<html><head><style type='text/css'>p { margin: 0; }</style></head><body><div style='font-family: Arial; font-size: 10pt; color: #000000'>It seems to me that FBL are good between MTAs. I'm not sure they are that good, to report back on mail originating from machines on a ISP network. You would have so many reports from zombies, it would not be funny. I think there are other tools to monitor your network activity and the presence of Zombies on your net.<br><br>no?<br><br>----- Original Message -----<br>From: "Chris Lewis" <clewis@nortel.com><br>To: "Anti-Spam Research Group - IRTF" <asrg@irtf.org><br>Sent: Thursday, 13 November, 2008 8:21:52 AM (GMT+1200) Auto-Detected<br>Subject: Re: [Asrg] FeedBack loops<br><br>Rich Kulawiec wrote:<br><br>> Incidentally, I recently concluded an analysis of nearly 5 years worth of<br>> feedback loop traffic from AOL. (Which is the first one I started using<br>> on a site of appreciable volume.) This analysis, partially automated<br>> and partially manual, arrived at the following interesting conclusion:<br>> the FP rate is 100.000%. Every single feedback loop report identifying<br>> traffic as spam was wrong.<br><br>On a similar note, we average about 2-5 spamcop reports a week. Only<br>two of them have been right in the past 5 or more years.<br><br>Rich's "result" is largely because he doesn't spam, his reports are by<br>fat-fingered TIS buttons.<br><br>Our "result" is because we have a /8, and therefore somewhere around 1%<br>of all forged Received headers are in our /8. Even spamcop's header<br>parsing goofs, and when it does, we get dinged. [Yes, we deal with the<br>true positives, and Spamcop is made aware of the parsing goofs. We are<br>satisfied with the current situation, despite the vast majority of the<br>reports being wrong.]<br><br>Our AOL FBL had a similar experience: 100% of our reports were wrong for<br>one of two reasons: fat-fingered TIS buttons, or, the fact that their<br>FBL generator couldn't cope with /8 declarations, and gave us reports<br>for someone else's allocation ;-) AOL eventually turned it off because<br>we mutually decided it wasn't worth the bits.<br><br>That is by no means to imply that FBLs are always wrong, or even wrong<br>most of the time. I'm sure that the vast majority of AOL's FBL reports<br>are absolutely right. It's just that neither Rich nor us see a<br>"typical" picture.<br><br><br></div></body></html>