|
There is an issue in the "transparent mining" that I can not see clearly: if someone hides an alternative chain, how does it identify what is the good branch and what is bad. I can not find the formula to do possible this identification...
The only thing I can think is that the two chains evolve by its side, both with their private 100% of the currency involved in the generation of each block, and it would be possible setting one fork or more of the network... Let we can say that the network is protected of attacks, but not for forks. I dont know what is worst...
Maybe I have not the data which is measured to determine the valid chain, I dont know if it is the greatest amount of currency, coin-age, or what...
You said: "this can be counteracted by some mechanisms of advanced consensus..." But I thing that in the moment that we talk about consensus, we have the 51% problem over the table again...
|
|
|
|
Pues ahí esta el quid de la cuestión. Veo que tú si que entiendes cómo funciona el tema. Resulta que , como el transparent mining no está todavía implementado ni se ha abierto su código, no se sabe cómo se ha diseñado ese aspecto. Tal y como tú dices, ambas cadenas tendrán la misma longitud y recuerdo que Come from Beyond dijo algo así como que se utilizarían mecanismos avanzados de consenso en redes descentralizadas para diferenciar la cadena legítima. Aquí te lo transcribo: Imagine someone is going to do a "51%" attack against Nxt and he owns 90% of all coins. The adversary must stop generating blocks for legit branch coz he won't be able to compete against 100% mining power with his 90%. So he decides to "skip" his turn to generate a block. The rest 10% of the network detects this and penalizes the adversary by setting his mining power to 0 and distributing it among other miners. Now the network is back to 100% power coz everyone got 10-fold increase. The adversary can mine other branch in a secret place but it won't be able to replace the legit branch. Of course, the 2nd branch will have 100% "hashing" power tied to it as well, coz the attacker will get his 90% bumped to 100% but this can be counteracted by some mechanisms of advanced consensus (still not revealed). Veo difícil que se pueda diferenciar y creo que esto puede ser un punto crítico del transparent mining... En el momento que haya un mecanismo de consenso, ya estamos con el 51% de nuevo... |
|
|
|
Voy a intentar explicarlo con términos mundanos, a ver si así se entiende mejor. (en otro post)
Hay un asunto en el " transparent mining" que no acabo de ver claro: en el supuesto de que alguien oculte a los demás una cadena alternativa, cómo identifican cuál es la rama buena de la cadena y cuál es la mala. No doy con la fórmula para hacer posible esta identificación... Lo único que se me ocurre es que las dos cadenas evolucionarían por su lado ambas con el 100% particular de la moneda participando en la generación de cada bloque, y podría establecerse una bifurcación o varias de la red... Digamos que la red estaría protegida para ataques, pero no para bifurcaciones... Quizá me falta el dato de qué se mide para determinar la cadena válida, no se si es mayor cantidad de moneda, coin-age, o qué... |
|
|
|
|
Una pequeña observación, que la verdad con el cúmulo de monedas, algoritmos y demás, puede perderse de vista...
Ahora mismo aunque NXT sea PoS, dado el numero de monedas PoW que hay, pueden obtenerse NXTs minando cualquier otra moneda PoW y comprando luego NXT. Con lo que la coyuntura actual posibilita que NXT se pueda obtener mediante el propio forjado del PoS, pero también minando con SHA256 o con scrypt... Lo cual favorece el reparto de la misma utilizando como vehículo el mecanismo PoW.
Por poner un ejemplo: con 5 tarjetas de video normalitas, que den 2500 Kh/s, minando por ejemplo FST, en el momento actual se producen unos 1200 FST/día, esto serían unos 250-350 NXT por día, que no está nada mal...
En fins, imagino que lo que comento es una obviedad, pero por si a alguien se le hubiera pasado el asunto...
|
|
|
|
To avoid abuse you could implement a system that uses lower fees (perhaps higher PoT-Paiout), the longer the coins have not been moved before.
What do you mean with lower fees? |
|
|
|
|
He visto un posible problema, que sería una cadena paralela oculta a la red, formada únicamente por dos direcciones que podrían generar una cadena fraudulenta con más transacciones que la cadena original. Con lo que luego al publicarla invalidaría a la original. Tendría alto coste, pero al estar oculta y formada por sólo direcciones de un mismo propietario se evita el perder los fees de cada transacción con lo cual se podrían generar transacciones sin límite en esa cadena alternativa...
Sigo viendo si este asunto podría ser subsanado de alguna forma...
|
|
|
|
|
Studying the issue in more depth I see a possible problem: it would be a parallel blockchain hidden to the network with only two addresses that could generate a fake blockchain with more transactions than the real blockchain...
I will think if is possible solve this issue...
|
|
|
|
Proof of Transaction = Dices would mine a lot.
What about proof of transaction*amount (or fee)?
it is possible... But in this way you reward the huge amounts of money and it would be like PoS. |
|
|
|
Surely the exchanges will always win the blocks because they have most transactions out of all the wallets??
It could be, only the withdrawals would be transactions that increase the transactions counter of the addresses of the exchanges. yep, this can see as critical point at the current moment when the coins are using for trading only... But in the future the exchanges will have less relevance. And this system promote the use of the coins, so perhaps it would accelerate the time of the adoption of crypto-coins in our day to day... The current situation with the exchanges could see as a support for the security of the system too. And they can reward the withdrawals like if they were pools...  |
|
|
|
Ok, then back to PoT vs PoS. Stake is more difficult to be faked than transaction numbers. Fees may help preventing average user to take advantage of high transaction numbers, but may be enough to stop 51% attacking. Moreover, usually the number of normal transactions of a new coin is very low, and there's not too much cost for a person to fake transactions to get the advantage in mining. If more and more people doing this, the system will be flooded with meaningless transactions and becomes both slow and bloated.
