Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 08:51:39 AM |
|
That's called 'substitution of notions'. You suggested this topic. I didn't address summation of probabilities in my original post. And I don't see how summation of probabilities is connected to it. This is my original post. Where do you see summation of probabilities? Why are we discussing it? Transition to Transparent Forging (at least the way it is described in the wiki) will ruin the axiom of "Rate of return to 1 Nxt is the same regardless how many Nxt you have." Here's why. Steps 3-5 in http://wiki.nxtcrypto.org/wiki/Transparent_Forging from the statistics point of view are equivalent to comparing random numbers uniformly distributed between 0 and 1/N where N is number of coins in the wallet (normalization coefficient is the same for all wallets and is omitted for clarity). Now imagine that you throw a dice with numbers from 1 to 6 and your opponent throws a semi-dice with three possible values from 1 to 3. Who gets the lower number wins the round. In half of the cases you will throw 4-6 and win. In the other half you will have 50:50 chances to win. Thus your chances are 1/2 + 1/4 = 0.75 and his chances are 1/4 = 0.25. Which is 3 times lower. For arbitrary ratio K between two wallets the probability ratio is 2K-1. For example, a wallet with 100k coins is 199 times more likely to forge a block than 1k wallet. I'm talking about three times probability increase for a two times larger wallet. What orphan blocks are you talking about? Here u say this: Now imagine that you throw a dice with numbers from 1 to 6 and your opponent throws a semi-dice with three possible values from 1 to 3. Who gets the lower number wins the round. In half of the cases you will throw 4-6 and win. In the other half you will have 50:50 chances to win. Thus your chances are 1/2 + 1/4 = 0.75 and his chances are 1/4 = 0.25. Which is 3 times lower.
|
|
|
|
jettico
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 80
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 08:51:57 AM |
|
Orphaned blocks don't make A and B not mutially exclusive...
They do and this is by design to make it impossible to predict next forging accounts 1440 blocks in advance. In TF it is possible by design to predict who will forge the next block. That's exactly what I address in this post.
|
|
|
|
lophie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 08:56:09 AM |
|
MPK is a public key generated using a wallet seed in a way so that you can generate other addresses right from the MPK without needing the priv keys. To see the utility of this check acceptbit.com, it is a POS system that do not require ANY private keys you just supply it an MPK and then it can generate addresses for your customers, imagine the implications, a cashless register, where transactions happen and only the administration can move the money.
Did I get u right? U state that there a function F() exists such as F(MPK, i) = PKi? Yes for bitcoin, Using electrum client, I was wishing someone can make something like this for nxt, I don't quite understand how it works but this is what I know so far. In electrum when you start it you can create a new wallet from a deterministic seed, You can use this seed to reconstruct the wallet later so you just have to backup the seed, the generation of the public keys for even larger and further addresses is almost instant procedure for any "i"(tried 1000000!) but when generating the corrosponding private key it will have to iterate throught the sequence of private keys until it reaches "i". for more information check this https://github.com/prusnak/addrgen, In electrum console try "wallet.accounts[0].get_address(0,1000000)", I am sill checking how it works to find out if I can generate the priv key for an "i" without having to calc all the priv keys in the sequence. - lophie
|
Will take me a while to climb up again, But where is a will, there is a way...
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 08:56:31 AM |
|
Orphaned blocks don't make A and B not mutially exclusive...
They do and this is by design to make it impossible to predict next forging accounts 1440 blocks in advance. In TF it is possible by design to predict who will forge the next block. That's exactly what I address in this post. Well, u should roll 2 dice, not 1 die. When u roll 1 die we get P(A or B) = P(A) + P(B).
