If Or When Will X12 Actually Materialize?

[nanonyme]

anyway a ligther implementation + encryt support + maybe ssl/tls support + and faster would be awesome, in fact, in theory if the protocol is properly created you could even render opengl remotely too.

Meh, everyone needing encryption for remote X is tunneling X over SSH anyway. And indirect rendering with OpenGL should work remotely over X protocol as far as I've been told. Dunno of AIGLX.

[Silverthorn]

The best approach, in my opinion, would be to select one feature and deprecate it. If the functionality is still needed, a new improved feature should be added to replace the old one.

This situation reminds me of what happened with OpenGL 3.0. It was delayed at least a decade. When it finally arrived they've focused on deprecation instead of adding new features.

[nanonyme]

The best approach, in my opinion, would be to select one feature and deprecate it. If the functionality is still needed, a new improved feature should be added to replace the old one.

Maybe, maybe not. Can you come about with functionality that should actually be deprecated?

[fabiank22]

Again, the real problem with "getting rid of all the bloat in X" is that nobody can agree on what that bloat actually is. For every person who says "feature X is crap, nobody uses it" there's another person saying "hey, wait, I use that all the time".

What problem are we actually trying to solve ?

Well the problem is that there are different states of progress/design, let's call them "layers" for the lack of a better term. For Example you have an old set of C-Bindings Xlib, which are still used frequently, then you have it's sucessor XCB which isn't used that often and is still in progress. Julien was happy about one(1) ported lib within all X and the Xlib-Code isn't even dropped yet. And then there are Bindings which work around the Problem alltogether by using Xlib but fixing it's shortcomings, like QT.

Now for all these layers we have different interests. Someone who is maintaining a lot of old code(let's say a LTS-distro) or someone who doesn't want to break things(Luc Verhaegen seems to fit that bill for Example) wants to keep Xlib for as long as possible, because a) It works okay, it's not the main source of your problems, other parts of the stack are and b) It doesn't need manpower. Apart from fixing bugs all work is done within toolkits which don't concern you right now.

Now everybody who is forced to actually work at the connections between Xlib and Toolkits oder Xlib and other stuff wants it gone asap. It's one of the reasons the knowledge-level needed to develop this stuff is so high, it's designed with hardware from 20 years ago in mind and there is an impenetrable wall of weird workarounds which make your job harder.

Now most users want the toolkits, because that's what they are using. 95% of all graphical users will use a system tray, but X over network? I'd be surprised if there were even 5% using this. So they won't care if the solution is ugly or breaks the principles of good software. Also most graphic-stack-stuff is moving at a pace where I can't fault the good of KDE or Gnome for working around it, even if it's ugly. What do you expect QT to do, pay somebody to port Xlib to XCB? Not if their developers can fix it on a sidenote in a month.

So can you say conflict of interest? The problem is - they all are right - they all have a valid point. However we already have parts of the stack were the layers are pretty clear - Gallium, KMS, new hardware. So I don't fault anybody suffering from "let's do it another day"-syndrome.

The solution is probably: If a developer or more are willing to step up and they can work around the layers, a step at a time, then great. But let's not try to get a whole community to waste their already spare manpower.

[cl333r]

I agree that X11 should be replaced by "X12" or WayLand, but I don't agree with some folks who say that writing X12 from scratch is as hard as sending a crew outside the Solar System, common, a guy from Red Hat did a good job with WayLand, imagine if a few folks from IBM/Novell/etc jumped in to help him, in a year it would be finished and then other folks could start writing all the main stuff over to WayLand/X12/whatever and in about 3 years we'd already have a pretty usable desktop already.

[nanonyme]

I agree that X11 should be replaced by "X12" or WayLand, but I don't agree with some folks who say that writing X12 from scratch is as hard as sending a crew outside the Solar System, common, a guy from Red Hat did a good job with WayLand, imagine if a few folks from IBM/Novell/etc jumped in to help him, in a year it would be finished and then other folks could start writing all the main stuff over to WayLand/X12/whatever and in about 3 years we'd already have a pretty usable desktop already.

*cough* I thought we do have a pretty usable desktop already...

[Silverthorn]

Maybe, maybe not. Can you come about with functionality that should actually be deprecated?

I'm not into X development so I wouldn't even know where to start looking

[Smorg]

Don't know about you guys but I really do rely upon NX for school and wish something like it were well integrated in to xorg since NX is incredibly flakey but does seem to perform better than anything else out there. I don't have a laptop nor would I want one since theres computers wherever I go. Sure I could get by with putty, but sometimes it's nice to be able to run anything including the terminal emulator I actually use (Konsole). Their java NX webapp is so nice... just browse to my server and I've got my desktop literally anywhere.

It still sucks in a lot of ways and plain X forwarding is totally useless unless you're on a gigabit LAN (and even then, NX is better). Really... if the only purpose of having an X "server" is as a VNC replacement with all the disadvantages that go along with it then it needs a total redesign because it fails miserably in that regard to the detriment of everything else (like running X locally.)

[agd5f]

Is Zaphod mode really gone? Maybe there used to be more to it but I still use a setup where I have DISPLAY :0.0 and :0.1.

Zaphod is still around. It's just not used as often as xrandr and is only supported in a few drivers. It's more of a ddx implementation detail than a specific xserver feature.

[cl333r]

*cough* I thought we do have a pretty usable desktop already...

I didn't mean the current desktop, but the future one - based on X12/Wayland/whatever which after a few years of work on it and on Gnome/cairo/etc stuff would make for a pretty usable desktop in the end.