FC: Politech debate on spam: Paul Bissex vs. Sonia Arrison

From: Declan McCullagh (declanat_private)
Date: Mon Apr 23 2001 - 19:02:16 PDT

  • Next message: Declan McCullagh: "FC: Antiporn groups tell Supremes that "morphed" child porn law is OK"

    [Paul Bissex sent a response to Sonia Arrison's message to politech 
    (http://www.politechbot.com/p-01935.html). I asked them to copy me on a few 
    rounds of their email exchange. Note at the end, they blame me for starting 
    their argument in the first place! --Declan]
    
    *******
    
    Date: Thu, 19 Apr 2001 16:52:02 -0400
    To: declanat_private
    From: Paul Bissex <pb@e-scribe.com>
    Subject: Re:FC: Canning spam through contracts instead of federal
      regulations
    
    Declan/Sonia,
    
    Really, though, what does this have to do with the spam problem as most 
    people experience it? Responsible corporations violating contracts aren't 
    the worst offenders. They'll learn to avoid spamming through customer 
    feedback and action. OTOH I've got no contracts with the originators of the 
    hundreds of unsolicited commercial messages I get each week. Perhaps "the 
    market" has a solution for this, but it sure isn't opt-out agreements.
    
    pb
    
    *******
    
    From: Sonia Arrison <sarrisonat_private>
    To: "'Paul Bissex'" <pb@e-scribe.com>
    Cc: "'declanat_private'" <declanat_private>
    Subject: RE: FC: Canning spam through contracts instead of federal regulat
             i ons
    Date: Mon, 23 Apr 2001 15:17:46 -0700
    
    Thanks for your comments, Paul.  We are almost in agreement (sadly, for
    politech!), but I would take issue with the idea that we need government
    regulation before people can charge for spam.  People already install
    filters for e-mail - all they need now is an added function on the filter
    that sends a message back to spammers saying that if they want their mail to
    go through, they have to pay Pay Pal acct #12345 25 cents. Yes, I know that
    breaks one of the golden rules of avoiding spam,(never respond to spam or
    they'll know your address actually exists), but just imagine if a bunch of
    people using computer programs started spamming the spammers back!  hahaha!
    Now *that* would be justice.
    
    Maybe the spammers would go for it, and you'd be paid.  Or, they wouldn't,
    but they'd recognize after a few attempts that your address is useless to
    them, and annoying if they have a real return address.  Or, maybe the
    address you reply to doesn't exist - but if someone is using a fake address,
    a law requiring them to use a real one won't help - they'd likely skirt the
    law anyway.
    
    The idea that people wouldn't use this type of system until the government
    sanctions it is kind of like thinking that no one would have used e-bay
    until the government got involved (the argument would have been: how would
    you ensure that you actually got the products you paid for?)  And, yes, I
    know that e-bay isn't perfect, but many people do use it with happy results.
    Do we get to fireworks here?
    
    Best,
    Sonia
    
    *******
    
    Date: Mon, 23 Apr 2001 20:15:54 -0400
    To: Sonia Arrison <sarrisonat_private>
    From: Paul Bissex <pb@e-scribe.com>
    Subject: RE: FC: Canning spam through contracts instead of federal regulat
             i ons
    Cc: declanat_private
    
    OK, maybe we're 70% in agreement. The thing is, spam that arrives with a
    real, deliverable address in the From: or Reply-To: headers is extremely
    rare, not just a "maybe." These guys send out millions of messages and they
    don't want  bounces or flak. So they're never going to get any little
    auto-reply invoice.
    
    Minus that element, the proposal becomes simple spam filtering. Not too
    hard to do right now. Lots of people do it. Spammers know people do it.
    Tens of millions of spams go in the bit bucket unread.  Has this deterred
    spammers on the whole?  No.
    
    I do agree that a user movement powered by a micropayment protocol could
    conceivably have an effect on *some* spammers. For instance, let's say a
    Eudora plugin exists that checks message headers for either 1) a known
    "friendly" address or 2) sign of a verified payment, whether that's a
    DigiCash-type token or a digitally signed affidavit from e.g. Paypal
    confirming a deposit.  Great!  You're right, there's no special regulatory
    role needed for this (though thorny infrastructural questions abound).
    
    However, while it might create some interesting new markets, and perhaps
    change the direct-email strategies of some responsible companies, there's
    no way such a thing would make a dent in the daily flow of spam. IMO
    spam-fixing models that put their faith in enlightened self-interest are
    fatally flawed.  A single idiot running an automated spam program can reach
    huge numbers of people. A dozen such idiots could probably annoy nearly
    every e-mail user on the planet.
    
    Therefore, you still need perfect filtering to keep out the idiots.  And
    once you have perfect filtering, I would argue, you've already created a
    technical solution to the problem and the invisible hand is superfluous.
    
    I think I'm nearly done talking about this, simply because I feel at this
    point I'm marking an unoriginal path on well-traveled ground.
    
    pb
    
    *******
    
    From: Sonia Arrison <sarrisonat_private>
    To: Paul Bissex <pb@e-scribe.com>
    Cc: "'declanat_private'" <declanat_private>
    Subject: Re:FC: Canning spam through contracts instead of federal regulati
             ons
    Date: Fri, 20 Apr 2001 18:59:40 -0700
    
    Hi Paul,
    
    Below are some responses to your comments.
    
    Cheers,
    Sonia
    ----------------------------------------------
    
     >Really, though, what does this have to do with the spam problem as most
     >people experience it? Responsible corporations violating contracts aren't
     >the worst offenders. They'll learn to avoid spamming through customer
     >feedback and action.
    
