On Sunday 16 July 2006 06:00, John Andersen wrote: > On Saturday 15 July 2006 08:49 pm, jdow wrote: > > Somehow I figure a better than 1200:1 scoring ratio is a pretty lopsided > > win for SpamAssassin. > > And yet, in spite of your statistics, there is more spam than ever. > Some estimates are that in excess of 95% of all email is spam. >
I think this is hype personally. I take 100k emails per day. About 30k of them are spam. Detection rates are in the 90%. (We do get some false positives, mostly far east languages. Chinese, Japanese, Korean). We reject a lot of mail, invalid addresses, sender & recipient. Those are easy. If I count each rejection, we MIGHT get up to 45-50% email spam. But that would be a lie because each email rejected, often gets rejected multiple times. (Obvious from IP, Sender & Recipient being the same). I do know that the AV detection totals dropped to only 1/3'rd of it's previous total when I did two things. 1. Verify addresses at the receiving MTA (Postfix). 2. Reject inbound email from my own domain. Including in SMTP headers. (Yes this breaks forwarding). Tanstaffl. (sp?) > If it didn't pay, no one would do it. Clearly spammers are succeeding. > > Spamassassin and Razor haven't made a dent in the amount of spam, > they just mask the problem. Further, I still pay for the bandwidth. > Reject more. Don't accept email fro non-existent addresses. Often people get annoyed because you won't accept mail from unverified addresses. Often some big organisations even refuse to send from a valid return path (Go figure. Apparently email is 'important' to them up till it leaves their servers, then suddenly not any more). BUt rejecting this way cuts down 2/3's Viruses & spam. Because they use harvested addresses. And the churn is obviously enough to reject a large percentage without too much trouble... Hamish.
pgp8TheJOmlCc.pgp
Description: PGP signature