|
|
|
|
jettico
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 80
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 09:00:39 AM |
|
That's called 'substitution of notions'. You suggested this topic. I didn't address summation of probabilities in my original post. And I don't see how summation of probabilities is connected to it. This is my original post. Where do you see summation of probabilities? Why are we discussing it? Transition to Transparent Forging (at least the way it is described in the wiki) will ruin the axiom of "Rate of return to 1 Nxt is the same regardless how many Nxt you have." Here's why. Steps 3-5 in http://wiki.nxtcrypto.org/wiki/Transparent_Forging from the statistics point of view are equivalent to comparing random numbers uniformly distributed between 0 and 1/N where N is number of coins in the wallet (normalization coefficient is the same for all wallets and is omitted for clarity). Now imagine that you throw a dice with numbers from 1 to 6 and your opponent throws a semi-dice with three possible values from 1 to 3. Who gets the lower number wins the round. In half of the cases you will throw 4-6 and win. In the other half you will have 50:50 chances to win. Thus your chances are 1/2 + 1/4 = 0.75 and his chances are 1/4 = 0.25. Which is 3 times lower. For arbitrary ratio K between two wallets the probability ratio is 2K-1. For example, a wallet with 100k coins is 199 times more likely to forge a block than 1k wallet. I'm talking about three times probability increase for a two times larger wallet. What orphan blocks are you talking about? Here u say this: Now imagine that you throw a dice with numbers from 1 to 6 and your opponent throws a semi-dice with three possible values from 1 to 3. Who gets the lower number wins the round. In half of the cases you will throw 4-6 and win. In the other half you will have 50:50 chances to win. Thus your chances are 1/2 + 1/4 = 0.75 and his chances are 1/4 = 0.25. Which is 3 times lower. You mean this? Thus your chances are 1/2 + 1/4 = 0.75 It is completely different '+' than the '+' from P(A or B) =/= P(A) + P(B) ![Grin](https://bitcointalk.org/Smileys/default/grin.gif) Here I solve a completely defined mathematical problem. Orphans should be taken into account in a different place. And by the way, shouldn't transparent forging eliminate orphans altogether?
|
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 09:01:39 AM |
|
I am sill checking how it works to find out if I can generate the priv key for an "i" without having to calc all the priv keys in the sequence.
Let me know if u figure out how they do this, plz. If u can create a new public key without knowledge of the corresponding private key... it's something awesome. Or insecure, coz 2 consecutive public keys would reveal ur private key.
|
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 09:03:34 AM |
|
It is completely different '+' than the '+' from P(A or B) =/= P(A) + P(B) ![Grin](https://bitcointalk.org/Smileys/default/grin.gif) I know. I say that u roll 1 ( one) die, not 2 ( two) dice. That's how I knew that u had assumed that P(A or B) == P(A) + P(B).
|
|
|
|
kunibopl
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 09:09:09 AM |
|
let's 3D print, engrave with QR-code or something to represent a colored coin. or just 3D print and deploy as marketing article.
Actually.. these could be even more easily produced from coloured transparent acrylic with high-pressure water jet cutting. could you provide 3D data for this cutting or CNC molding?
|
NXT: 5231236538923913892
|
|
|
lophie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 09:13:46 AM |
|
I am sill checking how it works to find out if I can generate the priv key for an "i" without having to calc all the priv keys in the sequence.
Let me know if u figure out how they do this, plz. If u can create a new public key without knowledge of the corresponding private key... it's something awesome. Or insecure, coz 2 consecutive public keys would reveal ur private key. asked that question last year and an electrum dev answered me here https://bitcointalk.org/index.php?topic=332500.0. granted that weakens the security of the keys but not significantly that they become insecure. I agree with you it is futile to search for a way to reproduce the priv of an arbitrary "i" instantly, that would mean the generating hashing function is basically broken, And that means such thing doesn't exist in electrum. Now for nxt that would be awesome it means we can make the passphrase generate numerious account numbers and the client can track them all BUT, there is the "gap limit issue". Basically you cannot know when the addresses will replicate but it is VERY unlikely, so how would you know which addresses are in use? What electrum does is you have to tell it that you use for example up to 100 addresses so when you regenerate the wallet from seed it generates the first 100, But what if you used the 101? The client will have no idea unless you tell it to look beyond 100 keys. So I got you interested? ![Wink](https://bitcointalk.org/Smileys/default/wink.gif) . Lets make this happen. Basically my exchange is delayed mainly because of this, right now NXT is hot on the server which I do not like, Bitcoin is 100% cold and withdrawals are automated with an external requests fetcher. This feature would be so awesome to make this easy for alot of things ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) .
|
Will take me a while to climb up again, But where is a will, there is a way...
|
|
|
jettico
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 80
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 09:14:56 AM |
|
Orphaned blocks don't make A and B not mutially exclusive...
They do and this is by design to make it impossible to predict next forging accounts 1440 blocks in advance. In TF it is possible by design to predict who will forge the next block. That's exactly what I address in this post. Well, u should roll 2 dice, not 1 die. When u roll 1 die we get P(A or B) = P(A) + P(B). 1) The bigger the number of faces on one die is the smaller is the probability of collision. And for sha256 it is about 10^76 faces. So the probability is something like 10 -762) We can redefine the events 'A' and 'B': instead of 'wallet forges a block' we can take 'wallet gets the reward' then the effect of orphans will be eliminated. 3) I don't know the mechanism of resolving an orphan. It should be something like 'let's try again' with really diminishing probability of two or more orphans going one after the other - then my analogy works without any corrections.
