Next-Generation Accessibility Architecture

I am very excited for this. There's a few things in here that are particularly exciting.

> We would prefer to extend Wayland rather than running in parallel with Wayland via D-Bus as AT-SPI does

I do not know if this is going to manage to make it past every compositor developer, and I'm particularly concerned that Mutter might not take well to it, but I think it is a good thing to put accessibility and automation directly into Wayland rather than off to the side. It should be first-class and front-and-center.

> This document uses the terms “client” and “provider” as defined by Microsoft’s UI Automation (UIA) API

I view it as another positive sign that Windows UI Automation is being used as any kind of inspiration, as even if it's a bit of a rat's nest of COM, it is a fairly complete API for accessibility.

Some of the goals go beyond what even Windows does, and that's a great sign. I really, really hope this project goes well.