    Indeed, it's true that responsible corporations aren't the worst offenders,
    but this example shows that when a corporation doesn't respond to consumer
    pressure, it still has to respond to the contract it made with its customers
    and various state laws.  New federal government regulations are *not* needed
    and might be detrimental.  Imagine if the FTC were to step in (as they did
    in the Toysmart case). They might decide to negotiate a deal where Kozmo can
    send mail to people who opted out if it fits certain FTC guidelines (similar
    to the deal they cut with Toysmart - allowing the company to sell data it
    said it wouldn't if it followed FTC preferences).
    
    OTOH I've got no contracts with the originators of
     >the hundreds of unsolicited commercial messages I get each week. Perhaps
     >"the market" has a solution for this, but it sure isn't opt-out agreements.
    
    There is a possible market solution for the problem of unsolicited
    commercial mail - start charging people for sending you mail.  I can easily
    see the day where someone writes a program that allows users to set
    parameters where:
    
    1) mail from family and friends can be sent to them free
    2) anyone you don't know has to pay a small amount (say, using Paypal -
    their already 7 million membership would skyrocket)
    3) any known spammers would be charged an outrageous amount.
    
    This type of solution would not only stem the spam problem, it might make
    you some money too.
    
    -Sonia
    
    *******
    
    Date: Mon, 23 Apr 2001 13:45:02 -0400
    To: Sonia Arrison <sarrisonat_private>
    From: Paul Bissex <pb@e-scribe.com>
    Subject: Re:FC: Canning spam through contracts instead of federal regulati
             ons
    Cc: "'declanat_private'" <declanat_private>
    
    Sonia,
    
    I think we're 90% in agreement.  Alas, no fireworks for the list!
    
    It was the subject line's reference to spam together with your claim that 
    "the market is working" that raised a flag for me. On reflection it seems 
    likely that the subject line was Declan's and that your claim was primarily 
    about consumer privacy after all.
    
    Still, I've added a few comments below.
    
    best
    
    pb
    
    
    At 6:59 PM -0700 2001-04-20, Sonia Arrison wrote:
    >Hi Paul,
    >
    >Below are some responses to your comments.
    >
    >Cheers,
    >Sonia
    >----------------------------------------------
    >
    >>Really, though, what does this have to do with the spam problem as most
    >>people experience it? Responsible corporations violating contracts aren't
    >>the worst offenders. They'll learn to avoid spamming through customer
    >>feedback and action.
    >
    >Indeed, it's true that responsible corporations aren't the worst offenders,
    >but this example shows that when a corporation doesn't respond to consumer
    >pressure, it still has to respond to the contract it made with its customers
    >and various state laws.  New federal government regulations are *not* needed
    >and might be detrimental.  Imagine if the FTC were to step in (as they did
    >in the Toysmart case). They might decide to negotiate a deal where Kozmo can
    >send mail to people who opted out if it fits certain FTC guidelines (similar
    >to the deal they cut with Toysmart - allowing the company to sell data it
    >said it wouldn't if it followed FTC preferences).
    
    I think you're making a valid point here, and the Kozmo story is important 
    -- but consumer privacy infringements by such corporations do not begin to 
    account for the majority of spam  people I know receive.  For a year or two 
    I archived every piece of spam I got; the portion originating from 
    reputable, accountable corporations was, and I think will continue to be, 
    vanishingly small.
    
    It's not that I don't think there's a threat of junk e-mail glut from 
    established corps, but the spam problem extends far beyond that.
    
    >OTOH I've got no contracts with the originators of
    >>the hundreds of unsolicited commercial messages I get each week. Perhaps
    >>"the market" has a solution for this, but it sure isn't opt-out agreements.
    >
    >There is a possible market solution for the problem of unsolicited
    >commercial mail - start charging people for sending you mail.  I can easily
    >see the day where someone writes a program that allows users to set
    >parameters where:
    >
    >1) mail from family and friends can be sent to them free
    >2) anyone you don't know has to pay a small amount (say, using Paypal -
    >their already 7 million membership would skyrocket)
    >3) any known spammers would be charged an outrageous amount.
    
    This is a good idea. IIRC, programmer Jef Poskanzer has been proposing 
    something similar for several years -- a sort of electronic postage which 
    is redeemed at the option of the receiver.  You waive the charge for 
    legitimate correspondents, while putting the economic screws to spammers by 
    cashing their tokens.
    
    Of course, writing "a program" wouldn't suffice -- this would have to be 
    embedded in the relevant protocols, and/or adopted by an overwhelming 
    majority of users, in order to exert the intended leverage. And it seems 
    likely that regulatory force would be called on to give "the market" a hand 
    with enforcement and collections.
    
    pb
    
    
    
    -- 
    ~~~~~~~~~~~~
    Paul Bissex <pb@e-scribe.com>
    E-Scribe New Media, 46 Main Street Northampton MA 01060
    Phone: 413-585-8095 ext. 1
    Pager: 413-290-5510 (toll-free in 413)
    
    *******
    
    
    
    
    
    -------------------------------------------------------------------------
    POLITECH -- Declan McCullagh's politics and technology mailing list
    You may redistribute this message freely if it remains intact.
    To subscribe, visit http://www.politechbot.com/info/subscribe.html
    This message is archived at http://www.politechbot.com/
    -------------------------------------------------------------------------
    



    This archive was generated by hypermail 2b30 : Mon Apr 23 2001 - 18:09:02 PDT