Search content
Sort by
Showing 15 of 15 results by Ekkio
Re: [ANN] Bancor | Protocol for Smart-tokens, solving the liquidity problem
One of the ways to think about reserves is as "liquidity pools". Reserves are essentially common pools that provide liquidity to their smart token holders. The current solution for liquidity is based on market orders on the different exchanges (and in some cases, in different currencies pairs). If you measure the size of these current "liquidity pools" (aka as "market depth"), relative to the market-cap of the currency, and compare the event where similar amounts are unloaded to the market through the exchanges vs. through a smart token with 10% reserve, you would probably find out that the market behaves in a much more stable way using the smart token, since there is a single common liquidity pool, which is growing relative to the market cap of the smart token.
It can be tricky to wrap the mind around it since this is a model for linking between currencies which is quite different than the one being used today and for a long time now, and I really hope my attempt to explain it was clear enough.
Anyway, I'll be happy to answer any other questions you may have!
And thank you for your interest
I went through all the math and tried out some basic modeling because I am excited and curious how it all works. One question though, the first step that allowed you to make a continuous formula out of the recursive: dR = PdS and then use that equation later, is this some known approach that helps to make continuous formula out of recursive, that I don't know about because of my limited math education, or is it a clever trick that you can credit to yourself?
This is basic calculus.
Alright, given A1, A2, A3 and recursive formula An+An+1/An+2 = An+3, eg A1=1 A2=2 A3=4 therefore A4 = 1 + 2/3 =5/3
Can you show me how to make continuous formula out of it f(x) eg f(4) = 5/3
They solve a simple differential equation. Check any calculus textbook to learn how to do this.
Sure, but first you need to get to the point of having simple differential equation, hence my question, can you get the simple differential equation out of recursive formula I provided?
You "recursive formula" makes a very little sense to me. Expression for bancor price is derived by integrating a differential equation which you pointed out in you first post, dR = PdS. This is not a rocket science, this is a level of homework for undergraduate students of technical schools. It is not possible to explain how calculus works in a forum post, just check any textbook, they explain all theory very well.
You don't seem to understand the question.
The original formula Price = Reserve / (Supply * CRR) is recursive.
It's calculated in sequential steps , add unit of Reserve, increase Supply according to price, recalculate price, rinse and repeat.
According to you making a continuous formula f(x) out of it is "basic calculus".
If that's that's true then it should be easy for you to make continuous formula out of any recursive formula.
Therefore please make continuous formula out of An+An+1/An+2 = An+3 .
You don't need no "recursive formula" to solve this problem. Derivatives and integrals of many analytic functions were derived using the limit theory long before we were born, you just use them when you need. Sorry for being boring, but I'd like to direct you to calculus textbook for undergrads again, where all such derivations are given in details.
I am going to try to clarify it one last time, if it doesn't work so be it.
I don't have a problem with derivatives and understanding their proof .
My question was not about the technique or how in this particular case they got their formulas.
Initially they had recursive formula Price = Reserve / (Supply * CRR) and using derivatives they were able to transform it to continuous formula Price(x) = ((1+ x / InitialSupply)^(1/CRR -1)) * InitialPrice
My question is there a general method to transform any recursive formula into continuous, or was it the clever trick that allowed to do that in just this particular case.
Re: [ANN] Bancor | Protocol for Smart-tokens, solving the liquidity problem
One of the ways to think about reserves is as "liquidity pools". Reserves are essentially common pools that provide liquidity to their smart token holders. The current solution for liquidity is based on market orders on the different exchanges (and in some cases, in different currencies pairs). If you measure the size of these current "liquidity pools" (aka as "market depth"), relative to the market-cap of the currency, and compare the event where similar amounts are unloaded to the market through the exchanges vs. through a smart token with 10% reserve, you would probably find out that the market behaves in a much more stable way using the smart token, since there is a single common liquidity pool, which is growing relative to the market cap of the smart token.
