 Show Posts
|
|
Pages: « 1 2 3 [4] 5 »
|
Hello Yj1190590, 1.According to my understanding, the approvals stored in each block-approval-block are the approvals for the current block. So where are the conflicted approvals stored? If they are not stored in the history blocks, how could you tell a stackholder approves conflicted blocks. In other words, according to what should you punish the double-approving behavior to prevent "nothing at stake" problem?
Multiple block creators publish block-approval-blocks for their own created blocks. While they are sent to all the nodes, block-approval-blocks are not stored themselves in the chain, their contents are. Conflicting block approval is detected by looking at all block-approval-blocks published on the network in the same slots. It is the next block creator's responsibility to check for these conflicts and not use contents of a block-approval-block that contains such a conflict. In addition, the approvers for the next block will also check for these conflicts and would not approve blocks that contains a conflicting approval from the previous slot. The punishment for a conflicting-approver is that they do not earn approval award since their approvals do not go in the chain. 2. About the instant finality. If you determine the realchain so quickly, how could you prevent the blockchain from spliting into different forks for ever when there happens something that separate the network temporaryly, like intercontinental optical fiber cable fracture?
Once there is >50% stake approval on a successor block, there can only be one and only one predecessor block under the honest majority stake assumption even with network outage. If in case a block cannot receive >50% approval, then no new blocks will get published. Those slots will go empty until a >50% approval is achieved. 3. About the history attacks. Why don't you just mention that the protocol has instant finality? Because that resolved all the history attacks if it works.
The protocol achieves finality after 1 block - not instant finality. PBFT and related protocols that build consensus on transaction level and publish a single block per slot have the real instant finality. This near instant finality helps with Bribe attack but not with History (aka Costless Simulation) attack. It is because the "finality" is in the short timespans where the stakes of the parties have not changed dramatically. History attack, on the other hand, is powered by dramatic stake changes which can only occur in the longer timespans. Hope it helps. Shunsai |
|
|
|
Hello Ix, If you're interested, you can check out my signature for a link to my whitepaper on the Decrits consensus algorithm which is relatively similar to yours (with an identical long range attack defense), and is 5+ years old.  I am very interested and will read it shortly. Thanks for the link. Shunsai |
|
|
|
Hello Paul, "Proof of approval is an approximation of a PoS chain in which every block is signed by all the stake in the system"?
How about this? Proof-of-Approval is a stake based protocol where every block is signed by the stake majority and every epoch is signed by almost the entire stake in the system. Regards, Shunsai |
|
|
|
Hello D5000, Only a little observation (and that's why I comment here in the old thread): The attack monsterer described (old majority stakeholder(s) taking over the chain with a double spend) is essentially the same attack than a profitable 50+% PoS attack. In a profitable 50+% attack, you will wait for a few blocks selling your stake first (most likely in an altcoin exchange as they confirm the "reception" of the coins relatively fast) and only then try to "take over" the new chain with a prepared attack chain. Otherwise, very likely your stake would decrease in value. So, also in this attack, in the moment you take over, you don't own stake anymore (or at least, you aren't forced to own any stake).
The difference between "long-range attack" and "profitable 50+% attack" be the time between the actual double spend and the "takeover" (when the attack chain replaces the "honest" chain) but the attack mechanism is the same. But monsterer is absolutely right and this is the reason why PoS protocols are to be considered - until this is solved - weaker than PoW protocols. I consider this attack unpractical in a chain with many stakeholders and reorg limits or "rolling checkpoints", and it's likely to be very expensive, but it may be considerably cheaper than a PoW 50+% attack.
