Mark [1] send me an email replying to my post about X11 [2]:
From: Mark Grosberg <XXXXXXXXXXXXXXXXX>
To: sean@conman.org
Subject: Of X11 and remote access…
Date: Mon, 8 Oct 2007 13:36:10 -0400 (EDT)
> Sean,
I read your blog post on how X11 is network enabled and thought you glossed over a very important point. See, X11 is network enabled in that it has this concept of a program is a client to a display. So yes, if I am at home I can run some program on my X11 machine at work and send the display over.
But VNC on X11 goes one step better. It's a bit like “screen” (a program I still use) but for graphical apps (something I rarely use). I can have my X11 desktop sitting in some virtual framebuffer somewhere, connect to it, interact with it, go home, re-connect and it is exactly where I left it.
The normal X11 networking model is such that I loose my program state every time I move around. At least for me, personally, I am the kind of person who maybe yearly reboots a computer such that I loose my working environment and I frequently move about to different locations.
So for me, VNC on an X11 desktop makes total sense (well, screen makes even more sense since my main UNIX box probably doesn't have enough RAM to make a virtual framebuffer).
Oh, and VNC is way faster too, especially TightVNC. You don't wait 6hrs for a window to pop up while they negotiate event masks and other nonsense that should not exist.
He does have a point, and it's one of the problems with X11—you can't redirect a window to another display on the fly. Or, at least, I don't know of a way to do that. It also depends upon how you work. Me? I don't tend to leave applications running for any great length of time (especially graphical programs) because I have this irrational fear that they'll just keep sucking memory up until the machine becomes unusable.