It can be tricky to wrap the mind around it since this is a model for linking between currencies which is quite different than the one being used today and for a long time now, and I really hope my attempt to explain it was clear enough.
Anyway, I'll be happy to answer any other questions you may have!
And thank you for your interest
I went through all the math and tried out some basic modeling because I am excited and curious how it all works. One question though, the first step that allowed you to make a continuous formula out of the recursive: dR = PdS and then use that equation later, is this some known approach that helps to make continuous formula out of recursive, that I don't know about because of my limited math education, or is it a clever trick that you can credit to yourself?
This is basic calculus.
Alright, given A1, A2, A3 and recursive formula An+An+1/An+2 = An+3, eg A1=1 A2=2 A3=4 therefore A4 = 1 + 2/3 =5/3
Can you show me how to make continuous formula out of it f(x) eg f(4) = 5/3
They solve a simple differential equation. Check any calculus textbook to learn how to do this.
Sure, but first you need to get to the point of having simple differential equation, hence my question, can you get the simple differential equation out of recursive formula I provided?
You "recursive formula" makes a very little sense to me. Expression for bancor price is derived by integrating a differential equation which you pointed out in you first post, dR = PdS. This is not a rocket science, this is a level of homework for undergraduate students of technical schools. It is not possible to explain how calculus works in a forum post, just check any textbook, they explain all theory very well.
You don't seem to understand the question.
The original formula Price = Reserve / (Supply * CRR) is recursive.
It's calculated in sequential steps , add unit of Reserve, increase Supply according to price, recalculate price, rinse and repeat.
According to you making a continuous formula f(x) out of it is "basic calculus".
If that's that's true then it should be easy for you to make continuous formula out of any recursive formula.
Therefore please make continuous formula out of An+An+1/An+2 = An+3 .
Re: [ANN] Bancor | Protocol for Smart-tokens, solving the liquidity problem
One of the ways to think about reserves is as "liquidity pools". Reserves are essentially common pools that provide liquidity to their smart token holders. The current solution for liquidity is based on market orders on the different exchanges (and in some cases, in different currencies pairs). If you measure the size of these current "liquidity pools" (aka as "market depth"), relative to the market-cap of the currency, and compare the event where similar amounts are unloaded to the market through the exchanges vs. through a smart token with 10% reserve, you would probably find out that the market behaves in a much more stable way using the smart token, since there is a single common liquidity pool, which is growing relative to the market cap of the smart token.
It can be tricky to wrap the mind around it since this is a model for linking between currencies which is quite different than the one being used today and for a long time now, and I really hope my attempt to explain it was clear enough.
Anyway, I'll be happy to answer any other questions you may have!
And thank you for your interest
I went through all the math and tried out some basic modeling because I am excited and curious how it all works. One question though, the first step that allowed you to make a continuous formula out of the recursive: dR = PdS and then use that equation later, is this some known approach that helps to make continuous formula out of recursive, that I don't know about because of my limited math education, or is it a clever trick that you can credit to yourself?
This is basic calculus.
Alright, given A1, A2, A3 and recursive formula An+An+1/An+2 = An+3, eg A1=1 A2=2 A3=4 therefore A4 = 1 + 2/3 =5/3
Can you show me how to make continuous formula out of it f(x) eg f(4) = 5/3
They solve a simple differential equation. Check any calculus textbook to learn how to do this.
Sure, but first you need to get to the point of having simple differential equation, hence my question, can you get the simple differential equation out of recursive formula I provided?
Re: [ANN] Bancor | Protocol for Smart-tokens, solving the liquidity problem
One of the ways to think about reserves is as "liquidity pools". Reserves are essentially common pools that provide liquidity to their smart token holders. The current solution for liquidity is based on market orders on the different exchanges (and in some cases, in different currencies pairs). If you measure the size of these current "liquidity pools" (aka as "market depth"), relative to the market-cap of the currency, and compare the event where similar amounts are unloaded to the market through the exchanges vs. through a smart token with 10% reserve, you would probably find out that the market behaves in a much more stable way using the smart token, since there is a single common liquidity pool, which is growing relative to the market cap of the smart token.
