31 ms·
Please consider listening to Bryan Cantrill's thoughts[1] on using Rust to, as you say, "get things done." -- In the presentation he explicitly calls out how le
by drbawb 8y ago
Please consider listening to Bryan Cantrill's thoughts[1] on using Rust to, as you say, "get things done." -- In the presentation he explicitly calls out how learning Rust by trying to bang out a "trivial" data structure is very much the wrong approach. He posits that to understand the value(s) of Rust you must be willing to approach it from first principles. Let that sink in: a kernel & distributed systems programmer w/ decades of experience had to resist the urge to just bang out a linked list & instead opt to learn the language from first principles. That being said he seems to be quite[2] happy[3] with the language.
[1]: https://www.youtube.com/watch?v=aWbGPMxs0AM https://www.youtube.com/watch?v=aWbGPMxs0AM
[2]:https://twitter.com/bcantrill/status/1035335403186806784 https://twitter.com/bcantrill/status/1035335403186806784
[3]: https://twitter.com/bcantrill/status/1038495464692146176 https://twitter.com/bcantrill/status/1038495464692146176
- glandium 8y agoAnd for kicks, here's a quote from that Youtube video you link: "So that was a surprise. (...) I've done zero profiling on this. I've made absolutely zero attempt to make this thing faster. None. In fact, you can pretty reasonably argue that I've gone out of my way to make it slower. And we are 40% faster than my C." Which doesn't mean that rust is faster than C, but that changing your approach can, in fact, lead to results you wouldn't have expected.
- oconnor663 8y agoQuote starting at about 1:05:00. Pretty nice :)