1) Random vs Encrypted -- Compression should make data fairly random, except for those annoying headers and markers in the resulting files that tell the software how to restore the data. Encrypted data should look fairly random except for the annoying headers and markers that make it easier for the software to determine if it has indeed decrypted the data correctly. And if we used raw streams... now there's an idea. Maybe you couldn't tell the difference.
2) Software vs Hardware cracking -- you can't implement algorithms in one that can't be implemented in the other. You give me a hardware algorithm, I'll code the software to do the same (but more slowly). Give me a software algorithm, and I'll have my friends build a little circuit that'll run circles around the software. Algorithms are algorithms no matter where implemented. And if they're rotten algorithms, then that's the weakness. If they're strong algorithms...well think about it.
Slashdot Mirror
Two things have been buggin' me...
1) Random vs Encrypted -- Compression should make data fairly random, except for those annoying headers and markers in the resulting files that tell the software how to restore the data. Encrypted data should look fairly random except for the annoying headers and markers that make it easier for the software to determine if it has indeed decrypted the data correctly. And if we used raw streams... now there's an idea. Maybe you couldn't tell the difference.
2) Software vs Hardware cracking -- you can't implement algorithms in one that can't be implemented in the other. You give me a hardware algorithm, I'll code the software to do the same (but more slowly). Give me a software algorithm, and I'll have my friends build a little circuit that'll run circles around the software. Algorithms are algorithms no matter where implemented. And if they're rotten algorithms, then that's the weakness. If they're strong algorithms...well think about it.