You make a valid point, but it is factually incorrect that midi was the driver for the changes. It was initially more seen as a test case of whether the framework was flexible enough. It just turned out sufficiently well that it seems to have some users. The gamepad module is an even more minimalist test. And any slider can be manipulated from lua and so by extension from another application using dbus. See a condensation of some of the discussions on irc that led to it here [FR] new pluggable input (not for drawing!) system for darktable · Issue #5063 · darktable-org/darktable · GitHub
The counterposition here is confusing. Nothing can be a priority unless there is a developer who is willing to scratch that itch (for free, in most cases). There is no list of priorities that people work off of. This is not how any of this works. Given that we are all “amateurs” (i.e. with varying availability), it is already unlikely that teams of developers would work together on something. We are lucky if we can find anybody willing to comment on proposals or code, let alone test what we have produced. And sometimes you need to do a lot of infrastructure work before being able to start the next phase. And then the lack of enthusiasm to engage or discuss “specs”/“requirements”/“wishes” can kill even the most annoying itch.
Of course we all know that this is not supposed to be how software development works; give us the budget of Blender and we’ll start work on changing it tomorrow! (Not joking here; I’m sure there’d be people willing to give up their day job for that kind of money. Even if it means writing unit tests). Which, in the end, is the argument AP makes; give him more money and everything will be better. His “rants” are basically a job interview. Not one he would pass in any organisation I have ever worked in, but good on him for trying!