[afnog] Re: AOL rejecting hosts with no rDNS?

[Brian Candler]

On Tue, Jun 29, 2004 at 12:58:58PM +0200, Alan Barrett wrote:
> > In the very best case scenario: all mail sent via AOL will end up with
> > @aol.com on the end of the MAIL FROM address. You won't be able to
> > distinguish AOL spam from AOL non-spam.
> 
> If AOL doesn't care about AOL spam, then this makes no difference.  If
> AOL does care about AOL spam (and I think that they do), then this
> will be be better than today's situation, because when you get spam
> purporting to be from user at aol, it will be more likely to really be from
> AOL, and AOL will be better able to cancel the spammers' accounts.

Essentially that's saying "it's easier to parse a partially-validated
Return-Path: header than to parse the Received: headers to see where it
*really* came from"

Maybe there is a small amount of value in that; but since the Return-Path:
is never 100% validated, you still need to parse the Received: headers
anyway.

I guess we could add to the problem set:

(5) People misinterpret Return-Path: headers and HELO/EHLO names recorded in
Received: headers, and misdirect spam reports to the wrong place

But to me, the problem is that people misinterpret what is there, not that
the information is bad in the first place. An EHLO name, in particular, is
just a cookie that the sender provides. It helps debugging, for the cases
where the host is being honest about its name. But it's not to be trusted as
a way of saying "this mail definitely came from host X".

> I am not sure if you are saying "SPF will not help at all against any
> kinds of spam", but if you are, then I disagree.

I am saying: "if SPF is implemented tomorrow, then it will only help against
yesterday's spam"

It is so trivial for spam to evolve to bypass SPF, that I don't see the
point of making the transition in the first place.

> You also seem to be saying that the existence of SPF records will make
> like more difficult for you, and I see your point, but I think that what
> will really make your life more difficult will be policy changes on the
> part of your IP connectivity provider, your mailbox provider, and the
> recipients of your email.  These policy changes are reactions to the
> spam problem, and only loosely correleted with the existence of SPF.

If it weren't for spammers, SPF probably wouldn't be in the picture, so you
could argue it's quite strongly correlated.

It just seems to me like a knee-jerk reaction, which will hurt normal,
honest E-mail users more than it will hurt spammers.

Regards,

Brian.