Quote from: dacoinminster on August 01, 2013, 03:20:33 PM

Quote from: thezerg on August 01, 2013, 03:11:05 PM

I'm the guy who earned 10BTC convincing bytemaster that his original system would not track the underlying asset.

Look, let's state the problem theoretically. Let us propose 2 independent entities "Digi", and "Commodity" which can be priced in a 3rd entity "fiat". Now the task is to devise a system to create a 4th entity "DigiC", using exclusively Digi that reflects the value of Commodity.

Since Digi and Commodity are independent variables, this is an impossible task. Any escrow system that holds Digi and issues DigiC is promising 1 Commodity for every DigiC issued. It is essentially SHORT Commodity. So lets say everything starts at 1, so we have N DigiC backed by N Digi. Now if Commodity goes to 10 fiat and Digi goes down .1fiat, the "expected" value of DigiC is 10*N, but the backing escrow is only worth .1*N. A factor of 100 difference.

This will happen in reality as well as theory.

As Cyprus learned when it gave people CyprusEuro (DigiC) backed by loans (Digi). When the value of those loans drops, the only solution is to give holders of CyprusEuro a "haircut".

If that doesn't convince you, look at the 7x appreciation of Gold 2000-2012. Or imagine trying to create a "MasterBitcoin" with a 10x appreciation in 1 year!? Or pretend that MasterCoin appreciates 10x like bitcoin did, while gold drops 20%. Now you have the opposite problem your backing is worth SO much more then your coins.

And these aren't even "Black Swan" events (or Black Swans are the new normal).

I think you know that it won't track. You use terms like "Given a reasonably stable MasterCoin". But Bitcoin has shown that to be VERY unlikely. Also, its not even a stable MasterCoin that matters, but a stable RATIO between MasterCoin and the commodity...

The sooner the powerful collective intelligence and creative force moves from this pipe dream to creating a system that simply allows identity-verifiable entities to back digital commodities with a legal framework to enable criminal prosecution if the digital commodity does not meet its contract (AKA no gold in the vault), the sooner we will have a workable system. Also, look at how the Winkelvoss' (and other) ETFs work to see how multiple backers could exist for a single DigiC.

...You can gain respect for Satoshi's intelligence both by what he did AND by what he chose NOT to do...

Hi TheZerg!

I followed the posts on bytemaster's thread with great interest, since obviously I have similar goals.

You are quite right that some currencies will fail. For instance, if I define a currency that appreciates at 10% a day, it will definitely not be able to track that. In fact, I plan to create some currencies of that type, just to watch them die. I think we will learn a lot about how to tune the parameters for real currencies from their deaths.

The point is not to make currencies that never fail, but rather to make currencies that can conceivably track their target values for a very long time before finally degrading in a graceful and predictable fashion when nobody wants them anymore.

I'm sorry to disappear for 3 days and then dredge this up, but I can see that you and I have a difference of opinion on the time frame here, and on the idea of the value of the "black swan" events in a commodity speculator's portfolio. But even if you continue with these attempts at a trustless representational currency, let me very respectfully ask you to also include the ability for people to create their own (trust-required) exchangeable entities, with a reference to a signed legal document describing the entity in its genesis block. These new "currencies" could be a mortgage, stock in a company, a commodity with real backing, etc. That is, add "native" support for colored coins within your system. Whether or not your trustless representational currencies work, these trust-backed currencies will be very valuable, will utilize most of the same software infrastructure as your trustless and would make MasterCoin still useful if your trust-backed currency system does not work. PM me if you need some impl help maybe I can find some time...