I think that right now if the spammer gets a bounced email message, they take it off their list. If you put this into effect, they would never take addresses off.

In the end I think that it would just end up clogging the network even more. Right now they are pretty much only sending to legit addresses... just imagine the load if they used every address that they could. Yikes.

And this feels, smells and tastes like a hack. If you start hacking on the archtecture level, that is pretty much a one-way ticket to hell IMO.

I the best way to deal with spam is not to go after the spammers.

Go after the corporations and business that write their fucking paychecks!

I mean seriously. This makes sense from *Every* perspective. Corporations are legal entities. They can be fined. They can be found (which, convieniently makes them remarkably easier to fine).

I think that whenever a spammer is actually caught, they should go through his list of 'clients'. Any 'client' that shows up on a set number of unsolicited spammer's lists could then be fined or forced to host a black-list server or something with corporate bandwidth.

(and the only reason I'm saying a set number is because you know that someone's going to use the 'we didn't know he was doing unsolicated spamming, that's not what we paid for' BS)

on another note, yet another spammer's snail-mail address made it onto slashdot today. ... maybe we should start calling this the Ralsky effect?

Haha, nice drama. Anyway, back to our regularly scheduled programming of semi-serious discussion. Spammers won't use bogus email addresses beause it *does* cost them money (bandwidth) to send emails, just not very much for one single one. When you deal with hundreds of thousands of addresses, the costs start to add up and your hit rate goes way down.

It's not so much of a hack as it is a solution to a problem. It's like not answering the door when you're actually home. The person outside doesn't know you're at home but your doorbell still works and you can still hear it.

True, it's not in the SMTP RFC but this is an Application level solution to a lower level problem. I don't see the 'hack' in that.

My opinion:

1. SMTP needs to have authentication enforced more. In fact, email needs to have some sort of authentication scheme in place. That is a massive task, which I forsee that legislation would actually be a faster solution. Yes, faster! Doesn't that sound crazy?

2. I don't like your idea because even though the spammers use up bandwidth. So do you...and Bandwidth costs money.

*sigh* back to the drawing board.