>> My question is, is `required=4.0' justified? How much score have the >> most wired bug report with garbage, as bug info? A wild guess is >> -0.0 .. 2.0, but 4.0? > > The question is at which point the false positives and false negatives > are appropriate. The lower the score goes, the more false positives we > have, which are more insidious and difficult to deal with than false > negatives.
Yes, it seems like very message body independent, but smtp/gmane magic dependent... > X-Spam-Status: No, score=2.1 required=4.0 tests=AWL,FORGED_RCVD_HELO, > IMPRONONCABLE_1,MDO_DATING2,SARE_MLB_Stock6,SARE_PROLOSTOCK_SYM3, > UNPARSEABLE_RELAY autolearn=no version=3.1.4 [] > Original-Received: (from [EMAIL PROTECTED]) by epsilon.donarmstrong.com > (8.13.8/8.13.8/Submit) id l69KC4lS027425 for [email protected]; > Mon, 9 Jul 2007 13:12:04 -0700 | X-Spam-Status: No, score=-2.4 required=4.0 tests=LDOSUBSCRIBER, SARE_MLB_Stock6, [] | Original-Received: from flower.upol.cz ([158.194.64.22]) by main.gmane.or > Additionally, Blars Blarson actually examines messages within that > range and adapts the rules to catch them in the future and deletes > them from the bugs as appropriate. > > So yes, in short, it is justified. So, hope this info will be helpful. Thanks. ____ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
