Monthly Archives: May 2003

Mr. Gates has proposed his solution to spam. Unfortunately, it is yet another idea that will not work.

The problem the MSFT solution aims to solve is the problem of distinguishing good spam from bad. The proposal has a clever (though I think dangerous) safe harbor provision to “create incentives for email marketers to adopt best practices, and to certify themselves as trusted senders who can be more easily identified by consumers and filters alike.” Presumably, if we know which marketers are “trusted senders” we can accept their mail, and block all the rest — spam and non-spam alike. Thus, email would become a more effective channel for trusted marketing — but little else.

The safe harbor provision could make sense if there were a background requirement that all spam be labeled. There’s a hint of that requirement in the letter Mr. Gates wrote to the Senate Commerce Committee (“participants would be entitled to avoid the burden of additional labeling requirements (such as “ADV:” )”). But the proposal doesn’t actually endorse a labeling requirement. And without it, the proposal does nothing to distinguish real email from HGH sellers. The proposal would help distinguish HGH sellers from, say, Amazon. Wonderful, but I didn’t know that was the problem.

The proposal does say lots about making sure ISPs and state attorneys general have the power to sue — again, like most (but not all) solutions, centralizing the enforcement function. But all such solutions will fail because a centralized system for enforcing spam regulations will never be enforced. ISPs and state attorneys general have better things to do than enforce spam regulations. They always have; they always will.

This is the key point: the enforcement problem. Whatever the requirement, if it is not effectively enforced — meaning that most spammers do not fear that they will be caught and punished for failing to obey a requirement — then it will fail. And if it is effectively enforced, then it will work even if its penalties are not harsh. Solve the enforcement problem, and a slap on the wrist will work. Fail to solve the enforcement problem, and even the death penalty would be ineffective.

It’s no surprise that Congress doesn’t get this. Congress gets points for “seeming” tough. If you seem tough, it doesn’t matter if your ideas work. So puffed-up “get tough” rhetoric tied to totally ineffective legislation is the norm.

But it is a surprise that a company as skilled as MSFT would make the same Washington (DC) mistakes. Mr. Gates has done extremely well in world where mistakes hurt profits. He is doing extraordinary good in the world where generosity (indeed, astonishing generosity) corrects for policymakers’ mistakes. But as a policymaker himself, he is still MSFT v1.0.

Let’s hope he gets to MSFT v3.1 soon. Continue reading →