CodesInChaos

less info
meta site favicon meta user | stack exchange network profile network profile
6,165 reputation
1927
bio website github.com/CodesInChaos
location Munich, Germany
age
visits member for 1 year, 11 months
seen 1 hour ago
stats profile views 112

1,192 Actions

suggestions reviews revisions comments badges posts accepts all
Dec
2
 comment Is Convergent Encryption really secure?
If your encryption primitive has weak keys it sucks, use a better primitive. It's also easy to implement convergent encryption rejects certain keys. This is not a valid problem in practice.
Dec
2
 comment SSL Key Exchange
The standard technique is trusting the installed software(in particular the installed root certificates) and CAs which sign a certificate only for the legitimate owner of a domain.
Dec
2
 comment SSL Key Exchange
"Assuming the messages can be read by a third party, but not manipulated." We generally avoid that assumption, since usually being in a position to read messages allows you to manipulate them as well.
Dec
2
 comment SSL Key Exchange
I can't say if a method is secure unless I get a clear description of it. The purpose of ephermeral key-exchange is to provide forward secrecy. So when you lose your long term private key, an attacker can't decrypt recorded connections.
Dec
2
 comment SSL Key Exchange
Generating one is easy in principle, but apart from client authentication(rarely used) and ephemeral key-change(for forward privacy) it's not that useful. IMO one should always use ECDHE where neither side creates the session key, and instead it's the result of a key-exchange operation.
Dec
2
 comment SSL Key Exchange
How would that work? You can't encrypt the session key with the client's public key, because it usually doesn't have one. There are different ways to avoid replay problems, using a nonce on both sides that's mixed into the session key is just one easy solution.
Dec
2
 answered SSL Key Exchange
Dec
2
 comment Ensuring integrity of a client side script
Impossible. It's just another variant of the DRM problem, where you don't trust the computer you're running on. Best you can do is throwing lots of obfuscation at it, and then pray that nobody bothers to break it.
Dec
2
 comment Webcam random number generator
@D.W. Sensor noise is a nice source of entropy. Obviously you still need to feed it into a standard PRNG.
Dec
2
 reviewed Approve suggested edit on What's a good P to use in ElGamal key generation?
Dec
1
 comment Weakness in using only one RSA key pair for two-way communication?
Your scheme is essentially symmetrical, so why would you use RSA over AES?
Nov
30
 answered What's a good P to use in ElGamal key generation?
Nov
30
 comment What's a good P to use in ElGamal key generation?
Any reason why you want to use ElGamal over Diffie-Hellman, DSA or Schnorr signatures? Are you talking about ElGamal encryption or ElGamal signatures?
Nov
29
 reviewed Edit suggested edit on AES timing attacks
Nov
29
 revised AES timing attacks
made link work + title
Nov
29
 reviewed Approve suggested edit on sha-3-competition tag wiki
Nov
29
 comment Message authentication codes construction
Please don't cross-post
Nov
29
 revised Message authentication codes construction
deleted 153 characters in body
Nov
29
 comment My Hand Cipher, Can some one tell if it secure
Your real problem is "By my other cipher", because I'm 99% sure that the other cipher sucks, and obviously you need to generate a new pad for each message. Like with most stream ciphers, pad reuse is absolutely fatal.
Nov
29
 comment My Hand Cipher, Can some one tell if it secure
Are your random numbers uniformly distributed between 00 and 99? In that case this part of the system should be secure but inefficient (ciphertext is twice the size of the plaintext). Simple modular addition is better. Essentially an inferior method for combining a keypad with a message, similar to stream ciphers/one-time-pads.
1 … 25 26 27 28 29 … 60