The original article explicitly says it's not about syntax or performance. It's about irrecoverably splitting call graphs when sync functions can't wait for asy

The original article explicitly says it's not about syntax or performance. It's about irrecoverably splitting call graphs when sync functions can't wait for async results. JS has this issue, but Rust and C# don't. In case of Rust the syntax difference is a deliberate design choice, because `.await` points can also return from the function. Rust handles errors with locally explicit `?` for the same reason.