Topic: Properties a crypto-currency requires in order to be self-stabilizing (Read 4015 times)

You are talking about price not value.

Value is purely subjective and arises when someone want something.

I believe in the near future, virtual economy will take over the physical economy and be much bigger in size, and the virtual product at that time will be good enough to encourage people spend more money than they did today on housing

So if you are still looking at the exchange price of the BTC, then you are on the wrong track, just like in a RPG game world you always talk about how many gold a certain item require, not $. Just ask a child today, is $ more valuable or his wow gold more valuable? I'm sure sometimes he will be willing to pay more $ for the gold, the problem now is just he could not get $ easily

I like gesell's ideas. Austrian economics proponents, however, don't instill confidence in me. http://www.cato.org/pubs/journal/cj19n2/cj19n2-4.pdf A very interesting review of Mises' thoughts on gold and how he often contradicts himself. http://mises.org/daily/1241 President of mises.org using price deflation due to efficiency gains as being an argument for a fixed supply, when the two have nothing to do with each other.

I think you get all the benefits of austrian economics by having it not backed by governments. A fixed supply is not required.


This is a weak argument, IMO. A $100,000 mortgage costs around $200,000 or more in interest on a 30 year loan. Anyone who can't afford the mortgage can't afford to pay anywhere near that amount for the points. Plus, in a system that is not backed by the government, it is up to the people to protect their own assets. If someone defaults on a loan, the investors in the bank are taking the hit.

And freicoin will still have interest. We know this. If you have to pay back the same amount, you're still paying interest. Unless you can get rid of the mindset that people want to buy what they can't afford.


I was referring specifically to your comment about how the system "determines price," when the system does no such thing.



Complicated, yes, infeasible, no. Basing the security of the protocol on hashing power is going to be the eventual downfall of bitcoin. A large part of the complication in my design is getting away from this. But all of the complication is a lot of simple formulas that add up to something more complex.


That is not the hardest part. Both Red's and my proposal require small but significant transaction fees. These fees can be used to guide all kinds of economic activity. I use giving merchants partial refunds as incentive to secure the network via reputation. Currently I have proposed that unrefunded fees are distributed as interest to those who hold ENC. They could just as easily be destroyed (and this was what I originally proposed). I think the effect is identical though. There will not be any instant return to whatever price people think is accurate, but it will happen over time. In the mean time, the lower price will encourage buying the currency, and those who bought in at a lower price do not feel any effects of inflation and are free to spend the currency. This is an oscillation, not a constant lowering of value as is present in traditional fiat. To keep lowering the value, faith has to be significantly lost and this is a problem for any cryptocurrency. At least with encoin, coins are not given out every 10 minutes regardless if anyone is demanding them.


Except that in June, how many bitcoins were for sale? It was restricted by the early adopters, there is simply no question. As 300 BTC per hour is all you're going to get, the market cap was actually around 100k BTC * $30 or whatever it was then. Probably less than that. If you look at the heat maps that farroh provided in another thread, there was virtually no BTC activity on the run up to $30. No one was selling compared to when the price started falling. If early adopters were to do this with Encoin, new users would be able to fill the need for coins in a matter of days. This means the incentive for those with coins to sell is very high when demand is high, not create a deflationary spiral--because you simply can't.


No, I mean the millions of coins that were mined for 0.001 cents or less. 1.6 million coins were mined at a difficulty of 1. 3-5 CPUs worth. Then the run up got people spending millions of dollars on hardware even causing GPU shortages. Then the sell-off started. This scenario simply won't apply to a stable value currency.


And I ask you: SO WHAT? I am not guaranteeing a value, only a stable cost to produce. If the currency loses all faith, of course its value will drop to nothing. This isn't mind-blowing news and it isn't an argument against a stable value currency. It's an argument for the barter system. Get rid of government manipulation, get rid of banker manipulation, get rid of early adopter manipulation, and you might just have a currency that has a lot of faith.


