Rob,
My intent, perhaps not well articulated, was that the SCT* submission
would use
the same name redaction mechanism you proposed, if they prove to be viable.
The step 4 submission would include that same data, the serial number,
and the
previously-issued SCT*. This would enable a log (doing more work) to
ensure that
the SCT it issues is consistent between the two submissions. It also
ensures that
the serial number is available for revocation when needed (which arises
in only
some of the attack scenarios).
Thus, whatever name redaction mechanism the WG ultimately deems suitable
should
work in my suggested two-phase protocol.
As Ben noted, there is a residual vulnerability with my proposal since
the SCT*
is not tied to the serial number. But, in the context of the attack
analysis I just
submitted, I'm not sure how serious this vulnerability is, relative to
the other
ones that I identified in the current CT design. We should discuss that
later,
once we have agreement on an attack analysis.
Steve
_______________________________________________
Trans mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/trans
