Impoverish a Spammer Today
esj at harvee writes "Recently the Camram project released its latest version of a hybrid sender-pays anti-spam system. The project has proven that sender-pays works and has demonstrated how to make it work with existing e-mail systems. Camram has developed hybrid sender-pays techniques that scale down to the desktop and up to the enterprise. It's a completely decentralized system that can put spam-fighting power in the hands of individuals. It gives you control of not only the current generation of spam, but also any future commercial spam -- why replace Viagra ads from a scam artist with Viagra ads from Pfizer?"
The problem is that I've seen no good way to stop non spammers from paying as well.
"Have you ever thought about just turning off the TV, sitting down with your kids, and hitting them?"
Others have mentioned that this will make it easier for the user to notice that their PC has been hijacked, but another side-effect is that it will perform a rate-limiting service on that zombie. If each zombie can only send 100 messages an hour instead of 100,000 then that is another important benefit.
It seems to me that one should need only one stamp generator. I receive a payment request containing a message encrypted with a short private key, and as "postage" I need to decrypt the message and return it. As computers get faster, the key length used to encrypt the message gets longer. The receiver can thus decide how much postage is required.
This way the stamp generator doesn't need to have any secret component, and could be written in any language. It could be part of the mail client.
Like whitelists and keywords, this is a special case of a token-based system. Token-based systems depend on the sender performing some action that is, at the time they send it, sufficiently hard to predict, unusual, or onerous for a spammer to bother with it.
For example, I have certain addresses that bypass my spam filter either partially or completely, and I have set up a scheme for my kids whereby a sender has to know a "magic word" to get in. Whitelists, of course, make the sender address the token.
Right now, these are good enough.
Spammers are beginning to respond to whitelists, though, and trying to guess sender names. It's only a matter of time before they start using the address books in their zombies to build up lists of probable whitelists, and start sending spam using pairs of addresses from the same address book the way viruses already are.
So the next spam zombie worm will just whitelist everyone?
Not all devices will have enough computing power available. My grandmother has an Amstrad E-mailer. How long will it take the 4Mhz Z80 in there to generate a stamp? How about the cpu in my phone?
From the Faq "You only generate a stamp the first time you mail someone." So when all 20 of the biggest spamhouses have generated a stamp for you, you are right back at square 1? Net cafes with changing clientelle pay a higher price than spammers? Forged headers cliaming to be from friends don't need a stamp?
A pizza of radius z and thickness a has a volume of pi z z a
I agree - worms are the biggest problem with this scheme. You can't hold the spammer accountable because the spammer is most likely not even sending the spam but using millions of zombie machines.
The best way to deal with the problem is follow the money then show up at 4am and stick a Glock in the face of the spammers and their family members. After they shit the bed give them the option to play nice or die anonymously. Harsh? Yes. But not quite as bad as prior reform methods such as the Pyramid of Skulls*. I may be biased, my computer system was compromised by trojans from those bastards last week and pretty much I am still pissed about it.
* Historical note on the making decortive yet functional pyramid of skulls (taken, I shit you not, from kids.mapzones.com): 1258 Baghdad was conquered and sacked by Hulagu, grandson of the great Mongol conqueror Genghis Khan. Hulagu killed all the scholars in Baghdad and erected a pyramid from their skulls. He destroyed the elaborate irrigation system that the Abbasids had established. Iraq became a neglected frontier area ruled from the Mongol capital of Tabriz in Iran. In 1335 the last great Mongol ruler of this region died, and anarchy prevailed. The Turkic conqueror Tamerlane sacked Baghdad in 1401, again massacring many of its inhabitants. He, too, built a pyramid of skulls. Tamerlane's invasion and conquest marked the end of Baghdad's greatness.
It may not seem fair to make everybody go thru a security checkpoint, just because of the actions of a few -- but you can bet your sweet ass it is necessary.
As an aside, I would wager that the percentage of your messages that are actually read by the recipient goes up, after this protocol is put into place. Because for the simple fact that your legit messages will no longer be lost in the noise of illegitimate ones.
Why can't they send out the messages via RSS or some simliar technology? You'd email your message to the list, & the list would RSS it to all the interested people. This has the advantage of letting people read without subscribing.
Seriously, does anybody know why this hasn't been done? I'm not an expert, so I wouldn't know of any limitations. I'm thinking of a cross between newsgroups & mailing lists.
testing out my trending skills