No, encoin holders are not losing money because there is no net sell off. For people to pay in encoins, they must hold encoins. Unless people decide to buy $50M of ENC only to use it all once, then yes it would be a problem. A realistic scenario would be one where there is a constant flow of money back and forth as there is with any normal currency. A realistic scenario for BTC would be that early adopters would have a mass sell-off and cause tens of thousands of others to lose value while only they gain. At least with ENC those who sold the coins originally were those who actually worked for them. 10-20% of the currency is not centered around when early adopters decide to sell.


I don't believe we're any better now. The sell-off yesterday was initiated by one person or group, and that group just happened to be around to buy up plenty of coins at around a dollar less per coin. In my estimation, they probably profited 20-40k USD and have about the same number of BTC. A $250k ask was up there for a matter of seconds (WHY it was up there I don't know, but I'm guessing it was a misclick). This person has a large sum in both currencies and is manipulating the economy. I highly doubt anyone who is wealthy enough to have $250k is dumb enough to invest it in BTC. This person has siphoned money off of the economy every step of the way and has made a gigantic profit. This is only possible because of bitcoin's batshit retarded coin distribution. We're talking about 50k BTC that has been used over and over and over again. And there is 7.5+M out there, with 5M of that that was mined for probably less than 20-40 cents on average. Everyone who continues to hold BTC is just going to be a loser. Unless bitcoin starts getting rapidly adopted, the only way for it to go is down. And if it does get rapidly adopted, you are just giving those with lots of BTC more and more power to do this same shit all over again until finally people wise up and GTFO.



If you're going to misapply economic theory, at least get the basics right. Prices do not drive production, demand drives production. I am not using labor theory of value because I am adding up costs and assuming this will apply to the final price. This is a gross simplification, just as is the original labor theory of value. You must be assuming that I have no concept of supply and demand or utility. I'm aware of these factors, and I'm aware that I have little control over them. All I can do is propose my best effort to making a usable currency and hope that other people feel that it is too. If you don't want to take the time to understand what I've proposed, then please don't waste my time and yours.

Can you come up with any actual arguments against my specific idea rather than generalizations that apply to all currencies?

I appreciate the effort. I really think that it is a great improvement over national fiats, but I think that same thing about bitcoin.
There's no academy economist that advocates for both demurrage and an unelastic supply of money. A synthesis between Gesell and Austrian economics would be needed. That's what I've been trying to accomplish here in the forums.
I know and like JAK, but it's not attacking the root of the problem in my opinion.
It's just replacing regular interest with "loaning points" that give you the right to loan free of interest later.
But if participants were allowed to sell those points, you would have just another form of interest.


I've just wrote that extreme example to ensure you take into account all the possible cases.
And yes it applies to Encoin, Gem or any other "stable coin".


Well, the constants in your explanation (6, 4.5) don't tell me much, but more or less I get the basic concept.
Some questions:

How the size of the groups is decided?
Do the groups have different difficulties?
Can someone switch from one group to another?
How is it guarantied that the "production cost" is the same in every group?
How the system determines when the supply must be restricted?
What if the users don't report the accurate price to their benefit?
How does the protocol know how much power are miners expending?

Maybe there are too many questions and I should just read the proposal. No offense, but I just expect a non-feasible complicated protocol that has little to do with bitcoin.
Don't get me wrong. I understand your intentions are good. I tried the same thing myself.


I see, more EC are produced so that it price drops. That should work (is the solution of most stableCoin proposals), but the hardest part is making it the other way around: destroying coins to increase its scarcity and price.


Well, I'll do it without very accurate number but more or less...
June
6 M btc * $30 = $180 M total market capitalization
Today
7.5M btc * $2 = $15 M total market capitalization
I don't know what you mean by market capitalization if not that.


You say that a single sell off affecting the price is less likely the more people using Encoin.
I'm just pointing out that the same happens for bitcoin.
Early adopters coins?
You mean coins mined when the price was under 40 cents ?
If I buy them today at $2, what's "easy" in them?   


