Topic: New musings for a stable currency (Read 3415 times)

I like the idea of a tax. However instead of a flat tax demurrage, let's take a percentage of the transaction itself and return it to the minable blocks. I don't even know if this is possibe, but it would serve two purposes. A 51% attack that attempted to reorganize the blockchain would have to pay a tax on every phoney transaction created. It would also be a form of incentive that would keep mining viable in perpetuity.

this is what I used:

(this was september '11)
15 THash/s divided by 250Mhash/s means approximately 60,000 miners. Multiplied by 150Wh, you get 9MWh of electricity. If we assume an average of 12 US cents per kWh, that means it costs $1,080 USD for 300 BTC, or $3.60 to produce 1 BTC


complain about my assumptions if you want, but 150Wh is probably really low considering that only takes the GPU into account and nothing else


edit: and again, there is no cost included for hardware purchased specifically to mine bitcoins

When did a coin cost $3.60 to create?

Using electricity produced from burning bundles of $100 bills and transmitted over platinum wires to datacenters cooled by drop shipped icebergs?

True, but probably have the least incentive to make it crash. But I agree (as I posted on the newbie forum) with the limited number of btc any sufficiently wealthy party can destroy bitcoin entirely. So let's get encoin rolling.
Looking forward to it!

But the thing is, those advantage are not bygones. I can't think of where the chart is off the top of my head, but someone posted a nice flash chart or whatever that let's you see how many bitcoins have been spent over time vs. when they were created. Nearly all of the first 2-3 million coins have *never been used once*. Look at the speculation forum to see various threads on the manipulator who is using around 50k btc (how many people have this much? several dozen probably) to continuously manipulate the price. Small market and all that jazz, but the thing is, the amount of coins available for sale will probably remain roughly constant from here on out--we're about to start awarding only 25 btc per block afterall. The satoshi group with its million+ will have the ability to crash the market at will essentially forever. Down the road this could be used as a coercive threat, who knows. Satoshi, who took so many pains to stay completely anonymous (what the hell for?) has disappeared without any indication as what he/they plan to do with these coins. Early successful bitcoin businesses like pool operators and mt gox especially will also have such a huge amount of power.

I'm not against the early adopters making a good chunk of change which is why I proposed coin reward multipliers, but I am vehemently opposed to the kind of unlimited earning potential and power over the economy that bitcoin (and real world government/banking) allows for.


I don't want to take away wealth or capitalism. I want to make it so that the wealthy cannot manipulate the money supply to suit their own ends though (getting more wealth). The wealthy still stand to benefit from all of the upsides of a cryptocurrency, and if the people choose encoin as their currency of choice, they are just going to have to deal with not being able to manipulate it.


yeah I'll bump the thread with my latest thoughts later today. I believe I've made a great compromise between anonymity of businesses and security of the network in the tradenet, and I think I've improved on the stability of coin creation on the supplynet side too. Not quite enough for a major revision, but enough to where I think just about every aspect is rock-solid.

Wow, you almost did. That was a long reply. And very clarifying, thanks.

You can't solve the pyramid problem - the early adopters of bitcoin got / will get rich no matter how many new projects are started.
So I wouldn't base any project on "solving" that. It's also a short sighted view, it's more important how the currency will look 10 years from now - by that time those early advantages are bygones.


My jibe was directed at your previous statement , which might lose you the support of some people (granted it might get you the support of others - but I'd prefer something neutral that everyone can use.


Exactly. I'm not that keen on the proposal in this thread, but I'm very keen on Encoin, which I think has great potential, if we're able to pull it off.
Maybe revive the encoin thread, or make a proposal v5 thread? I'd like to hear more about your new thoughts on the SupplyNet for example.

Ouch. I could sit here and write a book on all the thought processes I've had going into this project. I came up with the idea in like june last year, and since then it (encoin) went through 4 major, completely changed revisions. The reason why I wanted to come up with a stable value currency was because I ran the numbers and figured that the first 1 million or so bitcoins cost about 0.00016 US cents each. At their peak around that time, bitcoins were costing around $3.60 each without hardware costs included. About a 3 million % return on the original BTC at today's prices, and something like 20 million % at $30. The people paying $3.60 per coin in cost to produce were continuously enriching the earlier adopters. Pyramid, whatever.

I wanted to have all the great features of bitcoin like anonymity, unregulated, easy to send payments, etc. without all this crap. I also wanted to fix the scalability issues and waste of energy issues and the 51% attack problem.


It's understandable that you have a bitcoin-biased view of a cryptocurrency's value. Yes, the currency I design will not inherently be worth what it cost to produce plus a profit margin. People must demand it. Since it has all of the same benefits as bitcoin, it is possible that it could achieve the same level of demand. The whole point of this exercise though is that if people are not selling (as in the case of BTC's run up to $30) or demand has jumped, it becomes more and more profitable to mine for new coins. This in turn brings the price back in line near the cost to produce.

