Stroller wrote: > Hi all, > > There doesn't seem to be much interest in keeping the Portage tree > up-to-date with the current releases of Bogofilter, however > development of it is moving quite rapidly & it is at a stage now at > which I believe the ebuild *should* be updated, if it is intended to > remain in Portage. > > From what I have heard of Spamassasin it uses several methods to > detect spam (including blacklists based upon the reports of other > users?), and I choose Bogofilter because it's relies purely upon > Bayesian analysis of MY spam messages. So I would like to see > Bogofilter remain current within Gentoo. > > I have forwarded below an email from Bogofilter's maintainer to the > Bogofilter mailing list giving some outlines of why I think an update > is needed at present. It's particularly worth noting that the present > version of Bogofilter within the tree (0.13.7.3) uses two separate > wordlists for spam & ham. Current releases use by default a *single* > wordlist of tokens with a pair of spam/ham counts - the separate > wordlists are depreciated, and the code will be removed in 0.17.x, so > I think Bogofilter-0.16 may be a desirable step to reduce problems > during `emerge --update`s in the future. > > If any interested &/or helpful dev could take a look at > <http://bugs.gentoo.org/show_bug.cgi?id=37017> I would be much > indebted. To give you a head-start on the New Year's bug-fixing > league-table, I think that if Bogofilter-0.16.0 were to enter the > Portage tree, this bug <http://bugs.gentoo.org/show_bug.cgi?id=31800> > could also be closed. > > Many thanks for your time & bandwidth, > > Stroller. >
You could setup SpamAssassin to function in an identical way to bogofilter by weighting all other spam criteria to 0.0 points. I found bogofilter to be a pain to use and it is resource intensive. However, it has the obvious potential if it could ever be daemonized. Tom Veldhouse -- [EMAIL PROTECTED] mailing list
