[Asrg] FeedBack loops

Barry Shein bzs at world.std.com
Thu Nov 13 14:27:45 PST 2008


On November 13, 2008 at 16:54 rsk at gsp.org (Rich Kulawiec) wrote:
 > So maybe you're right.  Maybe they don't want to be on that particular
 > mailing list any more, or maybe they don't want any more mail on that
 > particular topic, or maybe they don't want any more mail from that
 > particular fellow list-member, or maybe something else entirely...
 > but there's no way for me to know which.  (Well, unless I ask them and
 > they answer, and given the volumes involved, that's not very practical.)

Ask 1% of them and report back? Or whatever sample size would seem
statistically significant? It needn't be very large.

 > So maybe you're not right.  Given the facts-in-hand, how am I to tell?
 > 
 > I have, by the way, in the case of *some* lists, tried the approach
 > of removing those reporting ordinary list traffic as spam.  One of the
 > things I've noticed is that a significant percentage re-subscribe,
 > then subsequently report more ordinary list traffic as spam.

They're going on vacation, or overwhelmed for a while. The SPAM button
has become the list membership management / nomail button. It works!

  I've even
 > noticed a handful of people reporting *their own messages to a list* as spam.

Accidents happen. Some of those accidents are the meat between the
keyboard and the chair. I have no doubt occasionally someone gets
frustrated with the volume or character of their email and just fire
thru their inbox holding down the spam button (assuming it
auto-repeats, but whatever.)

The first error in any observation about large groups of people is to
assume most of them will act rationally and in their own
interest. Even the dismal science (economics) has gave up on that a
long time ago.

 > 
 > Let me see if I can put this in a larger context.  Feedback loops
 > are often used as (partial) input to operational SMTP policies.
 > Those policies, along with HTTP and FTP and SSH and DNS and other
 > policies, are all part of a bundle I consider "site security policy".
 > I don't think it's a good idea to hand users a knob that controls
 > any part of that.

How about this:

They're mostly something dreamed up by marketing departments working
in concert with engineers to make the end-user think something is
being done about spam but at about zero cost to the organization.

The idea probably started with "hey, you know how hospitals have had
great success with giving people their own pain meds button? How
studies have shown they actually use *less* pain meds when they feel
in control and are generally happier about their pain control? How
about a SPAM button?"


-- 
        -Barry Shein

The World              | bzs at TheWorld.com           | http://www.TheWorld.com
Purveyors to the Trade | Voice: 800-THE-WRLD        | Login: Nationwide
Software Tool & Die    | Public Access Internet     | SINCE 1989     *oo*


More information about the Asrg mailing list