The whole free coin aspect which you seemingly take issue with is not some Marxist plot to redistribute wealth. If the demand of the economy requires, for example, 1,000 coins to bring the trading value of the currency back in line with its cost to produce, then either 1,000 coins can be mined at a cost of $X electricity, or 500 coins can be mined, 500 given away at a cost of $X/2 electricity. The people using the network win rather than the electric company. But the point is we do know that coins are demanded because people are using electricity to make them. I don't think I posted this in the encoin thread although I had hinted that I was working on ideas with this, but I think the best way to give those free coins away would be to people who send transactions (and hey maybe people will spend instead of hoard!). Random lucky draw gets 10 or 20 coins for making a transaction, etc.


The value is always determined on the open market. The system design doesn't query people and ask them how much it cost to create coins or how much profit they'd like to make. The proposal in THIS thread has to manipulate difficulty and supply because of the problems inherent to the bitcoin block chain design. With the encoin proposal, this is not a concern. This thread was more of just musings to make it work on the bitcoin codebase because no one had stepped up to help with encoin, and I knew I couldn't do encoin on my own.

The point is, the currency should not be some speculative, profit-driven machine. If the price is rising, demand is up, so new coins can be profitably made. It is that simple. Keep money creation constricted like bitcoin or gold-backed currencies, and you will have recessions galore. The wealth will get more and more concentrated to the top. Nowadays it's done in a different way with excessive inflation, but there is no downside (cost) for government to create debt-based money as they can just further inflate the currency down the road. The banks and the government win, the people lose.

edit: I later realized you may have been referring to Marx's labor theory of value because I use cost to produce as a baseline for the sell price of the currency. I've already been in a 5 page discussion on this topic so I don't really want to rehash it again. Simply, Marx did not come up with the LTV and it was the accepted economic theory for a long time prior to Marx's version.

Please note this summation of the prior theory from wikipedia:

The Smith theory of value was very similar to the later utility theories in that Smith proclaimed that a commodity was worth whatever labor it would command in others (value in trade) or whatever labor it would "save" the self (value in use), or both. But this "value" is subject to supply and demand at a particular time.


This is really not that different from the modern economic theory of marginal utility, except one has more neat expressions to go with it. The encoin design allows supply to meet demand in a very efficient way, nothing more, nothing less. It is to avoid the commodity-like currency (or currency-like commodity more aptly) problems of gold:



and bitcoin:

It does sound like a political agenda, and a marxist one at that. Which is a fine viewpoint to have, but politics should be solved with politics in my opinion.
Although a stable currency is a valuable goal, I see a lot of confusion between the value of an xCoin vs the cost of producing an xCoin.
For example, bitcoins are valuable to me personally because I live abroad, and it's a fast and cheap way to transfer money to/from friends and family back home, for stuff like birthday presents etc. The value I place on bitcoins has very little to do with how much electricity it cost to produce it.
For a multinational corporation or organisation, savings on money transfers could be huge, so they might value it a lot. Again, regardless of the production costs.
I see you want to adjust production costs by manipulating difficulty, and at the same time value by manipulating supply.... but how will a market react to a commodity that is designed to thwart market forces?
I would prefer to see supply meet needs for liquidity instead, and let ppl determine the value on an open market. And for environmental reasons try to minimize electicity waste to the minimum needed for securing the network.

Well my goal is a stable value currency, either in the money itself or via the changing account balances. If I were to add the ledger and just award 50 coins per block to infinity, this is not a stable value, it will still almost always be deflationary (in the price level sense) unless it loses faith, and you still get the problems of people with lots of wealth being able to manipulate the currency--IMO the primary problem of currencies today, one that bitcoin does not solve, it only takes the government part of the equation out of it. The wealthy will still have out-sized power over the currency by being able to withhold it and cause deflation and buy up cheap goods before the value resettles. On top of that, the competition for new coins will always be fierce causing people to spend excessive amounts of money on electricity (wasted value) or GPUs/FPGAs (wasted value). I want to make the people rich, not the Edisons and the ATIs. Perhaps you may call this a "political agenda" but the agenda is taking the power out of the hands of the wealthy/government and giving it to the people, and to do this the currency needs to be unmanipulatable.

I didn't set out to fix one problem with bitcoin and currencies in general, I set out to solve them all.
The tax and transaction fees are simply incentives to keep mining going as it needs to happen to secure the network. Again, if there is enough currency in circulation to fit the economy's needs, then no one needs to mine. Likely people will step in and fill the role out of necessity, but then the currency will devalue over time.
The coin bonus is there to create free money (where it does not need to be paid for to the electric company) when there is demand for more money. Why force the electricity usage when it isn't necessary? Also, it protects against the CPU->GPU type issues that will undoubtedly happen again in the future. If the block award is set, then this isn't as big of an issue, but then you've already given up on a stable value anyway.

