🚀 go-pugleaf

   Date: Tue, 29 Aug 89 15:55:46 EDT
   From: arnold@mathcs.emory.edu (Arnold D. Robbins {EUCC})
   Newsgroups: gnu.bash.bug
   References: <8908291715.AA12548@rose.vlsi.ll.mit.edu>
   Organization: Math and Computer Science, Emory University, Atlanta GA

   In article <8908291936.AA29762@aurel.caltech.edu> you write:
   >   but lots of people leave a terminal sitting at a shell prompt for
   >   days (or weeks!).
   >
   >So what?  They are not taking any cycles by just sitting there.  They
   >would be taking up more cycles if the shell was getting alarm clock
   >signals telling it that `autologout' time had passed.

   The cycles aren't the point.  The terminal lines are.  Around here,
   lots of our users come straight in on terminals, which are limited in
   number.  This is quite common in lots of places.  Autologout timers
   *are* useful.  I even added one to the csh here.  Also, KSH has it,
   the TMOUT variable if I remember right.

But what prevents users from unsetting the variable?  Additionally you
mentioned that current implementations only work at shell prompts, so
what prevents a user from being idle within an Emacs (my normal state :-).

Are you proposing that the system administrator would compile this
variable into the shell so that it was on?  Would you have to have
special privileges to turn it off?

If the issue is users wasting a resource, perhaps the users could be
made aware of that fact.  If the administrator wants to blow away idle
users, why not just have a daemon running that kills idle jobs?

If I can be convinced that this is a feature that a USER wants, then I
will glady put it into the shell.  But if it is an artifact to make the
lives of a select few better, I don't think I will be inclined to
include it.

Brian