On top of that, I would add that instead of porting yet another desktop, it could be awesome to polish one. This could actually help adoption or at least make life better for existing users.
TLDR:
- Working UI battery status
- UI Network configuration and status
- Sound control
- Screen locking and suspend working properly.
- Working on stubs instead of fighting against linux practices es imho is the best solution.
I've tried multiple DE on NetBSD (And OpenBSD) and none have these working correctly.
Long story :
I mean a lot of people do not want to have to spend time to configure a manual minimalist WM and learn all the shortcuts to use it, me included being the reason BSD are only on tinkering or router/server command line systems on my side.
Not that I don't know how to use the command line and open a manual.
But If I want to actually work and use my computer, I prefer to have a fully complete well integrated desktop (XFCE, KDE, Gnome) on which :
- Battery status
- Basic Network Configuration and status
- Keyboard/mouse settings
- Suspend settings
- Integrated Login manager (gdm on gnome and sddm, without having to fight to make it work)
- Proper sound control.
Works and focus only on my work, not on fixing the problems.
Also helping web performance is a must because nowadays most resources even for development are online and honestly I prefer using Linux on that because of performance.
Not saying BSD should adopt Linux practices for everything.
But for Desktop components, I really think that making stubs that make things work (libinput, elogind, udev stub, dbus, upower, and things I forget) Is the best way not to fall in history and keep being relevant.
I also say that for a developer perspective. Having better experience (including UI) is much more confortable to just use the laptop, work with it and maybe contribute.
My developments on NetBSD (despite being small contributions) are mostly done throughout ssh on a lid closed laptop because the experience is so much better from my linux laptop
(Either ssh with Helix, or by exposing an NFS share and literally developing on Linux and then build on NetBSD)
I really think NetBSD would win into improving common (sadly linux, but again I think it is that or becoming even less relevant on the desktop side of things, fighting against is IMHO counter productive) desktop components support.
Yeah in the end I think it is counter productive to port yet another DE.
I Like NetBSD being the reason I care.
Otherwise I would just "Stay on Linux"