Every time I tell another Linux person I'm building an RDP server, I get the same reaction in two parts. First, "RDP? Isn't that a Windows thing?" Then, a beat later, "Why not just use VNC?" Both questions are fair, and they're worth answering directly, because the answer is the whole reason the project exists.
Here's the position, and then I'll defend it: For a modern Linux desktop, RDP is the better protocol, and building it natively on Wayland instead of bolting it onto the old X11 path is the part that actually matters.
VNC is straightforward, and that's its problem
I'll give VNC its due, because it earned its place. It's simple, it's everywhere, and it does one thing you can explain in a sentence: It sends pictures of your screen and sends your clicks back. That simplicity is why it runs on everything from a fresh Raspberry Pi to a decade-old server, and why it'll outlive most of us.
The trouble starts when the desktop gets real and the network doesn't cooperate. VNC is built around shipping framebuffers, and no amount of clever encoding fully hides that. Put a full desktop on the far end of an ordinary connection, scroll a long document or drag a window, and you feel it. The protocol is moving pixels by the pound, and moving pixels by the pound is exactly what a remote desktop should be smart enough to avoid.
RDP grew up
RDP went a different direction. Underneath it is a system of virtual channels, which is a formal way of saying the protocol expected to carry more than pixels from the start: Graphics, clipboard, multiple monitors, audio, and device redirection, each in its own lane. The graphics path especially has had years of work poured into sending less data and sending it smarter, with real codecs instead of a framebuffer with good intentions.
That maturity is unglamorous, and that's the point. The hard problems in remote desktop aren't exciting. They're clipboard semantics, multi-monitor layout, and shaving bandwidth on a connection you don't control. RDP has fought those battles already, and I'd rather inherit that than fight them again.
SPICE solved a different problem
People bring up SPICE next, and it deserves a mention. SPICE is good at what it was built for, which is a virtual-machine console wired into QEMU. That's a different problem from reaching a real desktop on real hardware across a network, and the gap shows. I'm building for the second case, so SPICE was never really the contender people assume it is.
Native Wayland is the whole point
Here's the part I care about most, and the part that's easy to skip past. I'm not building this for X11. Wayland is the Linux desktop now, and the work worth doing is the work that meets it on its own terms.
On X11, a remote desktop could reach in and take whatever it wanted, because X11 let anything see everything. Wayland closed that door on purpose, and good riddance, but it means screen capture, input injection, and clipboard access are now separate, permissioned things, and each desktop has been working out how to offer them. That's more work for me, and it's the right way to do it, because the alternative is pretending it's still 2008.
So I'm building native to Wayland, talking to the actual capture and input and clipboard mechanisms the compositors expose through XDG Desktop Portal and PipeWire, instead of routing through a compatibility shim and hoping. Where those mechanisms are still immature, I'd rather feel it directly and help push them forward than paper over it.
The fair objection
The strongest counterargument is that this is harder, and it is. RDP is a large protocol, and the Wayland side is a moving target, with more than one way to capture a screen and no settled answer yet for input or clipboard across every desktop. Choosing RDP and choosing native Wayland means signing up for both at once.
That's exactly why it's worth doing once, carefully, in the open. The difficulty isn't a reason to reach for the simpler protocol and the older path. It's the reason the good version of this doesn't exist yet, and that's the most interesting reason there is to build something.
It's early, and the bet is made. I'll show you the pieces as they come together.