Domain: spambayes.org
Stories and comments across the archive that link to spambayes.org.
Comments · 7
-
Spambayes
http://www.spambayes.org/
Works pretty well, we have about 75 users. Doesn't use much of our P2-450 server running Slackware. -
Re:Seems like a good place to ask it...
I personally use Spambayes. It runs on my Outlook client and employs Bayesian heuristics and the few messages that do get through I mark as spam to even better train the system. Although I haven't tangibly recorded the success ratio I would estimate that it was about 95% effective the first week of usage. After training it that amount of time it has been about 99% effective. Rather than download and update all of the DNS blacklist stuff the Bayesian filtering seems to do the job. Even those weird random word messages or haiku deals get flagged.
-
Re:A different approach to a block listThere's another effective cross platform tool that I'm hooked on. It's called Spambayes and uses similar Bayesian filters. I would say that when the thresholds are correctly set it filters out about 99% of the spam that's out there. Even the haiku, random word, etc. variety. The more spam you get the better the Bayesian analysis becomes. If you're a Microsoft Lookout user you can just have the Junk Mail folder automatically empty out every x number of days and won't have to worry about most spam again.
Looking at all of the broadbased effects that spam has --- added network traffic, open SOCKS proxy exploits, open SMTP relay exploits, trojan host takeovers, lost business time/productivity, added storage allocation --- it really is high time that the standard governing organizations expand the SMTP protocol in to a stack that includes more sophisticated mechanisms to ensure message integrity. A sender verification token of some sort. Be it a PKI check, a site certificate, a challenge/response between sender and receiver mailhost, etc.
Since supposedly the spammers can hide their tracks well perhaps whatever commercial product being spammed should be targeted by the authorities. The websites and entities in question would certainly be less likely to hook up with spammers then I would think.
-
I don't quite see the problem...
Looking at my spam-box, I find the usual stuff:
From ukKimble@mailthat.net Tue Jan 13 00:43:36 2004
X-Habeas-SWE-1: winter into spring
X-Habeas-SWE-2: brightly anticipated
X-Habeas-SWE-3: like Habeas SWE (tm)
X-Habeas-SWE-4: Copyright 2002 Habeas (tm)
X-Habeas-SWE-5: Sender Warranted Email (SWE)
(tm). The sender of this
X-Habeas-SWE-6: email in exchange for a license for this Habeas
X-Habeas-SWE-7: warrant mark warrants that this is a Habeas Compliant
X-Habeas-SWE-8: Message (HCM) and not spam. Please report use of this
X-Habeas-SWE-9: mark in spam to .
Subject: Fwd: V|@gra, Vali(u)m, X(a)n@x. Prescribed Online and Shipped
... and finally, the real information as far as I'm concerned in in the last header:
X-Spambayes-Classification: spam; 1.00
So whether the spam is "legitimate" (is there anything like that?) or not, SpamBayes doesn't seem to have much trouble with it.
-
Re:How many?
Same here. I recently installed Spambayes, which is a freeware spam filtering add-on for mail clients. Works both on Outlook clients and the Linux platform. It filters out 99.9% of the spam I get. Really good for freeware.
-
All hail Bayes!
-
Re:CloudMark's SpamNet