RE: Sendmail white paper, SRS, and forwarding

On Sun, 2004-11-21 at 17:06 -0500, terry(_at_)ashtonwoodshomes(_dot_)com wrote:
> > It would be better to have SRS happen _only_ if the original source
> > address has IP-dependent mechanisms followed by '-all', and preferably
> > also only if the recipient is known to check SPF.
> How do you propose to know policies about the recipient?
It's not hard. Most people aren't actually throwing away valid mail by
rejecting for SPF failure -- there seem to be only a few who are willing
exercise their 'right' to do that.

If I get a report of a failure to deliver forwarded mail due to an SPF
failure, I contact the postmaster at the offending domain and explain
why their use of SPF is rejecting valid mail. 

Often they haven't really thought about it or understood the problem
properly -- they've just been taken in by the bogus marketing on of SPF.
If they don't fix it by turning off their SPF checking (or at least
whitelisting my boxes) I get to consider whether I really do need to
send mail to that broken domain, or whether I'm happy to let them
exercise their right to break their own mail. If I really want to work
around their brain damage, I add them to my 'spf-afflicted-domains'
list.

-- 
dwmw2