[cryptography] ciphersuite revocation model? (Re: the spell is broken)

2013-10-05 Thread Adam Back
You know part of this problem is the inability to disable dud ciphersuites. Maybe its time to get pre-emptive on that issue: pair a protocol revocation cert with a new ciphersuite. I am reminded of mondex security model: it was a offline respendable smart-card based ecash system in the UK, with

Re: [cryptography] ciphersuite revocation model? (Re: the spell is broken)

2013-10-05 Thread Natanael
Should we create some kind of CRL style protocol for algorithms? Then we'd have a bunch of servers run by various organizations specialized on crypto/computer security that can issue warnings against unsecure algorithms, as well as cipher modes and combinations of ciphers and whatever else it

Re: [cryptography] ciphersuite revocation model? (Re: the spell is broken)

2013-10-05 Thread Peter Todd
On Sat, Oct 05, 2013 at 02:29:11PM +0200, Natanael wrote: Should we create some kind of CRL style protocol for algorithms? Then we'd have a bunch of servers run by various organizations specialized on crypto/computer security that can issue warnings against unsecure algorithms, as well as