It is an interesting attack scenario, not sure if there is an official name for it, but let's continue calling it, profitable 50+% attack. Stake based and derivative chains are more susceptible to it because, unlike PoW, a large number of blocks can be built in a short time. The attacker must wait for a certain degree of finality (after his stake transfer transaction is placed in the chain) before launching this attack. The more uncertain a chain's finality is, the harder it is for the attacker since they have to wait longer. Proof-of-Approval's (v2) features that may help or hurt against this attack are - (hurt) Near instant finality - the attacker can launch the attack quickly after transferring - Max stake transfer is limited per block so that entire 50+% stake cannot be transferred in 1 block. If the limit is set to 0.5%of total network stake, it would take the attacker 100 or more blocks to transfer. - If the attack fork is larger than an epoch (epoch length vs stake transfer amounts are not yet finalized), then epoch approvals in the "real" chain would be larger than that in the attack chain and the real chain wins. - If the attack fork is smaller than an epoch, the fork determination procedure as currently specified in v2, chooses higher stored approval in the topmost block. This makes such an attack possible. A limit that has been considered and perhaps should be included in Proof-of-Approval is the max stake transfer per epoch. If that limit is set to something like 25% of stake, then this attack cannot succeed because of epoch approvals. So, I plan to add stake transfer limit per epoch. Shunsai |
|
|
|
Hello Paul, I think Abdulkhaliq123 was talking about from an incentive PoV - why be a block approver when you can get the same reward for doing nothing, essentially?
If you reward each upoc approver the same amount as each block approver, then no one will approve blocks. On the other hand, if you divide the epoc reward between all nodes, then the amount of the reward will be zero because you can just sybil attack it.
Block approval award is separate from epoch approval in approximately the same amount (over an epoch). For example, let's say 1. 100 blocks in an epoch 2. block approval award 1 coin * approver's network stake fraction 3. epoch approval award 100 coins * approver's network stake fraction So, if an approver has 10% stake in the network 1. Approving epoch awards 100 * .1 = 10 coins 2. In addition, if they approve 90 blocks in that epoch, they additionally get 90 * 1 * .1 = 9 coins 3. If they also create one or more blocks in the epoch, they get creator awards + transaction fees Hope it makes is clearer. Shunsai |
|
|
|
Hello Abdulkhaliq123, In that case, why approve blocks at all? There are two different problems that need solvin  Approving blocks provides another very desirable property - finality. In fact, the transactions are stable once a single block has been deposited on it. Regards, Shunsai |
|
|
|
|
They should have clarified their fees upfront before the start of the call. What they did is unethical.
You may be better off finding developers on your own from Github :-)
|
|
|
|
Hello Paul, I'd like some clarification on the epoc rewards; every node on the network will receive an award for approving an epoc and said award is the same as the block approval reward?
In that case, why approve blocks at all?
There are two different problems that need solving 1. Block approval - where faster is better. If it is slow, the blockchain will have low transaction rate. 2. Defense against History attack - where higher stake participation provides stronger defense. Since some parties will be on slow connections, they may never be able to approve blocks. Epoch approval solves problem 2. In addition, wouldn't this cause the currency to inflate out of control as everyone on the network receives the same award?
Yes the currency inflation rate would be higher - about 2x of what without the epoch approval. Determination of the exact reward amount needs additional work. Epoch award needs to be large enough for small stakeholders to feel that participation is worthwhile. Block approval award is mostly expected to go to large stakeholders. edit: also, wouldn't evidence of the epoc rewards result in a massive amount of data (needing to be stored), since every node on the network will participate?
