FC: Ray Everett-Church on systemic problems with SpamCop

From: Declan McCullagh (declanat_private)
Date: Mon Nov 04 2002 - 12:07:51 PST

  • Next message: Declan McCullagh: "FC: RIAA replies to Politech over Hilary Rosen and Oxford debate"

    Ray Everett-Church is a longtime Politechnical, an attorney, and a board 
    member of the Coalition Against Unsolicited Commercial Email 
    (http://www.cauce.org/).
    
    Previous Politech message:
    "Politech incorrectly blocked by SpamCop -- for the third time"
    http://www.politechbot.com/p-04121.html
    
    -Declan
    
    ---
    
    From: "Ray Everett-Church" <rayat_private>
    To: <declanat_private>, <politechat_private>
    Cc: <julianat_private>, <deputiesat_private>
    Subject: RE: Politech incorrectly blocked by SpamCop -- for the third time
    Date: Mon, 4 Nov 2002 11:43:22 -0800
    
     > "There is nothing wrong with the system and it may well happen again.
    Sorry."
    
    If you find this statement nonsensical and outrageous, you should. As a
    veteran of the spam wars for many years, I happen to think using
    blocklists can be a good thing. But there's quite a difference between a
    well-run blocklist and the poorly run blocklist at SpamCop.
    
    Julian's system is designed to have a hair-trigger response to any spam
    complaint from a SpamCop user -- no matter how wrong the complaint is --
    which translates through some arbitrary algorithms into a blocklist
    entry. Julian knows his system creates a situation in which any idiot
    can report valid mail as spam, which in turn can result in blocking that
    occurs quickly and without notice. Beyond the potential for outright
    malicious blocking by a determined individual, it unsurprisingly results
    in frequent problems with blocked legitimate mail. The errors are
    predictable and repeatable, but remember, "there is nothing wrong with
    the system!" It's a feature, not a bug.
    
    Julian has a standard jig he dances when confronted with the crazy
    situations this badly flawed system causes:
    
    - He defends the erroneous spam reports and demands proof that your mail
    isn't spam.
    
    - If you can provide that proof, he says, "Oh well, your listing has
    expired anyway so you have no reason to complain (until the next time)."
    
    - He reminds you that the "disclaimer" on his site recommends the
    blocklist not be used in production environments, as if that absolves
    him of responsibility.
    
    In short, SpamCop regularly turns erroneous reports into an erroneous
    blocklist. In doing so, he willfully harms legitimate mailers. And
    what's worse, he seems to relish it. As such, he's created a system that
    adds to the perception of anti-spammers as unreasonable and unconcerned
    with the damage that their ill-conceived decisions can inflict. This
    makes SpamCop a continuing embarrassment to those engaged in responsible
    anti-spam efforts.
    
    Meanwhile, the real problem is with brain-dead sysadmins who abdicate
    their responsibilities by letting such a demonstrably unreliable system
    make the blocking choice for them automatically. There are far better
    blocklists out there with greater reliability. In the end, though, if
    your ISP is using SpamCop's blocklist, you deserve to lose the mail that
    you are inevitably going to lose.
    
    -Ray
    
    
    
    
    -------------------------------------------------------------------------
    POLITECH -- Declan McCullagh's politics and technology mailing list
    You may redistribute this message freely if you include this notice.
    To subscribe to Politech: http://www.politechbot.com/info/subscribe.html
    This message is archived at http://www.politechbot.com/
    Declan McCullagh's photographs are at http://www.mccullagh.org/
    -------------------------------------------------------------------------
    Like Politech? Make a donation here: http://www.politechbot.com/donate/
    Recent CNET News.com articles: http://news.search.com/search?q=declan
    -------------------------------------------------------------------------
    



    This archive was generated by hypermail 2b30 : Mon Nov 04 2002 - 13:09:39 PST