On Thu, 01 Jan 2004 22:47:39 -0500 Joe Landman <[EMAIL PROTECTED]> wrote:
> Hi Folks: [...] > I did not want to monkey around with the base system scoring if > possible. I simply wanted to integrate another spam filter with its own > scoring into SpamAssassin. As I have been playing with bogofilter, I > thought this would be a nice idea to integrate bogofilter into > SpamAssassin, and amavisd-new. This is not meant to be a criticism of > SpamAssassin, just a note about a method to integrate two excellent > tools. I don't believe the core development team will be too receptive to this, based on what I remember from the the last time this was brought up on SAdev. I'm counting on being corrected by someone more in-the-loop but IIRC the arguments against including hooks to bogofilter, CRM114, etc. were mainly: - it's not clear who would maintain the SA<->external code interfaces - they'd rather not turn SA into an anti-spam Swiss Army knife (KISS) I believe the arguments were more managerial than technical; the limited resources of the core development team would be better spent fixing bugs, improving code performance, and refactoring rather than adding features that could be performed with procmail, et. al. This is not to say they couldn't be convinced (IIRC nobody seemed adamantly and vitriolically opposed to it) but I'm guessing you'd have to make a very compelling case, i.e. contribute a patch & documentation & a nice bribe... :) > I use a late model of Postfix, though I think it might be possible to > do this with other MTA. I have a large corpus (2500+) of spam and about > the same of real mail. > > Punchline: It works, quite nicely. I see the same results here with a slightly larger corpus (8k.) [...] > Why: Couldn't you just use procmail? Yes, though I wanted a single > point of scoring to run under amavisd-new. While I could keep hacking > something simple under procmail, I would much prefer to make bogofilter > just another SpamAssassin test. Agreed. While it's somewhat redundant, it would be a nice optional feature. Additionally, you could compare the behavior of various Bayesian classifiers. Besides, DCC, Razor, and Pyzor are all hooked into SA; I don't know why maintaining interfaces to bogofilter or CRM114 should be any different. Personally, I like the Swiss Army knife approach in this case but I also see how that direction may pose too much risk to the project. Of course, until I start paying the dev team or contributing patches, I'm not going to complain about the status quo. :) Regardless, you may have better luck with this on SAdev; you'll probably want to check the list archives before diving in. hth, -- Bob ------------------------------------------------------- This SF.net email is sponsored by: IBM Linux Tutorials. Become an expert in LINUX or just sharpen your skills. Sign up for IBM's Free Linux Tutorials. Learn everything from the bash shell to sys admin. Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