Encoins, like bitcoins, are backed by nothing.
You cannot derive the value of something from its costs of production.
Prices are what drive production and not the other way around.
You're using labor theory of value there.
I think you haven't understand Gesell critics to Marx. Austrians (well, almost anybody) also disagree with Marx here.
If the costs of production of an encoin is $1, its value on the market doesn't have to be the same.
Like any other non backed currency, the value of Encoin could drop to zero overnight and there's nothing that can be done about it.


You're assuming that the seller is an attacker.
The sell off can be caused, for example, by a sudden lost of faith in the currency or by the appearance of a competitor.
But let's take a less extreme explanation.
Say bit-pay and BTCinch start accepting Encoins.
Many shops start selling thing for encoins but they want to be paid completely in dollars.
In this context, a $50 M volume through these services represents a sell off of $50 M.
The services are actually making money through fees.
The shops are selling their products at the same price in dollars that they always do.
Only encoin holders are losing money.

Maybe that is part of what has happened to bitcoin since June. But, hey, I think we're better now. Hoarders have lost money (including me, although not much), but screw hoarders.
These services increase the "real" (non speculative) value of btc (by increasing its uses) and make it more stable by increasing the trade volume.
Anyway, this last thing is kind of off-topic.

Say the real market cap is $200M and the price of 1 ENC is stable around $2. It's impossible to say how much you would get buying $50M worth of coins. If only around 10% of the market is selling, you have around 10M ENC for sale, but obviously there are going to be people attempting to sell at $3, $4, and so on in case of a rapid network expansion, so even though 10M ENC is worth $20M, you are paying (roughly) $50M or more. Then what? Sell for $1.50 and lose $35M? Sell for $1 and lose $40M?

There will be no early adopter coins minted for hundredths or thousandths of a cent, so these can't do it. Someone either is going to have to spend $30-$40M mining or spend $50M in real cash value just to do what, drop the price? Unless faith has been lost by other means (cryptography problems or similar), everyone can see it still costs around $2 to mine 1 ENC. This means nobody is going to be mining until the value gets back up to around $2. Everyone selling high makes money off of this guy, everyone buying low only has to hope that the economy will continue to expand until these coins are sold through. Merchants may certainly raise their prices in the mean time, but once the price levels back off at $2, their prices can go back to what they were before. The overall system has not inflated. The "pump and dumper" has lost millions upon millions while those legitimately using the system have earned those millions.

Unlike bitcoin where the hoards of early adopted cheap coins are not obvious to the casual user, someone buying up the entire ask side of the market will be obvious. If the demand for new coins is still exceedingly high, everyone will win by the "free coins" that would be created by over mining. They can trade some of their coins for fiat and keep the original coins they had before, therefore now possessing both fiat and ENC. If the price of ENC then drops, they haven't lost anything. The market mover has lost many, many millions. Who is going to do this? Governments? No matter what, those existing users of the system profit, and the manipulator loses. This is not what is happening in bitcoin because all these manipulators are not people who have put real value in the system, but people who got tens of thousands of coins for a few dollars taking advantage of those who got a few coins here and there for many dollars.

The convincing argument is that whoever does this has to be insane, or has absolutely no care for money. ENC holders will be happy to take that money off of their hands, I'm sure. The people legitimately using ENC win, the manipulator loses.

I think this is a good test, but I think 1 and 2 are quite well addressed by Encoin already.

The problem is number 3. I don't have anything, and haven't seen anything convincing on that. Yet.

Well I've read what was available on Freicoin on your forums at the time (a month or two ago), plus a ton of your posts, plus some 100+ page novel that Gesell wrote, plus tons of other resources on the internet to understand the free interest theory and demurrage. I like that it fixes a big flaw in bitcoin; however, I don't believe it fixes anywhere near all of them, and most of the consensus I find is that it really won't accomplish much over traditional fiat. It is very difficult to find any economic theory on how it would interact with a fixed supply. And there is always the huge psychological issue of losing your money by holding it. I understand the benefits to the world of getting rid of interest, but this can be achieved without demurrage (see JAK bank).

