@brillenfux So, let me see if I understand you right.
In our current world, the developers -- as well as some of the most enthusiastic of users -- tend to show up to conferences and engage in email threads or community forums. But there tends to be a lack of actual, I guess, enforcement.
If users want X to happen (say, a new feature, or specific language changed in the interface to be more inclusive, or an endorsement from a toxic community dropped from the project website), they can engage online and try to get it to happen. Heck, even the smaller contributors could want it. But if the people in charge of the project don't want to, then it doesn't happen.
Under your recommendation of using co-ops, an enthusiastic user who wants to actually be part of the decision-making process could actually be somehow empowered to engage beyond just asking. And the same for the dozens of smaller contributors who don't own the project.
Am I interpreting you correctly, here?
If so, there would be no need for a user to really join a coop to use an OS or a word processor, unless they want to have power over the direction of that software.
CC: @cava