Rupert Goodwins' Diary
News just in from the States, where a company called Peppercoin has announced version two of its product. Its business is micropayments, the tricky business of extracting small amounts of money from people in return for small services or goods. Easy if you're a newsagent and can hand over a penny gum in return for a penny coin: impossible if you're an online vendor and have to give the credit card company 20 pence for the same transaction.
The original Peppercoin was very clever, very useful and almost totally ignored. It worked by passing digitally signed tokens around the place only some of which got redeemed, thus reducing the number of authorised transactions. Statistically, everyone got what they expected, more or less. Peppercoin was the invention of Ron Rivest and Silvo Micali, MIT professors with a track record in cryptographic cleverness, Rivest being the R in RSA, and it relied on seeing micropayments as being a tiny part of a statistically well-defined very large transfer of funds. Which it is, if you're a statistician.
Alas, accountants care not for statistics and cryptographics: they want the sums to add up to the last boring pfennig. So when Peppercoin said that version 2.0 fixed this -- everyone got exactly what they expected -- and the other drawbacks to 1, special software and having to sign up for an account, I got interested. Not least because I couldn't see how the original scheme could be modified to work like that: the press releases were less than forthcoming and nobody else seemed to know.
So I phoned up Peppercoin, getting a very loquacious chap who had his spiel perfectly polished. But no, nothing juicy. "But how does it work?" I asked a couple of times -- and eventually got the answer. All the clever stuff had been ditched, at least from the sharp end where the money moves around. Instead, the Peppercoin servers take a note of your credit card number the first time you do a small transaction and then get authorisation from the credit card companies for a much larger amount. They then act as authorisation proxies for your next few micropayments, letting you buy stuff without troubling the credit card company computers until you run out of authorisation, when they repeat the exercise. Result: fewer transaction charges, and cheaper transactions.
Which is fine, but hardly what Peppercoin had been trading on as their unique selling point. In particular, it won't scale very well and it's very vulnerable to the credit card companies deciding to implement a similar scheme -- which they could do very quickly if Peppercoin shows signs of eating their 50-cent lunch. No magic MIT cryptostats.
This was almost a news story, but I wanted a response from the credit card companies themselves. I got a very good lesson in news management from their PRs: I was promised a response, then a response from the US, then -- at 5:30 on the dot -- a phone call said "ah, actually, we don't comment on this sort of thing."
Now, that's class.