Comment by diggan

I understand that quoted part as "it's tricky" rather than "It's impossible because no one wants to do it", just like many collaboration-efforts in FOSS. But you're right that it's probably for the foundation to figure out, a lone compiler-optimization geek isn't gonna be able to come up with a project-wide solution and force it through.

Haven't the Rust team already implemented "grand features" that took many years to get across the finish line? For example, GATs didn't look particularly fun, exciting or sexy, but somehow after being thought about and developed for like 5-6 years eventually landed in stable.

Edit: Also just remembered a lot of the work that "The Rust Async Working Group" has done, a lot which required a large collaborations between multiple groups within Rust. Seems to have worked out in the end too.