AOH :: PT-1340.HTM

John Gilmore on DearAOL.com and user control of spam filtering

John Gilmore on DearAOL.com and user control of spam filtering
John Gilmore on DearAOL.com and user control of spam filtering



Previous Politech messages:
http://www.politechbot.com/2006/04/13/why-was-moveonorg/ 
http://www.politechbot.com/2006/04/13/aol-blocks-e/ 

-------- Original Message --------
Subject: Re: [Politech] Why was Moveon.org blocked by AOL? Did 
recipients want the email messages? [sp]
Date: Fri, 14 Apr 2006 01:53:05 -0700
From: John Gilmore  
To: Declan McCullagh  
CC: Politech  

> dearaol.com has this astroturf feature - load up a big list of email
> addresses and put in a pitch about dearaol, we'll mail it out for you
> 
> chances are very high that people will receive these without
> soliciting them - and they'll hit report as spam.
> 
> that WILL cause a block

I think this is called the "heckler's veto".  If a listener doesn't
like what someone is saying, just disrupt the communication enough
that nobody can hear it -- like shouting "spam!" in a crowded theatre.
The architecture of today's anti-spam systems is designed to automate
exactly this kind of censorship.  Two guys mark a message as "spam",
and a thousand recipients are censored from ever knowing that it existed.

> we had a similar block in place for a while before I removed it, given that
> the campaign does seem to have some legitimate, if highly uninformed and
> misguided popularity.

It's nice to know that Mr. Ramasubramanian has investigated whether
dearaol's messages are "legitimate" enough that his system's recipients
are permitted to see it -- after blocking them automatically.

Who appointed him as the censor in a supposedly open medium?

You did, dear readers: by insisting that "spam" was a plague and
telling ISPs to "just make it go away".  Most ISPs don't have anyone
with a degree in constitutional law, nor do they have great insights
about how to prevent communications in a medium designed to make
communications rapid and cheap.  The result was predictably ham-handed
censorship, which has naturally evolved into self-serving censorship.

ISPs get a lot of complaints about messages that arrive.  They get
very few complaints about messages that don't arrive, even if they
were interesting personal messages, because the recipient doesn't know
the message even existed.  So self-serving ISPs tend to err on the side of
censoring.

> reasonable to ask: (a) Is each and every address receiving alerts from 
> dearaol.com confirmed double-opt in?

First, is this "alert" a one-time message, or was each recipient added
to a "list" that they'd have to unsubscribe from?  Double opt-in is
foolish for single messages.

If I ask dearaol.com to send a note about AOL's email censorship to my
friend Declan, without telling him first, is that a "spam"?  For one
thing, he's a reporter, my presumption is that he IS interested in
unsolicited tips about newsworthy happenings.  (Should I stop sending
'em, Declan?  I know other reporters who'll be happy to get 'em instead.)

Or should dearaol.com have sent Declan a message saying, "I have a
message for you; would you like to receive it?".  Only if he replies,
would it send him a message.  Except oops, it already did send him a
message.  It might as well have sent him the real message instead of
wasting his time with a bogus interaction.  (If he doesn't want to see
the real message, he can ignore it just as well as he can ignore the
"Do you want to see it?" message.  And if he does want to see the real
message, then it's right in front of him without further hassle.)

> (c) Did a human at AOL 
> intentionally block dearaol.com messages because of the content of the 
> mailings or was it entirely automatic because so many AOLers were 
> marking the alerts as spam?

As anti-spam censorship gets more automation, it's getting harder and
harder to tell whether a censorship decision was made by an explicit
human command, or merely by a machine "learning" a pattern derived
from explicit human commands about "related" messages.  That's why
it's quite important that these human command decisions be made UNDER
THE CONTROL OF THE RECIPIENT.  Not by some intermediary like AOL,
which has its own axe to grind.

> (b) Did dearaol.com borrow lists 
> from some of its member organizations like moveon.org that may have 
> less-than pristine list management practices?
>
> From: Suresh Ramasubramanian  
> moveon.org has never been noted for good list management, like the case
> where they were inviting our ISP tech support staff, based out of hong
> kong, to participate in a rally outside the white house (what, and pay
> airfare + hotel for 8 people from hong kong to DC)?

Ahem.  Declan's last message to me, a San Francisco Politech member,
was about a security seminar today in an anarchist bookstore in Baltimore.
Report that man as a *s*p*a*m*m*e*r*!

Anti-spammers have a long history of making up totally arbitrary
demands, like blocking all your innocent emails if you happen to own a
domain name which appears in bulk messages sent by somebody else.

But calling you a spammer if your email list doesn't track your
recipients' physical location -- and avoid sending messages to each
person if responding would involve more than X kilometers of travel --
is utterly egregious.

	John Gilmore


_______________________________________________
Politech mailing list
Archived at http://www.politechbot.com/ 
Moderated by Declan McCullagh (http://www.mccullagh.org/) 


Make REAL money with your website!

The entire AOH site is optimized to look best in Firefox® 2.0 on a widescreen monitor (1440x900 or better).
Site design & layout copyright © 1986-2014 AOH
We do not send spam. If you have received spam bearing an artofhacking.com email address, please forward it with full headers to abuse@artofhacking.com.