|
|
|
|
jettico
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 80
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 09:18:48 AM Last edit: January 29, 2014, 10:50:05 AM by jettico |
|
It is completely different '+' than the '+' from P(A or B) =/= P(A) + P(B) ![Grin](https://bitcointalk.org/Smileys/default/grin.gif) I know. I say that u roll 1 ( one) die, not 2 ( two) dice. That's how I knew that u had assumed that P(A or B) == P(A) + P(B). Steps 3-5 in http://wiki.nxtcrypto.org/wiki/Transparent_Forging from the statistics point of view are equivalent to comparing random numbers uniformly distributed between 0 and 1/N where N is number of coins in the wallet (normalization coefficient is the same for all wallets and is omitted for clarity). Now imagine that you throw a dice with numbers from 1 to 6 and your opponent throws a semi-dice with three possible values from 1 to 3. Who gets the lower number wins the round. In half of the cases you will throw 4-6 and win. In the other half you will have 50:50 chances to win. Thus your chances are 1/2 + 1/4 = 0.75 and his chances are 1/4 = 0.25. Which is 3 times lower. For arbitrary ratio K between two wallets the probability ratio is 2K-1. For example, a wallet with 100k coins is 199 times more likely to forge a block than 1k wallet. Do you know how to count from 1 up to 2, or you hadn't read THAT far? ![Grin](https://bitcointalk.org/Smileys/default/grin.gif) Please try to concentrate and read all three paragraphs one by one. It is not SO difficult ![Wink](https://bitcointalk.org/Smileys/default/wink.gif)
|
|
|
|
NxtChg
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 09:51:47 AM |
|
That would be a nice paperweight. Or cookie cutter. Or nut crusher.
|
|
|
|
Emule
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 09:56:09 AM |
|
You can kick this pig, is going nowhere. One of your big stakeholders in all respect of NXT dumped 1M coins on Bter just to let you all see I can. I hope you all not surprised this pig is not moving. The big stakeholders are offloading for more than a weeks now and they will continue to do so for months so don’t expect price to rise in the coming months, but expect the worse, and don’t give me the“good coins getting wide spread” now sh**, cos coins going nowhere, 60 people own 80% of NXT.!!! You were still not able to release the “official” client (vaporware) so people will have to use other clients instead. Nexern is personally responsible for the price dip (3x he did not deliver) that will follow after my message. Just like CCinvestor I am a whale waiting to dump but as you do not succeed to produce anything I start dump right now. (and don’t give me the “good coins getting wide spread” again) Marketing is a disaster and a joke, you gave 1M to 1 guy that decides to give it out on stupid translations and stupid video. Nowhere have I seen any marketing effort of value! There is no master plan, no guidance and leadership like dogecoin did, lack of that will kill this puppy soon Meanwhile you all discussing stupid things like voting system and color of the coin, NXT get your act together and start do something cos ethereum and emunie will kill you instantly they hit the market. What you all got right now is hot air. It is arrogant to think you have something to offer that others don’t have, if you don’t start get NXT ibn to the open before ethereum and emunie might as well sell all your freaking coins right now cos rhey are worth sh***
and where is that f*** client!
|
|
|
|
lophie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 09:59:41 AM |
|
You can kick this pig, is going nowhere. One of your big stakeholders in all respect of NXT dumped 1M coins on Bter just to let you all see I can. I hope you all not surprised this pig is not moving. The big stakeholders are offloading for more than a weeks now and they will continue to do so for months so don’t expect price to rise in the coming months, but expect the worse, and don’t give me the“good coins getting wide spread” now sh**, cos coins going nowhere, 60 people own 80% of NXT.!!! You were still not able to release the “official” client (vaporware) so people will have to use other clients instead. Nexern is personally responsible for the price dip (3x he did not deliver) that will follow after my message. Just like CCinvestor I am a whale waiting to dump but as you do not succeed to produce anything I start dump right now. (and don’t give me the “good coins getting wide spread” again) Marketing is a disaster and a joke, you gave 1M to 1 guy that decides to give it out on stupid translations and stupid video. Nowhere have I seen any marketing effort of value! There is no master plan, no guidance and leadership like dogecoin did, lack of that will kill this puppy soon Meanwhile you all discussing stupid things like voting system and color of the coin, NXT get your act together and start do something cos ethereum and emunie will kill you instantly they hit the market. What you all got right now is hot air. It is arrogant to think you have something to offer that others don’t have, if you don’t start get NXT ibn to the open before ethereum and emunie might as well sell all your freaking coins right now cos rhey are worth sh***
and where is that f*** client!
This is not a get rich quick scheme, This is the history of decenteralized payment systems in the making, That includes etherium and other coins, want pump and dumps? This is not one sorry, So let the big boys discuss code and math and get the FUCK out of here.
|
Will take me a while to climb up again, But where is a will, there is a way...
|
|
|
gvans
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 10:00:03 AM |
|
let's 3D print, engrave with QR-code or something to represent a colored coin. or just 3D print and deploy as marketing article.
