Well, I think he has a point about the epoll/select - while I'm not sure about how good that is, the tokio library is relatively new - barely reached stable.

I do think that the majority of the resistance though is that the author could not get productive with Rust in 4 days. For that reason, even though the epoll and documentation issues might clear up, I don't think the author would ever select Rust, not matter how much time has passed.

Which as I write, I realize that I think that Rust would be better, lack of easy on-boarding aside.