In comp.sys.raspberry-pi,
Martin Gregorie   wrote:
> As others have said, everything with a GUI interface uses roughly the
> same component hierarchy to interface programs to graphical screens. The
> only difference is that Windows and Apple don't really admit that it
> exists or explain how it works, while UNIX has always made this explicit
> and so has Linux, which was initially written by Linus Torvalds as a
> cleanroom reimplementation of UNIX, has naturally done the same.

I'm not sure how "cleanroom" Linus' Linux was or is now, but that
doesn't matter. X11 was not "cleanroom" reimplemented for Linux, it was
one of several existing Unix GUIs that was ported to Linux. (I've also
"used" MGR on Linux, an alternative to X11 that predates Linux. I used
it for a couple of days, and never tried it again.)

X11 is vastly different from the windowing systems of the time it was
created. Under X Windows there's explicit support for:

* multiple users on multiple screens on one system (tricky to do under
  Linux, because of separating keyboard / mouse inputs). But if you want
  to, you can  to a new virtual console and start a new
  server (it will be :1 instead of :0)

* window server and applications running on different hardware
  -> different from mirroring a desktop, you could run just an xterm
     from a remote host, with everything else local
  -> I recall a guy in the mid-90s who ran a screen full of load
     monitors, each one running on a different system
  -> it also explains how systems like Xquartz can manage windows
     when it doesn't control the "desktop"

* complete separation windowing server from window manager

All this stuff makes X11 a complex environment and much different from
Windows or Apple's GUI.

Elijah
------
writes his own X11 start up scripts

--- SoupGate-Win32 v1.05