Hard to answer, Transparent Forging is not implemented yet. Without TF Ix's attacks would be successful.
Are you saying that after TF is fully implemented, the new guy can tell who is the bad guy from the blockchains? How? |
|
|
|
Right.
Got it. Thank you very much. |
|
|
|
The new guy (who has no actual blockchain yet) must choose "well-known nodes" to get the blockchain from. He won't accept ur secret chain coz he will get the legit one from other nodes. U have to "isolate" him, but the same attack is applied to Bitcoin as well and can't be counteracted if the new guy has no information about what nodes r legit.
The new guy may be not necessarily brand new, he may have the part of the blockchain before the bad guy forging the secret chain. So, if the new guy has chosen the bad guy as his "well-know nodes", he will go on with the secret chain, and the network will be forked. This is true for both PoW and PoS. right? |
|
|
|
Then what stops the adversary and tells the account that is offline during the procedure which branch is legit when it comes back online?
Explain ur attack step by step if u want to get a detailed explanation. Unfortunatelly, most of guys ask general questions and become upset when get general answers. Ok, I'll try later after giving it more thought. Well, just try to understand how an account choose the legit branch Assume the attacker has 51%+ of the stake Case 1: the attacker forges on both the legit and secret branch - Consider the attacker has two computer, computer A is connected to the internet and computer B is offline. - The attacker use his account to forge on the legit branch as usual with computer A, and at the same time he also use his account to forge on the secret branch with computer B. - After some time, he connects the computer B to the internet and broadcast his secret branch to other accounts. he can do this with computer A online or turn computer A offline. Case 2: the attacker skip his turn on the legit branch and forges only on the secret branch - The attacker use only one computer. - At first, the computer is connect to the internet, the attacker forges on the legit branch as usual. - At some time point, he turns the computer offline, begins to forge his secret branch. - After some time, he turns the computer online In case 1, because the legit branch has more stake involved in the forging than the secret one, every account has no problem to tell which one is valid. right? In case 2, though more stake is involved in the secret branch, every account that forges on the legit branch know that the attacker's account has been penalized, so they will not accept the secret branch. but for an account that was offline when all this happened, how does it know which account(s) has been penalized for forging on a secret branch when it comes back online? the secret branch can also be interpreted as the legit because all other accounts have been forging on a secret branch. |
|
|
|
Then what stops the adversary and tells the account that is offline during the procedure which branch is legit when it comes back online?
Explain ur attack step by step if u want to get a detailed explanation. Unfortunatelly, most of guys ask general questions and become upset when get general answers. Ok, I'll try later after giving it more thought. |
|
|
|
Now the mechanism that was stopping forging is disabled.
Then what stops the adversary and tells the account that is offline during the procedure which branch is legit when it comes back online? |
|
|
|
He would have only 15 sec to release his "secret" chain. 16 sec later and it's too late to do an attack. This can never be true because a previously unconnected or not-well-connected node will not have the "correct" notion of order and time--it must rely on other nodes. There is no guarantee that those nodes will be honest, and it would only take one dishonest node to present a conflicting viewpoint for which the unconnected node has no way of determining the correct chain. I think I understand lx's concern. It is related to the following description of transparent mining: 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).
Whether the adversary works on only the secret branch and gets penalized or works on both branches without being penalized, the two branches both have 100% "hashing" power. So, without the advanced consensus mechanisms, the two branches will have the same length and for an account being offline during the procedure, it cannot recognize the legit branch from the one generated secretly. This issue is supposed to be resolved by the advanced consensus mechanisms, Regarding the advanced consensus mechanisms, we know that, when one account is left on a fork alone, it will generating all the blocks, but after some blocks, it stops. Only after some other accounts join in, can it continue to generate blocks. I believe this is caused by the advanced consensus mechanisms. It is obvious that the adversary cannot attack Nxt by working on a secret branch using only a few accounts. But that's all what I know about it. How will the advanced consensus mechanisms counteract if the adversary introduces enough accounts for the secret branch? Can you explain it, CfB? |
|
|
|
if you have a lot of pos shares then you have less incentive to spend vs hoard thus mining more by hoarding
forging/mining income for PoS is only comparable to bank interest. how much incentive do you think bank interest gives people to hoard instead of spending? |
|
|
|
in your post: Everything built on top of Bitcoin must inherit Bitcoin's speed, its proof-of-work legacy of waste, and its transaction processing model. Nxt has ninenone of these, so everything built in top of Nxt is lighter, cheaper, more energy efficient, and faster. please fix it |
|
|
|
|
have we finished discussing the "Nxt only" thing on Nxt AE? (Sorry, I cannot follow every post right now)
I think it has been made very clear by some of the previous posts that we should not do the "Nxt only" thing
1. It would limit the max order subject to the market capitalization of Nxt
2. It would lead to a clone of Nxt without the "Nxt only" limitation. Think you are a trader, which one would you choose? I think a lot of people would go to the platform without any limitation.
We should make Nxt as free as possible, the freedom will gain Nxt a much more value in a long run. the limitation will harm Nxt. We should NEVER force our user to do something.
|
|
|
|
There's just 1 problem with a list of predefined categories though; it's centralized.
As far as I understand, the DGS (digital goods store) also won't have categories. CFB does not like this idea of centralization. (even though it would help the user)
how about storing the information about categories locally in a file, people can share the file if they like. This could again be a task for a service provider. He checks each asset and adds it to a category. not necessarily be the task for service providers, every one can tag the assets and share their version of the file. also good if there are service providers doing the task. |
|
|
|
There's just 1 problem with a list of predefined categories though; it's centralized.
As far as I understand, the DGS (digital goods store) also won't have categories. CFB does not like this idea of centralization. (even though it would help the user)
how about storing the information about categories locally in a file, people can share the file if they like. |
|
|
|
I could post yours! ;-)
you don't know mine  It's not a problem as long as you don't do an outgoing transaction in the clone blockchain.
then you cannot sell the Nxt clone for Nxt, what's the benefit? |
|
|
|
@CfB: Can we replay transactions from NXT in a clone and vice versa?
Yes, if a clone doesn't change transaction format. So anyone who uses the same account in both cryptos can lose NXTs.If this is really the case, it could be very dangerous for those who post their Nxt account there. |
|
|
|
That guy triggered one of the traps. It's sad that people paid money for coins that will become useless... I´ve provided an address. Let´s see if he will be able to send some coins to it  I posted a hash there, too. cost nothing after all. Let's see whether he can get it running at first. Not a single Nxt clone is actually launched yet, as far as I know |
|
|
|
|
I am doing some marketing on my own
hope it will help
|
|
|
|
what about adding a field called label/nickname (whatever) which is not required to be unique and can be changed.
We already have it - description field. just thought of that I agree that Ian's proposal should be implemented by client software not by the core |
|
|
|
I *gave* you a better solution.
Allow multiple Assets with the same name.
This will cause average Joe to go "what the fuck" and stop him losing his NXT.
Why do you need to ignore my solution (your bosses are "telling you to ignore it")?
I think ur solution is worse. People will see legit "CIYAM" and fake "CIYAM", they won't read numbers, coz numbers r much harder to "read". what about adding a field called label/nickname (whatever) which is not required to be unique and can be changed. |
|
|
|
BTW, when we switch thing to binary, how many TPS can we get? any ideas?
I would do it gradually: 10 tps, then 100 tps, then 1000 tps. See, guys, Nxt will be very successful. I never worried about the price of Nxt, In fact, I always hope it can hold back for some time so that I can buy more like what BCNext said, don't pay much attention to the price Yes, we indeed need more core devs. |
|
|
|
|
Show Posts
