Re: Requiring self-signed uids? (was Re: PoP & Signer's User ID subpacket?)

[Jon Callas <jon@xxxxxxxxxx>]

On 7/20/03 5:34 AM, "David Shaw" <dshaw@xxxxxxxxxxxxxxx> wrote:

> I'm not sure 2440 says that.  The relevant bit is in section 11.1,
> which says "In a key that has a main key and subkeys, the primary key
> MUST be a key capable of signing."
> 
> I took this, perhaps wrongly, at face value - that is, if a key had
> subkeys, the primary had to be able to sign (for the binding
> signatures, presumably).  The flip side of this is that if a key does
> not have subkeys (and there is nothing wrong with a V4 key without
> subkeys), the primary did not have to be able to sign.
> 
> Did I misinterpret the intent in 2440 there?  If "a key that has a
> main key and subkeys" was intended to mean "V4 key", then I strongly
> suggest changing it to say "V4 key" explicitly to avoid the confusion
> that spawned a good bit of this thread.

Uh, I thought that meant that the top-level key can't be an encrypt-only
key. So yes, I was quite sure that 2440 said what you wanted.

    Jon