Re: Re: Why SRS really sucks

On Mon, 2006-04-03 at 12:46 +0000, Julian Mehnle wrote:
> So if I take your message and resend it with all the headers unmodified, 
> just using my own envelope sender, then it is a fake?  No, it isn't!  It 
> is a perfectly legal case of the layer separation between RFC 2821 and RFC 
> 2822.

If _you_ (the user) resend it, then yes -- of course it's not a fake.
The accepted norm is to add Resent-From:, Resent-To, Resent-Message-Id:
headers when you do that.

If the message is just passing through a system in transit, and the
reverse-path is mangled from its original form to some new form, then
that's understood by many to be 'fake'. But you're just arguing about a
choice of words -- to suggest that he doesn't understand SRS or SPF just
because he considers SRS-mangled addresses to be 'fake' addresses is
somewhat disingenuous of you.

Of _course_ SRS provides 'fake' addresses. That's the whole _point_ of
it. They're not real addresses with a mailbox (and a person) behind them
-- they're fake addresses because you decided you didn't want to send
the mail on with the original and correct reverse-path. 

What I said (or at least what I _intended_ to say) was that you
shouldn't do this mangling except where it's actually necessary to work
around a problem which can't be fixed in other ways.

Mostly I find that the problem _can_ be fixed in other ways -- usually
by getting the offending recipient to stop using SPF to reject mail, or
by getting the offending sender to stop publishing SPF records with
'-all'.

-- 
dwmw2