Yes, epoch rewards will result in a large amount of data storage once per epoch. The epoch period does not have to be very small - it can be around 500-1000 slots. Regards, Shunsai |
|
|
|
Hello Aliashraf, I believe I have a good solution to this History attack (aka Costless Simulation attack) issue. While most PoS systems are relying on the so called "Weak Subjectivity" solution, this solution is purely objective. I am writing it up now and will post it here soon for everyone's review. I can't overstate the value of the feedback provided in this forum. Thanks everyone! Regards, Shunsai Hello Paul, In the following example, say I had a majority of stake at block A. Then at B0, C0 all the way up to head block H0, I sell off my majority of stake, understanding that there is an upper limit on stake transfer. A - B0 - C0 - D0 - H0
\ B1 - C1 - D1 - H1
Then, I create a double spend of my stake using the historical private keys, which I still own, thus sending it to myself in a fork continuing fork B1 - C1 - D1 that I create and approve myself (majority stakeholder). This is not a 50% attack, because I don't own the value held in the private keys anymore, the cost for doing this is nothing. Why doesn't the network just accept this fork as canon, given that its the same length (or longer if I chose) than the genuine fork? Any single person owning (or colluding with) a quorum (majority) stake is a problem because that person (or colluders) can rewrite the chain as they choose. This situation is essentially the same as an adversary owning a quorum stake which is the limit of the protocol. This is not how I would categorise the presented scenario. This is the NaS problem in it's essence. The attacker, as a historical majority stakeholder, can still take over the blockchain at any point in the future after he has sold his stake. So, at the current block height, said attacker owns 0 stake, yet he can still rewrite the entire chain due to this NaS problem. You are correct in pointing out an issue here which must be solved. I'd describe the issue slightly differently even without any single party holding a quorum stake. An adversary buys empty private keys (with zero current stake) that once upon a time did hold a quorum stake in the blockchain. That adversary can start rewriting blocks from the time the keys did hold the quorum stake to the present - completely rewriting all the transactions. I do need to think about this problem. Thanks for pointing it out. Regards, Shunsai Shunsai, I appreciate your hard work and feel deep sympathy with you, a person who knows something is wrong and current situation with crypto is not good enough because of the classical paradox between security and decentralization on one side and scalability on the other side. The thing is POS, even your 2nd phase commit version, is not proved to be an answer because of problems like Nothing at Stake mentioned by Paul here and a lot more. It is not just a simple 'issue' to 'think about' it has been discussed over and over and no answer available yet, other than complicating everything to make it harder for both adversaries and innocent participants and/or leading to a sophisticated protocol vulnerable to implementation bugs and a series of other limitations. Just ugly, far from elegant solutions too complicated, too fragile. I have gone through this before and have chosen another approach: forgetting about the answers and re-thinking questions. For instance, let's take a look at your 'permissionless' index (whether participating in a protocol needs permission from an authority or not): What if one needs permission but not from a centralized authority but a decentralized entity like a curated list which is maintained in a classic PoW base blockchain tp participate in another protocol that carries the much burden of the transaction load, being secured by a trivial, low cost algorithm ? See? A lot of possibilities out there and nothing is 'obvious' and, as I'm used to say, you shouldn't stick with common sense. |
|
|
|
Hello Slava79, Did you read Tendermint white paper ? I scanned article and the core of it looks quite similar to Tendermint's PBFT consensus. Could you please provide a comparison ?
Comparing Proof-of-Approval with Tendermint1. Tendermint is a PBFT ( http://pmg.csail.mit.edu/papers/osdi99.pdf) like algorithm where all nodes together agree on transactions and their order that would be put in a block. Therefore, all nodes together come up with a single block to be inserted in the blockchain. In Proof-of-Approval, just like Bitcoin, individual nodes bundle transactions on their own and the network chooses one of those bundles. 2. The "validators" in Tendermint volunteer to put some of their money as collateral that will earn them rewards but could also be confiscated. Validators are similar to those in Casper the Finality Gadget of Ethereum ( https://arxiv.org/pdf/1710.09437.pdf). This article compares the two https://blog.cosmos.network/consensus-compare-casper-vs-tendermint-6df154ad56ae. 3. Tendermint a single "proposer" who proposes the blocks. The proposer is selected by some process from the validators in round robin. 4. Tendermint transactions achieve instant finality while Proof-of-Approval takes 1 block to finality. 5. Tendermint is likely to have less "liveness" than a block creating protocol like Bitcoin or Proof-of-Approval. 6. The amount need to takeover Tendermint is 1/3 of the collateral amount while the amount to takeover Proof-of-Approval blockchain would be almost half (just under half) of the value of the blockchain. Regards, Shunsai |
|
|
|
Hello Bonzenboss, I read your technical paper and from what I understood it seems to be quite a good idea but of course as all protocols it first needs to be tested in action to see how good it actually works.
