From: patrick@Verity.COM (Patrick Horgan)
Date: Fri, 22 Sep 95 08:23:07 PDT
To: perry@piermont.com
Subject: Re: Seeds which depend on machine states
Message-ID: <9509221519.AA19310@cantina.verity.com>
MIME-Version: 1.0
Content-Type: text/plain


> 
> Miguel Diaz writes:
> > It is my suspicion that seeds which depend on machine 
> > states(ie state of your computer at a specific instance of 
> > time) would always be subject to scrutiny and de-cryption.
> > As long as the software used to encrypt is not self-modifying, 
> > the machine state can (through careful manipulation involving 
> > temperature, clocks, processes etc)always be replicated and 
> > fixed to an acceptable degree.
> 
> Try getting a human to type with the same timing, to microsecond
> precision, the same way twice.
> 
That assumes that you have someway of measuring the timing to microsecond
precision.  On most machines I've been on, if you get something time-
stamped, even if there is a microsecond portion of the timestamp it's
meaningless because it wasn't based on a timer with the required precision.
If a timer can only resolve milliseconds, the microseconds don't have any
meaning.

Patrick
   _______________________________________________________________________
  /  These opinions are mine, and not Verity's (except by coincidence;).  \
 |                                                       (\                |
 |  Patrick J. Horgan         Verity Inc.                 \\    Have       |
 |  patrick@verity.com        1550 Plymouth Street         \\  _ Sword     | 
 |  Phone : (415)960-7600     Mountain View                 \\/    Will    | 
 |  FAX   : (415)960-7750     California 94303             _/\\     Travel | 
  \___________________________________________________________\)__________/