using a revocation certificate

[Huels, Ralf KSV]

> 	Even that probably would never work.  Bruce Schneier was saying
> that someone had uploaded a fake "Bruce Schneier" key to the keyserver
> years ago (this was about 2 years ago that he was telling me about this)
> and he tried for ages to get it removed.  As fast as it would be removed,
> the other keyservers would resynchronized and add the key back.  Unlike
> pokeymon, he couldn't catch'm all.  Since he didn't have the private key,
> he couldn't revoke it either.  :-(

Well. Fake keys (i.e. keys with someone´s name in it that don't belong to
that person) should not really be a problem. Isn't that precisely what the 
Web of Trust is for? If someone's going to use a key only because it has
the name of a recipient in its ID, they're missing the whole point of the
Web of Trust. Even if it´s signed by other notables, all the keys could be
fake...

I was going to reply to this making a point about unwanted additions to keys
I thought I had previously made in de.comp.security.misc, but a strong
sense of dèja vu led me to find out that I actually made it here:

http://lists.gnupg.org/pipermail/gnupg-users/2000-July/005899.html

Has there been any progress in the developments subsequently discussed by
Werner and L. Sassaman (i.e. the use of an "owner-update-flag")? 

Would it be feasible to create a protocol to allow deletions
from unrevoked keys if they arrive at a keyserver in some kind
of packet signed with that key?
 
Tschüß,
Ralf

-- 
Ralf Hüls                                                  Bismarckplatz
KSV Kreditschutz-Vereinigung GmbH                           44866 Bochum
Score-Consult                                         Tel. 02327/9114-28
http://www.schufa.de/                                 Fax. 02327/8 40 27