-> He said it was too easy to read not that he didnt want anyone to find
-> it out quickly.... And any 'Serious Hacker' can decrypt anything if
-> they put there mind to it.
Well... I'm not sure about that! For example, there's an encryption
method called a "one time pad", which is thought to be absolutely
unbreakable. Basically, a truly random method is used to produce a
sequence of numbers which are printed in duplicate. One copy is used to
encrypt a message, XORing the numbers with the ASCII values of the
characters, or something similar, and the other copy is used to decrypt
the message at the receiving end. The sequence of numbers is used only
once. A brand new, truly random sequence is used for each new message.
Since the numbers are truly random, the encrypted message will also
satisfy any test for randomness. So there is no pattern that can be used
to break the encryption. Furthermore, a number sequence can be
postulated that will map the encrypted text onto *any* document of the
appropriate length. For example, one "codebreaker" might say, "Here's
the decrypted text, and here's the set of numbers that decrypted it",
but another codebreaker might make exactly the same claim with a
completely different text. And there would be absolutely no way to tell
who was right.
Using BASIC's RND generator in place of a truly random number sequence
makes this method a lot more practicable than it used to be, but it does
make it theoretically possible to break. After all, there are only a few
hundred million (I think) entry points into the RND sequence, so a truly
obsessive hacker might try them all. But for practical purposes the
method is still extremely secure. I'd still be prepared to entrust quite
valuable information to it....
                             dow
--- PCBoard (R) v15.3 (OS/2) 5
---------------