Attribution of the tested Rust implementation

[jix]

The linked paper attributes the "Rust" method tested in this repository to a "Rust 1.69.0 mathematical library", while it is in fact from the third party "roots" crate that is not part of the Rust language, the included standard library or of any other library that is included with the language distribution. The term "crate" roughly corresponds to what many other languages call a package and anyone with a GitHub account can publish a crate on the central https://crates.io/ registry. (I believe the Julia equivalent would be the "General" registry.) Note that https://docs.rs site provides documentation for all Rust crates published on https://crates.io/ and thus also doesn't indicate that the code found there is part of the standard library or otherwise provided by the Rust project itself. The documentation for Rust's standard library can be found at https://doc.rust-lang.org/std/index.html, but it does not include any method for solving quadratic equations.

A better attribution would thus be 'the Rust "roots" crate', linking to github.com/vorot/roots or https://crates.io/crates/roots, instead of linking to the official Rust project website https://www.rust-lang.org/. This would also be in line with the attributions for GSL, Boost and Numpy, none of which are part of the respective C, C++ or Python standard library.

[goualard-f]

Thank you for the clarification. However, that error is never made in this package itself, as the comments in rust.jl only refer to a roots mathematical library. I understand that crate might be more adequate than library in the context of the Rust language, but no reference is made to a standard library. Maybe you should have directed your remark directly to me as the author of the article, since it is not relevant to this very package?

I'll revise the text of the article linked to take your remark into account.

[jix]

Sorry for not having used the appropriate channel to contact you regarding this. Given that the paper links to this repository as implementation and that this repository links to the paper as documentation, I viewed them as part of the same project and since many open-source developers, including myself, strongly prefer people using issue trackers over getting contacted directly for something that has an issue tracker, I assumed this would be more appropriate.