Why is there always this deep desire to dog on Rust?
Yea. I donât like writing Rust that much, but I still think itâs a great language. I donât like the constant comparisons. Zig does not need to be everything to everyoneâŚ
Wait hold on, two comments on the issue that called out the needless comparison to Rust were just silently removed, what the hell is going on?
Image from a still open tab
I think the title is unnecessarily provocative and practically begging for drive-by comments but the issue tracker isnât an appropriate place to litigate these things.
I agree that itâs the wrong place but Andrew did that to himself and where can you even litigate this? Can the project BDFL really just shit-talk other languages and no one can object?
Iâm honestly more and more disappointed with the ZSF. You have a possibly good idea, and it gets burried down under this toxic behavior. Itâs hard to discuss on the merit of the proposal. Guys, please, do not make such a bad image for the Zig community. You created a great language, allow people to use it
Sure, it is an unnecessary addition to the proposal, but it isnât shit talk and the nature of the proposal would have invited rust proponents regardless.
Iâm not defending it, but it itself is not a big deal. Whether itâs becoming a trend for Andrew to seemingly consider less the interpretation of his words remains to be seen, and I would barely qualify this as such a data point.
Regardless that is a different topic, and it would be much more productive to talk about the meat of the proposal.
It stresses me. Good idea but oh look another controversy is attached to it. Literally no reason to mention Rust.
Not on the issue tracker, at least. You know what the âDâ stands for in BDFL, right? ![]()
I donât get it. Rust people often say Rust is better because of itâs memory safety and x, y and z. When Zig says it has something better (or will have something better) then noooo itâs super controversial. There are many competing system level languages and comparisons will happen all the time. I donât see anything wrong with it
Just because ![]()
I donât really see the problem here.
The proposal is basically to implement Fil-C.
Fil-C indirectly mentions Rust because of memory safety âŚ
Fil-C achieves full safety with no escape hatches. There is no
unsafekeyword in Fil-C that could be used to turn off protections.
So why is it bad when Andrew does the same?
Please donât be blind to the intent. Itâs mentioned in the title, to attract attention.
Rust could have very easily be mentioned in the issue, saying that it does not cover all the memory safety this approach would. But it was not done like that.
The âBâ is more important here, look up the meaning.
When Zig says it has something better (or will have something better) then noooo itâs super controversial.
This isnât what is happening here.
There are many competing system level languages and comparisons will happen all the time. I donât see anything wrong with it
This is true and healthy for the vaster ecosystem itself but again, that is not what is happening here.
The proposal is about Fil-C in Zig and doesnât ever compare to Rust or even mention Rust outside of trying to shit fling in the title by implying it is not actually memory safe.
I cannot know why Andrew wrote it like this outside of stirring controversy, no value was gained from doing this.
A benevolent dictator is still a dictator, the point being that Andrew is under no obligation to let people use the issue tracker to object to the wording of the issue.
I actually gained value from this because I learned that Fil-C exists and has a different approach to Rust where checks happen at compile time.
Well ⌠fil-c is definitely unlike Rust.
There was literally no reason to mention Rust like that. Simply stating that the proposal covers safety areas Rust doesnât would have sufficed, hell, ânot even Rustâ would still be fine if Andrew really wanted to brag. The âactually memory safe (unlike Rust)â comment is purely in bad faith.
@Jan200101 I strongly disagree with your characterization. In my reading, the issue in question has a clear-cut factual statement:
- That Rust programs in practice rely on
unsafeandCffi, which creates potential for UB - That Fil-C like mode, which doesnât have an escape hatch, guarantees memory safety even for non-cooperative code.
Now, this factual statement can be taken into two directions. First, we can debate whether the tradeoff is worth it: what is the slow down? wouldnât this defeat the purpose of unsafe (unsafe abstractions, and another)? Can this be really implemented in water-tight fashion such that, eg, you could load untrusted drivers into your monolithic kernel, to yield the second actually memory safe language (the first one being JavaScript)? Does it makes sense to use Fil-C mode for new code, or are you better off just writing Go?
(To ensure that factualness gets through, does everyone get it how JavaScript, unlike Go, Java, C#, Python, Haskell, Rust, and what else, is the only non-esoteric programming language which stays memory safe, even if the code in question is adversarial?)
OR we can take this factual statement and open a local chapter of crossover fan fiction between âRust VS. the Worldâ and âZig VS. the Worldâ universes, by reading it in the most inflammatory way possible. And the Internet at large is surely going to do just that, as it feeds on drama. And thatâs really fine â people enjoy negative emotions, they love a good flame, whatever the fuel, such is the human nature.
But I would really prefer if here, in our local Zig sub community, we tried to push back against this aspect of human nature, and try to keep the discussion a bit more level-headed, trying to read what actually was written, rather than inventing the most controversial take.
It appears it is practically impossible today for people to survive the title of a thing, let alone the thing itself. People are pearl clutching over some (2, literally 2) words in the title of an issue, canât make this up.
