----- Original Message ----- From: "R. Scott Perry" <[EMAIL PROTECTED]>
> > > There is a new interim release (1.76i28) at > > > http://www.declude.com/release/176i/declude.exe that changes the way that > > > the weight is calculated (in i27 it would count negative weights, but no > > > longer will), and adds logging at LOGLEVEL HIGH that should help determine > > > if there are other issues. > > > >Just for a bit of further clarification, when you say it will no longer > >count negative weights, does that mean the score will not be reduced by > >negative weights, even if they are list first in a filter file? > > That just refers to weights in the last column in the test > definitions. For example, the IPNOTINMX test ends in "... 0 -3". That -3 > was originally used in counting the weight (if the E-mail did not or had > not yet failed the test), but no longer is. I'm sorry, Scott, I must have a mental block here. Are you saying that the -3 for the IPNOTINMX test is not deducted before the filter tests are run, but would still be deducted after all of the filter tests have been run, and before final disposition of the message? Sorry for being dense, I just want to make sure I understand the new scoring process. Thanks, Bill --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type "unsubscribe Declude.JunkMail". The archives can be found at http://www.mail-archive.com.
