Re: #23 [*] KE interoperability (section 6.8)

At Wed, 09 Feb 2005 08:43:07 -0800,
Michael Thomas <[email protected]> wrote:
> 
> Let me clarify my position: I think that a KINK implementation
> MUST deal properly with receiving a KE payload from a sender.
> I'm perfectly happy if the receiver somehow rejects the KE
> payload and proceeds from there (either by sending an error,
> or not performing the DH and not returning its KE payload).

I understand, thank you.  If "rejecting" is allowed, I'm
happy with it.
Regarding how to reject, I like the idea of the absense of KE.


> I believe this gives you what you're asking for: the ability
> to not have to run the big number arithmetic... But even
> if we required implementation (and I'm not recommending this),
> nobody says that the DH arithmetic has to be _fast_ :)

Indeed.  :-D

-- 
KAMADA Ken'ichi <[email protected]>