Bitcoin Forum

I was navigating through different projects and a new consensus called "Proof of Honesty" (PoH).

In simple words, in this consensus users select in which nodes their transactions will be validated, rather than PoW and PoS where nodes determine the validity of the transactions.
The idea is that users do not select the "dishonest" nodes as they let only the "honest" validators to validate the transactions, resulting to nearly 100% BFT.


Do you think that this idea may improve the security of blockchains?

But how does Proof of Honesty prevent collusion in how to decide "who" is honest and dishonest? Plus, more importantly, how cheap is it to collude?

Proof of Work works because there is no need to trust one another. A miner simply provides his "power" to hash and leaves his opportunity to win a block depending on how much power he gives. Plus colluding in the network would be expensive because the miners would be risking their investment in mining equipment and electricity, making it more profitable to be honest.

I agree on your points. However, in PoW hasing power is concentrating to 1-2 maximum 3 mining pools so in my opinion this dosn't make them "honest" 100%. Of course, there is a possibility to do a fork if they attack the chain, but wouldn't that affect the reputation of the PoW consensus?


As for "who decides who is honest or dishonest", the answer is "Users". People who would be affected by a wrong validation, in other words.

If discrepancies in a block validation may be noticed, then the user can inspect them and then decide with fork of data he wants to use for his transactions. I cannot though define the cost as i'm not a programmer in this project (www.Geeq.io). 


I just read about it and wanted to bring it there in order to exchange opinions about the implementation of a new consensus that may be beneficial to current projects as well.

I have checked the link and the "POH Explainer Video" but nothing's really explained but the possible positive effects of using their newly developed "just nodes coin".
Noticeably, they allotted more time explaining the downside of POW coins like Bitcoins than explaining their system.
Just like the quoted reply.

Not to mention, Proof of Honesty is the worst name for a system that was built for security.

Is there a better resource where "Proof of Honesty" was better explained?
Please provide some links.

Bitcoin and most cryptocurrency supposed to be trustless and use consensus method which enable trustless, so IMO this consensus isn't practical at all. Why don't we back using fiat/bank if high degree of trust is required in here.
I also check their explanation (https://www.geeq.io/the-most-secure-blockchain (https://www.geeq.io/the-most-secure-blockchain)) and they vaguely explain the PoH which don't help at all.


That also applies to most consensus method, but in PoW and some kind PoS, there's punishment/wasting resource such as big amount of staked coins taken away or require million USD for ASIC/electricity when someone attempt to attack network.


Just with this sentence, there are few critical problem such as :
1. How do they find out whether the user is real or don't have multiple votes/identity within the network?
2. Even if assuming they magically find way to solve 1st problem, people can be bribed, manipulated or have their device hacked to manipulate the vote.

Here is a link that may help: https://youtu.be/m5zL2euruYQ?t=20m47s

it is also a provided example of how this can work. I know they are still in an initial stage and I want to see something "tangible" (I.e. code) in order to let us have a clear opinion about what Proof of Honesty really is.

Sounds pretty much like Ripple's and ByteBall's consensus algorithm, which suffer from either centralization, vulnerability to Sybil attacks or both.



Let's look at their whitepaper.


Centralized it is, then.


Facing the problem of hard forks by simply ignoring it is... brave, I guess?



GBBs are unfortunately never explained, neither in their whitepaper, nor on their website.

I'm going out on a limb here, but I assume it's something you'd have to purchase from GeeqCorp? If so, that means: First you buy a genesis block, then you buy consensus access? Is that what it would look like if EA created a cryptocurrency?

It may defined as Centralized, but it will guarantee that blockchain is running smooth and with the same features as any other in the ecosystem.




I would not agree to this point. I can see it as a democratic option and people can decide where to put their money depending on their fork preferences.




I want to see more on GBBs as well. I don't think that you have to purchase something but maybe you have to gain access to that somehow and It could be limited to a specific number of users.

There's a difference between hard forks as a democratic option (see blocksize debate) and chain splits as a regular occurance. A blockchain that hard forks regularly without internal consistency becomes unreliable as users end up on different forks and transactions get "lost" due to users following different chains.




Why use a blockchain then? If the network is centralized and permissioned one might as well just the classical banking system and the likes of VISA and PayPal. Same difference.

The mentioned consensus seems to be very interesting especially when it comes to who users have the authority in the validation process.

