On Sat, 11 May 2002, Daniel Pittman wrote: > On Fri, 10 May 2002, Bart Schaefer wrote: > > On Fri, 10 May 2002, Craig R Hughes wrote: > > [...] > > >> using the Bcc field instead of To or CC then you'll want to manually > >> adjust the score the GA assigns. But you might also want to > >> re-evaluate your email practices -- do you *really* need those > >> messages to be bcc'd? > > > > It's not just Bcc's -- it's any mail that comes from, for example, a > > sendmail alias expansion. Mail that comes from just about any mailing > > list is also affected -- this one included. > > Bart is quite right here, Craig. This isn't going to be useful ... > especially as it's going to catch a reasonable number of mailing lists > as well.
I've been using this test for a while, and personally find it fairly useful with a manually-assigned 1.0 score. Much like the zillion other tests that catch lots of mailing-list traffic, this one will too. The trouble with it is that you can only GA-optimize it if your corpus all has the delivered-to headers. So it probably needs to be a manually-scored rule. But it probably needs to be manually-scored and enabled anyway, because everyone has different support for this header. -- Charlie Watts [EMAIL PROTECTED] Frontier Internet, Inc. http://www.frontier.net/ _______________________________________________________________ Have big pipes? SourceForge.net is looking for download mirrors. We supply the hardware. You get the recognition. Email Us: [EMAIL PROTECTED] _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
