bradhd

I've tried a couple of setups including LibreELEC on a Raspberry Pi and self-compiled Plasma Bigscreen on x86, but my current and favourite setup is just stock Plasma with some configuration.

• First, get yourself a cheap "air mouse" remote. I have a WeChip brand one, works like a remote control crossed with a Wiimote, with a little keyboard on the back. There's a handful of different brands, a quick search on Amazon will show a variety of different types. Having a regular wireless keyboard + mouse (like those cheap Logitech ones) is helpful if you want to do more complicated things than watch media.
• Starting with a stock Plasma desktop, I turn up the desktop icon size and make the bar taller, so it's easier to click on things accurately. There's also an alternative launcher menu in Plasma that goes fullscreen (instead of the start menu-style default).
• Next I make desktop shortcuts for the main apps/services I use. For Jellyfin I make a shortcut to the client, for commercial streaming services I make URL shortcuts that open in the default browser. I also like to set the desktop wallpaper to a solid colour.
• In Plasma's system settings, make a window rule for the default browser to make it open in fullscreen. If you need, you can do the same for other programs (my Jellyfin client has its own setting, so this isn't needed).

I use an x86 machine for this to avoid having to deal with getting widevine working for browser streaming. I use Librewolf as my browser and it works well. I know with some commercial streaming services there are video quality restrictions when watching through a browser on Linux, but that can't be avoided without using a locked-down platform anyway. These days I'm mostly using Jellyfin and it works great.

I used to really want the "polished" HTPC UI, but over time (and after yet another botched Ubuntu version upgrade), I realised that this setup gets you 90% there with way less hassle, and the last 10% is just aesthetics. Using a common, well-supported desktop environment makes troubleshooting a lot easier too.

As far as I can tell, it's more about what the final output is used for. If it's to be rendered on a screen, then the convention is that the z-axis is depth (since the x-y plane represent the location on-screen). So game assets and VFX tend to prefer Y-up. If it's a digital representation of a real object (like in CAD programs), then the more "physical" convention is the useful one.

Frankly I'm surprised how many other systems are Z-up.

Right-handed Z-up is the default orientation in mathematics and related disciplines (like physics) when working with 3D Cartesian coordinates, which I guess is why most of the software in that quadrant is used for modelling real-world objects (compared to digital-only game assets etc).