Please rip apart this idea, but please also explicitly detail the reasoning behind your critique with evidence if possible. Please do not comment on the economic assumption that the quantity theory of money is more or less correct, but please comment on the ideal rate that it should be calculated with detailed explanations for your reasoning: rolling as available, rolling 24 hours, rolling hours, rolling instants, or decreasing intervals as popularity thus data increases.
I'm not looking to personally profit from this but will work on it until someone else beats me, an easy feat.
My qualificationsI'm about a week young at studying cryptocurrencies, have been programming for a few decades, and have studied economics for about a decade.
FlatteryIt is my opinion that cryptocurrencies are on the road to totally displacing conventional national currencies. I believe that all involved, even malicious miners, have worked to make cryptocurrencies more superior by the day. I believe that cryptocurrencies will soon be closer to the ideal currency, totally secure cheap fast stable & ubiquitous, than conventional currencies, so I'd like to contribute my small knowledge to that effort.
General IdeaI think cryptos should be able to provide both instantaneous transactions and high confidence transactions for small rapid purchases like in a rush at a vending machine and large slower transactions like bank to bank clearing a billion USD equivalent respectively, and I think that the way to achieve stability is by setting price inflation to 0%.
Mining no longer issues new supplyI think the supply of the crypto should be exclusively determined by price inflation, detailed next. To that end: there should be no rewards for mining, but a transaction should be rejected if the fee is not > 0.
Total stability through 0% price inflationI think this can be effected by including Fisher Index data, quantity per unit price & good/service identifier, into the transaction, and applying a currency multiplier, the inverse of the Fisher Index, indicating at what rate all account values should be multiplied. My concerns are the obvious: how can the master list of the goods/services identifiers be maintained, and how can the price reporting be made honest? I think but am unsure that using the hybrid PoW/PoS model that Peercoin does could make dishonesty costly since one must become a large holder to manipulate the supply at low cost: there is no reward for mining, transaction fees are > 0 so probably goes to someone else if one's stake is small, only a large % of total transaction value can manipulate the Fisher Index thus the supply. Also, any benefit a price manipulator gleans for oneself is also distributed to every other holder equally, so someone wishing to get more coins by falsifying a price deflation would get them at the exact same rate as everyone else, nullifying any advantage over other holders.
Total price stability could be where cryptos begin to exceed traditional currencies in traditional quality, and I would be ecstatic if every crypto adopted it. Assuming that the price reporting is correct on average, the currency could correct itself instantaneously. This is impossible with banknotes and metal coins but could happen somewhat if monetary authorities gathered debit & credit transactions. With cryptos, there are no banknotes, so the price inflation could be always calculated from the start. I confess I'm unsure at what rate the index should be calculated, but the interval could be as low as the data provides. Seconds? Deciseconds? While some part of me believes that rolling 24 hours would capture all of the daily fluctuations, another part tells me that velocity & production would become more constant at higher frequencies allowing for a closer link between supply & prices, and please comment on how to make the price reporting system more secure.
Varying transaction verification speed & confidence through user-defined difficultySince mining no longer creates new supply, difficulty is now unlinked from creation rates.
In the real world, a bartender doesn't necessarily check for counterfeiting especially on a busy night on every bill, but a bank will employ a complex clearing process to determine that a one million USD wire transfer is legitimate before permitting a withdrawal. To give the kind of transactional flexibility that traditional currencies have, should it be effected with a sender-defined difficulty? Since difficulty is no longer linked to creation rates, should it be instead linked to transaction speed or confidence? If the user wants a beer, they set the difficulty low for a quick confirmation. If the user wants to buy an island, they set the difficulty to some extraordinary height.
"Proof of presence" to thwart secret chains on high speed verificationIt has been explained to me that with high speed verification, malicious miners can hold back chains, and then publish them so to double spend etc. I think that by using the same underlying methods as proof of share, that I'm calling "proof of presence" for expediency's sake, miners could be further prevented from monopolizing the blockchain thus this type of attack by forbidding consecutive blocks.
I'm sure this could be enhanced by someone with more expertise than me, but I think this could at least add a small amount of cost to this kind of attack.
SummationI think a cryptocurrency can be created with total stability and transaction flexibility, with varying preference for speed or confidence by:
- Issuing/retracting supply not by mining but by a price index
- Allow senders to determine a transaction's difficulty to trade-off between speed and confidence
- Make transaction fees mandatory
- Prevent consecutive block additions by a single account to dampen the risks of high speed verification
Final appealI am still fuzzy on the technical aspects of cryptocurrencies, so please correct any technical idea. Please explicitly state how the protocol should be effected to achieve the above.
I also understand that the price index is wide open to manipulation and will remain a huge risk, so any suggestion on how to make the reporting more secure would be appreciated.
Actual code is more than welcome!
