Tom, Monday, July 29, 2002 you wrote: T> I see, so there are a few out there. I guess I could use another T> list with a weighing system for the BOUNCE address. What are your T> thoughts, if any?
"bounce" is less common now in lists than a short time back. I suppose that's why so much spam has it. Lists are a terrible problem. It is really difficult for me to see a good pattern for programmatically determining what constitutes a list and then what constitutes a valid subscription to a list. It is even more of a problem than normal spam because the user frequently has paid money for a subscription and certainly doesn't want it deleted. Most of our false positives continue to be list subscriptions. But they are really important so we continue to manually inspect for them. I've toyed with trying to parse a subscription address and then compare to an imail user. But the parsing is difficult because every conceivable method is used for this and frequently it is just a number with no correlation to anything else. Sometimes it is encoded. I don't have good ideas for you. I just know I see quit a few legitimate subscriptions using bounce in the mailfrom. Terry --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type "unsubscribe Declude.JunkMail". You can E-mail [EMAIL PROTECTED] for assistance. You can visit our web site at http://www.declude.com .
