David Jablon <[EMAIL PROTECTED]> writes:
> If you're thinking about building or modifying a system to take
> advantage of the latest methods, and your goal is to provide both
> strength *and* convenience, take a look at the zero-knowledge
> password protocols. These were specifically designed for
> strong mutual authentication based on a short secret.
> EKE, SRP and SPEKE come to mind.
Thanks for the advice. I have heard a little about those protocols;
enough to suspect that they may be suitable for ssh, but not enough
to really know what they are like.
> These are available in a variety of forms, and can be used
> where the host knows either a password, a hashed password, or
> a public-key that corresponds to a password.
> The page at www.IntegritySciences.com/links.html lists most
> of the research in this area.
Thanks. I'll try to read up. One quick question: Are any of these
methods free from patent-restrictions? As I'm writing free software, I
really can't use algorithms unless they are either patent-free, or
available on very liberal licensing terms. I suspect IETF-bias is
similar, although I in no way speak for the IETF.
Best regards,
/Niels M�ller
