1. 58
  1.  

  2. 5

    They’re signing praise. That sounds like a big achievement, especially for a TLS library.

    1. 6

      It’s probably only because Rust is so unreadable that they didn’t find anything. /s

      On a serious note, no matter what you think about Rust, more diversity in the realm of TLS libraries is a good thing. Just like BearSSL, rustls offers a way to escape the factual OpenSSL-monoculture.

      1. 2

        It’s probably only because Rust is so unreadable that they didn’t find anything. /s

        I see the /s, but is that a common criticism? I genuinely do not know how Rust is perceived other than the hype/enthusiasm.

        1. 3

          Speaking from experience, there’s technically nothing wrong with Rust’s syntax. In fact, there’s lots of great stuff about it, like types that remain human-readable even when they’re complex (nested arrays and function pointers are easy). Greppable fn keyword for function definitions is very handy too.

          However, Rust tries to look like C, but has syntax details significantly different from C. I suspect it gives an “uncanny valley” impression to users coming from C-family languages. Rust doesn’t need as many round parens, but requires more braces: if true {}. Rust has generics, which sometimes sprinkle the code with lots of <T>. This might affect overall aesthetics of the code, but I don’t find anything that would be objectively unreadable about that.

          1. 1

            Ive seen many say it’s hard to learn but don’t see much of that claim. All the changes going on in the Rust ecosystem, esp libraries, suggests folks can read the code fine. Suggests, not proves.

        2. 2

          This is great news! This is enough to convince me to switch to rustls in my side projects.