On Wed, Feb 5, 2014 at 6:50 PM, Tony Arcieri <[email protected]> wrote: > On Wednesday, February 5, 2014, Robert Ransom <[email protected]> > wrote: >> >> If you're talking about authenticating an interactive connection > > > I am not. There's SRP or PSK for that. > > I'm talking about using a randomly generated password in lieu of public key > fingerprints for authenticating public keys.
Hi Tony, Yeah, you're talking about the case where Alice and Bob can manually exchange a secret to bootstrap their communication (scribbled password on a napkin, QR code, etc). That's a very interesting case, particularly when you consider leveraging that secret to try to hide even the fact of their communication (like Pond's "PANDA" protocol), or authenticate a key agreement that sets up forward secrecy. But that's a whole other thing than the public-key fingerprints this thread is considering. I also think that fingerprints and "introduction secrets" may be complementary instead of mutually exclusive. Anyways, I suggest we let the fingerprints discussion run its course but discuss that "PANDA" scenario later (maybe next week)? Trevor _______________________________________________ Messaging mailing list [email protected] https://moderncrypto.org/mailman/listinfo/messaging
