We Have Mir & Wayland, But There Still Could Be X12

[Delgarde]

The core X11 protocol is. But sure, no one uses it anymore.

So all those X apps I regularly run remotely... I'm imagining it?

[JS987]

it's the right way to get the job done, it's like supposed to be 10 year ago.

It's more like we will return 10 years back with performance.

[xeekei]

It's more like we will return 10 years back with performance.

You're suggesting that Wayland is slower than X?

[Vim_User]

So all those X apps I regularly run remotely... I'm imagining it?

Nope, as long as you don't run ancient Motif apps X is just acting like a very shitty VNC.

[halfmanhalfamazing]

Well that is the case already. Wayland basically is the new X. It's basically X12, except that it's entirely different in design than X11 and doesn't maintain compatibility with the X11 codebase... so it's not called X12.

What would the advantage be of renaming Wayland to X-something?

I do not mean renaming.

Currently, Xorg consists of multiple pieces, xserver being just one of them, using the X11 protocol. X RandR is another component.

Let's just take x.org 7.6 for example, which has in it xserver 1.9, xrandr (it was version 1.2 IIRC), and other pieces - piece3, piece4, piece5, etc.

A new X12 would consist of wayland, xrandr, and other pieces. piece3, piece4, piece5, etc.

Now, I realize that all of these said pieces would most likely be re-written from scratch, with whole new parts. But all I'm asking is why could not Wayland sit at the heart of the new X12 considering how far along it already is? Especially considering that Xwayland (should) come along nicely as a good X11 compatibility layer.

It doesn't have to be called x-anything..... ever. But with continued advancements, as the X.org crew creates what it sees as its new vision for a new X12 protocol, they could bolt on/bolt in the new pieces onto the already completed Wayland.

[xeekei]

I do not mean renaming.

Currently, Xorg consists of multiple pieces, xserver being just one of them, using the X11 protocol. X RandR is another component.

Let's just take x.org 7.6 for example, which has in it xserver 1.9, xrandr (it was version 1.2 IIRC), and other pieces - piece3, piece4, piece5, etc.

A new X12 would consist of wayland, xrandr, and other pieces. piece3, piece4, piece5, etc.

Now, I realize that all of these said pieces would most likely be re-written from scratch, with whole new parts. But all I'm asking is why could not Wayland sit at the heart of the new X12 considering how far along it already is? Especially considering that Xwayland (should) come along nicely as a good X11 compatibility layer.

It doesn't have to be called x-anything..... ever. But with continued advancements, as the X.org crew creates what it sees as its new vision for a new X12 protocol, they could bolt on/bolt in the new pieces onto the already completed Wayland.

Wayland doesn't need those extra pieces. That's the whole point. Wayland is the replacement for X11. Wayland uses OpenGL ES or Pixman instead of XRender. The replacement for the Xserver would be either Mutter, KWin, Weston, Openbox (if they want to port them). Wayland is the protocoll to make the parts of X.org that we actually use, to talk to each other, instead of going through this X middleman.

[Delgarde]

Nope, as long as you don't run ancient Motif apps X is just acting like a very shitty VNC.

But it works, and it works well. I can, and regularly do, ssh onto other machines to run complex apps like Eclipse or sometimes Firefox, or various in-house development tools.

[sarmad]

But it works, and it works well. I can, and regularly do, ssh onto other machines to run complex apps like Eclipse or sometimes Firefox, or various in-house development tools.

While that is currently true, it's not where software development techniques are heading to. Apps that run across the network should actually consist of a client part that runs on the client. In other words, the applications themselves should be network transparent, rather than the rendering stack.

[mrugiero]

I think the right question to ask is who will work on it, in case it exists. I mean, the experienced X guys are mostly centered on working on Wayland, and that's their bet for the future. X12 would probably require a development time similar (greater, actually, because even when they have most of the infrastructure done and most of the work is shrinking the protocol, it still requires some protocol design, and THEN actual coding) to the one Wayland still needs to be mainstream.

On the Wayland as heart of X12, I think if it would exist (is there any need for it?) it would make sense to reuse the Wayland protocol (and library) as the buffer manager of the X12 protocol. Of course, an X12 would require more than just Wayland, as it's proposer said.

[nachoig]

We Have Mir & Wayland, But There Still Could Be X12

Why? Who cares? Who needs? What developer talked about it in this week?