From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: To: 9fans@9fans.net Date: Tue, 21 Apr 2009 17:13:52 -0500 From: blstuart@bellsouth.net In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Subject: Re: [9fans] Adventures of a home user Topicbox-Message-UUID: eebd2740-ead4-11e9-9d60-3106f5b1d025 > i'm speculating on the design of the auth system. i wasn't > there so i could be wrong. but in order to have a terminal > that many people could log into would require either > (a) killing off the original factotum on logout and changing > eve back to bootes or something. and beware the 1001 places > that stash eve somewhere. I wasn't there either, but I do have some fuzzy memories of running 2nd edition. Factotum came along later than that. It may have been new with 4th edition. The general impression I always got was two-fold. >>From a philosophical point of view, whoever is logged into a terminal is the owner of the hardware. After all, physical access is the ultimate privilege. So if you just don't bother creating a logout mechanism, then the only time the hardware doesn't have an owner is from reset to authenticating a user. The practical side means that a lot gets simpler and a number of concerns go away. If the system reboots between users, there's nothing of the first user left on the terminal when the second user comes along. I don't have to be careful to clean up all the processes, etc left behind; rebooting does that. The old login trojan horses also go away. ^t^tr is grabbed by the kernel and we reboot. It seemed a little strange to me at first, but the more I worked with the system, the more logical it seemed. BLS