And note that the values in the OP were just early musings. Transaction fees and the flat tax can be lower and there are other things I have thought of to go with it.

However, if we did do something together I'm leaning back towards the encoin design again. I've come up with a minor epiphany that solidifies my idea for the TradeNet side of things with a balance between anonymity and security. I am still working on perfecting the SupplyNet side, but I don't know that there is a perfect solution there to keep the value of 1 ENC always the same or thereabouts. But I'm willing to settle on a bit of a hybrid between this thread and encoin that I think will work well. All in all, I think the effort is absolutely worth attempting.

Ah, clever.  That should work.
I'd say that alone (ledger / balance sheet, instead of a huge blockchain of transactions), together with removing the upper bound of coins created, are good enough improvements upon bitcoin to be worth trying out.
I'd prefer no tax, no coin bonus and no forced transaction fees though. A currency should be a currency, neutral, and not a vessel of political agenda.

The balance would be kept separately and available on request. There's no need to repeat the same data over and over again in each block. The hash can't be faked, so the client knows if the balance block is correct or not. Instead of needing to download the entire history of all transactions, a new client will just download the balance sheet and block headers.

I see, you just calculate your new balance next time you make a transaction. That should work.
Regarding ledger / balance sheets - I saw you propose this for encoin. But surely the block would need to contain the actual balance, not just the hash of it?

There doesn't need to be any extra transactions. Instead, when the wallet is used in a transaction, the balance will be brought up to date. This is slightly complicated as a cache of coins created in each block and how many coins existed in each block will have to be maintained, but after that it is just math operations. However, I'd rather use a balance sheet or ledger of accounts or whatever you want to call it anyway. Each block has a hash of the balance of all accounts, and transactions can simply refer to the balance sheet instead of previous transactions. Slight anonymity boost, no need to download a gigantic block chain (but headers would still be easy), and so on.

As far as the effect of inflation, look at it this way: instead of people mining for 2x the coins, 1x the coins are mined and 1x are given away. Less electricity and energy need to be put into the system to reach price equilibrium. No more coins will be created than necessary. Additionally, this does keep account value up in the event that coins become much easier to mine because of a leap in hashing power. The trading value of the coins go down, but everyone has more of them. If there were no protection for this, the situation is as I described in the third post between 1mil/4mil coins could happen. The goal is to keep a steady account value, not balance.

This would be technically difficult to do. If there are a million users, each time a block is created there needs to be a million extra transactions in the block.

The economic effect of this inflation I wouldn't know how to predict.

Because that favors miners over existing balances, at least in the short term. If transaction fees are awarded to miners, it makes it profitable for more people to mine, thus increasing the difficulty, thus increasing the coin award. Destroyed fees increase the value of existing coins (since there are less of them with the same amount of demand), thus encouraging continued mining to replace them. In the end, it probably makes little difference but the value of individual coins will vary more wildly I would think, without any real advantages that I can think of. I think keeping the energy impact on coin creation lower is better, especially if heuristic block selection works. Miners will still be encouraged to include as many transactions as possible because if a "better" block comes along around the same time, they will lose.

I still don't understand how destroyed fees encourage anything.  Do you mean that they're regular fees given to the miners?




Why not just increase the rewards?

That's the point, to keep the value stable, not allow for manipulation. The other side is assuaged by the 1% flat tax.


It doesn't have to be perfect. If it is underperforming the curve, the lucrativity of mining will increase because the value goes up. If it underperforms by too much and isn't profitable, then eventually the 1% tax will be enacted that increases the value of mining. If it overperforms the curve, by the time enough coins are mined to have a significant impact, all existing coins have earned interest enough to account for it.



To encourage continued mining in a network that has sufficient money. The fixed minimums and maximums are to prevent spam on the one end and to prevent costly transactions on the other when a large transaction costs no more to the network than a small one. But these are definitely up for debate. It is possible these measures aren't necessary with the flat tax, although some minimum I think would be necessary to prevent spam.



Trying to accomplish keeping the value of creating new coins stable with existing coins. If creating new coins is not cost-effective, no one will do it and the network will fail.



Take the example of CPU->GPU mining. In a non-limited currency, this would quickly invalidate all the work of anyone prior to the switch. That is unacceptable to maintain a stable value of an account. Say in month 1, 1 million coins were created with CPU mining. In month two, 4 million coins were created with GPU mining for the same relative cost. The first million are now worth about 75% less unless you triple the amount of coins in the account.



It may not be.



