Re: Spam ... that discussion has been way to long. Hasn't it?

Harvey Newstrom (
Sat, 6 Feb 1999 00:11:52 -0500

"Michael S. Lorrey" <> wrote:
> However, why can't mail servers that the recipient belongs to
> merely check the validity of the emails header addresses and origins of every
> message, bouncing the ones that look specious?

Actually, this used to be more commonplace in many e-mailers. The problem is that to have this work, you have to allow outside people to query your machine and see if the username sending the e-mail is a real username on your machine. Most large ISPs, such as AT&T, specifically block queries into their domain from the outside. Thus, you can't do the lookups anymore. When this started happening, people started complaining that valid e-mail from certain domains was being blocked by the e-mail servers. So they took out the features.

A compromise used in some mail servers was to produce a warning header that shows a warning if the user could not be verified, or a warning if the sending machine is not the same as the "From:" field machine. Most products don't do this, because the warnings sounded scary and were often applied to legitimate mail.

Today, I think most people have given up trying to autodetect spam or false mail. Many ISPs offer filtering programs that leave the spam filtering up to their users, so that the ISP is never blamed for doing the wrong thing. There are pros and cons to every scheme of spam blocking, and I don't know any that are worth it to me to use. The side-effects of the spam-blocking usually are worse for me than handling the spam after I get it.

Harvey Newstrom <>
Author, Consultant, Engineer, Hacker, Researcher, Scientist.