Now, after the many hours I've spent understanding your point of view, perhaps you can see why it frustrates me to see you say it "can't be done" after asking questions that do not apply to what I, Red, or others have proposed.

Bear in mind that mining does not secure the network. Merchant reputation forms the "51%". They gain reputation only by paying transaction fees, and it is harder and harder to gain as more merchants secure the network. Little to no electricity required, but still a real, actual cost.

The finer details are in flux, but the simple summary is this:
1) Coins are competed for by small groups of people using processing power in the same vein as bitcoin (the amount of coins each peer is awarded is based on how well they did, but the total award per block is fixed)
2) There can be any number of these small groups, so there is no limit to the supply
3) As processing power increases, the difficulty increases just as with bitcoin, but on a per group basis
4) The supply is occasionally restricted by lowering the block award. Half of the time 6 coins per block, the other half 4.5 coins. Software will automatically adjust output down to 75% when the award drops to 75%, OR the built-in client calculator can automatically adjust based on a market price that is supplied by the user. What this does is allow people with efficiency advantages (either cheaper electricity or GPUs that use less watts) to adjust their clients up from 75% to be able to grab a larger portion of the restricted supply. This will increase the time to make coins.

There is an example in the wiki.


No value of any kind is programmed in, it is based on competition and market price that the USER supplies. I am still working on the details of block awards and difficulty adjustments, but it would be phenomenal if other people would chime in with ideas. Red seems to have given up due to a lack of interest.


It is the absolute opposite of bitcoin. When the price of BTC rises, more people mine to bring the cost of mining 1 BTC in line with the sell price. With Encoin, more people mine to bring the cost of 1 ENC in line with its production price.


You're going to have to supply more details, because as far as I can tell, bitcoin's market cap is about the same, there are just more coins in circulation. I really hope you don't think that because BTC was $30 USD that that meant that bitcoin's market cap was $200 million, because that was obviously not the case. It was likely less than $2M then, it is less than $2M now. Yet people are still dumping in 10-20k of electricity per day, plus all the hardware and associated costs for mining.


Early adopter coins, on the other hand, will always stay easy. I really don't see what point you are trying to make.


Because it will eventually get back to being near the cost to produce. Speculation is never going to be a significant portion of the economy. Coins will only be created as they are demanded, so the amount of coins will stay small until it is necessary that they grow. The coins are supposed to be backed by electricity + processing cycles + time (ROI). If there is a large sell off that brings the value below the cost to produce, there are other ideas to keep relative stability--interest based on unrefunded tx fees for one, the fact that it still costs more to produce than they are selling for for two (opportunity to buy), and Red had an idea to charge a tax on the person making the tx, but that was based on his proposal where mining was still required to secure the network. Plus something I borrowed from Red where if people are intentionally creating too many coins that "free coins" would be created and distributed to those showing economic activity. The details of this are still being swirled around in my head though.

Sorry, I've read and discussed many proposals on stable coins but I'll wait for a summary to read yours. The fact that it only aims a constant cost of production is a good start to understand the proposal.
I don't think people trying this are morons: I've tried it myself.


Oh, I see. Just like bitcoin.


Compare bitcoin's market cap in July with what it  has today and you'll see that's definitely possible.


Yes, it will become more difficult (like point 2) the bigger the bitcoin economy is.


I'm glad we can agree that it can't be done.
Sorry I didn't know that Encoin wasn't about a stable value coin.
If it's not to have a stable value, why is so important to have a constant cost of production (that you just admitted it's not possible within my extreme case)?

How many times does this same thing have to be asked? The point is to keep a consistent cost to produce. Whether it's based on competition (encoin) or a constant increase in difficulty (gem), there doesn't need to be some "vote" for the price if you actually read the proposals. I mean these are central questions that you think might be addressed, unless you just assume everyone putting time and effort into this stuff is a moron.


