Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site hropus.UUCP
Path: utzoo!watmath!clyde!cbosgd!ihnp4!houxm!hropus!ka
From: ka@hropus.UUCP (ka)
Newsgroups: net.micro.pc
Subject: Re: Re: Copy Protection - a case study
Message-ID: <180@hropus.UUCP>
Date: Tue, 31-Dec-85 15:10:38 EST
Article-I.D.: hropus.180
Posted: Tue Dec 31 15:10:38 1985
Date-Received: Wed, 1-Jan-86 01:06:32 EST
References: <3624@think.UUCP> <131300002@ima.UUCP>
Organization: Bell Labs, Holmdel, NJ
Lines: 21

With respect to hiding a serial number in each program, Timothy Thomas says:

> Even if you did this, you could get two copies of the program, and compare
> them byte by byte.   The areas in which they differ are obviously those
> which contain information specific to that disk, namely the serial number.
> It would be a simple matter of substitutine 0's or nulls in place of those
> differences.  

You could encrypt the code for a few subroutines using the serial number,
and decrypt these routines before they were called.  If the encrypted
code for these routines were replaced with 0' or nulls, the code would no
longer work.  Alternatively, you could require that the serial number had
odd parity, and have the program crash the system (after optionally trash-
ing the disk) if the serial number had even parity.

>There will never be an answer to the software problem, except good software
>at reasonable prices, like Borland.

This is probably true.
					Kenneth Almquist
					...!houxm!hropus!ka