I just checked the reverse DNS entries of about 10,000 legitimate E-mails we have here. Of them, 2.5% meet your regexp expression that you posted earlier.
You mean you checked the PTR hostnames of MTAs that sent you 10K legitimate emails?
So, you selected legit mail and said 2.5% would be blocked by my regex. You are selecting the data to fit your position.
For 1OK IP hostnames that meet my regex,
1. how many IPs are sending spam?
2. how many spam msgs (or rejects) are sent?
3. how many are legit mailers? (the rest)
4. how many legit messages are sent?
I consider false positives to be percentage of total rejects, not a % of total legit. I don't care about legit mail. It's not the problem.
The FP issue is how many REJECTS are really legit?
iow, of 10K rejects, how many are false positives?
I expect for you to get 10K legit mails, you actually received 20K totals msgs MINIMUM, ie, 10K rejects, but for many ISPs, rejects are running at 75+%, so to get 10K legit mails, they actually have to receive 40k total msgs, and reject 30K. you your inflated 2.5% number is now cut down by 2/3, to 0.8% FP in 30K total rejects.
But even that is still too high.
The issue we are discussing: for 10K rejects using my cable/DSL filter, how many of those are false positives?
Len
To Unsubscribe: http://www.ipswitch.com/support/mailing-lists.html List Archive: http://www.mail-archive.com/imail_forum%40list.ipswitch.com/ Knowledge Base/FAQ: http://www.ipswitch.com/support/IMail/