Pretty simply, you don't. But when the price has risen well above the cost to produce, more people are encouraged to mine. This will bring the price back down near the cost to produce over time.


You can't. But I think you'd be hard pressed to find someone willing to invest $50M to then be guaranteed to lose whatever amount was sold off before you sold it back. Even so, this is a market cap issue that will become more and more difficult after a long enough time. And it certainly affects bitcoin or freicoin in the exact same way.


Yeah, highly informed and unbiased opinion.

Good way to describe it.
Some people around freicoin see this as an necessary requirement, but after thinking about it for some months I've come to the conclusion that it is impossible.
My extreme test for anyone trying to solve this problem...

1) Your currency has a total market capitalization of $5 Million. How does the system knows the price of the coin in a decentralized and secure fashion?
2) People suddenly spend $50 M buying the currency. How do you prevent the currency from rising?
3) The people has bought X coins in 2. Now they sell the X coins back. How do you prevent the currency from dropping?

I definitely think it just can't be done.

Interesting, you'd effectively be slowing the velocity of money.  I'd hate it if I really needed to spend that money, and it would create some odd change transactions.  Instead of freezing a percent of all coins, you could increase the waiting period before transactions are confirmed.  If value goes down, a coinbase transaction doesn't mature for 240 or more blocks, a normal transaction doesn't get confirmed until 12 blocks have passed.  I wonder how long you could freeze the funds before it had a negative impact.  I know you're thinking of enCoin which has different mechanics, but I think it would apply to either (with some variation).

Yes, it does make more sense, and was proposed by Red in his GEM thread. In Encoin, I have proposed that all transaction fees that are not refunded get spread out evenly as interest among coin holders. It doesn't counter any immediate drop in price, but it encourages holding while others may sell. Also, I am working on ideas that if X% of the total coins gets created in a day that any coins above that % will be doubled and potentially split first among merchants (with higher tx fee refunds for securing the network), then possibly randomly to a number of accounts that made transactions, then if there is still more distributing it as interest again. This "free money" is given to people who believe in and use the system, so it is less of a drain on resources (electricity), and they are going to benefit anyway by selling when the demand is higher than the supply.

There are ways to do this, and it is going to be a lot more appealing as a store of value to "later adopters" who actually paid real money for coins, as well as for merchants who won't have to worry about accepting them and needing to instantly convert it to cash.


Interest, my man.

I really think this is a system that will get more users and merchants on board. And it won't have to be reduced to just being used as a payment processor with just as many fees as CCs.

I just had an idea. If it seems the value of something is declining, why not freeze assets and therefore reducing supply temporarily? Say, x% of all money stored in addresses cannot be used for the next y blocks. If an address recieves money, that money is also portion-frozen immediately to slow down economic activity. These methods should cause temporary reductions in supply and economic ability, but nobody actually loses any money.

Alternatively, a sales tax can be implemented for transactions to cut economic activity also. This sales tax approaches 0% when the currency is moving up in price, and can go up to 100% if the currency was devaluating so fast the economy needed to be halted. This might make more sense than freezing money.

In order to prevent speculators anticipating these events and creating a massive selloff, the economy should reward keeping coins when by some metric the market looks like it is devaluating. This can be implemented as a percentage gratuity when the system determines the devaluation is over, but only if the coins are kept and not used.

None of this actually refutes anything I said; all it is is rhetoric.

Supply is not value, therefore your statement of falsehood is incongruent with the argument.

The exchange rate is an indication of relative valuation, not the inherent value of an item.

Value is not purely associated with demand; it arises from the interaction between supply and demand.

Correct me if I'm mistaken, but you seem to be suggesting management of unit value in relation to external forms of wealth.

Manipulation? One child learning how to ride a bike more quickly than another could be considered "manipulation" due to greater skill. Does that mean the bicycle is faulty?

Terrestrial gold is a finite resource, no different from Bitcoin. What happens when gold runs out? The argument is absurd, as is the suggestion that Bitcoin is a static system incapable of change.


