LFCS 2012: X and Wayland

Another interesting article! I haven't had time to look at Wayland yet, but I kept an eye on earlier efforts. I'm looking forward to more articles of this kind, especially to know more about what long standing design issues they tackle.

For example, X security leaves a lot to be desired from the setuid server to the all-or-nothing model where any app can be a keysniffer. It would be interesting to hear what is being done about this. Another example is the somewhat convoluted events model which long made it difficult to use multiple heads and multiple pointers. (And let's not even mention the screen lockers...)

I am one of those who actually uses X's features, but it's clearly crufty (in more ways than the above) and I think Wayland sounds like a solid base on top of which X can continue to evolve. The developers seem to understand the complexities involved and actually strive to make something simple. That can only be a good thing.