Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site sdcrdcf.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!oliveb!hplabs!sdcrdcf!markb
From: markb@sdcrdcf.UUCP (Mark Biggar)
Newsgroups: net.crypt
Subject: Re: Encryption using compression
Message-ID: <2140@sdcrdcf.UUCP>
Date: Thu, 11-Jul-85 12:22:22 EDT
Article-I.D.: sdcrdcf.2140
Posted: Thu Jul 11 12:22:22 1985
Date-Received: Wed, 17-Jul-85 04:40:49 EDT
References: <5992@duke.UUCP> <11379@brl-tgr.ARPA> <5993@duke.UUCP> <78@oberon.UUCP> <2591@randvax.UUCP>
Reply-To: markb@sdcrdcf.UUCP (Mark Biggar)
Distribution: net
Organization: System Development Corp. R+D, Santa Monica
Lines: 9
Summary: 

Another thing that could be done would be to scramble the initial subistution
table that compress uses (compress starts out with a table giving
subistutions for the ascii char set and add subs for longer strings as it
is compressing) so that even if the final subistution and translation are
known there are still 127! possible decompressions.  Of course distributing
the initial table has all the key transmission problems as well.

Mark Biggar
{allegra,burdvax,cbosgd,hplabs,ihnp4,akgua,sdcsvax}!sdcrdcf!markb