On Monday, November 6 at 03:16 PM, quoth Dean Anderson:
But no one (significant) blocks email based on SPF,
There, unfortunately, I beg to differ. In my experience,
verizon.net blocks mail based on SPF (which REALLY bugs me, because
some of my clients have me forward their email to their verizon.net
accounts, and my server is left holding the bag for all these mails
verizon won't accept). They aren't hotmail.com, but they're still a
major ISP in the US.
How recently? I haven't had any problem with verizion.net. Though,
verizon.net has a history of doing stupid things for a short while.
This past spring was when I first found it out, and according to my
log files (which record Verizon's servers rejecting some email with
the error message "You are not permitted to send mail."), continues to
this day. More accurately, what I've been told that they do is respect
the descriptions contained in SPF records; i.e. if mail from domain
example.com comes from a server that example.com's SPF record doesn't
authorize, the mail is rejected.
Blocking non-spam email is a violation of federal anti-trust law
(participation in unlawful group boycott) and also state and federal
electronic privacy laws (no authorization to block non-spam email).
Some letters to Verizon may be necessary.
Hmm. That would be rather surprising. There's enough collateral damage
from using DNS-based blacklists (sorbs, spamcop, etc.) to make most
Fortune 500 companies (among others) into criminals. Between the
Standord Wallace v Compuserve
(http://www.netlitigation.com/netlitigation/cases/compucase.htm), T3 v
McNicol (http://t3-v-mcnicol.org/), and EmarketersAmerica v SPEWS.org
et al (http://www.linxnet.com/misc/spam/slapp.php and
http://research.yale.edu/lawmeme/modules.php?name=News&file=article&sid=1102)
lawsuits, I think the legality of blacklist spam blocking (even with
its false-positives) is well-established.
BTW, if they are using SPF to prevent email outsourcing or to link
IP access services with email services for domains they host, that
would also probably be an additional anti-trust violation, distinct
from the participation in an unlawful group boycott, as well as
torts of interference with contract, unfair competition, etc.
I have no idea *why* they're doing it; I rather assumed it was for
anti-spam purposes. In any case, the idea is that the sending domain
published the SPF record for the express purpose that recipients could
use it to make decisions about accepting or rejecting email. Given
that any blocking is done, ostensibly, at the behest of the sender, I
don't think John Gilmore's argument about DNSBLs violating antitrust
law (which, to my knowledge, has never been confirmed by any court in
the world) applies.
Duress, conversion, and extortion are also possible, since they've
already contracted with the customer to provide services which
include communications with your servers, and therefore have no
honest grounds to demand further payments.
No demand for payment of any kind has ever been made. I think it's
just a good reason not to use Verizon for your email services.
~Kyle
--
I beseech you, in the bowels of Christ, think it possible you may be
mistaken.
-- Oliver Cromwell
pgphk9OGwhAO2.pgp
Description: PGP signature
|