Right, they took the real money out of the system, and now all that's left is some worthless tulips/stamps/digital tokens of worthlessness/etc.


How sound will it be when the hash rate drops to 1/10th of what it used to be?

Aren't you kind of giving the exchanges the power you seemed to worry about? Or at least assumed was inevitable. It seems as if the door is wide open under this system. What if a large conglomerate of businesses decide they want to change the protocol? Change to their system or your coins are worthless. Taking power away from the people and putting it right back into the hands of corporations.

And why work toward hacking this on to bitcoin when bitcoin can't scale? Without deflation, this system will scale even less well. Since the value of coins don't go up, people will continually use the same amount of coins, which means each transaction may have hundreds or thousands of threads to follow. So GEM visa will have to be created and now you have corporations controlling what does and does not get approved, if they so choose.

I'm thinking this quote should follow every post immanuel makes

Aren't there like 16 other threads where you could post this?

a taste of the next proposal:

TD-1) Determining Speed of Currency Creation; Adjusting for Processing Power
Coin-Hours are used to determine how quickly currency has been created.

Coin-Hours = Number of Peers Per Mint Block * Time Between Mint Blocks

If there are 35 users paid in a given Mint Block and 3 hours have passed since the last Mint Block from that CoreNet, the Coin-Hours for this block is 35 * 3 or 105 Coin-Hours.

Coin-Hours are added to a value stored in the Consensus Block. The number of Coin-Hours and the number of coins created is kept track for a period of 30 coin-creation CBs. These 30 CBs are called a Coin Creation Period (CCP). 10 of these such periods are maintained in each CB with the newest overwriting the oldest. For a CB to count as a coin-creation CB, the following formula is used:

Minimum Coins to Count for CCP = # of Coins Minted in Last 10 CCPs / 300 * 0.50
OR
Minimum Coins to Count for CCP = 1,000

Whichever is higher. By multiplying by half in the first formula, the amount of coins required to increase the difficulty can go down over time so that it is not trivially easy to avoid increasing the difficulty by staying just under the coins necessary to do so. However, even during CBs when not enough coins are produced, the Coin-Hours and coins are still recorded so that a maximum number of coins can be created before changing the difficulty.

Maximum Coins Before Difficulty Change = # of Coins Minted in Last 10 CCPs / 10 * 1.5

To determine the increase in processing power, it is simply a matter dividing the number of coins by the Coin-Hours to get the number of hours it took for an average computer to make one coin, then compare it to the previous CCP. The increase in this power is determined with a simple formula:

Processing Power Increase = (Previous CCP Coin-Hours / Previous CCP Coins) / (Current CCP Coin-Hours / Current CCP Coins)

If this number is less than one, one is used so that the difficulty can never go down. To avoid intentional, large increases in difficulty as an attack on the network, a weighted system based on the last 9 difficulty increases is used as follows (with example increases):

Oldest (10th-9th) CCP weight: 0.05 x 1.06 increase (0.053)
9th-8th CCP weight: 0.05 x 1.09 (0.0545)
8th-7th: 0.05 x 1.04 (0.052)
7th-6th: 0.05 x 1.00 (0.05)
6th-5th: 0.10 x 1.08 (0.108)
5th-4th: 0.10 x 1.07 (0.107)
4th-3rd: 0.10 x 1.03 (0.103)
3rd-2nd: 0.20 x 1.07 (0.214)
2nd-current: 0.30 x 1.25 (0.375)

A very large increase was used as the last amount to show what might happen if someone were to attack the difficulty of the Network: a 1.1165 final number, or an 11.7% increase in difficulty. Unless this attack is sustained, the difficulty will not increase for some time afterwards.



The difficulty attack will probably go in a separate section, but there it is for an example. At bitcoin's estimated $10,000 cost to run per day, this attack would require $300,000 of electricity just to equal the network, so whatever increase it was trying to achieve would be cut in half unless it does more than half the network, then it would only be weighted 30% (perhaps less, first run of numbers). Much of the coins would be distributed to other people in the same Net, so most of that electricity is lost too.

Why, pray tell, would I care if your totally vague statement is right or wrong? I could point out various posts on this board that claimed bitcoins would be worth $100 by now or the end of the year. I don't care if you think a ponzi scheme is an investment, but I will be doing my best to educate newcomers that it is not--even though the price of bitcoin has done a pretty good job by itself.

So how does anyone other than satoshi bear that risk for bitcoin? How do people joining a website or mining some coins for a couple dollars in electricity bear a risk? ohnoos I wasted $3 to mine 50k coins...


