what Ghostty does probably requires a decent chunk of "unsafe" that would likely hide a bug anyway.
To me, the tradeoff of a demonstrably small number of bugs in a system software language in return for the demonstrably faster developer velocity (Zig compiles WAY faster than Rust, Zig wraps C/C++ libraries way easier than Rust, Zig cross compiles way more easily, etc.) is a decent tradeoff for my use cases.
For me, programming is about corralling motivation more than anything else. Rust saps my motivation is ways that Zig does not.
I love what the Oxide folks are doing. Having someone pwn your big server because of a memory bug? Yeah, I'd reach for Rust in their use cases. Or, if I'm handling credit cards, I'll have a very different set of tradeoffs that swing against Zig. (Although, in that case, I'll probably reach for a GC language in that case so that I don't have to think about memory management at all.)
2) Rust is to C++ like Zig is to C.
Zig is simply a much smaller language than Rust. I use Zig because my brain isn't big enough for either C++ or Rust.
I'm not a 10x programmer, but I still want to get stuff done. Zig and C enable me to do that without bending my brain.
> For me, programming is about corralling motivation more than anything else. Rust saps my motivation is ways that Zig does not.
Yes, agree with you a lot! Maybe our brains are just wired differently, for me, no other language (until now) gives me as much motivation as Rust, as it's type system and docs make me feel really good.
> Zig is simply a much smaller language than Rust. I use Zig because my brain isn't big enough for either C++ or Rust.
Disclaimer: haven't really tried Zig yet. IMO you don't need to keep the whole of Rust in your brain to use it, I usually can just focus on the business logic as if I make a stupid mistake, the compiler will catch it. That (and the type system) is what makes me more efficient with it than other langs. I also stay clear of lifetimes unless I really need them (which is almost never for application code). An extreme example of the opposite is C, where I need to take care about anything I do as it will just accept anything (e.g. auto casting types) so I need to be vigilant about everything.
All of that said, there are patterns that will just be harder to reason about in Rust, mostly with self-referential things, and if you area already used to using them a lot, this can be a hassle. If you're used to smaller data structures and data-oriented programming, it will be a lot easier.
This is not trying to convince you or anyone else, just showing a different perspective. If you feel better with Zig, use it! Everyone has their own experience.
> All of that said, there are patterns that will just be harder to reason about in Rust, mostly with self-referential things, and if you area already used to using them a lot, this can be a hassle. If you're used to smaller data structures and data-oriented programming, it will be a lot easier.
Also an underrated chance to be forced to learn about Arenas, most of the time.
Simple data structures are easy to handle though, KISS rulez. You need rust when you write something comically complex like servo, that's when you need all the checks.
Yeah, not everyone is mitchellh -- I'd argue that his intuition was right (expecting to find more issues), there's a reason we don't all write C.
That said, it's great to hear that others are having success with Zig and aren't writing very many bugs.
> To me, the tradeoff of a demonstrably small number of bugs in a system software language in return for the demonstrably faster developer velocity (Zig compiles WAY faster than Rust, Zig wraps C/C++ libraries way easier than Rust, Zig cross compiles way more easily, etc.) is a decent tradeoff for my use cases.
Reasonable points! This is one of the reasons I have on my list to consider Zig at all (that and wanting to write more C), it seems people that like it are wonderfully productive in it.
I really like a lot of the design decisions and watch the streams than Andrew and Loris, Tigerbeetle folks and others do to see the decisions they make. Their managed IO approach is certainly interesting and quite powerful.
Dislike some small petty syntax things (whyCamelCaseFunctions()!?), but that's a me thing.
> For me, programming is about corralling motivation more than anything else. Rust saps my motivation is ways that Zig does not.
Agree with this, anecdotally.
> I love what the Oxide folks are doing. Having someone pwn your big server because of a memory bug? Yeah, I'd reach for Rust in their use cases. Or, if I'm handling credit cards, I'll have a very different set of tradeoffs that swing against Zig. (Although, in that case, I'll probably reach for a GC language in that case so that I don't have to think about memory management at all.)
Yup, Typescript works for me here, and usually the hard parts are offloaded to an external service anyway (e.g. Stripe)
> Zig is simply a much smaller language than Rust. I use Zig because my brain isn't big enough for either C++ or Rust.
>
> I'm not a 10x programmer, but I still want to get stuff done. Zig and C enable me to do that without bending my brain.
I wouldn't put Rust and C++ in the same league there as far as brain size requirements, but point taken!
> Yeah, not everyone is mitchellh -- I'd argue that his intuition was right (expecting to find more issues), there's a reason we don't all write C.
I'd argue this is also down to the debug memory allocators in Zig which run a bunch of checks for things like use-after-free. Again, one of the nice things about Zig is that debug mode isn't dramatically slower than release so you're not inclined to turn it off too soon.
However, one thing that does shock me is that no "reference to stack allocated thing escapes" errors existed. Zig does not help you very much on this front, and I'm stunned that none existed.
> I'd argue this is also down to the debug memory allocators in Zig which run a bunch of checks for things like use-after-free. Again, one of the nice things about Zig is that debug mode isn't dramatically slower than release so you're not inclined to turn it off too soon.
Ah another good point -- also excited for what opens up with regards to this approach with managed/explicit I/O. Pretty obvious that permutation testing of async behavior gets unlocked which is nice. The equivalent in Rust would be loom.
> However, one thing that does shock me is that no "reference to stack allocated thing escapes" errors existed. Zig does not help you very much on this front, and I'm stunned that none existed.
TIL -- is everyone just bolting on asan here and sticking to the clang backend for now?
1) Rust, in practice, is "safer" than Zig but doesn't seem to be "much safer".
See: https://mitchellh.com/writing/ghostty-gtk-rewrite
> Our Zig codebase had one leak and one undefined memory access. That was really surprising to me (in a good way). Our Zig codebase is large, complex, and uses tons of memory tricks for performance that could easily lead to unsafe behaviors. I thought we'd have a lot more issues, honestly. Also, the one leak found was during calling a 3rd party C API (so Zig couldn't detect it). So this is a huge success.
Take that as you will. And what Ghostty does probably requires a decent chunk of "unsafe" that would likely hide a bug anyway.
To me, the tradeoff of a demonstrably small number of bugs in a system software language in return for the demonstrably faster developer velocity (Zig compiles WAY faster than Rust, Zig wraps C/C++ libraries way easier than Rust, Zig cross compiles way more easily, etc.) is a decent tradeoff for my use cases.
For me, programming is about corralling motivation more than anything else. Rust saps my motivation is ways that Zig does not.
I love what the Oxide folks are doing. Having someone pwn your big server because of a memory bug? Yeah, I'd reach for Rust in their use cases. Or, if I'm handling credit cards, I'll have a very different set of tradeoffs that swing against Zig. (Although, in that case, I'll probably reach for a GC language in that case so that I don't have to think about memory management at all.)
2) Rust is to C++ like Zig is to C.
Zig is simply a much smaller language than Rust. I use Zig because my brain isn't big enough for either C++ or Rust.
I'm not a 10x programmer, but I still want to get stuff done. Zig and C enable me to do that without bending my brain.