Ideally we would have something that's more dehierarchized, but until someone figures out how to do that, all open source repositories have a group of maintainers (or some other name for those who have commit access) who need to (individually) approve of contributions. As it stands, open source is indeed shared effort, but shared effort that has to be approved by a member of a cabal. There's a reason stewards are often called https://en.wikipedia.org/wiki/Benevolent_dictator_for_life (which is indeed a slightly different concept, but hopefully you get my point that chief maintainers have lasting influence and power).

It's still much better than other levels (i.e. proprietary). When maintainers slog behind community PRs and wishes for too long and/or persistently screw them up, a critical mass of users leaving will build for alternatives created, accelerated by the fact that most of your project's users are going to be enthusiasts accustomed to switching and upholding open source principles.

In summary, yes, maintainers make decisions to make users (or some shared vision) happy and are supposed to backtrack when enough of the users object. This is true whether the change is antagonistic or not. What you said in your first paragraph is the opposite of that. (I guess my initial question was what do you think should be the reason to backtrack other than users aren't happy. "Users are hurt" is the same thing as "Users aren't happy".)

As for Audacity? It's been four years with no attempts, so Tenacity doesn't look like it's going to pick up momentum. Until the next incident, Users won't believe they're being screwed (and I agree with them) and aren't building that critical mass. (I also object that the opt-in telemetry PR was antagonistic.)