From: Chuq Von Rospach <chuqui@xxxxxxxxxxxxxx>
Date: April 14, 2006 9:35:44 PM PDT
To: Kevin Marks <kevinmarks@xxxxxxx>
Subject: Re: [IP] Why was Moveon.org blocked by AOL? Did recipients  
want the email messages? [sp]
FWIW, I deal with the AOL folks some. That was the other scenario I  
wondered about (but I decided to keep it simple....)
We have about 122 coalition members on the list, which is enough  
to trip AOL's
volume filters, which I understand are set at about 100 mails  
from a single IP
address.  I sent out a mailout to our coalition around noon  
yesterday. I found
out that AOL was bouncing any mail with our URL in it at around  
4.45pm - one
of our coalition had mailed a friend at AOL with a note about our  
site, and
received a bounce.
that's one of their automated systems. FWIW, they use a ratio of  
about 2000 per million emails from a site as the boundary for spam.  
For a small mailing from a previously unknown site with a common  
URL that generates a few spam complaints... It's not that hard to  
trigger it.
received yesterday's mail (carefully avoiding the D*arA*L.com  
word).  A few
had; the majority had not, which leads me to believe that the ban  
occurred
somewhere in the middle of the mailing run.
and anything in process of being delivered gets flushed.  
blackholed. you don't get a bounce, the user doesn't get a chance  
to accept it.
Ploughing through the error logs, I have found one person on the  
list whose
error message indicates that he does not want to receive mail  
from my address.
Whether he is simply set to only receive mail from friends or  
whether this is
a specific ban is  unclear: but he's the only indication I have  
that anyone
complained about the mail.
But it's unlikely to be the only complaint. You're seeing the AOL's  
whitelist/blacklist setting. You won't see anything back from the  
"I call this spam" button, unless you set up a special relationship  
with them to receive them (which I have. I get all of those for  
apple; used to get them for my personal domain. it's -- eye opening)
Many AOL users treat the AOL client's "spam" button, rather  
sensibly, as a "I
don't want to receive any more of this mail". I suspect this  
person was
unsubscribing by hitting this button.
Unfortunately, AOL's semantics are rather different: they take it  
as meaning
"treat this mail as suspect for everyone else".
one has to be VERY careful generalizing AOL users. About all I can  
say is a user didn't want to see the message. Many of them don't  
understand (or care about) the concept of unsubscribing. You can't  
assume they understand these concepts at all; many don't. Or that  
they're paying attention. They don't.
(This is one of the practical
problems of having intermediaries attempt to make decisions about  
end-user
email delivery without adequate feedback or transparency. Fixing  
this semantic
gap is one of the ongoing challenges of fighting spam: a  
consistent standard
for confirm and unsubscribes may well go some way to fixing it.)
Probably not. but that's a different discussion.
Anyway, AOL clearly doesn't view the mail as spam in a strong  
sense, because
they haven't banned my email address or IP. What they did, it  
appears, is
check out the mentioned URL.
no, they just used the URL as a signature check. Doubtful any human  
ever saw anything. Given the propensity of spam to come from  
zombied relays, a common URL is a more effective block than an IP  
block (it's also a common attack by Brightmail, etc...)
Somehow - and this is what AOL's tech support folk told me when I  
called them
this morning - they identified www.dearaol.com as a "morpher".  
This is a site
that redirects user clicks to many different sites.
It's true: www.dearaol.com has round-robin DNS.  I plead guilty to
load-balancing of the most heinous kind.
Far be it that I defend AOL's anti-spam system. It's pretty broken.  
Here's a great example of how. They have a really tough job, to be  
sure; Many of their choices are, in my mind, sub-optimal.
That includes, incidentally, people mailing themselves the URL.
that includes mailing a piece of mail to your postmaster contact at  
AOL askign "why are you banning this?" -- that makes for fun  
mornings trying to unwind a mistake, their postmasters are behind  
the walls, too.
AOL's spokesman told reporters variously that that there was a  
software
glitch, a technical glitch, and finally a hardware glitch that  
affected dozens
of web addresses.
I find all of these hard to believe.
when all else fails, blame the computer. not the people who  
programmed it or run it. of course.
working days. EFF has received reports of these kind of URL bans  
before.
Bennett Hasselton, of the free speech group PeaceFire, has  
documented many
innocent groups who find all mails discussing their URLs removed  
from
AOLspace.
Never attribute to conspiracy what is better defined by stupidly- 
built systems. UNless. of course, paranoia feeds your cause.
This appears to be a private AOL ban list. Goodness knows how  
many URLs or how
long they are held. I suspect if I hadn't received that mail from  
a friend, or
put out a press release, www.dearaol.com would still be banned  
from 20 million
user's private communications, and would remain so until I made  
that call.
yup.
This is exactly the kind of overreaching, black-and-white anti- 
spam filtering
that goes on all the time among ISPs and is largely unnoticed by  
their
customers - for the simple reason that nobody notices a mail that  
never
arrives.
well, depends on the ISP. the major problem I have with AOL -- it  
doesn't block, it doesn't use a junk folder, it blackholes. And the  
user has no ability to whitelist around an AOL blacklist.
on the other hand -- so do I. But I choose my data sources  
carefully. AOL uses its users as a primary data source, and as  
Spamcop long ago showed, the larger the population, the less  
reliable the data.
And that's why we're concerned about Goodmail: it rewards ISPs  
for such bad
filtering, because with such large problems, large companies will  
pay a great
deal to avoid those filters. And no market forces can come into  
play to fix
this failure to deliver when the symptoms themselves are so hard  
to detect.
nah. people have just misunderstood goodmail from day one.
And frankly, few companies will pay very much for goodmail's  
services. It impacts the profit margin. The people who are running  
around screaming about this don't understand the business at all.  
they're too busy being paranoid. I took one look at goodmail's  
revenue setup, and it makes no sense for a typical marketing  
organization. the margins are too tight.
To say market forces can't come into play only proves they don't  
undersatnd the situation. For one: goodmail feeds are still  
required to stay under that 2000/million spam complaint rate. If it  
goes over that, it can lose its goodmail certification. It's not  
carte blanche, that argument's a factless strawman.
(I could go on. maybe at some point, I will on my blog. it's a  
witchburning. even more, it's a waste of energy, since goodmail  
will likely fail on its own merits, without all this help; it's  
really a very limited-use system, the only mail streams likely to  
take advantage of it are transactional ones, not marketing ones.  
There IS more to the world than marketing email and mailing lists,  
you know....)
I'm more disturbed that Suresh had a similiar block, which he  
finally deigned
to remove because he believed us to have "legitimate" popularity.  
Suresh's
company manages filters for over 40 million users.
um... Who will watch the watchers?
Which is a fight I had with Dave Rand many years ago over MAPS,  
back when MAPS kept blacklisting my personal domain because I  
happened to have a subscriber to one of my mailing lists that lived  
on his personal server, which decided to arbitrarily start  
declaring mail spam when it had keywords he didn't feel appropriate  
for mailing lists.
What all this comes down to is an absolute lack of transparency  
among the spam police. Which is even worse than the fact that we  
can't even define what spam is in terms more specific than the  
nation's war on porn (I know it when I see it, and local community  
standards). That, frankly, is even more important than "what is  
spam?" standards, and woefully missing in all of these discussions.  
AOL is just worse at blackholing stuff without any response or  
appeal than many.
chuq
(and yes, if you feel it's worth passing back up the reply chain,  
you may...)
--
Chuq Von Rospach, Architech
chuqui@xxxxxxxxxxxxxx -- http://chuqui.typepad.com/
He doesn't have ulcers, but he's a carrier.