Joe Z.
Reputation
243
Top tag
Next privilege 250 Rep.
 Dec 21 awarded Yearling Dec 6 comment A proof-of-work random number generation system for Pokémon The paper is here, by the way: joezeng.com/papers/pokegen.pdf Mar 28 comment Why are RSA key sizes almost always a power of two? "But the difficulty of computing the RSA public and private functions grows largely stepwise as n increases (more on why later)." Exponentiation by squaring, right? Feb 19 comment Are asymptotic lower bounds relevant to cryptography? Now, to define h(y) without using the key x at all would theoretically turn it into a one-way function; doing so is a field known as "white-box cryptography". However, in the traditional one-time pad, h(y) requires the key x to define. Feb 19 comment Are asymptotic lower bounds relevant to cryptography? Ah, but that's where we differ in what we are talking about. The key, along with the encryption "function", creates another "function" specific to that key. If you're familiar with currying, the one-time-pad f(x, y) can be curried into g(x), which takes the key x and returns the "function" h(y) that I was referring to, that takes the plaintext y and turns it into ciphertext z. Feb 19 comment Are asymptotic lower bounds relevant to cryptography? @WilliamHird: In the case of the one time pad, you don't know the function that was used to encrypt the data. That's where all its security comes from. A one way function is one where you can know the function and it will still be hard to invert. Feb 19 comment Are asymptotic lower bounds relevant to cryptography? @WilliamHird I think you are misunderstanding the nature of a one-way function. Any symmetric encryption algorithm (such as the one-time pad) using the same key to encrypt and decrypt is not a one-way function. A one-way encryption algorithm is a cipher that you cannot decrypt, even given all the information that was used to encrypt it (such as a public-key function). Jan 27 comment Why can't one implement bcrypt in Cuda? For a moment I misread "Cuda" as "Cuba" and thought this was an encryption import problem. Jan 26 comment Cryptographic Symmetric Stream Cipher Are S and P inverses of each other? What are you trying to do here? Jan 26 awarded Critic Jan 26 comment Cryptographic Symmetric Stream Cipher We'd need to know more about these S() and P() functions in order to tell you how secure they are. Jan 25 awarded Teacher Jan 24 answered What is the most secure way to let Alice “input” her private key, transfer it to the server and decrypt the message? Jan 24 accepted Why does Rijndael with a 256-bit block require the bottom two rows to be shifted one more space than usual? Jan 24 revised Why does Rijndael with a 256-bit block require the bottom two rows to be shifted one more space than usual? A small grammatical nitpick. Those things bother me. Jan 24 revised Why does Rijndael with a 256-bit block require the bottom two rows to be shifted one more space than usual? Going to re-tag it as well. Jan 24 asked Why does Rijndael with a 256-bit block require the bottom two rows to be shifted one more space than usual? Jan 23 comment Does it take brute force to find a pair of plaintext and ciphertext that each follow a certain condition, given an AES encryption key? In the application I'm using it for, the block size is 256, and about 112 bits are fixed on each side. Jan 22 revised Does it take brute force to find a pair of plaintext and ciphertext that each follow a certain condition, given an AES encryption key? Apparently this is what I really meant. Jan 22 comment Does it take brute force to find a pair of plaintext and ciphertext that each follow a certain condition, given an AES encryption key? So basically the "security assumption" is based around the number of operations required, in which case, supposing I used AES-256 with 128 bits on each side, it would be. It's not that I actually need it to be a security assumption; I was really just initially wondering whether there was a better way than brute force to find plain/cipher pairs.