elective-stereophonic
elective-stereophonic
Some thoughts on arguments of PoW guys  
Please login or register.

Login with username, password and session length
Advanced search  

News:

Latest Stable Nxt Client: Nxt 1.12.2

Pages: [1] 2  All

Author Topic: Some thoughts on arguments of PoW guys  (Read 5641 times)

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Some thoughts on arguments of PoW guys
« on: May 28, 2014, 09:49:06 am »

From https://bitcointalk.org/index.php?topic=584703.msg6982574#msg6982574

Quote
You haven't eliminated the fact that there is absolutely no reason not to sign all possible chains in event of a fork.
Can a penalty fix this problem? It seems to me it can.

Quote
Bitcoin in it's short history had had ~40M txs. Adding 32 bytes to each transaction would have bloated the blockchain by another 1.2 GB.
So PoW guys were saying that current blockchain size was not an issue but now they want to save 1.2 GiB? It's even funnier that SPV is not mentioned.

Quote
One can perform a reorg with no cost by simply acquiring private keys which at one time had (past tense) a majority of the stake.
I wouldn't use the word "simply", unless someone could show 2 cases when ppl sell their old private keys.

Quote
There is a chance you will out mine/mint/forge the rest of the network.
Sure, odds r the same as to win 1 million dollars in a lottery.

Quote
In NXT when you forge a block which is the best at current height but doesn't enable you to forge the next block (due to incompatible block signature) there is no reason to not use computing power to try and find an alternative which enables you to forge the next one as well.
Could someone show an example? Right now the quoted phrase looks as the beginning of a Sci-Fi novel.

Quote
The chance may be low but with no cost it would be foolish to not attempt it.
No cost? Right, 1 (one) SHA256 operation costs nothing, 1015 * [nothing] is still nothing. Bitcoin network spends no resources, neat.

Quote
With no cost, forks will last longer, and reorgs will be deeper.
Penalty makes it very costly to extend multiple forks.

Quote
So saying "put a hash of a recent block in the tx and nothing at stake is a strawman" misses the larger context.  If there is no cost to attack the network there is no reason to NOT attack the network.
How many times should I repeat that there will be a penalty?

PS: A bonus thought. Nxt consensus rule solves the problem of Bitcoin Fork 2013. Nxt to Bitcoin as Einstein's theory to Newton's theory.
Logged

Fatih87SK

  • Hero Member
  • *****
  • Karma: +127/-36
  • Offline Offline
  • Posts: 2206
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #1 on: May 28, 2014, 10:04:38 am »

Can we share this URL to that thread at Bitcointalk to invite people over here to discuss further?

And can we quote this post of you to that thread as well?
Logged

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #2 on: May 28, 2014, 10:11:12 am »

Can we share this URL to that thread at Bitcointalk to invite people over here to discuss further?

And can we quote this post of you to that thread as well?

I see that we can't discuss this issue with PoW boys. They won't admit that N@S "problem" is solved. This is their last bastion. In the OP I showed what tricks they use.

PS: U can share this URL, of coz.
Logged

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #3 on: May 29, 2014, 06:47:30 am »

Quote from: jonald_fyookball link=https://bitcointalk.org/index.php?topic=584703.msg7003704#msg7003704
The penalty suggestion would need to be much further defined if you want a serious discussion about it.

There is only 1 penalty for forgers - they r not allowed to forge for the next 1440 blocks. Main goal is not to punish but rather to "disable" them.
Logged

mczarnek

  • Hero Member
  • *****
  • Karma: +68/-4
  • Offline Offline
  • Posts: 898
    • View Profile
    • Nxt Place - Craigslist for Nxt
Re: Some thoughts on arguments of PoW guys
« Reply #4 on: May 29, 2014, 12:41:03 pm »

Quote from: jonald_fyookball link=https://bitcointalk.org/index.php?topic=584703.msg7003704#msg7003704
The penalty suggestion would need to be much further defined if you want a serious discussion about it.

There is only 1 penalty for forgers - they r not allowed to forge for the next 1440 blocks. Main goal is not to punish but rather to "disable" them.

Let's say I'm attacking the network and decide to split my Nxt across 1000+ accounts.  Meaning anytime I forge from one and am banned, it barely effects me because I'll just forge from another one of the accounts next time.

Someone buys 1% of Nxt and tries this, you only ban 0.1% of his Nxt if split across multiple accounts, barely affecting him and I could see it causing problems.  That's why I would vote for a higher cap for the minimum amount of Nxt someone needs in their account in order to forge from it.  Or is there another solution?
Logged
NXT Organization: Tech
Donations greatly appreciated: NXT-DWVJ-G89C-RHNL-6QW6Q

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #5 on: May 29, 2014, 12:52:22 pm »

Let's say I'm attacking the network and decide to split my Nxt across 1000+ accounts.  Meaning anytime I forge from one and am banned, it barely effects me because I'll just forge from another one of the accounts next time.

