Newsgroups: comp.windows.x
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!think.com!snorkelwacker.mit.edu!bloom-beacon!dont-send-mail-to-path-lines
From: mouse@lightning.mcrcim.mcgill.EDU (der Mouse)
Subject: Re:  Making a display secure?
Message-ID: <9103201028.AA24936@lightning.McRCIM.McGill.EDU>
Sender: daemon@athena.mit.edu (Mr Background)
Organization: The Internet
Date: 20 Mar 91 10:28:29 GMT
Lines: 33

> I have a short question.  Is it possible to make a display on a
> machine secure to input from other users on the same machine?

In X terms, yes, or at least as secure as your server supports.
(Typically proof against accident and casual attempts, but not against
a reasonably motivated person with some crypto-knowledge.)

> This is of course rather against the whole philosophy of being able
> to display anything anywhere, but we're having probelms with certain
> users who start up software like Mongo, with it configured for
> sunview ('cos of a 3/50 on the network belonging to a prof who
> can't/won't get into X) and that wipes out the screen on :0.

This sounds as though the offending user is starting SunView, which
starts arguing with X over who owns the framebuffer/mouse/keyboard.
(Starting a SunView application without SunView running just produces
an error message, doesn't it?)

If so, this is out of the realm of X and straight into operating system
arbitration of access to the hardware.

Since you talk about sunview and a 3/50, I assume you are using Suns.

What I would probably do is remove SunView support on all machines
except that -3/50 you mentioned, probably by deleting the relevant
binaries.  You also might be able to get away with deleting /dev/win*,
though I don't know whether that is sufficient.  (They are not needed
for X.)

					der Mouse

			old: mcgill-vision!mouse
			new: mouse@larry.mcrcim.mcgill.edu