Op vrijdag 05-01-2007 om 08:26 uur [tijdzone +1000], schreef Bill Hely: > As a long-time SpamBayes user who finds that SpamBayes seems to > be relatively helpless against image spam, I'm looking for > informed comment on the following, received from Sunbelt > Software: > > Start Quote > > Does your Antispam solution catch Image Spam? Ninja Does! > > Rather than using possibly unpredictable heuristics or > computationally expensive optical character recognition, Ninja > uses fingerprinting algorithms that generate identities from a > single piece of content. Leveraging the power of the Cloudmark > engine, fingerprinting algorithms offer the most efficient and > robust Image Spam protection to date! > > Stop Image Spam today, with this free 30 day Messaging Ninja for > Exchange trial: > http://www.sunbelt-software.com/SunbeltMessagingNinja-Download.cf > m
When I read how Cloudmark fingerprinting works, I get the impression it does the same as most bayes-based spamfilters, only on a much larger scale. Anyway, it's for Exchange (mailservers), and Spambayes isn't: it is for the end user with his individual mail client. > Can SpamBayes ever conquer image spam to ant substantial degree? It already does that for me. Doesn't it do that for you? What version do you use? -- Amedee Van Gasse [EMAIL PROTECTED] _______________________________________________ [email protected] http://mail.python.org/mailman/listinfo/spambayes Check the FAQ before asking: http://spambayes.sf.net/faq.html
