Re: DRAFT minutes, SACRED WG, Minneapolis IETF

[Tom Wu <tom@xxxxxxxxx>]

Radia Perlman - Boston Center for Networking wrote:
> 
> It would be nice to at least come up with a single MUST implement, as
> well as a bunch of other choices, should the MUST alg wind up cryptographically
> broken or with someone showing they have a patent on arithmetic which
> covers it.

If one of the strong password methods like PDM or SRP becomes a MUST
implement, though, how does that help implementors that want to deploy a
different topology, like a Ford-Kaliski multi-server configuration? 
Their servers cannot support a "regular" strong password protocol, so
does that make their implementation non-sacred-compliant by design?

> Radia

Tom
-- 
Tom Wu
Principal Software Engineer
Arcot Systems
(408) 969-6124
"The Borg?  Sounds Swedish..."