Wow. This is great. Paying for posts on a topic that I am interested in is a really fun use for bitcoin.

I will try to be concise, but this will probably be tl;dr for some of you.

Quote from: Stephen Gornick on July 23, 2011, 02:22:28 AM

Are you aware of Open Transactions?
- http://en.bitcoin.it/wiki/Open_Transactions
- http://forum.bitcoin.org/index.php?topic=20425.0

Also:
- http://forum.bitcoin.org/index.php?topic=28841.0

I read a bit about OT when it first came out, but I don't claim to understand it completely. At first blush it seems way too complicated for the average Joe to understand and use, and it is not clear to me who is being trusted. With bitcoins, I know I only have to trust an algorithm implemented in Open Source software.

What would be interesting to me would be the "marketing copy" for OT (like how WeUseCoins makes the case for bitcoin). I'd be interested in reading anybody's attempt at marketing copy for OT: a convincing story about who would use OT, under what circumstances, why they would trust it, the "killer app" that makes everyone start using it, how it could change the world, and most importantly (to me) how it could be used to create a black market for currencies, commodities, stocks, bonds, etc.

Quote from: Olomana on July 23, 2011, 04:20:32 AM

Consider gascoins / anti-gascoins. The problem with pegging a gascoin to a gallon of gas like a futures contract is that the holder of the anti-gascoin (the person who is short one gascoin) has unlimited liability, and he can't place an infinite number of bitcoins into escrow. (The price of gas can't go below zero, but there's no limit to how high it can go.) There would have to be some way to manage risk, issue margin calls, etc.

An Intrade-style prediction contract might work better. For example, a contract might settle at 1 BTC if the price of gas is $5 or greater at the end of the year, or settle at 0 BTC otherwise. Contracts are created out of nothing whenever a buyer (long) and seller (short) agree on a price. Maximum escrow is 1 BTC for the short. Contracts are destroyed whenever a short buys back a contract, or at the time of expiration when all contracts are settled at 1 BTC or 0 BTC.

A small fee on each transaction could keep the miners going.

This is a very good summary of the biggest problem with my suggested coin/anticoin model: how to make sure there are enough bitcoins in escrow to cover all possible market actions. I will discuss this more below.

I would LOVE to see a "distributed intrade" powered by bitcoins. There would be a lot of technical challenges, but it may in fact be a better way to create a black/grey market for speculating on tons of different currencies, stocks, commodies, world events, etc than what I have posted about above. The only problem is you need a stabilized currency (otherwise your profits/losses from your bets will be dwarfed by your profits/losses from holding bitcoins to make those bets!)

IMHO, the next major step for bitcoin MUST start with a price-stabilized version of bitcoin, backed by bitcoin value, but providing a way to separate those who wish to hold bitcoin as a speculative investment from those who want to use it as an unvarying store of value for commerce, bets, etc. My guess at one way that might work is here: http://forum.bitcoin.org/index.php?topic=30741.0 which was an idea I then extended into this thread.

I am very, very interested in how a price-stabilized extrapolation of bitcoin might be created, with a very simple user experience, and the protocol working behind the scenes to match stabilized bitcoin holders to leveraged bitcoin speculators. This would not (and must not) be a bitcoin replacement, but would have to build on to the existing bitcoin infrastructure.

Quote from: cunicula on July 23, 2011, 12:27:23 PM

This is a good idea, but how is the infomation about exchange rates fed into the blockchain. You need I thinlk two items of third party data: the current USD exchange rate and the exchange rate between current USD and whatever commodity you are pegging value to. Presumably, humans need to supply this data. How do you incentivize the humans to supply honest data?

If you want a distributed currency pegged to ANYTHING, this is one of the biggest technological problems to solve. However, consider this: there are also attack vectors on bitcoin that involve fraudulent timestamps. Bitcoin uses a distributed timestamp protocol, where nodes reject timestamps that differ significantly from what they think the time is. I believe the same logic can be extended to exchange rates. If somebody lies about the exchange rate, other nodes will reject that block. Consequently, I consider the problem of distributed exchange rates a (mostly) solved problem.

