RE: [TLS] Re: Review of draft-ietf-tls-openpgp-keys-08

Eric Rescorla wrote:

> Rather than arguing about whether such rules are possible (though 
> I think they are in practice), why don't we see if Pasi wants to
> try to write something?

As far as I know, there is no RFC that really describes how PGP key
validation works (RFC 2440 describes the on-the-wire formats, but not
much more). Thus, even the current version of
draft-ietf-tls-openpgp-keys 
can't do any better than saying "The key validation procedure is a 
local matter outside the scope of this document".

So if we're really worrying about unspecified semantics, that's
certainly one huge missing part...

If the key validation rules would be described somewhere, I think
handling multiple keys/certificates could be as simple as saying "the 
sender MAY include additional keys/certificates it believes to be useful

to the recipient (how this is determined is a matter of local policy and
configuration)" and "the recipient incorporates all the received
keys/certificates (not just the first one) in the key validation
procedure".

Best regards,
Pasi

_______________________________________________
TLS mailing list
TLS@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/tls