It can be tricky to wrap the mind around it since this is a model for linking between currencies which is quite different than the one being used today and for a long time now, and I really hope my attempt to explain it was clear enough.
Anyway, I'll be happy to answer any other questions you may have!
And thank you for your interest
I went through all the math and tried out some basic modeling because I am excited and curious how it all works. One question though, the first step that allowed you to make a continuous formula out of the recursive: dR = PdS and then use that equation later, is this some known approach that helps to make continuous formula out of recursive, that I don't know about because of my limited math education, or is it a clever trick that you can credit to yourself?
This is basic calculus.
Alright, given A1, A2, A3 and recursive formula An+An+1/An+2 = An+3, eg A1=1 A2=2 A3=4 therefore A4 = 1 + 2/3 =5/3
Can you show me how to make continuous formula out of it f(x) eg f(4) = 5/3
Re: [ANN] Bancor | Protocol for Smart-tokens, solving the liquidity problem
One of the ways to think about reserves is as "liquidity pools". Reserves are essentially common pools that provide liquidity to their smart token holders. The current solution for liquidity is based on market orders on the different exchanges (and in some cases, in different currencies pairs). If you measure the size of these current "liquidity pools" (aka as "market depth"), relative to the market-cap of the currency, and compare the event where similar amounts are unloaded to the market through the exchanges vs. through a smart token with 10% reserve, you would probably find out that the market behaves in a much more stable way using the smart token, since there is a single common liquidity pool, which is growing relative to the market cap of the smart token.
It can be tricky to wrap the mind around it since this is a model for linking between currencies which is quite different than the one being used today and for a long time now, and I really hope my attempt to explain it was clear enough.
Anyway, I'll be happy to answer any other questions you may have!
And thank you for your interest
I went through all the math and tried out some basic modeling because I am excited and curious how it all works. One question though, the first step that allowed you to make a continuous formula out of the recursive: dR = PdS and then use that equation later, is this some known approach that helps to make continuous formula out of recursive, that I don't know about because of my limited math education, or is it a clever trick that you can credit to yourself?
Re: [ANN] Bancor | Protocol for Smart-tokens, solving the liquidity problem
when tokens are traded on the exchange relative to BTC, understandable pricing mechanism. is the flagship and the others were compared. Here, we propose the creation of crosscourse. Prices will be based on courses that are on the exchanges? If so , then not all tokens are liquid and the prices of some are very different in the exchanges, will average?
The network will only work with tokens which are received by smart contracts on everyone, crosscourse to zec, litecoin, Dash will not be used? what will happen to the tokens of other platforms different from etherium?
The network will only work with tokens which are received by smart contracts on everyone, crosscourse to zec, litecoin, Dash will not be used? what will happen to the tokens of other platforms different from etherium?
If I understand you're question correctly, you're asking how the inevitable price differences between smart tokens internal prices & their prices on exchanges.
The answer is the same as the solution to price differences between exchanges: arbitrage bots. When there is profit to be made by a difference in price, these bots automatically come and make trades until they've evened out the prices (and so it is no longer profitable).
To be clear: smart tokens' contracts can buy/sell themselves. No second party needed. That is how every single currency on the network can be fully liquid from day one.
Yeah, right, but since you are going to run a fractional reserve, actual liquidity is going to limited by initial supply. For example, at 20% reserve ratio change in supply by a factor of 2 in either direction will change the price by a factor of 4. This will limit the amount of tokens we will actually be able to buy or sell at reasonable price. Liquidity would be really infinite in two cases: either the initial supply is very large; or reserve ratio is close to 1.
