[Spambayes] RE: [spambayes-dev] Suggested Feature?

[Tony Meyer]

> In future versions of SpamBayes, do you think it would be a 
> good idea to have a central repository that clients could 
> connect incrementally to get updates on what spam is?  I was 
> thinking the server could have some sort of way to collect 
> and find relationships between all user data and form rules 
> based on all data from all users.

There are two main problems with this:

 * SpamBayes works by gathering an understanding of what spam looks like
*and* what ham looks like.  So you'd really also need a central collection
of ham, as well as spam.  There are a lot of privacy issues with that, among
other problems.

 * The strength of SpamBayes (and similar filters) is that it builds up an
individual understanding of what mail you want to see and what mail you
don't want to see.  This would negate that strength.  (It would also make it
simple for a spammer to download the central corpus and remove all words
that score highly).

=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.