Supply expands to the extent that it is demanded; the phenomenon known as elasticity. There is no need to mine in order to maintain value in Altcoins, a simple transfer from Altcoin to Bitcoin (or any other asset that maintains wealth during deflationary periods) works for preservation of wealth. During contraction in demand below what would justify existing supply, the Bitcoin side becomes more attractive and wealth flows in greater volume out of Altcoin and into the former. This is no different from the current shift of trend from holding wealth in fiat to real assets.

A potential concern might be transaction verification time based on participating nodes. In that case, system participants in need of such a service (e.g. retail merchants) will have to provide additional processing power to maintain transaction speed or contract with providers of such services. There is more to an economy than just the monetary base.


I'll let 5,000+ years of history know you disagree. In the meantime, reality just left a message stating that I should tell you an ounce of gold from 2,000 BC is still an ounce of gold.

The dual currency system worked quite well until central banks began managing (read: distorting) supply. Bitcoin doesn't have humans at the helm controlling the supply; the known and verifiable point of failure has been removed from the equation (it isn't perfect - there are other concerns). This is the core of what makes Bitcoin so powerful. Please try to understand this, otherwise further meaningful discussion will be very difficult.


As stated in my comment prior to this: management. Central banks' control over the supply in combination with heavy influence in the rules and regulation thereof. Without the human element now present in existing fiat systems, the one century to collapse we see now might take an order of magnitude longer.

Regulation has been protective of those in positions to manipulate the money supply. Government is complicit in the illegitimate expansion of the monetary base which is used as a store of wealth.

Bitcoin has removed the government, banks, regulators and any other direct form of control over the money supply. This is the critical aspect that allows it to function similarly to fiat, only better. It is market participation, not the market itself, that gives rise to "manipulation" as those whom are more highly skilled at accrual of wealth in a closed system will accumulate more than those that are not as good.

Just as gold removes the possibility of perpetual expansion, Bitcoin's monetary base limitation precludes subversion of existing wealth. The method that Bitcoin's system enforces the limit is the same way laws become entrenched: distributed agreement upon a tenet.

Interaction being cyclical, the dominant factions of a closed system eventually expend more resources maintaining and defending their share than accumulating, leading to smaller participants being able to make headway. The "manipulators" are offensive until they can no longer accrue past a threshold, then they become defensive and the smaller participants become the manipulators.

As an example: for decades, the large banks such as JP Morgan and Deutsche have been the "manipulators" and gamed the system. Recently, the Occupy Wall Street movement has been eroding the dominance exerted by the former that began its decline in the early 2000s. The dominating banks are on the defensive while the smaller participants are building their offense. That is the cyclical nature, and new dominant players will emerge from this populist movement.


Exactly. There is no perfect solution, only a balance of interaction between opposing forces. It's an observation that has been made repeatedly throughout history, mostly via religion.


This is an understandably confused perspective. It is not functionally better than fiat: its supply is just governed more reliably. It is not "better" than Bitcoin: it is complementary.

The key to this concept is the interaction between components, even more so than the components themselves.



You don't "pay" because of the return banks gain from holding your money for you. There are other ways the banks make money from holding your money so that they can entice you into entrusting your wealth to their safe-keeping. Checking accounts can be "swept" overnight to earn a return for the bank, and savings accounts are generally limited to a set number of monthly transactions because interest is paid to the client, precluding the bank from "eating" too many transaction fees behind the scenes.

Costs may be hidden from you, even buried, but there is always a price to pay. Bankers and politicians are particularly adept at disguising true costs. Remember: TINSTAAFL.



From a trading perspective, hedging bets is prudent. In a mature system, the pace of change is glacial and obviates the need to stare at a screen just to protect one's assets. Having personal wealth in any currency existing today is speculation anyway, so the point is largely moot.

When travelling abroad, use of traveler's cheques has been preferred to most other forms of transaction (modern digital equivalents notwithstanding for argument sake). This is not the case when conducting general transactions on a day-to-day basis. People use traveler's cheques for certain types of transactions, but they don't store their wealth using them.

