Maybe it turns out we didn't need network transparency for graphics, but do for audio? Least ways, I myself think I'd stream audio from one server to another a LOT more frequently than I'd run an app.
Opposite directions or not, both of those directions seem to be meeting my needs better.
That's nice.The latency of fiber optics is 5 microseconds/km.
That's funny because there are remote desktop solutions that can deliver just that.I don't need to watch full screen 1080p movies over X.
That's fine. I don't see how that is relevant. Even if you think your talking about Compiz or whatever it's not really relevant as graphics acceleration and compositing should be happening on your local machine.I don't need my remote windows to jiggle like Santa Clause.
Connecting to a X application running on a server in the room next to you is uninteresting.Even a 10 millisecond latency over copper is more than sufficient for a normal remote session.
I use X at work and running something as simple as a browser on a desktop on a different Vlan is painfully slow a times.
Meanwhile I can run Citrix with Microsoft Windows hosted in a Xen virtual machine 50 miles away over a busy internet connection and response time is immediate. Draw times are instantaneous.
X Windows Networking is obsolete. It CAN be improved and it CAN be fixed, but I don't see anybody working on X12R1 yet.
Also it's irrelevant when discussing Wayland since Wayland can run X11 networking also. Just like Microsoft and OS X's display managers can.
X11 sucks for networking. If even the guys who designed and wrote the original protocol (e.g. Keith Packard) are getting behind Wayland, you have to wonder if maybe -- just maybe -- there's some truth to what people say about X11's performance on modern networks.
The funny thing is that many of the same kinds of clueless people who think X11 is great for networking are the ones who claim that X11 sucks for local displays because of networking overhead, which is untrue. (Wayland, Windows GUI, OS X, etc. all involve a local IPC communication channel between applications and the rendering/display server backed by direct-rendered acceleration... just like modern X11.)
Wayland really is just about cleaning up the cruft. The use of EGL and KMS and DRI2 for core rendering could be dropped in X.org and all the DDX drivers could be replaced with that common backend and an OpenGL-based renderer for the core drawing commands. That isn't something Wayland has that X11 cannot have.
Wayland is just a cleanup of the protocol and architecture. It removes all the legacy cruft of the X11 protocol (including, yes, transparent protocol-level networking). Its architecture relies on a display server doing the compositing and core window management (like OS X or Win7 do). It expects applications to do more of the work in terms of window decorations and interactions rather than relying on a third process (the window manager) to do those things. It requires applications to do their own rendering. It expects a proxy display server to be used for networking when you want it.
Really, the only advantage the X11 networking model theoretically has is that the display server can do all the rendering while the application client just issues commands. That means that in a thin client environment you let the thin clients themselves do the rendering instead of having the mainframe do it. Unfortunately, that's not actually all that well supported in this day and age, because GLX only handles modern OpenGL with proprietary vendor extensions. Plus, modern OpenGL apps require very low latency during two-way communication with the GPU, meaning that the network separation between the application and the display server imposes a huge latency/performance hit. It's simple better to get thinner thin-clients without heavy-duty GPUs and beef up the server with as much GPU horse power as possible.
X11 does well over networking with what it was originally designed for - that design is not really equivalent to current desktops.
Ik think that the entire need for the networked apps is useless nowadays. Here's why I'm thinking that:
-Computers nowadays are fast enough to run any app on the planet;
-For server-client configs that are usefull in the internet age we already have the cloud/webapps like gmail, facebook, etc.;
-Remote filesystem mounting makes it even more useless as you can simply run your own app local on the networked data as if it was local data;
-Existing crappy remote desktops solutions (MS Windows style) are good enough to just turn on a remote app in case of remote scientific calculations that need to be put in action on a powerfull remote 'super'computer;
-In case of special cases, such as audio and video there are dedicated protocols and apps for that already. Think of VLC for networked music and video and 'that' HD video streaming on devices such as a Playstation 3 for which there are also apps available that can use this protocol on your computer.
Now X11 protocol is not a solution that enables you to do things you couldn't do otherwise anymore. But the problem is that while the X.org implementation doesn't exclusively creates advantages, it still fully imposes all (or most/many) of its downsides.
So Wayland is not such a bad thing to considder.
Actually I think that people are using networked applications far more nowadays then they ever did in the past. I use apps every day. 99% of everybody at my work uses remote apps... but they _don't_know_it_. That is how seamless it is.Ik think that the entire need for the networked apps is useless nowadays.
examples of network applications:
It's just that they are not using X11. They couldn't use X11 even if they wanted to. They use web apps or ICA or whatever.