Someone buys 1% of Nxt and tries this, you only ban 0.1% of his Nxt if split across multiple accounts, barely affecting him and I could see it causing problems.  That's why I would vote for a higher cap for the minimum amount of Nxt someone needs in their account in order to forge from it.  Or is there another solution?

The penalty is not for punishing the attacker. It's retroactive and simply invalidates all blocks of the forger including both the blocks belonging to contradicting branches.
Logged

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #6 on: May 29, 2014, 04:37:44 pm »

Quote from: DeathAndTaxes link=https://bitcointalk.org/index.php?topic=584703.msg7016078#msg7016078
A penalty for an inactive but otherwise non malicious forgers is useful but it doesn't help in a 51% attack.

By definition in a 51% attack the attacker IS NOT mining the main chain so a penalty that doesn't allow him to mine on the chain he already isn't mining isn't much of a penalty is it?  On the other hand in the attack chain it will be the legit miners who failed to mint a block and they will be subject to the penalty.  When the attack chain is longer and the attacker broadcasts it, then it becomes the longest chain and some or all of legit miners will be penalized for up to 1440 blocks.

In Nxt both the chains (legit and hidden) will have the same cumulative difficulties coz forging power of penalized forgers is delegated to the others and total power is bumped back to 100%. But the hidden chain won't have transactions of the economic cluster and this is where extra consensus rule becomes handy.

The attacker has ~10 minutes (may be changed after we collect stats) to reveal his chain. After 10 mins all transactions will be set in stone and 51% attacks won't be possible at all. If the attacker combines 51% attack with eclipse attack then the victim will notice that the recent blocks don't have transactions of the cluster (actually there could be some coz of time difference and corrupt participants of the cluster but the ratio will be much lower than threshold).

Original design assumes that every block there will be 2-3 forgers who have to compete at certain block height. So every block at least 1 forger will be penalized. Number of forks within 10 min window is supposed to be high and users shouldn't rely on low number of confirmations.

Some forgers will attach their blocks to dead branches but they shouldn't reforge blocks after blockchain reorgs coz other forgers may see contradicting blocks (signed by the same account but belonging to different branches) and report them by including both the blocks into their own chains.
Logged

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #7 on: May 29, 2014, 05:14:41 pm »

Quote from: DeathAndTaxes link=https://bitcointalk.org/index.php?topic=584703.msg7017917#msg7017917
Over time but not instantly and thus the chain with the majority of the stake will be the longest.  If it wasn't true then someone with 1% of the stake could form a chain as long as the network with 99% of the stake (99% of the stake is penalized for not mining on the attack chain and eventually the 1% has 100% of the active stake).

Over time, right, but in the end 1% will have 100% forging power. If a malevolent forger creates a longer chain within 10 min window then his chain will be adopted and he will be considered a benevolent forger. In other case his chain will be ignored.


Quote from: DeathAndTaxes link=https://bitcointalk.org/index.php?topic=584703.msg7017917#msg7017917
This is where you get into trusted node voodoo.  The attacker can include tx from the "honest" chain.  He doesn't need to (nor can he) double spend all transactions.  The attacker can broadcast double spends which are picked up by the main chain and then include the legit tx in the attack chain. 

The idea that 100% of the nodes will always be online, and always know which chain is the best is dubious.   If that were possible then you wouldn't even need PoS or minting at all.  Nodes would simply agree which tx are valid and confirmations would be nearly instantaneous.  So what happens when there are two chains A & B.  In A BTC-E is double spent a small fortune and in B BitStamp is double spent a small fortune?  One of them is just going to take a massive loss for the team?  No in a scenario where longest chain is ignored and the EC doesn't reach 100% consensus either you have a network split or the attacker is able to double spend anyways.  Sorry BTC-E your tx was in the longest chain but the board of trusted elders has decided that the inferior chain is the "best" and you lose to the double spend. 

Not trusted node voodoo, it's economic cluster voodoo. Without it any money makes no sense at all.

U need PoS to answer such questions as "Who will get transaction fees", "What transactions will be included into blocks", "What order the transactions will be processed on". Without PoS we would get Ripple (lol).

Double-spending is not possible (almost) if u wait long enough. In Bitcoin function of reliability of transactions is smooth - it grows over time depending on number of confirmations. In Nxt this function jumps from 1% straight to 99% when number of confirmation exceeds 10 ("1" and "99" used as an example and depend on network topology). Double-spends will always be possible if network splits and this is applied to Bitcoin too. Unlike Bitcoin, Nxt has an advantage coz it can detect such splits by analyzing recent blocks.
Logged

mthcl

  • Hero Member
  • *****
  • Karma: +96/-8
  • Offline Offline
  • Posts: 562
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #8 on: May 29, 2014, 07:03:06 pm »

It's clear that the PoW guys are discussing something else, not the Nxt TF. On the other hand, there is no reason for us to disclose all the concrete details of TF until it's really ready.
Logged

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #9 on: May 29, 2014, 07:20:46 pm »

It's clear that the PoW guys are discussing something else, not the Nxt TF. On the other hand, there is no reason for us to disclose all the concrete details of TF until it's really ready.

