> -----Original Message-----
> From: exim-dev-bounces+msk=cloudmark.com@??? [mailto:exim-dev-bounces+msk=cloudmark.com@exim.org] On Behalf Of Todd Lyons
> Sent: Monday, March 19, 2012 9:19 AM
> To: exim-dev
> Subject: [exim-dev] (no subject)
>
> I guess the real question is, how many other valid options to mail (or
> I guess rcpt) are there that are only a singleton, not key/value pairs?
> In my searching, I've not really found any definitive list, and those
> that I do find, I've only seen key/value pairs, no singletons.
>
> I'm hoping for a suggestion of what would be the best way to have this
> non key/value option to get accepted.
I'm interested in seeing the results of your project. Since PRDR was never standardized, it's in a phase where input from actual implementation attempts (plus, of course, live operation) would be excellent feedback. And the IETF really likes to see running code when it publishes something, especially as a Proposed Standard, rather than just documenting an idea or theory.
The various people that work on sendmail and postfix so far haven't come up with any sort of agreement on how they would add support for PRDR to milter, so chances are they won't be adding it to their MTAs in the near future either. That's a barrier to true interoperability testing; you'd want to run exim against some other MTA that supports PRDR. But I don't think that means you should stop. Your anecdotes will be useful either way.
I'd be willing to help move PRDR toward publication once you're done with your project and have that feedback available.
-MSK
