It's very easy to add to the standard library but once added, it needs to be sustained forever, deprecation isn't trivial. People come and go so that responsibility falls on the core team. They are picky for a valid reason. Meanwhile, backwards compatibility is a huge issue that one cannot take for granted.
Meanwhile, nimble packages are not limited in any way compared to the standard library, there are many fantastic packages out there with dedicated developers. Those devs have full freedom to iterate and move quickly, independent of Nim and can support multiple Nim versions. You can use an old release with the latest Nim, or vice versa. Fusion is somewhere in the middle and an experiment to fill that gap - how it will pan out is open since it is early times. It's not that the core team isn't receptive to the community's need, it is just being realistic with the resources at hand. As for compiler experiments, `--gc:arc` came out of legitimate limitations in the existing gc's - working with C/C++ libraries or across threads or DLLs is not possible today. We can all debate whether it was a good use of time but it is exactly the kind of thing that makes Nim special.
