I'm not arguing with you (aren't you just summarizing somebody else's viewpoint?). I think it's important to see the non-technical factors that may cause a new

I'm not arguing with you (aren't you just summarizing somebody else's viewpoint?). I think it's important to see the non-technical factors that may cause a new scheme to lose out to even technically-inferior alternatives. How much value is there in a NIST or NSA rubber stamp? Well, so far MS and Mozilla have only implemented Suite B curves. How hurtful is the ECC patent FUD? Well, look at how much work Red Hat did to strip all ECC code from all the software they ship in their products. How likely is it that there will be a standard that takes off without being distributed by either Microsoft or Red Hat (or Oracle or CentOS)? What would it take to get Curve25519 used in Firefox? Well, there's not much reward for implementing a curve that isn't going to be implemented by many other browsers or servers, so it's hard to justify spending time to consider it, even ignoring the NIST/NSA/IETF/IANA angle.