Gonçalo Rica Pais da Silva

> Is there a reason why this shouldn't be just a third party crate? There was interest in a first-party solution, and the defacto standard for RNG in the rust...

> What about using the hasher as a source of entropy? > > https://matklad.github.io/2023/01/04/on-random-numbers.html `getrandom` is a bit more extensive in the support of platforms than what is on the...

As a further follow-up, `RandomState` relies on `sys::hashmap_random_keys()`, which for unsupported platforms, [is not random](https://github.com/rust-lang/rust/blob/8c6ce6b91b172f77c795a74bfeaf74b865146b3f/library/std/src/sys/unsupported/common.rs#L34). And browser WASM is an unsupported platform that will yield non-random state (you need WASI...

> I don't see where, especially given the RFC seem to justify itself (given its name) on the basis we need a _deterministic_ RNG, then goes: The rationale being that...

TBF, one of the biggest cases for Deterministic RNG is being able to test systems that deal with random output. Being able to enforce the correct patterns for this is...

In the mean time then, I've published https://github.com/Bluefinger/bevy_rand which is basically this crate, but with some updates. So now it exists as a third-party solution to battle-test what this crate...

> When building distributed systems (which is what bevy is structured as, it just happens to run on one machine), there are certain things you make sure to control to...

> There is a chicken and egg here. I'm not sure why entropy / randomness has a higher bar than previous plugins with desirable functionality that were included in core...

This updated image broke my CI WASM tests which make use of Firefox. I know this is incoming, but it's been a while since it weas stated that this is...

Awesome! Good to know and thanks for the update