1. Get the latest CVS version of SA following the instructions on the
sourceforge website.  You don't need to install it, just get it.
Instructions at:

  https://sourceforge.net/cvs/?group_id=25457

2. Now cd to the "masses" directory in the checked-out CVS code tree.
3. Read README to gain understanding of what mass-check does.
4. Run mass-check against your ham mail archive.
5. sort -rn +1 ham.log | head -20
6. Check each of those 20 messages by hand to make sure they're not spam that
   slipped through, or a forward of a spam message.
7. Repeat #6 until the top 20 are "clean"
8. Repeat steps 4-7 for your spam archive until they are "clean"
   (except you do sort -n +1 spam.log to look for low scoring spam)
9. Run a mass-check for ham and spam together (one mass-check run)
10. rename ham.log and spam.log to ham-yourname.log and spam-yourname.log
    (ie for Craig Hughes: ham-craig.log and spam-craig.log)
9. rsync -CPcvzb ham-yourname.log spam-yourname.log belphegore.hughes-family.org::corpus

That should do it.  In the future when you re-run this procedure
(starting from the top to get latest CVS rules), the rsync bit should
make the transmission of the results a lot faster, since it should only
send diffs (more or less) of the file, which probably won't change much
from run to run.

Thanks for your help,

C
