Stories
Slash Boxes
Comments

SoylentNews is people

posted by Fnord666 on Saturday February 10 2018, @08:12AM   Printer-friendly
from the maybe-Y-will-be-better dept.

Chris Siebenmann over on his personal web page at the University of Toronto writes about X networking. He points out two main shortcomings preventing realization of the original vision of network transparancy. One is network speed and latency. The other is a too narrow scope for X's communication facilities.

X's network transparency was not designed as 'it will run xterm well'; originally it was to be something that should let you run almost everything remotely, providing a full environment. Even apart from the practical issues covered in Daniel Stone's slide presentation [warning for PDF], it's clear that it's been years since X could deliver a real first class environment over the network. You cannot operate with X over the network in the same way that you do locally. Trying to do so is painful and involves many things that either don't work at all or perform so badly that you don't want to use them.

Remote display protocols remain useful, but it's time to admit another way will have to be found. What's the latest word on Wayland or Mir?

Source : X's network transparency has wound up mostly being a failure


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 3, Insightful) by jmorris on Saturday February 10 2018, @08:10PM (2 children)

    by jmorris (4844) on Saturday February 10 2018, @08:10PM (#636113)

    Guys, stop defending X. The world moved on, the assumptions behind X's design have changed and that means it is going away.

    Those assumptions were 2D image prims that were easy to send over the network. Look at old screenshots and it is easy to imagine how practical that was on 10mb ethernet. We can rage about the reality that everybody is doing local font rendering and bitmapping literally everything but it won't change. Then we get GL and the explosion of textures and other crap. Yes GL is network transparent but have you done the math lately? Video cards come with 8GB and rising. The other issues in X could be addressed and solved, the round trip delays, etc. But the math of bandwidth is simply brutal and remorseless. Now compare to doing all rendering on the same machine with the application and sending out a high quality x265 stream of the final product. A video stream takes at most 20mbps for a really high quality full screen full HD with sound. And being a video stream it can suffer lost packets, make use of RTSP and such to make it more robust. And no long delays to fill gigabytes of textures on scene changes.

    What we want is a protocol to get there that allows the things we care about:

    1. Send an individual application window, not whole desktops.

    2. Fully open protocol. Probably want pluggable video codecs so we can use VP9, etc. while the corporate world will insist on X265.

    3. Like X, applications run on the app server without needing to be displayed anywhere but do somehow share access to the GPU and custom video compression hardware.

    4. A standardized way to share local resources from the display side, like storage, misc USB devices, etc. This gets close to crossing streams with virtualization but it is a popular feature and whoever has it will beat all comers who do not.

    Fighting a rear guard defense of X means we aren't going to get a seat at the table and will end up reverse engineering whatever Microsoft, Apple, Google, RedHat and such impose from above.

    Starting Score:    1  point
    Moderation   +1  
       Insightful=1, Interesting=1, Overrated=1, Total=3
    Extra 'Insightful' Modifier   0  
    Karma-Bonus Modifier   +1  

    Total Score:   3  
  • (Score: 0) by Anonymous Coward on Sunday February 11 2018, @05:09AM (1 child)

    by Anonymous Coward on Sunday February 11 2018, @05:09AM (#636266)

    1. Single windows. Check. Also capable of shadowing full desktops if you want.
    2. GPL/Expat (MIT?)/BSD licensed code and nothing to hide, lots of image/video codecs to pass data around. Check.
    3. Maybe you mean "unlike". Check anyway, you can keep them running or kill them. They think they are using plain X11, then get shown somewhere else (with zoom... HiDPI for old apps for free). Check.
    4. Partial check, it can connect some things like pulseaudio or dbus or printing. And patches for other things (like X11 tablets... I guess full XInput) are welcome. Storage is probably better done via FUSE SSH or plain old NFS.

    What is not there is the will of the powers that be to use all this "old tech" that solves issues now with lots of backwards compability, because they don't control it, or if they do, it means less jobs rewriting everything and creating a huge new mess that needs support contracts.

    We are not getting a seat, that is for granted. You can play their bitch all you want, but they will not give a seat. Just like in the past, and the solution is probably the same: don't bend over, go our own way.

    • (Score: 2) by jmorris on Sunday February 11 2018, @06:23AM

      by jmorris (4844) on Sunday February 11 2018, @06:23AM (#636283)

      Ok, that looks fun. Except the manpage says all software video encoding so only expect VNC level performance. And when I just installed it and tried it locally I "lost" bigtime. Tried "xpra start-desktop" then "xpra attach" and got a black box full of nuthin. So did "xpra exit" and X went bye-bye. It was still there somewhere, music was still playing so the player hadn't lost contact with X, just the screen had went to a text console. Other text consoles didn't have sound, back to tty1 and sound. Weird. A quick drop to single user and back seems to have cleanly stopped and restarted the desktop.

      Playing more, started firefox in a single app mode window since it is one of the examples in the manpage. SLOW, even running on the local machine. Not looking like a contender, at least not on Fedora.