And your subjective belief is completely unclouded by the fact that for that reward to increase, bitcoin needs more suckers who need to be convinced that they too will earn this wonderful return.

<ripper furiously responds with a link to the bitcoin wiki that adamantly states that bitcoin is not a pyramid/ponzi scheme as proof that it's not!>

There was a risk to joining facebook in 2004? There was a risk to joining bitcoin in 2009?

I'm proposing that it is avoidable in the newest version of my proposal.


Not sure what you mean. By putting their money on the line, I mean they must maintain a certain balance in their account to get transaction fees refunded. It is basically security against approving bad spends. The penalty for attempting to subvert the network is losing whatever balance you have.


It will be less decentralized in that each node in the network is aware of what other nodes are approving transactions. It will be less anonymous in that nodes that are approving transactions will have to sign those transactions. But these trade-offs mean that the network does not rely on hashing power and it will ultimately be far more scalable and far more adjustable than bitcoin. The endgame for bitcoin is that bank-like entities will be running the show as regular nodes have no hope of actually being a part of the network (8Gb/s bandwidth, a terabyte of space every few days). It puts the power back into the hands of the elite.

Not to mention that any direct competitor to bitcoin that is forked from bitcoin is going to have the major issue of 51% attacks on the network. See fairbrix, SC2, namecoin, etc.

Well, I'm still speculating here. I don't have anything set in stone.
We can't assume koomey's law will hold, so we have to let the market figure it out.
To account for Moore's Law, I believe the difficulty should adjust every 25k coins (up for debate) based on coin-hours. Coin-hours = # of users in a block * # hours between blocks. Though I haven't released proposal 4.0 obviously, this can be determined correctly based on the way my new design for freenets/corenets will work. People won't be able to just join or leave willy-nilly.

Then, to account for Koomey's Law, the award will adjust every 250k coins (up for debate). How the award will adjust I'm not sure yet. It can't be from 6->3 because that will just bring in the reign of FPGAs and kill GPUs. So I'm thinking from 6->4.5->3->6 or 6->5->4->3->6 (probably this). If it goes too fast, we are likely to give low wattage high efficiency miners an even bigger advantage by making GPUs totally unprofitable way before their time--thus killing the supply and causing a permanent increase in the CTP before its time. So "on occasion" is sort of referring to the fact that when it happens is not very predictable. "if it lowers over time" is just me speculating on how this scenario would work.

Perhaps # of coins shouldn't be the only factor though, because if ENC explodes in popularity, 250k coins could go pretty fast. Then again, I dunno, because people might just be mining like mad because the CTP is much lower than 1 ENC. So perhaps it has to stay rigid. Then again, perhaps the change in award could be based on how long it took to get there.

All this is assuming that the difficulty never goes down which I think needs to be part of the plan. Difficulty is based on coin-hours, not amount of computing power, so I really don't think it should be an issue unless someone is trying to grief the system or intentionally make coins cost more to produce so their existing hoard is worth more. To do it will cost them a lot of money, but I'm trying to think of the far-future.

Some of it could be applied, sure, but certainly not all of it. Bitcoin is secured by hashing power. This is a joke that needs to be eliminated. I've spent too much time thinking how to get away from how bitcoin works, so no offense, but I'm not spending any effort on trying to figure out how to hack the encoin idea in.

Right, this is why the award would gradually (or perhaps not so gradually) lower on occasion, to weed out the less efficient.
If FPGAs or low-wattage GPUs become prevalent, these devices are making coins at a lower cost to produce than previously possible.
If the award is dramatically lowered to 3 instead of 6 coins, inefficient cards have to lower their output while efficient cards can remain at the same output (assuming they use about half the electricity). This provides a new baseline that I mentioned. Inefficient cards will have to quit, and efficient cards will have to compete against each other instead of siphoning off the inefficient cards. Then, after a period of time, the award and difficulty can double back to 6 so that the process can start over.

Since, in a stable economy, coins may not need to be produced all that much, I came up with the idea of doing this by the number of coins produced rather than time. Perhaps the award could adjust every 250k coins or something.

But this provides a potential attack. Supercomputers and/or pools could get in and intentionally raise the difficulty if not too many people are making coins. This could be alleviated by having a minimum amount of time before producing a block and having a maximum difficulty increase per chunk of coins produced (this figure will be hard to determine though because we have both Moore's and Koomey's Laws we need to account for). It's a band-aid though. Intentionally raising the difficulty means the coins would start trading for more than their CTP, and this gives an opportunity for hoarders to make a lot of money. So it might encourage them to raise this difficulty. That's why I'm thinking a bigger chunk like 250k coins, it would cost a lot of money to do it. But by making it so high, this means that coins may be produced for a long time at less than the CTP, but at least there is a cap on it.