That is entirely possible. It is just a simplified way of marking the incentive for early adoption.



Noted, this is obviously an area that will require some thought. Bitcoin's philosophy has been to always allow the longest chain to win, but this is not necessarily the best way to do things. Most of this is hinged on "what happens if the network splits" but in reality I think that this case is far less likely than a 51% attack. In either situation, there will still be transactions that were approved and removed, so nothing has been lost by using heuristics rather than longest chain wins. (and transaction fees would make an attack of this type even more costly than normal.)

This will cause high inflation when the value rises.  That could work well, although I'd be concerned about what happens on the other side.




That's a pretty big assumption.  Hashes per joule has been fairly stable except during the transition from CPU to GPU.  There is some improvement, but overall it will probably underperform that curve for a while, then leap ahead with the next technology jump.





How do destroyed fees promote security?  Why do you need the fixed minimum and maximum?




What are you trying to accomplish?  A 1% tax on all wealth just causes all coins to increase in value by about 1%.  The QUANTITY held in a wallet will change, but the VALUE in the wallet does not.




And in reverse: You haven't increased the value of anyone's wallet, just the quantity.




In my opinion the cost of merged mining isn't worth the benefit.




Why not merge this with your "Koomey's Law" curve?  Just make the initial curve extra-steep.




I like the priority on transactions.  I suggest making the priority of a new block (sum of chain's (difficulty * total value of transactions)) or (sum of chain's (difficulty * bitcoin days destroyed)).  This will prevent 51% pools from DOSing the network.

Resolved: Design a cryptocurrency based off of Bitcoin where the purchasing power in any given account remains relatively unchanged over time

Difficulty

• The number of coins awarded per block is based on the difficulty (for example: difficulty / 4)
• The number of coins awarded based on the difficulty does change as a factor of time (on a per block basis) to primarily take into account "Koomey's Law" where the number of computations per joule of energy dissipated has doubled approximately every 18 months.
• The difficulty typically does not go down, only remains the same or rises.

Taxes, Fees, and Bonuses

• Transaction fee: to put pressure on network security and creation of new coins, all transactions will carry a mandatory transaction fee of 0.25%, with a minimum of 0.05 coins and a maximum of 2.0 coins. These fees are destroyed.
• Flat tax: If over a significant number of blocks, the block creation rate averages over 20 minutes (vs. the 10 minute goal), a 1% flat tax will be applied to all existing coins and difficulty will be reduced by 25%.
• Coin bonus: For each new block of coins, every existing coin will gain a small amount of interest so that the total of all interest earned will be equal to the total amount of new coins created.

Bootstrap

• Merged mining: Merged mining with Bitcoin to promote initial interest. A clean break several hundred thousand blocks down the road.
• Coin award multiplier: 2x, 1.75, 1.5, and 1.25 multipliers for the first X blocks to encourage early adoption.

Blockchain Defense

• Heuristics: All clients agree that competing blocks will have priority weight based on number of transactions, average age of coins in transactions, and other factors.
• User interaction: Users will be notified of odd behavior occurring or competing block chains and will be encouraged to investigate.

The coin bonus is used as defense against attacks on SHA-256 or something similar to the CPU->GPU change in the block chain to cause minimal long-term disruption. Since the distribution of coins is not limited, coins earned in the "harder days" will not lose value (over time) as those account balances will increase.

The flat tax is used as defense against a shrinking (or non-growing) network to maintain transaction confirmation stability. Additionally, those with wealth in coins will be pressured to partake in the security of the network to avoid the tax if necessary.

Heuristics' use may be complicated, may be easier than it seems, but it will help prevent or eliminate 51%-type attacks. Since coins are not limited and coin awards are doubled (via coin bonus) it is expected that less demand for new coins will exist under this system, therefore it must have more than no protection at all against 51% attacks.

Other notes: for the first several thousand blocks, an initial account will make a transaction every so often that must be included for a block to be valid (to prevent "small penis big operator" merged mining take over). P2Pool will be included with the software to discourage traditional pools. Flat tax and coin bonus will probably not apply until after a large number of blocks have been mined to promote initial stability. A separate cache will be maintained on clients to keep track of flat taxes and coin bonuses so that it is easy to compute and verify an old amount to a new amount in a transfer (this is when coin bonuses and flat taxes will actually be realized, but the client will adjust the amount as necessary). Transaction fees could potentially be reduced or eliminated (especially for microtransactions) if the receiver is a miner in a block. Special transaction type to transfer coins to the next major version (potentially something like Encoin).

While the stability of value in one actual coin will vary over time, an account balance should maintain similar purchasing power and in fact a ratio of inflation/deflation would be easy to calculate over time and merchants may use this to automatically adjust pricing; employees/employers could potentially use it for consistent salary negotiation.


Thoughts, comments?