Another entry in the internet security hall of shame....

[Anne & Lynn Wheeler]

Trei, Peter wrote:
> Ironically, Peter's message above kicked off warning
> dialogs from MS Outlook, since it was signed using a keypair
> signed with Peter's own self-signed root, which was not in 
> MSO's list of trusted
> roots.
> 
> Self-signed certs are only useful for showing that a given
> set of messages are from the same source - they don't provide
> any trustworthy information as to the binding of that source
> to anything.

basically somebody may eventually load the public key from the
self-signed digital certificate into their local trusted public key
repository ... possibly based on some out-of-band trust process.

that isn't any different than almost every certification authority
public key that is in use today. almost every certification authority
public key is represented by some sort of self-signed certificate and is
loaded into the trusted public key repositories of relying parties.

in that sense, it is frequently possible to show (from an information
theory standpoint) that such digital certificates are redundant and
superfluous  ... they however may not be not useful (double negative?)
or may not be unsueful.

given that there exists deployed software that thinks that it requires
some sort of digital certificate in order to perform some processing ...
then even if the digital certificates are redundant and superfluous
(from an information theory standpoint) they can still serve some useful
when attempting to have compatibility with existing deployed software.

recent posting in sci.crypt on slightly related subject
http://www.garlic.com/~lynn/2005o.html#31
http://www.garlic.com/~lynn/2005o.html#33


---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at metzdowd.com