You are 100% correct that the testing with implementation is required. It turns out that even that isn't sufficient since an adversary will keep his/her attack vectors hidden till the blockchain actually succeeds when the attacks can earn him/her maximum profit. I think a peer review (what's being conducted here) can bring out many issues that a testing without serious adversarial attacks may not find. Could you maybe explain what permission, adversary and adversary tolerance means in the comparison chart?
Permission - A party (node) needs some kind of authorization before it can join the protocol execution. In permissionless environment (e.g. Bitcoin), any party can download code and start a computer to join the protocol execution. Adversary - A party or a group of parties that are trying to take unfair advantage of the blockchain e.g. spending the same coins multiple times. Such adversarial actions are called "attacks" and blockchains should be able to deter them. Adversary tolerance - Power of the adversary's attacks increases as he/she accumulates more and more of the resource critical to the blockchain. For Bitcoin, this resource is mining power. For Proof-of-Approval, this resource is the stake in the network. Adversary tolerance is the maximum amount of adversarial power a blockchain can deter. Regards, Shunsai |
|
|
|
Hello Paul, In the following example, say I had a majority of stake at block A. Then at B0, C0 all the way up to head block H0, I sell off my majority of stake, understanding that there is an upper limit on stake transfer. A - B0 - C0 - D0 - H0
\ B1 - C1 - D1 - H1
Then, I create a double spend of my stake using the historical private keys, which I still own, thus sending it to myself in a fork continuing fork B1 - C1 - D1 that I create and approve myself (majority stakeholder). This is not a 50% attack, because I don't own the value held in the private keys anymore, the cost for doing this is nothing. Why doesn't the network just accept this fork as canon, given that its the same length (or longer if I chose) than the genuine fork? Any single person owning (or colluding with) a quorum (majority) stake is a problem because that person (or colluders) can rewrite the chain as they choose. This situation is essentially the same as an adversary owning a quorum stake which is the limit of the protocol. This is not how I would categorise the presented scenario. This is the NaS problem in it's essence. The attacker, as a historical majority stakeholder, can still take over the blockchain at any point in the future after he has sold his stake. So, at the current block height, said attacker owns 0 stake, yet he can still rewrite the entire chain due to this NaS problem. You are correct in pointing out an issue here which must be solved. I'd describe the issue slightly differently even without any single party holding a quorum stake. An adversary buys empty private keys (with zero current stake) that once upon a time did hold a quorum stake in the blockchain. That adversary can start rewriting blocks from the time the keys did hold the quorum stake to the present - completely rewriting all the transactions. I do need to think about this problem. Thanks for pointing it out. Regards, Shunsai |
|
|
|
Hello Fresheneesz, Comparing Proof-of-Approval with 2-hop Blockchain (2-hop)1. 2-hop uses PoW consuming resources (in addition to PoS), Proof-of-Approval has no PoW. 2. 2-hop's alternate blocks are different, one is from PoW and the next is from PoS. All Proof-of-Approval blocks are the same (but each block has 2 parts to it). 3. In 2-hop, during PoS round, the network choose one stakeholder for PoS block. In Proof-of-Approval any (or all) stakeholders approve blocks. 4. In 2-hop, fork selection is based on the longest PoW chain. In Proof-of-Approval fork selection is based on a better approval score. 5. Proof-of-Approval has 1 block finality. 2-hop's finality properties seem to be similar to PoW chains. The analysis here is more difficult since neither PoS nor PoW can be analyzed independent of each other. The PoS part of the analysis matches Cardano's Ouroboros ( https://eprint.iacr.org/2016/889.pdf) static case. Since the static case isn't a reality (stakes will change during chain's lifetime), it is unclear if the PoS does add any additional security to the chain. Hope this comparison helps. Regards, Shunsai |
|
|
|
Hello fresheneesz, How would other nodes detect that they're intentionally not approving blocks in order to blacklist them? Isn't anyone allowed to not approve blocks if they don't want to?