Actually, the supply can be (and for most smart tokens, we think it will be) much smaller than 20%. The reason that smart tokens can still be liquid with tiny reserves is that, when you buy/sell a smart token (or its reserve tokens) the price is calculated as if you sold an infinitely large amount of micro transactions that equal the same total amount. For every micro transaction you do, the price for the next goes up (as with every supply/demand system). So if supply were to grow smaller, the price of each smart token would decrease proportionally. The token would stay liquid, it would just decrease in value.
If demand were such that the price is no longer 'reasonable' to enough people, demand would go down and said price would stop rising (or fall, if demand went low enough that there are now more people selling than buying). In other words, the price of a smart token is kept reasonable by the market, like any other freely traded commodity.
A price of your token is a power law of its supply relative to initial supply. You need to have a fairly large initial supply if you don't want the price to raise/drop by crazy amount after each trade, unless your reserve ration is close to 1.
This is true, a deeper market depth (which is a main function of a reserve currency in this type of smart token) makes for less volatility. The market depth on any given exchange is usually <1%, so a 20% market depth can be considered quite large.
I am sorry, but you are comparing apples and herrings here.
You can't have 5 BANCOR issued initially, which is reserved by 1 ETH. If you do this, your token will behave as any illiquid token, a price will be a random number, which varies significantly with every trade of a fraction of ETH. If you run at low reserve ratio, such as 20%, you need to have a huge initial supply, which is much larger than projected daily volume.
The ratio between Token market cap and Reserve have to always stay the same, so by dropping just 1 ETH in 5 to 1 contract you are increasing market cap by 2 and change the price for one Token from 1:1 to 1.74:1 which is the whole 74% price increase, consider that you just dropped a liquidity bomb that changed Token market cap from 5 ETH to 10 ETH, what would have happened to an exchange if you tried to execute order of that magnitude.
Re: Hashers are not miners, and Bitcoin network doesn't need them.
Ekkio, do you think the Bitcoin Foundation should
A) Make the issue of 'centralisation of mining' a higher priority
B) Make it the no.1 priority
C) Literally stop everything else and just focus on that alone, till there is some solution in place?
A) Make the issue of 'centralisation of mining' a higher priority
B) Make it the no.1 priority
C) Literally stop everything else and just focus on that alone, till there is some solution in place?
I certainly would like to see people spending more time looking for a political or technical solution to the problem but it can't be number one priority for Bitcoin Foundation because it's an academics and it's almost impossible to speed up the process no matter the amount of additional resources.
Re: Hashers are not miners, and Bitcoin network doesn't need them.
The main purpose of this thread is to create a healthy discussion of the problem by using better definitions. So I don't have an opinion about what would be a sound solution, but what I do know is that current agenda of Bitcoin Foundation and involved core devs will give us centralized mining.
When you say 'agenda', do you mean that, you believe that the Bitcoin Foundation and involved core devs are deliberately not trying to give us decentralised mining? If so, do you have any simple proof of that, that I would understand?
As far as I can tell there is no significant resources allocated from Bitcoin Foundation group to make mining more decentralized and they say that mining pools are ok as long as they don't misbehave, is that what you mean by "deliberately not trying to give us decentralised mining" ?
Re: Hashers are not miners, and Bitcoin network doesn't need them.
Is there anything that can be done about it?
First of all it is important to have people understanding the difference between Miners and Hashers, very few know what the difference is and why it is important to decentralize verification process. So at the moment the real question is how to make people understand that those who participate in pool mining are not miners.
Please elaborate on what you believe to be a sound resolution. Specifically, one that maximizes earning potential for individuals as well as bitcoin network security.
The main purpose of this thread is to create a healthy discussion of the problem by using better definitions. So I don't have an opinion about what would be a sound solution, but what I do know is that current agenda of Bitcoin Foundation and involved core devs will give us centralized mining.
The next statement requires leap on intuition, I predict that centralized mining will have regulatory safeguards built around it and Bitcoin will become Banking 2.0 system, and I would like to hear why or why not it could be true.
Re: Hashers are not miners, and Bitcoin network doesn't need them.
Is there anything that can be done about it?