Actually.. these could be even more easily produced from coloured transparent acrylic with high-pressure water jet cutting. could you provide 3D data for this cutting or CNC molding? I could! It's propably iges-format you're looking for. Which size, three options: paperweight 10x40x90 mm, sidetable 100x400x900 mm or modern art landmark 10000 x 40000 x 90000 mm? ![Grin](https://bitcointalk.org/Smileys/default/grin.gif)
|
|
|
|
LiQio
Legendary
Offline
Activity: 1181
Merit: 1002
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 10:15:48 AM |
|
... I am a whale waiting to dump but as you do not succeed to produce anything I start dump right now. ...
prove it - in the meantime don't behave like a mule ![Wink](https://bitcointalk.org/Smileys/default/wink.gif)
|
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 10:18:37 AM |
|
Now for nxt that would be awesome it means we can make the passphrase generate numerious account numbers and the client can track them all BUT, there is the "gap limit issue". Basically you cannot know when the addresses will replicate but it is VERY unlikely, so how would you know which addresses are in use? What electrum does is you have to tell it that you use for example up to 100 addresses so when you regenerate the wallet from seed it generates the first 100, But what if you used the 101? The client will have no idea unless you tell it to look beyond 100 keys. So I got you interested? ![Wink](https://bitcointalk.org/Smileys/default/wink.gif) . Lets make this happen. Basically my exchange is delayed mainly because of this, right now NXT is hot on the server which I do not like, Bitcoin is 100% cold and withdrawals are automated with an external requests fetcher. This feature would be so awesome to make this easy for alot of things ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) . In Nxt u could do it via Account Control with the rate of SHA256 hashing (100'000 accounts per second). Not sure though if it's the same functionality as in Electrum...
|
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 10:20:42 AM |
|
1) The bigger the number of faces on one die is the smaller is the probability of collision. And for sha256 it is about 10^76 faces. So the probability is something like 10-76 2) We can redefine the events 'A' and 'B': instead of 'wallet forges a block' we can take 'wallet gets the reward' then the effect of orphans will be eliminated. 3) I don't know the mechanism of resolving an orphan. It should be something like 'let's try again' with really diminishing probability of two or more orphans going one after the other - then my analogy works without any corrections.
1. It's ~1/100 coz of orphane blocks 2. Then u can't rely on assumption that ROI ~ balance 3. Mechanism is stochastic and heavily depends on network topology
|
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 10:22:54 AM |
|
Please try to concentrate and read all three paragraphs right one after another. It is not SO difficult ![Wink](https://bitcointalk.org/Smileys/default/wink.gif) I can't pass the 2nd paragraph coz writing the code at the same time. I let someone else to continue the discussion instead of me.
|
|
|
|
Emule
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
January 29, 2014, 10:23:39 AM |
|
You can kick this pig, is going nowhere. One of your big stakeholders in all respect of NXT dumped 1M coins on Bter just to let you all see I can. I hope you all not surprised this pig is not moving. The big stakeholders are offloading for more than a weeks now and they will continue to do so for months so don’t expect price to rise in the coming months, but expect the worse, and don’t give me the“good coins getting wide spread” now sh**, cos coins going nowhere, 60 people own 80% of NXT.!!! You were still not able to release the “official” client (vaporware) so people will have to use other clients instead. Nexern is personally responsible for the price dip (3x he did not deliver) that will follow after my message. Just like CCinvestor I am a whale waiting to dump but as you do not succeed to produce anything I start dump right now. (and don’t give me the “good coins getting wide spread” again) Marketing is a disaster and a joke, you gave 1M to 1 guy that decides to give it out on stupid translations and stupid video. Nowhere have I seen any marketing effort of value! There is no master plan, no guidance and leadership like dogecoin did, lack of that will kill this puppy soon Meanwhile you all discussing stupid things like voting system and color of the coin, NXT get your act together and start do something cos ethereum and emunie will kill you instantly they hit the market. What you all got right now is hot air. It is arrogant to think you have something to offer that others don’t have, if you don’t start get NXT ibn to the open before ethereum and emunie might as well sell all your freaking coins right now cos rhey are worth sh***
and where is that f*** client!
This is not a get rich quick scheme, This is the history of decenteralized payment systems in the making, That includes etherium and other coins, want pump and dumps? This is not one sorry, So let the big boys discuss code and math and get the FUCK out of here. You have no right to speak, I know you dumping millions of coins below market price so STFU. You personally ar the reason price dont get up!
|
|
|
|
|