Ed Foster classifies Comcast's inability to properly handle blacklists as a preview of non-neutrality. The problem:

A reader who is a Comcast broadband customer had a disturbing experience recently. "I'm at a total loss about how to handle this situation," the reader wrote. "An e-mail to me from a friend got bounced apparently by Comcast. He resent it to my G-Mail account so I could see it. It said that his message was "Blocked for abuse. Please send blacklist removal requests to blacklist_comcastnet@cable.comcast.com ' among other stuff. So apparently there exists a Comcast blacklist that I cannot control that stops e-mails and that requires my correspondents to ask to be permitted to send me messages."

I like his summary better than his introduction - there, he figures that this is almost certainly incompetence. I wouldn't be surprised if it's fear of a lawsuit. Say customer A gets a message (or more than one) from someone, and he complains to Comcast that the message is "hate mail". I can definitely see scared management talking to overly excitable lawyers first, and blocking an entire domain. I can see the same thing happening with overly aggressive spam filters.

Heck, here at Cincom, the spam filters have been improperly tuned - they were junking all mails that had snippets of Smalltalk code in them. That made things real easy for support, I can assure you.

The root of this problem has very little to do with neutrality, and everything to do with an over-reliance on automation. Spam detection tools are just that: tools. I think too many shops just crank up the settings and never bother to look at the results - they never realize what the false positive rate looks like. That's a management issue, not a technical one.

Technorati Tags: net+neutrality