1997-01-18 - One time pads, Next question?

Header Data

From: AaronH4321@aol.com
To: cypherpunks@toad.com
Message Hash: 628f3bb5a157aaf3426a83e3c8cda774c1dfb31057a61484da3a88c7f8ce5c03
Message ID: <970118173710845991204@emout17.mail.aol.com>
Reply To: _N/A

UTC Datetime: 1997-01-18 22:38:05 UTC
Raw Date: Sat, 18 Jan 1997 14:38:05 -0800 (PST)

Raw message

From: AaronH4321@aol.com
Date: Sat, 18 Jan 1997 14:38:05 -0800 (PST)
To: cypherpunks@toad.com
Subject: One time pads, Next question?
Message-ID: <970118173710_845991204@emout17.mail.aol.com>
MIME-Version: 1.0
Content-Type: text/plain


Okay, with the intel I have gotten from the responses let me modify and
clarify. 

So, I now know that you can't use a PRNG to create a OTP. If I did have a RNG
that could create a "cryptographically" random, million character OTP and I
physically loaded the data onto each computer I wanted to communicate
with(e.i. group of laptops all loaded at the same location, same time), how
cryptographically secure would the data be. I know that physical security
would be a nessescary to the system. 

If I add some type of error checking to make sure someone that knows the
plaintext doesn't modify specific characters, an attacker knows what the
message is (e.g. "Send $10 to Bob") the message can be altered (to "Send $99
to Bob") even though it is encrypted. 

So, lets recap. If the keyig material is kept secure, and if the OTP is
cryptographically random, and if error checking is implemented then would
this be considered a secure(however encumbering it may be) system. 

Aaron.....





Thread