Hi, I am shamelessly cross-posting this news from Slashdot : http://linux.slashdot.org/story/10/11/05/137212/Ubuntu-Dumps-X-For-Unity-On-Wayland http://www.cio.com.au/article/367050/ubuntu_risky_leap_unity_wayland/
I am posting it here because this announcement from Shuttleworth is already attracting a lot of criticism and debate. The CIO article itself seems to be heavily biased against it -with a lot of inaccurate info. On the technical side, X-server has been the backbone of Unix/ Linux graphical interface since mid-80s. Several attempts to supplant it has simply failed. Now Canonical seems to be trying that again- the difference being that Ubuntu is kind of a king maker now. With their large customer base, they may be able to decide GNU/Linux's future themselves. This is despite the big resistance they always face when trying something new. There are some good examples from the past. First was the introduction of pulse-audio (the system was very unstable at the time and caused a huge ruckus). Another was the renovation of the notification system (which everyone hated at the time, but eventually found itself reimplemented in Gnome). Now they are planning to sideline Gnome-shell for unity interface and now replacement of X server with Wayland. Canonical is always blamed of introducing half- baked features into ubuntu. But as I see it, Canonical forces everyone to adopt new features and eventually improve it, rather than waiting for a new feature that never matures on its own. So, how does Wayland migration look as of now?: Pros: 1. Wayland is very much less bloated compared to X-server. Any software 2 decades old is bound to be bloated 2. Wayland takes advantage of latest GPU and kernel features - like DRM directrendering 3. Compared to X-server, fewer time consuming 'context switching' is required. Here is it in more detail: http://wayland.freedesktop.org/architecture.html 4. Wayland can be much more responsive 5. Wayland doesnt make as much mistake as X-server when doing 3D and desktop effects (ie, compositing). Here, the server itself will be the compositor. Cons: 1. The biggest point of contention is that 'ssh -X' (Xserver ssh tunneling) wont be possible anymore. Wayland and its client (ie, your application) communicates through shared memory buffers in the GPU. This is unlike Xserver-client communication which can be routed through network. The first casualty that I see is the 'Linux Terminal Server Project' which some of our group members use with relation to Edubuntu. Frankly, I dont want to miss that feature - RDP and VNC doesnt even come close to X tunneling. This has already become the subject of the biggest debate. 2. Every X-based application will have to be rewritten or run in compatibility mode. In compatibility, the X-server will itself run as a client on top of Wayland. In case of GTK/Gnome and Qt/KDE apps, they will be fine as long as the toolkits (QT and GTK+) can be ported for Wayland 3. Getting hardware vendors to supply Wayland drivers instead of X- drivers will be hard (I am unsure of this part). Nvidia users (like yours truly) are the ones the most worried (Dont ask me why!). Some of you may have played around with X-server. What do you think of this move? Regards, Gokul Das -- "Freedom is the only law". "Freedom Unplugged" http://www.ilug-tvm.org You received this message because you are subscribed to the Google Groups "ilug-tvm" group. To control your subscription visit http://groups.google.co.in/group/ilug-tvm/subscribe To post to this group, send email to ilug-tvm@googlegroups.com To unsubscribe from this group, send email to ilug-tvm-unsubscr...@googlegroups.com For details visit the google group page: http://groups.google.com/group/ilug-tvm?hl=en