Not trying to detect anything; trying to foster competition among the more efficient so that it can happen automatically.


I think if you encourage hoarders to hoard during inflation (via interest), this works itself out. As I said, new buyers won't care if prices are inflated because they paid less for their coins. If 1 ENC costs 0.5 ENC from a year ago, they have no issue with paying 2 ENC for a loaf of bread. They shouldn't be penalized for buying in during inflation. It sounds like you are trying to prevent an inflationary spiral which I think is awfully unlikely with the monetary policy. No one is going to make coins during inflation, so the economy should eventually recover, so this will encourage people to buy coins and remove them from circulation.


But they are minimized. Merchants will be providing a service to the network by confirming transactions and so on, and in doing so they are receiving a large chunk of these transaction fees back. Hoarders receive a small interest payment and this payment could be used to sell on the market. May as well sell now before the coins get back to CTP.


I think it is more fair to go to the hoarders as they are people who already believe in the system. Distributing these fees to miners only encourages more people to mine when the price is high, sell, and be done with it.


I think it is just too hard to predict what will happen by doing this.

If the award lowers over time, people have to adjust their output or risk making coins unprofitable. This is kind of annoying because people who want to go full blast still can (supercomputer/pool problem). Certainly the software can automatically adjust when the award changes, but each time every person will have to look at their mhash compared to their watts compared to the market price to see if what they are doing is profitable (going full blast is likely going to subsidize others as described before). But I think this is the only way to do it and really let the market decide. 1 ENC tends to 1 ENC.

If prices of electricity increase faster than fiat inflation, people will have to wait until more efficient hardware can catch up. In the mean time, hoarders will have the opportunity to smooth out deflation by selling. They will make a profit, but not by their "early" investing, just by an unpredictable consequence of the world.

If fiat inflation increases faster than the price of electricity, there will be more competition to make coins and this will cause the difficulty to go up and the market price to go up, as it should.

If the price of electricity falls or hardware becomes more efficient faster than expected, the lowering of the award causes renewed competition that will increase the difficulty faster than normal (it's profitable for me to mint at full blast, so I will).

The *only* problem I see is, like I said, someone with the time and the means to intentionally try to drive up prices. If there are always people minting it is a lot less easy to do, but during times of inflation it would be even easier to accomplish. I'm not sure how to safely account for this yet.

So you didn't get in early enough on bitcoin? what is the point of this? At least some of the other currencies tried to do something different.


faster transactions and scrypt is innovative?

lemme hit you with a big ol book called a dictionary:

innovative [ˈɪnəˌveɪtɪv]
adj
 using or showing new methods, ideas, etc.

Yes, you're right. However, it is essentially the same thing when the money supply is fixed and the population increases.


"This proves that the structure of money has an impact in our way to value things over time." - I very much agree with this point. I don't think it proves anything, but I agree with it. It would be interesting to see how it plays out, but not at a bitcoin-esque level.


How can a bitcoin bank not have fractional reserve? What bitcoin and any other digital currency that isn't controlled by the government solves is things like the FDIC and 1.5 trillion dollar bail outs. Banks do have to compete, and they have to take the loss themselves on any bad loans. Bitcoin has no way of stopping fractional reserve, just like gold-storage banks originally did it (because of the profit from demurrage, lol).


You can't predict price inflation from monetary inflation. If the monetary inflation rate is an ever reducing amount compared to the total amount of money, it is very unlikely that prices would inflate anywhere near what you suggest. The currency would eventually be price deflationary assuming a steady economic/population growth. This is the same thing that will/has happened with bitcoin/freicoin except it will hardly benefit early adopters since the money supply is not so heavily weighted towards the beginning of the distribution.


Good luck. IIRC you guys argued about having a reduced amount of demurrage for X years or whatever to start. Don't let the greed go to your heads!
And BTW, I think constants are a terrible idea.


It isn't possible to predict exactly how this will play out, but when you take that money, you are owing the fee plus interest in demurrage. In inflation, it is the inflation plus interest. As far as actual currency goes, there won't be a difference. But where I quoted you above I believe a difference will be made. I think combining demurrage with a fixed supply of money is not the solution though. It's putting a band-aid on bitcoin.