Quote from: jtimon on July 23, 2011, 07:08:47 PM

Quote from: cunicula on July 23, 2011, 12:27:23 PM

That's exaclty what I was going to ask.
How the block chain knows if prices of gas went up or down?

If you solve that problem, I have another proposal.

Your link led me to this very related thread ("Achieving stable prices through a reference currency"): http://forum.bitcoin.org/index.php?topic=11614.0

Quote from: deslok on July 25, 2011, 02:10:59 AM

"Backing" bitcoins or any currency with a single resource is never a very good idea since any single comodity can be controled. if a single wealty enough individual wanted to they could inflate and deflate the vaule of bitcions at will. look at our market cap it's less than a hundred million dollars one or a group of sufficently rich people could control the entire market fincially. If bitcions were to be backed by any single comodity the same thing could happen right now however they are treated as a comodity themselves not so much as a currency.

If bitcoin is successfully extrapolated to provide a stabilized version backed by bitcoins, and then is further extrapolated to allow distributed speculation on stocks, commodities, currencies, etc, the bitcoin black market will be massive. The only ones rich enough to influence it will be the early bitcoin adopters, who would be trillionaires (by USD valuation)

Quote from: Desolator on July 25, 2011, 04:02:44 AM

Backing bitcoin-like currencies with commodities seems like an insanely bad idea. It will take everything wrong with systems like speculative investments in oil causing gas prices to go up and give it an alternate anonymous underbelly. That's insane. I still don't quite get the logic on how bitcoin can be merged with anything. I don't get why they would do it or how they would mathematically make it work. . . .

Regardless of how insanely bad the idea may be, I also think it is inevitable. There is just too much money at stake for bitcoin holders to ignore this potential million-fold increase in bitcoin values (Here's how I do the math: http://forum.bitcoin.org/?topic=7985.0).

As for how it could be done, that is what this thread is about. Maybe it is not possible, but nobody would have thought bitcoin was possible a couple years ago. If it is possible, mark my words: someone will do it. Then all hell will break lose.

Quote from: jtimon on July 25, 2011, 06:35:26 AM

Quote from: Desolator on July 25, 2011, 04:02:44 AM

Backing bitcoin-like currencies with commodities seems like an insanely bad idea.

I think is impossible to "back" a currency without introducing centralization.
In this case, he want to use option contracts but I see a few problems.

1) As far as I know, option contracts are not fungible. I didn't get the Gascoin/antigascoin thing.

-Suppose oil is at 10 btc right now. How many btc to buy the oilcoin, a much for the antioilcoin?
dacoinminster, can you elaborate a little bit more on this?

2) If you could peg a chain currency to a commodity using contracts, the options market still needs an arbiter to determine what the price of the commodity is at a given moment. How do you make that arbiter decentralized?

For #2, see above my comments on distributed exchange rates above.

For #1, I am not sure that gascoin/antigascoin is the way to go. It is possible that an intrade-like system as discussed earlier might be a better way to do it.

I will say that I imagine gascoins/antigascoins would have to have some sophisticated rules to prevent the bitcoins held in escrow from running out. I'm guessing the rules might involve some of the following ideas:

1) All bitcoins held in escrow could be used as needed when payouts happen, not just the bitcoins from one coin/anticoin pair
2) Anticoins would be fungible. If you bought one, some of your bitcoins might go into escrow if that anticoin was last sold when the commodity had a lower price. Some bitcoins might come out of escrow (to the seller) if the commodity had a higher price at the last sale.
3) If an anticoin was in danger of going "in the red", the protocol might force a sale, similar to a margin call
4) In the event that all the entire escrow fund went "in the red" (like if there was a "run" on the funds similar to a bank run), bitcoins could be created "out of thin air". I recognize this would violate the 21M hard limit, and that would probably never fly with the community. Perhaps instead of real bitcoins, they could be some kind of "IOU" from the escrow fund, redeemable when the fund went back in the black.

Regardless of whether people are trading something that looks like a future's contract (coin/anticoin), something that looks like intrade, or something else entirely, the first problem to solve is how to create a stabilized extrapolation of bitcoin.