[Spambayes] (no subject)

[Tony Meyer]

> Do the version upgrades improve on the programs
> ability to detect Spam (not that it needs improvement)
> or is it just to add additional features and fix bugs?

It depends on the update.  The best way to tell is to read the little
announcement email (the one that gets posted to spambayes-announce and
sometimes here) and if there are changes to the scoring system it'll say so.

The last few releases have been building up to a proper "1.0" release, so
have only had bug fixes (and really minor ones at that).  The next release
will be 1.0 and (unless something pops up in the next few days) will be
exactly the same as 1.0rc2.

After that things will settle down somewhat, and be more predicable.  Any
1.0.1, 1.0.2, 1.0.3 (etc) releases will *only* contain bug fixes.  It's
possible that a bug fix could improve the scoring, but unlikely.  The first
1.1 release (1.1a1, probably) will definitely have new features, and some of
those could make improvements to the scoring.

Two other points (in a rather long version of the "it depends" answer):

  1.  There are some experimental features that are disabled by default in
the recent versions.  These are experimental because we're not sure whether
they are good across the majority of mail streams, though we are sure that
they are good in some cases.  You can (with the Outlook plug-in or
sb_server, or whatever) enable one or more of these, and that might improve
detection.  Or it might make it worse.  Either way, it's easy to turn on or
off (although to get the full benefit will require retraining with many of
them).

  2.  It's likely that the biggest gains to be made are in finding effective
training techniques.  While we plan to make it easier to use different
methods with the various SpamBayes applications, you can experiment with
different ways of training with the existing releases, if you want to.

Hope this helps!

=Tony Meyer

---
Please always include the list (spambayes at python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes. This
way, you get everyone's help, and avoid a lack of replies when I'm busy.