"For every complex problem, there is a solution that is simple, neat and wrong."
- H.L. Mencken

That is one of my favorite quotes because Mencken, as always, hit the nail on the head. We tend to select solutions that have worked before, which makes those solutions appear to be simple and neat, but they are often wrong.

Here's an example. A few weeks ago Howard Anderson, a fellow Network World columnist, argued for one of the worst ideas in the history of fighting spam: An electronic post office that charges you for each e-mail message you send. Anderson's scheme calls for a simple flat rate - he suggests $0.01 per message - to be applied to messages sent internally and externally.

Now I respect him immensely. He is incredibly knowledgeable about the network industry and a really smart guy, but maybe because the pay-to-send and post office ideas are familiar he misses the many reasons why this is a really, really bad idea.

For some reason lots of other smart people have been seduced by the bad idea too. Included among the champions of the pay-to-send model are Esther Dyson, the first chairman of ICANN, and Brad Templeton, chairman of the Electronic Frontier Foundation (although Templeton later renounced the idea after backing it in the mid-1990s).

But the latest luminary to publicly support this ridiculous idea is none other than Bill Gates. At the recent World Economic Summit in Davros, Switzerland, Gates discussed a variation on the pay-to-send model that he calls "payment at risk," and went so far as to say that "spam will soon be a thing of the past."

In Gates' scheme, the message recipient would get to set a price to be paid by the sender if the recipient rejects a message as unwanted. Of course there's no product yet to back up this vision, but I'm sure that Gates will provide one. For a fee.

What the pay-to-send supporters seem to forget is that you don't change the way the 'Net works without major consequences. The first consequence would be consumer dissatisfaction.

If a pay-to-send system is forced on Internet users consumers will abandon the 'Net in droves and in the process kill the retail e-commerce golden egg.

Second, a whole billing infrastructure would be required to manage the payments and someone would have to pay for that infrastructure. It is unlikely that ISPs would foot the bill and I can't imagine any of the ISPs happily working with the government to create an infrastructure.