What's wrong with wayland?
martyj19 at comcast.net
Sat Feb 19 04:20:11 PST 2011
On 02/19/2011 12:59 AM, microcai wrote:
> I used to think wayland is better.
> But now, I don't.
> What is Wayland ?
> Let me tell you:
> Final Wayland = X11 that rip out everything but GLX.
> Current wayland still missing Input framework like XIM. And batch of
> other stuffs. Adding it will led wayland to another X11.
> Wayland forces every app uses OpenGL as drawing facility. Then they
> said, hey , mine are faster than you!
> Of-course wayland is faster! if X11 app also switch to OpenGL
> completely , X11 is also faster!
> needless to say , wayland has only one working driver, but X11 has a lot!
> That's what I'm now thinking about wayland, maybe I am wrong. And I hope
> I am wrong.
> Thanks for any comments.
> wayland-devel mailing list
> wayland-devel at lists.freedesktop.org
To expand on what Sam said a little, everything that the Wayland compositor does is being done now by the window manager, only it is being done in a far less efficient way because the drawing is done into regular-memory buffers instead of directly into video memory where the GPU can accelerate it, and it is all mediated by several trips through the X server instead of being done directly in the client process.
If you had input methods before, you will have them then. It would be client side, like most all other input processing. I am hopeful we can do them in a way that is not bolted on to the side. I continue to think that just a little shared state would be very helpful.
The new way gets video device drivers down to lean and mean, doing the things that device drivers should do: you know, bus enumeration, programming the registers. Get them out of the business of implementing anything else at that level. There is little point in having a lot of device drivers if they are for hardware that is all in scrap heaps. We have, I think, over the last week gotten the Big Three video drivers all functioning. Nothing stops someone from writing a DRM driver for a piece of hardware that has an existing X driver if one were needed; and it will be very much simpler than the X driver.
If remote clients is your thing, instead of forecasting doom because the old way may not work any more, first off, nothing stops someone from writing a thing that listens on port 6000 and acts just like a remote X server only it is a Wayland client, and second off, nothing stops someone from redesigning and rethinking what the proper remote protocol is, using modern encryption and compression techniques and whatever else is needed to get a VNC like solution that performs well and is secure. Let's not penalize every local client in order to get remote clients.
I encourage everyone to relax, pay a little less attention to the day-to-day details, and let the development go on. In a project of this size, not everything can get done first, and it is still very early. It may not be obvious what is scaffolding and what is done, but I think we have the best available people on the planet working on this and paying attention to it, and I am confident it will be polished until it comes out good in the end.
More information about the wayland-devel