First of all it is important to have people understanding the difference between Miners and Hashers, very few know what the difference is and why it is important to decentralize verification process. So at the moment the real question is how to make people understand that those who participate in pool mining are not miners.
Re: Hashers are not miners, and Bitcoin network doesn't need them.
You're trying to demoralize a perfectly legitimate behavior on part of the "hashers", but this only diverts the discussion from the real issue: the protocol itself.
The protocol was designed to encourage profit seekers, or "hashers", to.. hash.
Satoshi didn't foresee this centralization behavior.
But I can guarantee you, that if he did, he wouldn't seek a moralizing argument, but a technical solution.
The whole point of Bitcoin is to be moral-agnostic, and trust-free. It's a cryptographically secured public ledger, not a charity fund or a society equalizer.
Calling names is pointless. We need technical solutions, not moral preaching.
The protocol was designed to encourage profit seekers, or "hashers", to.. hash.
Satoshi didn't foresee this centralization behavior.
But I can guarantee you, that if he did, he wouldn't seek a moralizing argument, but a technical solution.
The whole point of Bitcoin is to be moral-agnostic, and trust-free. It's a cryptographically secured public ledger, not a charity fund or a society equalizer.
Calling names is pointless. We need technical solutions, not moral preaching.
What I am trying to do is different, good definitions have a much greater chance to lead to intelligent discussion and this is why it is important to recognize difference between Miners and Hashers. Opening up a good discussion is a first step in finding solution.
Believe it or not OP, miners, hashers, are in this to generate bitcoin. Do you believe anyone would buy a jupiter or example if they were told they might not generate a block, ever? People join one pool over another not because they are evil parasites, but because they're new and they don't yet understand everything there is to know about the protocol.
Helping new people to understand the difference between Miners and Hashers is important. Miners are different from Hashers because they honor the social contract with Bitcoin community, in exchange for the reward they provide security to Bitcoin network. I think that development of tools that would make it easier for Hashers to become Miners would be a good thing.
Re: Hashers are not miners, and Bitcoin network doesn't need them.
I completely agree with you OP. Hashers, as you call them, are just another example of profit mongers that are using Bitcoin to reap whatever short term reward there is out of Bitcoin and convert it to fiat. That's Bitcoins biggest problem as far as I can tell. This community is plagued by them and even has devs that seem to fit in that category. I've always seen pool sysops as profit mongers with the exception of a couple of them. Pool operators don't help the network as much as they help themselves to network power and control. If P2Pool was implemented before the pools took control it could have captured and kept majority control of the network because individuals would have been required to learn its use to reap the reward. Sadly, it's now too late for any meaningful change.
Thank you for your support, I think if we can explain the problem and why it is important, then it would be possible to convert Hashers to Miners.
Re: Hashers are not miners, and Bitcoin network doesn't need them.
no longer hashers have to run bitcoin software and govern Bitcoin network instead they just point their hashing power to pools and delegate their enforcement power to small group of people, and that makes that hashing power useless in terms of securing Bitcoin network.
Yes, I see your point. I agree. Miners are trying to get the best reward for their work, pools can help achieve this. But it was bitcoin that offered this reward and encourage miners to chase it.There is a clear distinction between Miners and Hashers. Miners verify validity of transactions and blocks according to Bitcoin protocol but Hashers do no such thing they are happy to hash anything while looking for magic number. I would like people to understand it because that will help to get the message out.
Bitcoin network follows bitcoin protocol and any message sent to bitcoin node has to follow strict rules that enforced by cryptography. If a bitcoin node has a hashing power attached to it, then it's not only verifies and relays new blocks and transactions but also creates new blocks and forces protocol rules into blockchain. So it's those bitcoin nodes that have hashing power and knowledge about protocol govern Bitcoin network and make it secure.
