[ On Friday, November 7, 2003 at 16:34:52 (+0000), Alan J. Flavell wrote: ]
> Subject: Callouts, was Re: [Exim] OT: Problem sending mail to verizon.net
>
> On Fri, 7 Nov 2003, Greg A. Woods wrote:
>
> > To me the issues are so black and white that I can see no way or
> > need to explain them any further.
>
> To me the issues are so multi-faceted, and inevitably calling for
> heuristic behaviour at many different levels to offset the
> misbehaviour - in ways that were never envisaged when the protocols
> were originally defined - of participants over whom we have no direct
> control nor even much indirect influence, that there seems little
> point in trying to discuss problems with someone who sees the issues
> only in black and white.
Ah, well, there you go. You're making this far more complicated than it
needs to be.
It is sufficient to understand that SMTP is a store-and-forward based
protocol, fundamentally and by design, to understand that using SMTP
itself, in-band, to verify parameters given in an incoming SMTP
transaction, is doomed to failure. Anyone with even a hint of
understanding of communications protocols, interoperability, and
feedback loops should see the problems as clear as day. Sure, it will
work some of the time, but it is also very clear that it will cause
serious interoperability problems with even the slightest purturbance
from the "expected", and these problems have been discussed and
documented to death in this and similar forums.
In-band use of a protocol like SMTP to try to test itself is foolish and
ultimately doomed to failure in far more cases than you might realize.
You are doing yourself and the Internet community as a whole a
disservice if you think SMTP callouts for sender address verification is
worthwhile and "works" in enough cases for your purposes. You can keep
your rose coloured glasses firmly obscuring your view if you please, but
don't expect the rest of us to condone your actions.
> While we're on this topic: I have the impression that we're seeing
> considerably more callouts to us than had been the case in earlier
> times.
That is no doubt true. The sheep are flocking.
--
Greg A. Woods
+1 416 218-0098 VE3TCP RoboHack <woods@???>
Planix, Inc. <woods@???> Secrets of the Weird <woods@???>
