Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!lll-crg!hoptoad!gnu
From: gnu@hoptoad.uucp (John Gilmore)
Newsgroups: net.crypt
Subject: Re: randomly adding bits/bytes (compressing text before encryption)
Message-ID: <1030@hoptoad.uucp>
Date: Fri, 22-Aug-86 01:02:56 EDT
Article-I.D.: hoptoad.1030
Posted: Fri Aug 22 01:02:56 1986
Date-Received: Fri, 22-Aug-86 07:22:05 EDT
References: <8608042018.AA04376@ucbjade.Berkeley.Edu> <7024@utzoo.UUCP> <1144@ttrdc.UUCP>
Organization: Nebula Consultants in San Francisco
Lines: 16

I wish people would take the time to read the compress program (this
includes you, Henry!) before flaming here.  The reason you don't
see anything in the "od -c" output is because the output of compress
starts off with 9-bit codes (which you are breaking up on 8-bit boundaries,
hence the gibberish).  These 9-bit codes mostly contain letters from the
plaintext; occasionally they contain pointers to strings of letters
built out of the plaintext.  As the file goes on, there are presumably
more pointers than letters (since otherwise it would be turning 8-bit
bytes into 9-bit codes and EXPANDING the file).  When it runs out of 9-bit
codes, it starts outputting 10-bit codes, and so on up to 16 bit codes.

I suspect that a lot could be done to decrypt the output, if the cryptanalyst
knew that "compress" was being used on the plaintext.
-- 
John Gilmore  {sun,ptsfa,lll-crg,ihnp4}!hoptoad!gnu   jgilmore@lll-crg.arpa
		     May the Source be with you!