Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!husc6!mit-eddie!genrad!decvax!watmath!atbowler
From: atbowler@watmath.UUCP (Alan T. Bowler [SDG])
Newsgroups: comp.mail.uucp
Subject: Re: cancelling mail messages
Message-ID: <3700@watmath.UUCP>
Date: Mon, 1-Dec-86 20:06:58 EST
Article-I.D.: watmath.3700
Posted: Mon Dec  1 20:06:58 1986
Date-Received: Tue, 2-Dec-86 03:27:19 EST
References: <2718@bu-cs.BU.EDU>
Reply-To: atbowler@watmath.UUCP (Alan T. Bowler [SDG])
Organization: U. of Waterloo, Ontario
Lines: 16

Assuming a reasonably secure mail implementation which keeps track
of wheither a message has been read or not, there is no real reason
why you can't cancel any message that has not yet been read.  As
Mark Brader observed the mail systems at Waterloo have allowed this
for years.  We have switched the implementation technique recently
and will allow messages to remote machines be to be cancelled provided
the cancel command reaches the recipient's mailbox before the the message
has been read.  (This ability of course stops when we read a gateway
to a mail system that does not support a cancel).
  Some people seem to think that cancelling a message is immoral,
but I really can't see that there is really any problem with it.
It certainly is useful if a sender can retract an unread message
whose content is no longer meaningful.  I know I send enough messages
like "comming for supper?" that have no read meaning if the person
doesn't see them soon enough.  Being able to retract them avoids
bothering the person with unneeded mail.