Merchants get whatever fees that they can back based on how much money they are willing to put on the line. Anyone else just loses their fees. These lost fees get redistributed to people hoarding currency. So in down times, instead of selling, people can hoard and earn interest on those who do sell or trade at inflated prices. No one who buys cheap is going to care that they're getting inflated goods, only people who saved would care. So instead of freaking out, savers can just sit back, earn interest, and wait for the economy to get back.

And as an addendum to the other stuff, instead of basing the ENC reduction on time, we could do it based on the number of coins produced. This means if we get to a period where things are too cheap, it will last a shorter amount of time while ENC gets to a new baseline. HOLY SHIT.

I think I've fricken solved it.

Keeping the "value" stable over significant periods of time.


Right I'm not too worried about the details of how it changes, just how to adjust based on missing the expected curve. But I didn't think about doubling the difficulty and doubling the ENC, it's so simple that it passed me by apparently. We get it down to 2 ENC per block so that there is a baseline of the current, most efficient machines in the 50W range. Then we have a new baseline for bringing it back up again. If difficulty was the same at 75W as it is at 50W, then we know people have been cheating (or some other world circumstances have changed--or hardware has a new lower range on wattage output). Then we 3x the difficulty and 3x the award back to 6 and do it again. Damn, this might just work. The implications this might have on the network could be a bit tough to figure out though. More thought required.

But essentially, if common hardware uses only 50W in the future, when tripling it back up to 6, earning blocks will take 3x as long. It doesn't matter if it's 50W--it will be whatever equivalent is most efficiently possible to keep a reasonable profit and stable prices.


same difference! A baker doesn't care about the customer's cost to acquire a gold coin either, he cares about its trade value.


What do you think about my post on giving interest?

A deflationary, demurrage-based concept is going to suffer from a whole lot of the same ills as bitcoin, imo. A wild, swingy ride all depending on who farted which way that day.


This is simply not true. If inflation was kept in strict agreement with population and GDP increases, the prices of goods would remain level. Of course, fractional reserve lending basically means there is never any hope of that happening.


Since when are inflation and demurrage exclusive concepts? You can have (monetary) inflation with demurrage, and you can have (monetary) deflation with demurrage. Your proposal for demurrage is merely a hackjob for fixing some of the problems with a deflationary currency. Invest instead of hoard, let other people pay the price for your money. Woot, so earth-shattering. I don't see how this fixes fractional reserve.


I assume we're talking about a bitcoin-like currency. Using your example starting with 1 million coins in year 1, we have 286 million in year 50, in which 11 new million coins get released. 14 million would have to be released for a 5% inflation of the monetary supply. The inflation rate is ever-reducing instead of the amount of money ever-reducing. Would actually give the world time to adapt to the currency rather than forcing millions on those innocent early adopters.


Inflation is not exclusive of demurrage. So these equations are all fine and dandy, but they don't mean anything. You are arguing for deflation, not demurrage. The cost of borrowing will work out to be exactly the same.

The next version of the proposal already has bits of a first section that goes into this market effect because nobody really got it. I understand that it wasn't clear now. So yes, the first section is basically all just economy and why I think the system will work.


We've established this. Now HELP ME TRY TO FIX IT! If we establish a baseline of 150W instead of 200W and a gradual decrease in the award from 6 ENC to 2 ENC, the baseline goes from 150W to 50W. If the difficulty doesn't increase on the order of koomey's law, then the people are adjusting lower because technology is not keeping up. But once we hit whatever the final award is, 1 or 2 ENC, we can no longer adjust lower, only higher. Could this somehow be based on a year-to-year system? I wonder. -- Though in trying to keep coin production down in a stable economy, there might be very few miners at times from which to draw conclusions. This could lead to mistakes.


I really don't think it does. I even mentioned this concept in one of the Q&As (on ukraine electricity). If I want to trade 1 ENC for a loaf of bread to the baker, the baker has some idea of what it took to make that ENC, just like he has a good idea of what it took to make his bread. Bakers in ukraine charge more because electricity is cheaper. Bakers in western Europe charge less because electricity is more expensive. Obviously there is little way of knowing if the coin was made in france or ukraine, but there are costs involved in that transport such as currency conversions.


I don't know how it will work out either. We'll have to wait and see.

Your example ignores a growing economy and a growing population. Not to mention that 5% inflation on the amount of coins awarded each year is not a 5% inflation of the money supply. Not even close.
Banks are not doled out fresh money by the fed in *coins. People (pools) spread the coins out evenly from the start. So there is little advantage to being "first."

What was the reason for the 7.7 million number? Did 7.8 seem like it would be too much? 7.6 not enough?