On Sat, Aug 25, 2001 at 01:11:52PM +0200, Florian Weimer wrote:
David Shaw <dshaw(_at_)akamai(_dot_)com> writes:
Here are two suggestions to help with resolving multiple user IDs
marked primary, as well as resolving multiple self-signatures with
This should probably go into a separate RFC. Currently, RFC 2440 and
RFC 2440bis deal only with syntactic issues (apart from a minor glitch
in RFC 2440bis, 'A revoked certification no longer is a part of
True, and it even says that in the Abstract. There is an exception
made for security issues: "It does not deal with storage and
implementation questions. It does, however, discuss implementation
issues necessary to avoid security flaws."
Offhand, I can't think of a security implication to having multiple
UIDs marked primary (though I'm sure someone here can). My concern is
with the security implications of having multiple conflicting
self-signatures. Without some suggested way to resolve the conflict,
there can be security implications. If it is truly a security issue,
then it is appropriate in 2440bis. (Obviously, I think it's enough of
a security issue to mention - I'd like to hear what others think.)
Self-signatures can carry subpackets that definitely affect the
actions that may be taken with a key. To use one of my examples from
last night, if/when a symmetric cipher or hash is broken, the user can
simply announce that cipher or hash is not accepted (via a "preferred
symmetric algorithms" or "preferred hash algorithms" subpacket).
Without a way to resolve which self-signature is the one to follow,
the broken cipher or hash may be used, which could compromise the
security of the message.
On the other hand, If such additions are accepted, I've got a long
list of them...
Care to work on a "Implementation Suggestions for OpenPGP" with me?
David Shaw | Technical Lead
<dshaw(_at_)akamai(_dot_)com> | Enterprise Content Delivery
617-250-3028 | Akamai Technologies