One of the big internet disappointments for me (so far) is that micropayments haven’t taken off. I wish I could tip a blog author 10 cents for a blog post that I like, but unfortunately it seems that the transaction cost of sending him 10 cents is greater than 10 cents.

So I started thinking about why this is, and I realised that it’s because when we’re dealing with money, we tend to be super-careful. To make that electronic transaction of 10 cents, we need secure servers, encrypted communications, encrypted databases, audit trails, firewalls, redundant backups, uninterruptable power supplies, etc etc. All these things are costly and generate those high transaction costs.

So my crazy idea is what if we could have a payment system that wasn’t quite so reliable? The problem is that we’re treating 10 cents as carefully as we would treat a million dollars. But who really cares if 10 cents goes missing here or there? If we could somehow build a less secure, less reliable system that was limited to small amounts, maybe micropayments could work.

A problem is that if you put enough 10-cent transactions together, you can get a million dollars. So if the system has weaknesses, it they have to be unsystematic. If it’s possible to defraud the system, it shouldn’t be possible to do so in a systematic way and turn it into a money-pump. Is this even theoretically possible? I don’t know. I also have no clue how to implement this vague and possibly stupid idea. But I really hope that someone can come up with a way to make micropayments work because the advertising revenue model is getting dull …