As discussed earlier, value is arrived at from the interaction between supply and demand. This carries to a point. Ultimately, the value of an item in foreign denomination is determined by the seller who is choosing to accept (or reject) a currency in exchange for an item with inherent value.

A relationally-static currency (closed system) will still experience "manipulation" as with any other. The distinction is that any additional methods of determination to arrive at the "appropriate" monetary base introduces unnecessary complexity. Simplicity of individual components precludes catastrophic failures while reserving complexity for the interactions between components. This is the same principle that has allowed computer technology to thrive - simple components that interoperate, the interaction of which gives rise to emergent properties (e.g. that internet thing). Monolithic designs easily approach scalability limits and incur the catastrophic failure that simplicity minimizes.

Bitcoin is elegantly simple and effective. Altcoin (or whatever the inflationary complement might be called) simply removes the unit base limit, maintaining simplicity in favor of interaction.


Really? So politicians actually saved the banks out the goodness of their hearts, and not to keep their own jobs? Right...

You're on the right track with the rest of your assessment, but reversed as to the order. The 2008 crisis was due to over-leveraged financial institutions failing. The most recent has been due to government intervention attempting to prevent widescale bankruptcies in addition to its own demise due to loss of public confidence. The European crisis erupting now is similar to the 2008 crisis of risky financial institution leverage, but that's been triggered early by acceleration of economic destabilization incurred from US gov't intervention. With the confluence and overlapping of numerous factors, the effects are amplified - the perfect storm.

Headlines mean nothing; they are reactionary, responding to undercurrents that have already occurred well before. Follow the money.

We can certainly agree that greed (or disparities in perception/skill, whatever you want to call it) will find a way to develop - whether it's given a way or not.


It may not be arbitrary, but it is unnecessary (even undesirable) for low level implementation. A better place for Encoin's methods to be introduced would be at one level of abstraction higher than the monetary base: i.e. the exchange level.


The deflationary and inflationary monetary cores are universal and need to be kept as clean and simple as possible. Abstracting one layer to the exchange level allows all manner of operations to be performed utilizing the underlying base. If that means providing a service that keeps a steady currency based on external factors, great. What could occur is widespread adoption of the derived exchange-level currency instead of the underlying components.

Risks are present, as centrally-managed currencies like these have shown. However, it should also be possible to construct a decentralized derivative crypto-currency based off of an Altcoin/Bitcoin platform the same way an exchange would operate, thus retaining the benefits of distributed control. An exchange would just be technically easier to implement, especially during early stages of development and testing.

Arbitrary feedback. pfft. The point is to come up with something that isn't arbitrary. I know that this is possible may come as a shock to many, but calling something that is not arbitrary, arbitrary, does not mean that it is.

How about renaming the thread to "Properties I can add to a stable crypto-currency to make its price diverge"?

Seriously. The point of speculation is that it corrects price. Adding an arbitrary feedback that goes against the correction -- or magnifies it, for that matter -- is a pretty absurd idea.

If the goal is to emulate a central bank, altering supply to achieve some murkily defined goal, I have a hint: don't. The last thing we need is more strange dynamics nobody understands that blow up the currency at random. I would probably have used a Bitcoin+demurrage system to keep the amount of circulating coins constant at all times, but Bitcoin is good enough. Run it for a few years, and speculators will make it increasingly stable. Voilà, there is something usable, without the threat of some failure in system dynamics suddenly causing a hyperinflation or whatever.

You don't see why there'd be huge fluctuations? Is the rise up to $30 completely anomalous to you then?

The supply is finite, but for the near future, the supply will continue to increase at a predictable rate.  It seems to me if bitcoin was widely used, we would see a slow, steady deflation, but I don't see why there would be huge fluctuations (up and down).

If BTC were as widely accepted/used as, say, CAD, why would you expect to see more fluctuation in BTC than you currently do in CAD?