Hehe, I noticed that we discuss different things. But I keep talking to get more reviews of BCNext's ideas.
Logged

Evan

  • Sr. Member
  • ****
  • Karma: +18/-2
  • Offline Offline
  • Posts: 333
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #10 on: May 31, 2014, 03:54:06 am »

After 10 mins all transactions will be set in stone

What does this mean exactly?
Logged
8897015223734827770
NXT-BYRU-8NKZ-PEYJ-9E6YQ

Evan

  • Sr. Member
  • ****
  • Karma: +18/-2
  • Offline Offline
  • Posts: 333
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #11 on: May 31, 2014, 04:01:00 am »

Quote
One can perform a reorg with no cost by simply acquiring private keys which at one time had (past tense) a majority of the stake.
I wouldn't use the word "simply", unless someone could show 2 cases when ppl sell their old private keys.

I don't understand the reorg. Let's say someone get some private keys, what can he do with them? Blockchain is a chain, he cannot just change some blocks without invaliding later blocks. right?
Logged
8897015223734827770
NXT-BYRU-8NKZ-PEYJ-9E6YQ

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #12 on: May 31, 2014, 06:39:00 am »

What does this mean exactly?

It means that nothing will change blocks older than 10 min.
Logged

Evan

  • Sr. Member
  • ****
  • Karma: +18/-2
  • Offline Offline
  • Posts: 333
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #13 on: May 31, 2014, 10:01:48 am »

It means that nothing will change blocks older than 10 min.
what makes it different from blocks younger than 10 min?

If there is two branches, both include blocks older than 10 min since they fork, then what will happen? how do they different from forks younger than 10 min?
Logged
8897015223734827770
NXT-BYRU-8NKZ-PEYJ-9E6YQ

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #14 on: May 31, 2014, 10:18:01 am »

what makes it different from blocks younger than 10 min?

If there is two branches, both include blocks older than 10 min since they fork, then what will happen? how do they different from forks younger than 10 min?

After 10 mins "longest chain" rule will be replaced with "economic majority" rule. There can be millions branches with blocks older than 10 min, but only one will be accepted by the cluster.
Logged

Daedelus

  • Hero Member
  • *****
  • Karma: +230/-12
  • Offline Offline
  • Posts: 3280
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #15 on: May 31, 2014, 10:21:23 am »

what makes it different from blocks younger than 10 min?

If there is two branches, both include blocks older than 10 min since they fork, then what will happen? how do they different from forks younger than 10 min?

After 10 mins "longest chain" rule will be replaced with "economic majority" rule. There can be millions branches with blocks older than 10 min, but only one will be accepted by the cluster.

Out of interest, are you still working from BCNexts notes/roadmap with EC?
Logged
NXT: NXT-4CS7-S4N5-PTH5-A8R2Q

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #16 on: May 31, 2014, 10:28:12 am »

Out of interest, are you still working from BCNexts notes/roadmap with EC?

Yes.
Logged

Evan

  • Sr. Member
  • ****
  • Karma: +18/-2
  • Offline Offline
  • Posts: 333
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #17 on: May 31, 2014, 10:48:00 am »

After 10 mins "longest chain" rule will be replaced with "economic majority" rule. There can be millions branches with blocks older than 10 min, but only one will be accepted by the cluster.
I thought the "economic majority" rule only replaces the "longest chain" rule at a much longer time period. Is there any consideration to make it as short as 10 minutes? doesn't that make Nxt rely on the "economic majority" too much?
Logged
8897015223734827770
NXT-BYRU-8NKZ-PEYJ-9E6YQ

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #18 on: May 31, 2014, 10:50:06 am »

I thought the "economic majority" rule only replaces the "longest chain" rule at a much longer time period. Is there any consideration to make it as short as 10 minutes? doesn't that make Nxt rely on the "economic majority" too much?

10 min is just an estimated period, only stats will tell us what number we should use. For convenience I talk about "10 minutes".
Logged

Evan

  • Sr. Member
  • ****
  • Karma: +18/-2
  • Offline Offline
  • Posts: 333
    • View Profile
Re: Some thoughts on arguments of PoW guys
« Reply #19 on: May 31, 2014, 11:01:42 am »

10 min is just an estimated period, only stats will tell us what number we should use. For convenience I talk about "10 minutes".
I thought it is something like 1440 blocks or at least 1 hour. At my first thought, 10 minutes are really very short. Maybe I missed something.
Logged
8897015223734827770
NXT-BYRU-8NKZ-PEYJ-9E6YQ
Pages: [1] 2  All
 

elective-stereophonic
elective-stereophonic
assembly
assembly