Field separators

[Paul Crowley]

Brad Fitzpatrick wrote:
> It'd still work:  The server would just have to return:
> 
>    openid.i_see_your_p_and_g_and_used_them = 1
> 
> Then the consumer-of-the-future looks for that and knows to proceed with
> either the "version 1" p and g defaults, or the ones it sent.

Yes, that's what I thought at first.  But if I used my p to generate gx, 
and you interpreted gx by your p, then it won't work out at all; 
essentially, gx becomes a public key to which I don't know the private 
key, so I can't make use of the result you send me.
-- 
   __
\/ o\ Paul Crowley, paul at ciphergoth.org
/\__/ http://www.ciphergoth.org/