RE: compression and encryption

Bill Hunt <bill@vpnet.com> Thu, 07 November 1996 20:06 UTC

Received: from cnri by ietf.org id aa19506; 7 Nov 96 15:06 EST
Received: from portal.ex.tis.com by CNRI.Reston.VA.US id aa19628; 7 Nov 96 15:06 EST
Received: (from majordom@localhost) by portal.ex.tis.com (8.8.2/8.8.2) id PAA10231 for ipsec-outgoing; Thu, 7 Nov 1996 15:00:33 -0500 (EST)
Message-ID: <01BBCCBB.B8CB4320@bill.scli.com>
From: Bill Hunt <bill@vpnet.com>
To: "ipsec@tis.com" <ipsec@tis.com>, 'Steve Bellovin' <smb@research.att.com>
Subject: RE: compression and encryption
Date: Thu, 07 Nov 1996 14:55:30 -0800
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Sender: owner-ipsec@ex.tis.com
Precedence: list

good analysis.  one minor note:  If a burst is compressed, certainly
uncompressing packet k requires receipt of packets 1..k-1.  But if
packet k is lost, is it necessary to retransmit packets 1..k-1?  Or
can the history be reset and the process begun again with k set
as 1?  The latter would allow a higher packet loss, since the
"average" packet loss would be lower.

----------
From: 	Steve Bellovin[SMTP:smb@research.att.com]
Sent: 	Wednesday, November 06, 1996 5:54 PM
To: 	ipsec@tis.com
Subject: 	compression and encryption

I've been thinking a lot about doing compression over a lossy medium
such as IP. 

[...]