I am not agree with this, because if you try to fake transaction numbers this has a cost for you and when you reach to the desired reward your transactions counter will reset to 0. And your desired reward can be very little, it depends of fees on the block you had discovered. Then you has spent large money to obtain a privileged position in transaction number counter to nothing. With respect 51% attack, suppose that you have one address that achieves the 51% of transactions numbers, then you discover a block and your transactions counter go back to 0. Your 51% has been useful to one block only. If you want to continue with the attack, you will have to obtain the 51% again, and spent money again in fake transactions with their fees to achieve it... The system self protects itself. With this system the number of normal transactions would be high, because they are rewarded... |
|
|
|
I am quite curious about how timestamp control can avoid a client to make multiple hash in a second. It may avoid a client to SUBMIT more than a hash a second, but hashing is done locally and no other node will know how many times I have tried.
I can see that it is simple: in PPC the Nonce is a Timestamp of one second. |
|
|
|
|
Yep, in NXT with transparent mining fix the node, I dont know how do it...
But fix the hash to one per second is possible.
|
|
|
|
|
In PoS making more hashes, you would obtain more benefits too... It has a kind of timestamp control to avoid this.
PoS do two things:
1.- Fixed hash to one hash per second.
2.- Difficulty is not fixed, it is variable and less for the addresses with more number of coin-age (coins*time_in_wallet).
PoT would do two things too:
1.- Fixed hash to one hash per second.
2.- Difficulty is not fixed, it is variable and less for the addresses with more number of accumulated transactions.
|
|
|
|
How to fix everyone's hashing rate to one hash per second? Relying on no one changes your source code, or is there any method to ensure this in the decentralized network?
For example, PPC already do it. And NXT I think too... Every PoS coins do it. A kind of timestamp control... |
|
|
|
Imagine a person is doing transactions between two address to increase his transactions counter to obtain high probability to get a block, when he achieves the block, his transactions counter resets to 0, and until this moment he has spent a large amount of his coins in fees to increase his transactions counter. And his reward can be very little, because imagine that he discovers a block with a total fee of 1 coin... He wont try the fake activity again! |
|
|
|
|
With respect point one:
I think fees would avoid this, because if you make a transactoin, you will pay fees and this fees can get back to you if you confirm the transaction, but other can confirm it then you will loose fees, then would not be profitable to do transactions between oneself...
I do not understand this point: 2) How do you spend a new coin? Nobody is going to accept it
why not? Only the input address resets to 0 its transactions accumulated count.
And about distribution problem, I think here there is no solution yet...
|
|
|
|
|
Perhap better name would be Proof of Use (PoU)...
|
|
|
|
Hey guys, Yesterday talking with Zesterer in fastcointalk.org forums, he told to me that he with a friend talked about to make a some coin that rewards its use. Thinking on this idea, the system Proof of Transactions (PoT) came to my mind... It is a system where the blocks are discovered by who has more transactions. It is simple: Hashing would be fixed to one hash per second for everyone. The difficulty would not be equal for everyone, it would be variable and less for addresses with more transactions. And when somebody discovers one block, his transactions count would be reset to 0, and the count would start again until the next discovery. The reward would be the fees of transactions in the block. It is very simple, but I think that system promotes the use of the coin, as Zesterer said. And avoids 51% attack because in each discovery resets the discovery probability to 0 to the account did it. I think it would be a good system to implement a future version of a coin to promote the use and distribution more equity and faster of the coin of our day to day... It does not reward the huge amount of money, neither who buy more technology and spend more energy... simply rewards to people spend their money. This coin would not have mining, all amonut would be generated at the begining, so it would be necessary a good distribution system (equity and fair)... It is a today idea, I dont have details about it yet... so I hope your opinions and pros and cons that you see it... Thanks in advanced...! |
|
|
|
|
La recompensa serían los fees de la transacción simplemente... Falsearla sería absurdo, ya que podrías confirmarla tu y te volverían los fees o confirmarla otro y perderías los fees...
Eso si irías acumulando transacciones en tu haber, pero a costa de perder los fees si te las confirma otro, eso lo que te permitiría es aumentar la probabilidad de confirmación de bloque; pero en una red con un uso masivo, no sería muy significativo ese aumento, resultando en que tal vez, no merecería la pena arriesgarse a perder los fees para conseguir numero de transacciones que no te darían ventaja muy significativa a no ser que hicieras muchas con la consecuente perdida de muchos fees. Se podía poner un límite en tu cuenta particular de transacciones que sería algo así como cuando lleves 100, tu contador de transacciones se resetea a 0 y vuelta a empezar, o que se resetee cuando confirmes un bloque...
Con esto del reseteo hasta evitas el ataque del 51%, porque cada vez que confirmaras un bloque el 51% iría bajando... Parece como si el mismo sistema se auto-protege del ataque...
Estoy tratando de pensar como sería un pool de PoT y me parece que podría ser bastante parecido a los pool de PoW. Lo cual no está mal tampoco...
Bueno pensándolo mejor, podría haber pools de dos tipos como los de PoW y como los de PoS... Cada cual tendría que elegir en cual le convendría más estar...
Me quedaría por ver si favorece el reparto de la masa monetaria...
No me parece mal sistema la verdad...
Si alguien que también sea programador se anima, podemos intentar hacer una moneda experimental basada en PoT...
|
|
|
|
|
Show Posts
