Spectrum Wayland

An aim of Wayland is to have a desktop windowing protocol that's designed from the ground up for security. But so far, the security benefits Wayland makes possible have largely not been realised, as compositors have focused on compatibility with their X11 predecessors rather than implementing new security features. Even in Spectrum, where applications run in isolated VMs, Wayland applications have a level of access to the compositor and each other that might not be expected or wanted by users. Wayland's solid fundamentals make a more secure approach to window systems possible, but there is still work to be done before end users get a lot of Wayland's security benefits. We are going to solve this problem.

We will make it possible to implement flexible application-specific permissions in Wayland. To do this we will need to design new protocols, participate in the standardization process, and work with lots of other projects on implementations. Our aim is not to develop a differentiating feature for Spectrum, but to design a solution that can be implemented across the Wayland ecosystem, increasing security for all isolation technologies, whether Spectrum, Flatpak or something new.