(bug?) Revoked keys and past signatures

Daniel Kahn Gillmor dkg at fifthhorseman.net
Tue Feb 10 19:01:17 CET 2015

On Tue 2015-02-10 08:37:38 -0500, Hugo Osvaldo Barrera wrote:
> Also, I see no reason why I should not be able to assign a trust to a revoked
> key - I might trust it even if the author revoked it as superseded:
>   $ gpg --edit 1BFBED44
>   [... info on revoked key ...]
>   gpg> lsign
>   Key is revoked.  Unable to sign.

fwiw, you said "assign trust" above, but then in your example, tried to
do "lsign", which is an entirely different operation from assigning trust.

> I believe the reason matters. I can even sit down with the owner of the key and
> verify his ID and fingerprint and sign it, meaning "this key belongs to this
> person, but was superseeded a week ago". If actually influences the validity of
> anything he signed up to a week ago.

your certifications (whether local or exportable) themselves have a
timestamp in them.  It would be silly to certify a key and its user ID
after it was revoked by the owner; you'd be claiming "i believe that
right now this is the correct key", which is not the case.

I understand the semantics of what you're trying to do, but i'm not sure
that OpenPGP has syntax to represent it.  The closest OpenPGP comes
would be to forge a certification yourself from *before* the revocation.


 gpg --faked-system-time 20100105T153023 --lsign 1BFBED44

This isn't exactly the same semantics (it says "on January 5 2010 i
thought that this key was correct") but it's close.


More information about the Gnupg-users mailing list