Nelson B wrote: > So, since your NR cert has no EKU extension, we have no reason to believe > that it is not usable for SSL client auth. It is not a CA cert, so > cannot be used to sign certs. It has a key usage that excludes use for > any sort of encryption. But it allows NR, which is a type of signature. > And signature is the key usage required for client auth. So, we conclude > that this cert can be used for ANY signature application whatsoever. > (Some signature applications require NR, but AFAIK, none forbid NR.)
OTOH, usage of the NR certificate might/should require explicit user approval (might: from the device itself, should: from the application): It appears that applications associate a "contract-signing" meaning with the NR bit, assuming that the user is always explicitly aware of the usage of the NR certificate (otherwise, the user could repudiate having signed a message, if that mesage was never presented to him). This is surely inconvenient for SSL applications: you shouldn't have to authorize signing each individual message if you said "select automatically" before. So I think it would be best if the browser detected that there is a better-suited certificate (one which doesn't need explicit user interaction); the browser should then also invoke explicit approval if the NR certificate is used even though "select automatically" was configured (explaining that this specific certificate is a formal signature). Regards, Martin _______________________________________________ dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto
