spf-discuss
[Top] [All Lists]

RE: Re: RFC 2821 and responsibility for forwarding

2004-12-07 11:33:30
On Tue, 2004-12-07 at 12:51 -0500, terry(_at_)ashtonwoodshomes(_dot_)com wrote:
You keep saying that, but all I have seen you mention have been refuted.

All solutions can be discarded if you're willing to accept arbitrary
limitations along the lines of "cannot configure MTA" or "cannot publish
TXT records". In fact, if nobody is allowed to change their MTA then SPF
can't work _either_, because you need to do something at the receiving
side.

Please do this:  Respond with THE ONE alternative solution which you
have evaluated to be THE BEST alternative to SPF. 

For me, that's most definitely SES.

My partial implementation of SES is _purely_ in Exim's configuration
language; no actual modification or plugin libraries were required. A
standard Linux installation can do it.

Many domains, like sourceforge.net, stopped accepting faked mail from
dwmw2(_at_)infradead(_dot_)org 9 months ago. And I stopped receiving bounces to
mail I didn't send, too.

If you want to impose the arbitrary limitation that it must be
implemented solely in the MUA, then -- bearing in mind that your
objection is about as valid as it would be for me to claim "but you
mustn't require the ability to publish TXT records" -- I'd probably
select DK or IIM instead. As it is, I'm happy enough with SES that I
haven't bothered to implement either of those two yet.

But if you have to implement something _now_ you're going to have to do
some work anyway. And it's surely better for you to get your hosting
provider to implement SES than to attempt to get the world to change its
forwarding practice? The former is feasible; the latter is tilting at
windmills.

-- 
dwmw2


<Prev in Thread] Current Thread [Next in Thread>