Since I had to completely turn off bayes for now, I think this is a good idea. I'll help how I can. I have a few free systems sitting around I'll try testing, devel, etc. on if I have time.
Justice London -----Original Message----- From: Eric "Shubes" [mailto:[EMAIL PROTECTED] Sent: Thursday, October 26, 2006 10:41 AM To: [email protected] Subject: Re: [qmailtoaster] dspam on the toaster [EMAIL PROTECTED] wrote: > Citando [EMAIL PROTECTED]: > >> Hi guys, i've read something about getting dspam on the toaster instead >> spamassassin. >> >> I don't know how dspam works but, shouldn't this be a very traumatic change? >> >> I think a nice approach about this will be options. >> >> You can use spamassassin or dspam ( can be used BOTH ?) >> >> AFAIK. This is the same as clamav, you can use clamav or TrendMicro (via >> trophie). >> >> >> This could let spamassassin and dspam fans be happy together :-P >> >> Just my 2 cents (euro-cents :-P) >> >> >> >> > > I think that BOTH is the right way. I'm still not very convinced about DSPAM. > > When I've the time I'm going to read the docs and setup a test server, but meantime I'm sticking to spamassassin. > > But I think the best is to integrate DSPAM support into simscan. > > I've read a post about it somewhere, maybe on this list???. > > > After reading the DSpam FAQs, I'm eager to try this out. I'm particularly impressed with the DSpam's ability to adapt to per-user situations. I've always believed that adaptive learning is best applied at the user level, not at a domain or server level. I also like DSpam's aspects of (claimed) high performance and low administrative requirements. These philosophies fit well with the toaster. I think that SA and DSpam should not be used together though. Per the FAQs, "If you are using blacklists, other spam filters, or any other form of training guides, you may be doing more harm than good. If the tool is less accurate than DSPAM is, then you will be training DSPAM to be just as inaccurate." As far as RBLs go, I'd be inclined to continue using a small number of loose RBLs in order to save resources. I'd like to see some test results before making a call on this though. So where do we go from here? I'd like to see a team of volunteers formed to work on a DSpam-toaster project. I'd like to participate in some fashion. Anyone else? P.S. I'm still going to attempt to fix my SA bayes DB problem in the meantime. It's urgency just dropped a bit though. -- -Eric 'shubes' --------------------------------------------------------------------- QmailToaster hosted by: VR Hosted <http://www.vr.org> --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- QmailToaster hosted by: VR Hosted <http://www.vr.org> --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
