Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10 5/3/83; site gsg.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!bellcore!decvax!gsg!kathy
From: kathy@gsg.UUCP (Kathryn Smith)
Newsgroups: net.news.adm
Subject: Re:ading other peoples' mail
Message-ID: <182@gsg.UUCP>
Date: Wed, 16-Apr-86 10:07:54 EST
Article-I.D.: gsg.182
Posted: Wed Apr 16 10:07:54 1986
Date-Received: Thu, 17-Apr-86 05:33:56 EST
References: <703@frog.UUCP> <12400018@uiucdcs> <2410@jhunix.UUCP>
Organization: General Systems Group, Salem, NH
Lines: 37


	With respect to the original posting, I suspect the systems where the
SA has time to randomly read the mail going through are VERY few and far 
between.  Our systems here at GSG are small compared to most of the sites
on the net, and I certainly don't have that kind of time.  The only reason I
would look at mail going through the system is if our mailer is having fits, or
I have reason to think someone is trying to break in here/elsewhere with it.
Otherwise, it is not my business.  
	
	On a more general level, however, I think if you went around to most
companies and talked to the accounting/administrative people who don't really 
know how UNIX works, they would be shocked to discover that someone with the
root password could look at all their files trivially.  They just don't realize
that anything you put in a computer is fundamentally insecure with respect to
a core group of people.  

	Finally, if you want to be really sure that no one else is reading 
your mail, there are lots of encryption and decryption programs out there.
True, very few, if any, of them are fool proof, but how many of you really
think someone wants to read your mail badly enough to bother breaking an
encryption program to do so?  This will cure 99.9% of the 'Peeping Toms'.

	Personally, I could wish more SA's paid attention to the mail that 
goes through their system.  I spent about two months last year trying to 
find out why a previously working mail path between here and mit-eddie stopped
working.  I must have sent out more than thirty mail messages trying different
paths, most of which neither got through to their destination, nor came back
as undeliverable.  They must have gone somewhere, but apparently none of the
SA's involved either noticed, or cared enough to help me solve the problem by
answering the pleas in the messages to tell me where it was going, if it 
wasn't reaching MIT.  I think an SA who pays some attention to what goes
through his/her system is a great improvement over this situation.

						Kathryn Smith
						General Systems Group
						Salem, NH 
						(...decvax!gsg!kathy)