Financial Cryptography

You do a great job explaining this subject. The whole idea of writing a formal contract, along with issue & issuer keys and signatures, is great because it creates a recognizable, meaningful, immutable, and evaluable object. Taking the hash of the thing creates the spectacularly concise yet distinct thing you need in software transactions. Threading that hash into each transaction with signatures freezes the meaning of that transaction for all time.

It's an excellent idea, and sorely needed in today's DGCs. Pecunix is just _starting_ down that road by adopting PGP signatures. Perhaps one day they will create a Ricardian contract for their issue and incorporate its identifier into their payment system.

I'll get back to you with more comments soon.

I remember deciding what sort of data structure to use to represent currency in ecashlib in 1996 or 1997, and you saying that observing this process led you to an epiphany about payment systems. I've never understood what you were on about until now.

Wow! I wonder what the epiphany was :-) Certainly, the experience of your design review formed the bulk of the scratch draft paper that had been hanging around since 1997. I had written up those experiences in (much more) detail back then.

At the time, I recall being willing to discuss it all, but I was faced with the normal blockage of your company not wanting to accept any ideas from outside. This process was highly illusory, in that I knew it wasn't because people weren't smart enough... The interesting thing was that it was a process, as all people in DigiCash did it, and what's more all people in other companies do it - including our own! I even have a policy of totally ignoring other people's payment systems until they become challenging.

So, all these good ideas don't spread out until they become successful. We are losing all the really good ideas because there is no easy way to share them - marketing seems to stop our brains accepting the "competition."

How to solve this? I don't know.

( I think the epiphany was that the Ricardian Contract solved the numbers allocation problem that DigiCash had in those days. But, we had arrived at the design by a different path - that of bond contracts, so we had never really appreciated how widespread the problem was. )

You said something about how seeing this issue come up again in ecash's context made you realize that it was a fundamental, but ill-understood issue.

At the time, I ill-understood it and didn't see why it was important.

I agree that DigiCash was closed.

I enjoy working on an open source project -- Mnet -- where I feel free to give and receive ideas with strangers, as much as I like.

I think that matches my dim dark memories.

In '95 I designed the Ricardian Contract (with Gary as the sounding board) from the pov of bonds. For which, I had finance classes, examples, and books to go on. Also, I think I had one great weapon - I was unpoisoned by the knowledge of the City (/Street), which I've later come to realise is so deep and intricate that few insiders can swim in it nor do they realise the colour, texture, and bouquet of their surroundings.

Initially our concept was bonds only. It wasn't until I went on a currency dealer's course in January of '96 that I had an earlier epiphany about cash being just another instrument, and the markets being nothing more than a more sophisticated fruit&veggie market. Up until that time, I had a specialist-style market, which was simply not working. After the course, I was able to do a direct market, and other instruments fell into place as yet more contracts.

But, I guess I still didn't understand the full relevance of the invention. Indeed, I recall being totally in agreement with (the other) iang when he called it "obvious" at the first FC in Februart of '97.

Your design review was the start of a rethink along "it may be only obvious once it is carefully explained." To be fair, I never explained the Ricardian Contract to you at all, instead I tried to redevelop it from DigiCash's pov, from first principles, assuming that its obviousness would fill in the rest. If it hadn't been for that design review - the outstanding case of a company opening itself up to external participation - I'd probably still be thinking that it's so darn obvious, someone is bound to get it soon.

(My response to the normal paranoia at the time, as displayed by all companies in our field, was to think, well, we are above that foolish and destructive secrecy thing, and we will share and trade ideas, but we'll stop short of doing the work for other companies.

Hence my comments about ideas transmission and so forth; what I'm really wondering myself is, if we had so many good ideas, and couldn't share them even when we tried, what about all the other companies? Did their good ideas sink without a trace? Are they lost until rediscovered? Why is it that our own little "obvious" idea took a decade to get into paper?)