From owner-freebsd-current  Fri Feb  9 06:47:50 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id GAA13529
          for current-outgoing; Fri, 9 Feb 1996 06:47:50 -0800 (PST)
Received: from time.cdrom.com (time.cdrom.com [192.216.222.226])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id GAA13521
          for <current@freebsd.org>; Fri, 9 Feb 1996 06:47:48 -0800 (PST)
Received: from localhost (localhost [127.0.0.1]) by time.cdrom.com (8.6.12/8.6.9) with SMTP id GAA20408; Fri, 9 Feb 1996 06:47:08 -0800
To: Poul-Henning Kamp <phk@critter.tfs.com>
cc: Julian Elischer <julian@ref.tfs.com>, terry@lambert.org (Terry Lambert),
        current@freebsd.org
Subject: Re: FS PATCHES: THE NEXT GENERATION 
In-reply-to: Your message of "Fri, 09 Feb 1996 15:18:42 +0100."
             <319.823875522@critter.tfs.com> 
Date: Fri, 09 Feb 1996 06:47:08 -0800
Message-ID: <20406.823877228@time.cdrom.com>
From: "Jordan K. Hubbard" <jkh@time.cdrom.com>
Sender: owner-current@freebsd.org
Precedence: bulk

> I want to be able to define a policy for permissions in /dev, and no
> form is more unix-like and suitable than
> 
> 	chmod 644 tty*
> 	chown root.dev disk/*

Actually, when we stood around discussing this, we agreed that the
"journaling" mechanism would have to deal with the addition of
wildcard rules like this.  I understand the need to define permissions
for entire classes of devices, not just single ones.

It's also not a question of smart or not smart, it's a question of
upholding the Principle of Least Astonishment and also not opening the
can of worms any farther than it has to be opened.  By preserving the
old semantics, all your various shell scripts and system admin hacks
survive and you don't have the "multiple incarnation of /dev (say for
chroots) initialization problem" to worry about, either.

					Jordan