45 reputation
15
bio website
location
age
visits member for 2 years, 10 months
seen Oct 31 at 2:08

Oct
25
comment Efficiently show/prove knowledge of third party secret without leaking information
Regarding "not every time authentication takes place, but only when Bob asks Alice for a certificate": the first part is correct, but the second I've tried to clarify a bit more in the edit. Alice is supposed to show a 'certificate' up front, agnostic of its uses later on. Also the identities of the parties are assumed to be known, so any secret information can be passed on a different channel without MITM attacks.
Jan
15
comment How can one securely generate an asymmetric key pair from a short passphrase?
I think I'll go for this approach. Like Thomas and Ilmari said, it's possible to derive a key pair for RSA (which is the easiest for me to deploy), but, since only the user needs to know his password to decrypt the files' private keys (and whatever personal info) and to encrypt new private keys, it's just as secure to use AES (again, easiest to deploy for me) with a simple KDF.
Jan
15
comment How can one securely generate an asymmetric key pair from a short passphrase?
"For RSA, this is less cheap and simple, because the key generation process entails generating random integers until primes are reached. This is still doable." Especially thanks for clarifying this; I feared it to be impossible to derive a prime from a random number (easily). Salting with the user's mail address is quite a good idea, too. (just like the generated salt in e.g. SMF, it should be unique for every user)