Hi Scott: I'm not certain that a country "black list" would be sensible (even though it's tempting). Who can say for certain that they will NEVER get a legit email from someone in China - EVEN if it is just from someone logging in from a hotel and sending a trip report from their vacation, or a valid Postmaster or Abuse complaint or reply.
However, I think a weight factor per country (or a weight factor for countries in a "list") would be extremely helpful. Of course, if multiple countries are used (origin and routing), then the one with the highest weight factor should be used as the overall test result. Finally - I wonder if it would be too difficult to relate originating country with the ISO country code (which is used in the country specific TLDs). If an email ORIGINATES in China and the mail from and reply-to is indeed a .CN domain, then I'm more willing to consider this VALID email - than if it is sent through China but uses a generic TLD. Of course, another option would be to have a customizable country/weight list that is used inside the SPAMROUTING test. In other words, it would allow us to define which countries are more likely than others to trip the SPAMROUTING test. Best Regards Andy Schmidt Phone: +1 201 934-3414 x20 (Business) Fax: +1 201 934-9206 -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of R. Scott Perry Sent: Monday, March 18, 2002 01:16 PM To: [EMAIL PROTECTED] Subject: [Declude.JunkMail] Identifying countries Several people have requested a way for Declude JunkMail to identify E-mail from certain countries that commonly send a lot of spam. We did not work on this in the past, as most companies can not block all E-mail from certain countries. However, with the weighting system, this may be a useful test. As those who follow the IMail Forum mailing list know, we are working on a way for Declude JunkMail to identify the country/countries that an E-mail travels through to reach its destination. Rather than rely on the host name of the sending mail server (which is often incorrect, or a .com domain), we wanted to find a way to determine the origin of the E-mail based on the IP address. We searched for a long time to find a way to do this, thinking it was nearly impossible, but have finally found a way to do it (without having our customers have to pay over $300/year for the IP/Country data). To see it in action, you can go to http://www.DNSstuff.com and use the Reverse DNS Lookup or IPWHOIS lookup, both of which will return the country the IP is located in. So, the question now becomes, what to do with this data? Right now, we are adding a "%COUNTRYCHAIN% variable that allows you to have a header such as "X-Country-Chain: Poland->China->Destination" (for an E-mail originating in Poland and routed through a mail server in China). I'm guessing the best use would be a way to have a weight set up for certain countries. For example, an E-mail from China or Korea may start off with a weight of 5, whereas an E-mail from Switzerland could start off with a weight of -2 (which would be useful for mail servers located in Switzerland, who likely do not get much spam directly from other mailservers in their country). Is there a better way to handle countries (such as adding an option for a blacklist, similar to the sender blacklists), or other ways the country information might be useful? -Scott --- [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 . --- [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 .