Node A is allowed to not approve another node's (say B's) blocks. That also means that B is allowed to not approve A's blocks. It works both ways. To detect if it is intentional or just network, point-to-point ping and bandwidth can be measured to guess number of blocks that should have reached within the time-slot. But isn't only including your block as the only approved block the best way of gaining an advantage? This is known as the "bullet-voting" problem in approval voting, ranked voting, and range voting circles. https://en.wikipedia.org/wiki/Bullet_voting . While in normal voting, this shouldn't usually be much of a problem, it seems like in Proof-of-Approval there is a big incentive to bullet vote. Yes, block approval should be incentivized. I am working on it. Thanks, Shunsai |
|
|
|
Hello Aliashraf, Just read the article, nice work but, as one of the guys have mentioned earlier, the 'proof of something' discourse is exhaustively saturated and it is hard to attract enough attention from the community these days for a new one.
I choose Proof-of-Approval name simply because it implies to have something to do with approvals. I could have chosen "Takahashi Protocol" name (similar to Ripple Protocol or Ouroborous) but that name wouldn't have told the reader anything other than the creator is named Takahashi. I'm not satisfied yet with the protocol itself as it seems not to be secure against the bootstrap poisoning attack...
Protocols that use "older" stakes (stakes from many blocks ago) are susceptible to bootstrap poisoning attack. In this protocol, since the stake used is at the current block, I have not been able to come up with a successful bootstrap poisoning attack scenario. ...I have not found strong enough game theory and incentive analysis (why people should approve each other's blocks) and I do not like the idea of zero cost block generation..
Multiple people have mentioned it and I am analyzing giving rewards for approvals. The way I see the situation in crypto world, we will be stuck in wasting resources and jeopardizing performance and scalability as long as we are seeking a solution for this problem: consensus.
The reason the crypto world is focused on consensus is because in absence of a good protocol, all other blockchain properties are compromised. Bitcoin is said to have breakeven of $8050 due to its resource consumption requirement just to achieve consensus. Therefore, blockchains need to be solve consensus to actually achieve that we all know they can. Yes, I'm crazy enough to don't pay attention to anything other than game theory and pure mathematics and to this point...
Game theory and pure mathematics - just the way I like it So, my advice, just forget about consensus protocols, they either do not exist or they are bad protocols.
Here I'd differ from you a bit - I do feel that a good consensus protocol (that works and doesn't consume resources) is needed for blockchain success - not just as price of token but their actual impact on the world. Thanks for your comments, Shunsai |
|
|
|
Hello Fresheneesz, Am I right that the goals of this protocol are primarily to reduce the finality time in comparison to Bitcoin as well as reduce the cost of running the system?
Yes, the goals are to 1. Improve security by achieving near 50% adversarial stake tolerance. Near 50% tolerance would offer deterrence even more than the corresponding mining power. Only Cardano achieves that (not counting TON because I have no knowledge of their protocol). 2. Improve finality. The fact that Ethereum is trying to get Casper FFG shows how important finality is (although Casper FFG only improves tolerance to long-range attacks). Cardano's finality degrades continuously with rising adversarial stake. 3. Reduce cost. AFAIK, Bitcoin's breakeven (cost of mining vs its price) is around $8050. That's a lot of expense for the node owners. Thanks for pointing it out - I had not seen that before. I will definitely offer comparison even if they are simply superior Your comparison table claims Proof-of-Approval is fully finalized after a single block, however, you also talk about how there may be multiple top-level blocks where the "preferred" chain is unresolved. Doesn't this imply that your transaction being inside a valid block doesn't mean its been fully finalized? How do you reconcile that?
