That's not the case. The job of language designers is to say "no" all the time and Walter is certainly a model with how to speak with users.
I don't think it's intentional but the argument always begins with explaining some detail as if you didn't know it existed even though you'd have to do be able to bring it up in the first place.
This is not just me, I've had this discussion with a few people. One of a short list of complaints of this kind it must be said.
There also has to be an escape hatch to make this work.
E.g., Linus can say that the __is_constexpr preprocessor hack is the product of a demented mind, and he'll merge that absolute monstrosity thanking the contributor.
As a leader you get the contributor community that you get. You can either make the development process workable for them, or you can risk the project languishing/forking.
Put another way: if a leader's concept of what their project ought to be/become veers too far from what the community is actively coding up to be merged, it's not going to go well.
No idea if this applies to D. But I'm absolutely certain Python's dev process has had plenty of escape hatches for resolving development issues. (And I'd guess there are also projects that have too many escape hatches, but I'm guessing that probably doesn't apply to D.)