Re:Internet Draft: SMTP Strict Transport Security
by yuhong on Sunday March 20, 2016 @01:12AM
(#11321910)
I forgot to mention that one of the known problems of DNSSEC is the DNS zone information disclosure problem. Online signing would allow the problem to be eliminated and reduce complexity by the protocol no longer having to be designed for enumerating DNS records. Of course, DNSSEC2 would probably only allow ECC signing too (I suggest both NIST curves and Curve25519 as options).
[ Parent
| on HN ]
Re:Internet Draft: SMTP Strict Transport Security
by mned on Sunday March 20, 2016 @02:45AM
(#11322058)
Online signing and white lies are already in DNSSEC. What you've described so far doesn't seem to be different to the status quo. Even deprecation of older cryptographic signatures is being worked through at the moment.
Re:Internet Draft: SMTP Strict Transport Security
by yuhong on Sunday March 20, 2016 @02:46AM
(#11322062)
I know about this. That is why I mentioned the complexity that comes from the original protocol design being designed for offline signing.
Re:Internet Draft: SMTP Strict Transport Security
by mned on Sunday March 20, 2016 @03:56AM
(#11322191)
You have said you have been thinking of a DNSSEC2 proposal for a while, that your protocol differs in multiple ways and that it is easier to implement. Given this you should be able to concretely articulate how your protocol differs and where its advantages are gained. Thus far you have not shown the ability to do this.