From dbh@doc.ic.ac.uk Fri Oct 17 18:36:51 2003
X-Coding-System: undecided-unix
Mail-from: From dbh@doc.ic.ac.uk Fri Oct 17 18:36:51 2003
From: Dave Gordon <Dave.Gordon@bigfoot.com>
To: Denis Howe <dbh@doc.ic.ac.uk>, Dave Gordon <Dave.Gordon@bigfoot.com>
Subject: FOLDOC entry for "Multics"
X-Host: fluorine.eu.sun.com
X-Url: sugcha
X-Browser: Mozilla/5.0 (X11; U; SunOS sun4u; en-US; rv:1.0.1)
    Gecko/20020920 Netscape/7.0
Sender: Denis Howe <dbh@doc.ic.ac.uk>
Date: Fri, 17 Oct 2003 18:36:48 +0100

FOLDOC entry for "Multics"

Apropos the following paragraph in the current entry:

The Jargon file 3.0.0 claims that on some versions of Multics one was required to enter a password to log out but James J. Lippard <lippard@primenet.com>, who was a Multics developer in Phoenix, believes this to be an urban legend. He never heard of a version of Multics which required a password to logout. Tom Van Vleck <thvv@multicians.org> agrees. He suggests that some user may have implemented a 'terminal locking' program that required a password before one could type anything, including logout.

Some of the more interesting features of password control on Multics were: 

passwords were normally (or only?) changable at login time (login -cpw), not (AFAIK) *during* a session

you could be *required* to change your password after a specific number of logins (somewhat like UN*X password aging); this could of course be set to "1", requiring a password change on *every* login

you could let the system generate a new password for you (login -gpw) rather than choosing your own.  The password was generated as a random string of syllables fro a table, and was thus guaranteed pronounceable, in the hope that this would also make it memorable

the administrator could *enforce* generated rather than user-chosen passwords.  Naturally, this could be combined with the enforced password change; thus the most extreme case would be that on *each* login, you would be issued with a one-time password to use on your *next* login!

PS: my credentials for this information are that I was a user (and would-be hacker;-) of the Multics system at the University of Bath from the time when it was installed in 1979 until 1982 ..