In order to be fully robust, the fork selection has to work on all conditions that may be produce in normal operation. It is not a detriment that the fork selection actually works even beyond those conditions. It's just a good design practice to design for more than needed (within cost constraints of course). Since any node can produce a candidate block, this seems like it would potentially cause unbounded network traffic with everyone and their mother suggesting blocks to vote on. How is this resolved in the case of, say, an adversarial ddos attack?
Traffic can be a problem but only the headers (which are few hundred bytes) are being transmitted. Still, a node would need to validate the blocks of headers it receives (after downloading full blocks). Having a smaller time-slot (as opposed to larger) will reduce the number of headers received by a node (assuming the software's low level code simply drops headers received in mismatched time-slots). An adversarial ddos attack may look like network fragmentation and may prevent blocks from being added to chain temporarily. But I have the same question as monsterer2, why would anyone with stake approve the block of anyone without stake? In fact, why would anyone with stake approve *anyone* else's blocks? It seems there's no incentive to mine on top of someone else's blocks. This could manifest in more subtle attacks where you have stakeholders who refuse to mint on top of a competitor. It could also manifest in deadlock as nodes try to maximize their probability of winning the minted block.
Yes, and there may be a need to reward block approvals. I will work on adding block approval rewards. They can be added to approval-block since the content-block has already been created. I have a hard time seeing the purpose of having nodes pay attention to timestamps and comparing them to their receive-time. At its core, the mechanism is just voting, right? I don't see why the protocol doesn't just let nodes vote for whatever blocks they want. I see the appeal of a simple stake-voting process for block creation (similar to DPoS), but the Proof of Approval method seems like rational profit-seeking minters would make things devolve into the top 50% stakeholders passing around permission to mine the blocks, completely excluding the other 50%.
Nodes are allowed to rank in any order they choose. The time is just the default order built into the software. Also, could you provide a glossary of variables you use in this paper? Its hard for me to keep track of which variables mean what and find their definition in the prose when I need to refer to what they mean later.
Good point, I will add glossary. Thanks, Shunsai |
|
|
|
Hello Monsterer2, In the following example, say I had a majority of stake at block A. Then at B0, C0 all the way up to head block H0, I sell off my majority of stake, understanding that there is an upper limit on stake transfer. A - B0 - C0 - D0 - H0
\ B1 - C1 - D1 - H1
Then, I create a double spend of my stake using the historical private keys, which I still own, thus sending it to myself in a fork continuing fork B1 - C1 - D1 that I create and approve myself (majority stakeholder). This is not a 50% attack, because I don't own the value held in the private keys anymore, the cost for doing this is nothing. Why doesn't the network just accept this fork as canon, given that its the same length (or longer if I chose) than the genuine fork? Any single person owning (or colluding with) a quorum (majority) stake is a problem because that person (or colluders) can rewrite the chain as they choose. This situation is essentially the same as an adversary owning a quorum stake which is the limit of the protocol. In another scenario, some stakeholders may only approve their own blocks in an attempt to tilt the winning chances in their own favor. This is a real possibility. A possible mechanism to deter such an "attack" is to simply "blacklist" such stakeholders temporarily. An honest stakeholder can tell within a few blocks with certainty if another stakeholder is not approving his/her valid blocks.
My question is, what does the stakeholder have to gain by approving anyone's blocks but his own? If he has an edge by not doing this, there is a very real problem. My assumption, which could be wrong, is 1. Stakeholders will approve (through approval message) more than one blocks or risk blockchain failure or being blacklisted by other node members. 2. Stakeholders will try to rank more candidate blocks created by other higher stakeholders lower in order to gain advantage in block creation. Of course, these are just assumptions and need to be proven in real life. Perhaps some coins should be given for approving a winning block in order of their ranking of the block. I am open to suggestions here. In the meantime I will analyze the if incentivizing block approval does have the desired effect. Thanks for pointing it out - this is exactly what I am looking for from this forum. Shunsai |
|
|
|
|
