Cable Forum - View Single Post

Matth · 04-10-2004, 00:53

Bouncing is WORSE than useless, as the sender address is invariably forged.

As I understand it, with Spampal, you can delete IMAP mail within spampal, but not POP.

If you are CERTAIN of your spam tagging, you could use a delete from server rule, but I'm not entirely certain if that would pass correctly through, or if there is a risk of skewing the numbering.

One thing, seems to be that the typical spam/bulkmail message-id formats are quite good predictors, and only need a small degree of cross-checking for legitimate bulkmail - or perhaps, secondary checking for known rotten IP ranges, like 24.0.0.0/8 , and perhaps the entire APNIC controlled range as well.

Mark it differently, and delete only that marking from the server.

Reporting:
To be honest, I used to report, now I've thrown a fairly solid blocker up - the days when spam had traceable sources are long gone, it's usually from an open proxy, on a "don't care" ISP, pointing to a website in "don't care" China.

Actually it looks like NTL might be despamming, as since I threw the block on my old Freeserve account (I used THEIR spam marking, with exceptions for any known good senders that were getting marked, and delete from server), I haven't noticed ANY spam at all - the ones that used to come through on both, don't seem to be coming through at all.

04-10-2004, 00:53	#95
Matth Inactive Join Date: Mar 2004 Services: BB:M, TV:XL, Phone:M, Loyalty Posts: 2,516	Re: Spam spam spam spam spam spam Bouncing is WORSE than useless, as the sender address is invariably forged. As I understand it, with Spampal, you can delete IMAP mail within spampal, but not POP. If you are CERTAIN of your spam tagging, you could use a delete from server rule, but I'm not entirely certain if that would pass correctly through, or if there is a risk of skewing the numbering. One thing, seems to be that the typical spam/bulkmail message-id formats are quite good predictors, and only need a small degree of cross-checking for legitimate bulkmail - or perhaps, secondary checking for known rotten IP ranges, like 24.0.0.0/8 , and perhaps the entire APNIC controlled range as well. Mark it differently, and delete only that marking from the server. Reporting: To be honest, I used to report, now I've thrown a fairly solid blocker up - the days when spam had traceable sources are long gone, it's usually from an open proxy, on a "don't care" ISP, pointing to a website in "don't care" China. Actually it looks like NTL might be despamming, as since I threw the block on my old Freeserve account (I used THEIR spam marking, with exceptions for any known good senders that were getting marked, and delete from server), I haven't noticed ANY spam at all - the ones that used to come through on both, don't seem to be coming through at all.