naturalnews.com printable article

Originally published March 11 2004

Intelligent spam filters simply don't work: the real solution is computation cost for email senders

by Mike Adams, the Health Ranger, NaturalNews Editor

You've no doubt noticed this in your own email inbox: an endless flow of babble introducing a spam-like message. It's the latest trick of spammers, designed to get around the so-called "intelligent" spam filters. By packing email messages with harmless words and phrases that don't contain spam-sounding keywords, spammers can get email past the spam filters with ease.

It just goes to show us all that spam filtering is simply not the answer. The only way to stop spam is to create economic friction in the process of sending email. And the best way to do that is to add a computational cost to the sending of each email: the Microsoft "Penny Black" solution.

It's also called the puzzle solution or "good faith" solution to stopping spam: every email sent by every mail server on the planet must be subjected to 10 seconds of CPU-intensive calculations in order to be stamped as a "good faith" email message. By spending 10 seconds of CPU time, the sender is proving that they have some economic investment in sending the email. It may sound like a small investment, but it's far too large for spammers to bear, since it would slow a mail server down to 360 emails per hour. Spammers need to send millions of emails per hour to make a profit. Sending just 360 per hour would halt their operations outright.

If we're going to tackle the spam problem, we have to stop thinking about receiver solutions and start thinking about making changes on the sender side. Receiver-based solutions will never work. As this article shows, even the intelligent spam filters that used to do a decent job are now useless thanks to spam babble. The ultimate solution to stopping spam now rests on technical solutions that essentially slow all mail servers. It's sort of like requiring all senders and email marketing firms to pay a bit of postage for their mail, which is exactly what happens in the world of snail mail. The sender buys a stamp and pays for the mail. Imagine: if all direct mailers could send an unlimited number of mail pieces each day at zero cost, your physical mailbox would be just as filled with meaningless messages as your email inbox.

So it's postage that keeps direct mail in check. Likewise, in the email world, senders need to be paying a sort of "computational postage" to associate some cost with their outbound email. That will make bulk email far more responsible: suddenly permission email marketers have to think hard about who they're emailing.

Some people think that charging real money to send email is a better idea, but paying money to send email is actually a terrible idea, since it would become a political nightmare trying to figure out who pays what, to whom, and who controls it all. That's not to mention world currencies, dealing with non-profits, poor countries, and a long list of people who simply can't afford a penny per mail. Are you going to take away their right to send email?

Of course not. That's why the only viable solution is to require senders to pay in terms of CPU cycles. By requiring computational time to send email, you are indirectly creating an economic disincentive for spammers without creating unreasonable barriers for everyday folks to send email. And that's the solution to spam, plain and simple.



Good news for spammers, the smart filtering software used to catch spam can be beaten. The discovery has been made by anti-spam researcher John Graham-Cumming who studies the novel ways spammers try to defeat the technologies used to stop junk mail. To cut out the junk, many e-mail users have turned to a technology known as Bayesian filtering to spot and stop spam before it reaches their in-box. When trained to spot what is spam and what is legitimate mail these smart filters can catch, in many cases, more than 99% of junk messages. But Mr Graham-Cumming, who is a member of the Sophos Anti-Spam Task Force, has found a way to beat Bayesian filters that guarantees a message will get through every time.


All content posted on this site is commentary or opinion and is protected under Free Speech. Truth Publishing LLC takes sole responsibility for all content. Truth Publishing sells no hard products and earns no money from the recommendation of products. NaturalNews.com is presented for educational and commentary purposes only and should not be construed as professional advice from any licensed practitioner. Truth Publishing assumes no responsibility for the use or misuse of this material. For the full terms of usage of this material, visit www.NaturalNews.com/terms.shtml