> I guess until the community manages to communicate, linux will remain mostly used on servers and by interested parties that want to configure everything.
So what? Why would (for example) the authors of the many different window managers specifically benefit from there being only one WM that fits them more-or-less well but allows more people in general to use Linux?
Or, put another way: Why should someone who is presumably an expert user (as they consider developing their own toolkit/WM/DE/audio server/foo) consider the implications of his decision on the mass-adoption of an existing piece of software he is trying to replace?
Using window managers as an example was a bad choice, they are like clothes, people like different styles.
But why do we need more than one good way of displaying stuff on screens (X, Wayland, Mir, etc)? Competition is good, but not if we end up having to have them all installed and running at the same time to be able to use a linux desktop environment.
Because they are like clothes: People like different styles. Even if it may be clothes that are not visible to you.
And I think you're unnecessarily pessimistic - even if you end up having more than one running now and again, there's little reason why end users would need to notice. E.g. Wayland at least can target X - I don't know if Mir can, but there's no good reason why it couldn't be made to run on top of X. Both Wayland and Mir will have ways of running X apps. At least the Mir guys have indicated that making a layer to make code that targets Wayland run directly against Mir would be possible (after all one of the stated technical reasons for Mir is to have a core that is independent of a specific client protocol/APIs as well as indpendent of specific targets). There are X servers that will run on X for that matter.
In other words: Until the dust settles, you might occasionally find that apps written for one or the other will fire up some version of one of the the others that targets whichever system owns your display, but chances are you won't even know it is happening unless you specifically look for it.
So what? Why would (for example) the authors of the many different window managers specifically benefit from there being only one WM that fits them more-or-less well but allows more people in general to use Linux?
Or, put another way: Why should someone who is presumably an expert user (as they consider developing their own toolkit/WM/DE/audio server/foo) consider the implications of his decision on the mass-adoption of an existing piece of software he is trying to replace?