Viewing Single Post From: One Time What?

Viewing Single Post From: One Time What?
Donald	Jun 22 2006, 08:16 PM
Elite member Posts: 454 Group: Trusted Member Member #8 Joined: September 2, 2005	Pulsar posted a tutorial about one time pads HERE I talked about them a little bit in this topic (Yeah, topic started off about books, we wandered a bit) And there I cover the question of WHY a one time pad is proveably secure. To sum up, the main issue is that a one time pad, DONE CORRECTLY, can be decrypted to be ANY text you want, depending on the key. And so, as long as you don't know the key, you can't know anything about the message. This isn't TRUE with normal block ciphers, like say, AES, or even something simple like a Caesar or Viginere. Take any of those ciphers, use a key to encrypt some plain text into a crypt text, and The odds are right next to impossible that you could find any key other than the correct key that would decrypt the crypt text back to a reasonable looking plain text. So when you find a key that decrypts the crypt text into plain text that makes sense, you know you've found the right key. BUT, with a one time pad, I can find keys that will decrypt to any plain text you choose. So all plain texts are equally likely. That said, it is NOTORIOUSLY hard to properly implement a one time pad. They are only usable in very very rare situations. It's NOT a practical cipher. And if anyone tries to sell you and encryption system that mentions "one time pad" anywhere in the description, don't walk, RUN AWAY! For an example of what can go wrong with a one time pad, check out the two time pad challenge. Rot13's Two Time Pad Crib Dragger, As well as my Java two time pad tool Did that help?


One Time What? · Debates