The basic security premises of bitcoin concept was that there would be decentralized network of bitcoin nodes each with small share of hashing power to ensure that there is no single person or small group who can control Bitcoin network. If it's not true then quite a lot of bad things could happen, here is the link with some examples of what kind of new rules can be enforced if someone have majority of hashing power under control, this is also known as soft-fork, to name a few: shorter block time interval, coin divisibility, native color coin support or require AML registration for every address and turn it into surveillance coin.
Bitcoin value comes from it's utility, it is much easier to store and transfer ownership of value by using Bitcoin network. There is a one way relationship between Bitcoin network value and amount of hashing power, the more value Bitcoin network has the more amount of hashing power would be bought by miners. But the opposite is not true, if tomorrow Bitcoin hashing power drop by two nothing bad would happen, after maximum of 2014 blocks network hashing difficulty would readjust, it would be twice easier to find a block and everything would go back to normal, bitcoin market price would not drop either.
People who create bitcoin infrastructure, provide services and innovate are those who make Bitcoin network useful and increase it's value, but via inflation hashers leech that value out of Bitcoin, in exchange they were suppose to help to enforce Bitcoin network rules in decentralized manner. That premises was broken by pools, no longer hashers have to run bitcoin software and govern Bitcoin network instead they just point their hashing power to pools and delegate their enforcement power to small group of people, and that makes that hashing power useless in terms of securing Bitcoin network.
Concerns about pools and how they erode bitcoin security were first raised more than two years ago when Deepbit acquired significant share of bitcoin hash power, but most people agreed that in case a pool would be used for nefarious actions hashers would just point to another more honest pool. The reality didn't match those expectations, Ghash.io was used for double spending. And yet two months after the incident their hashing power share was 15% bigger, and it's only dropped after they stopped accepting new hashers. This is happening because hashers do not care about Bitcoin network security and from that point of view they are evil.
Hashers are not miners, they just hash previous block looking for magic number, they don't care where this block came from or how it was created. Bitcoin network doesn't need hashers, they are parasites who are leeching value from Bitcoin without providing anything in return.
The basic security premises of bitcoin concept was that there would be decentralized network of bitcoin nodes each with small share of hashing power to ensure that there is no single person or small group who can control Bitcoin network. If it's not true then quite a lot of bad things could happen, here is the link with some examples of what kind of new rules can be enforced if someone have majority of hashing power under control, this is also known as soft-fork, to name a few: shorter block time interval, coin divisibility, native color coin support or require AML registration for every address and turn it into surveillance coin.
Bitcoin value comes from it's utility, it is much easier to store and transfer ownership of value by using Bitcoin network. There is a one way relationship between Bitcoin network value and amount of hashing power, the more value Bitcoin network has the more amount of hashing power would be bought by miners. But the opposite is not true, if tomorrow Bitcoin hashing power drop by two nothing bad would happen, after maximum of 2014 blocks network hashing difficulty would readjust, it would be twice easier to find a block and everything would go back to normal, bitcoin market price would not drop either.
People who create bitcoin infrastructure, provide services and innovate are those who make Bitcoin network useful and increase it's value, but via inflation hashers leech that value out of Bitcoin, in exchange they were suppose to help to enforce Bitcoin network rules in decentralized manner. That premises was broken by pools, no longer hashers have to run bitcoin software and govern Bitcoin network instead they just point their hashing power to pools and delegate their enforcement power to small group of people, and that makes that hashing power useless in terms of securing Bitcoin network.
Concerns about pools and how they erode bitcoin security were first raised more than two years ago when Deepbit acquired significant share of bitcoin hash power, but most people agreed that in case a pool would be used for nefarious actions hashers would just point to another more honest pool. The reality didn't match those expectations, Ghash.io was used for double spending. And yet two months after the incident their hashing power share was 15% bigger, and it's only dropped after they stopped accepting new hashers. This is happening because hashers do not care about Bitcoin network security and from that point of view they are evil.
Hashers are not miners, they just hash previous block looking for magic number, they don't care where this block came from or how it was created. Bitcoin network doesn't need hashers, they are parasites who are leeching value from Bitcoin without providing anything in return.