The spam problem part 3: Objections to attention bonds

In part 1, I enumerated the approaches to spam eradication I was aware of, and explained my conclusion that the only approach which will work is an economic approach. In part 2 I discussed various options for tackling spam economically, ending with the one I think would actually be acceptable and useful: attention bonds.

Now I’ll run through (and shoot down) a few of the objections commonly brought up when the possibility of involving actual cash in e-mail sending is raised.

Continue reading

The spam problem part 2: The dismal science

In Part 1 I took a “from first principles” look at the spam problem, and concluded that the only way to actually solve the problem was to make people pay to send e-mail.

Now, it’s time to look at what I mean by that—because there are almost as many ways to implement “pay to send” as there are ways to implement filtering.

This is going to be a bit more technical than part 1. I’m going to assume you know basically how SMTP e-mail works. If not, there are tutorials available.

Continue reading

rejoice amatory butt isinglass

[2004-03-02] Well, pobox.com’s new spam filtering system picked up 2,982 spams in the last week, and 1 false positive. And that wasn’t really a false positive—it bounced a newsletter from sudhian.com because they’re apparently too incompetent to set up their MTA to provide a proper HELO hostname, so their SMTP request was invalid (as per the RFCs). I sent them e-mail to warn them, and it bounced because their newsletter reply address was invalid too. I’ve forwarded the bounce back to postmaster, what’s the betting they’re violating that RFC as well?

I have no problem with bouncing mail from anyone that incompetent, and 99.99% accuracy is plenty good enough, so I’ve switched the filters over to full automatic, and now they’ll reject the spam e-mail during the SMTP attempt. It won’t even reach my second-line adaptive bayesian filters.

[2006-03-09] About two years on, and the spam rate remains more or less constant: 2,840 spams in the last 7 days.

Contrast this with the claim from the FTC that the CAN-SPAM Act has been effective, and that consumers are receiving less spam than they used to.

It’s quite possible that consumers are receiving less spam, but from my numbers it seems clear that the amount of spam being sent hasn’t gone down. Instead, filtering for the average person is getting more effective.