We currently have only one case where native is the most practical way to accomplish what we need. You can read the full discussion of that with Robin here. Other use cases were mentioned in the linked reddit thread, and more could be found in other searches.
Regarding optimization, native modules are still usable by Elm package contributors, and are also subject to optimization. So I doubt optimization is a blocker to using them.
I don't have a problem with removing native access. But the timing was when the API is still known to be missing important cases. And Evan's post about it reads like a punitive decision. Between this and the community issues (which have the same root cause), the risk for us in using Elm has gone up.
We're a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.