On Mon, Oct 14, 2002 at 03:40:04PM +0100, Justin Mason wrote: > b) nowadays, there's quite a lot of worry being expressed about false > positives from over-zealous spamcop users nominating non-spam. We're > already talking about zeroing out SPEWS, and Spamcop is #2 (as far as I > can tell) in terms of blacklists with reported FPs. So I wouldn't be > too keen on enabling spamcop by default due to this.
There is a still more common source of FPs with Spamcop: skewed statistics. Spamcop weighs spamreports against normal traffic. But for many sources, Spamcop doesn't usually see the normal traffic, as the paying users (that generate it) are not evenly distributed across the internet. Example: The outgoing relays of the largest German hosting providers are regularly listed. With a few million domains, you of course get a bad customer from time to time, and even if you get rid of him soon, some *correct* complaints will arrive at Spamcop. The enormous normal traffic will usually stay in Germany and not come across Spamcops sensors. I still give it a 1.0, though, with required_hits at 7.0. Jost -- | [EMAIL PROTECTED] Please help stamp out spam! | | Postmaster, JAPH, resident answer machine am RZ der RUB | | Pluralitas non est ponenda sine necessitate | | William of Ockham (1285-1347/49) | ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