My concern can be to the fact that this game theory of honest and dishonest nodes can bring confusion to the chain and cause latency in the transactions.
I have my doubts as well on the implementation of the consensus, but I think we should not criticize the project as it is still in the beginning of developing and we haven't seen proof of Honesty in practice.

Federated trust as decribed by GeeqCorp is precisely how humanity has done monetary transactions for centuries. Merely throwing in a blockchain for good measure doesn't change anything. Getting rid of federated trust is what made Bitcoin so revolutionary. Returning to it seems rather regressive.



That's how public discourse works though. Otherwise they should have just worked in the shadows and presented the world with a finished product.

My opinion is that an authority on the blockchain should be present in order to avoid fraudelent and malicious actions. See what's it going on right now on the Ethereum blockchain and ERC20 tokens which is full of scams.
And also, if people are not so much in fond of federated trust, how is it possible that Ripple has so much support by users making it 3rd regarding Market Cap?


It can work to make things better, this is the reason why none works in the shadows. And I believe that there is a great exchanges of ideas in this thread that can lead to a positive outcome.

If you want an authority, why do you need a blockchain?  Do you see a particular usecase that blockchain solves in a situation where you are fine with a trusted authority anyway?  In my opinion, a blockchain would just be a very inefficient database and/or buzzword in such a situation.

Authority exists in every blockchain. Do you think there is not high volumes of authority in Bitcoin, when Satoshi holds a huge amount of BTC and mining pools the majority of the hashing power? They get the rewards of the transactions so the value is accumulated on them and coins are spread among few users.

I had a look at this project because your discussion made me curious about it. GeeqChain is pointing towards serious issues on the blockchain like scalability, security, interoperability and inexpensive transactions. I am very interested in seeing how this network is going to solve all of them.

Reading the whitepaper, I can understand the point of Honest and Dishonest game theory and it seems very exciting. Many corners are still shady because we don't have the code and check it.

Personally, I am fed up with the other projects that try to refine their chains that many times leading to forks and community confusion that lowers the value of the tokens.

If GeeqCorp can give an end to that I will definitely support them from the start till the end of the development.

But it would be unprofitable for the miners to also be dishonest because their investment in mining equipment and electricity would be wasted. Proof of Work eliminates the need for participants to be honest and dishonest. The protocol is everything.


Are you suggesting that we should trust a collective third party? I believe that is a security hole.


This cannot work. If a transaction or a block is invalid, it should be rejected by the nodes immediately, but what would stop them from rejecting double spends if all the network stands on is your "honesty"?

Exactly, you got the point. This is why this consensus defines that the protocol is secure. None will ever want to be "dishonest" because of what you mentioned above.

This collective party is the people who have their money in the chain, so there is not a security hole because if they try to manipulate a decision they will simply lose their money.



In this consensus users decide. How can you trust a node to "judge" if a transaction is invalid, or it hasn't been invalid on purpose?
Because of the feature what users decide, nodes should try to avoid validating invalid transactions due to double spending as you said before.


I believe that there is an interesting concept behind Proof of Honesty consensus and I want to see how it is actually working in such cases.

Any distributed p2p, digital currency which relies on trust to make it work is nothing more than a horrible mistake a best, and at worst it's a scam.

As soon as you make that trust sacrifice, you might as well throw it all in the bin and use VISA, which has none of the drawbacks associated with crypto, and is already accepted everywhere.

Alt coin and token scams are what makes it all the more important to call out questionable buzzwords such as Proof of Honesty. Dressing up federated trust as a trademarked term called "Proof of Honesty" is borderline deceptive. Selling a centralized solution as "100% Byzantine Fault Tolerant" is like selling asbestos free milk -- it's good to know but the least that I would expect.



I'm not saying that people are not fond of federated trust. I'm just saying it's a regressive and naive approach at cryptocurrencies.



So we do agree after all that even in the conceptional phase valid criticism is productive :)



Many corners are still shady because the whitepaper claims to solve various problems without proposing concrete solutions. The implementation is irrelevant if the very concept is flawed.



The problem of hard forks is effectively ignored on the protocol level.




Owning large amounts of currency does not equate authority. Large amount of hashing power do not equate authority.

Anyone can participate in the network. No one can block or censor transactions. Block subsidy is not something that gets handed out by a central authority but is what pays for electricity, hardware and infrastructure, keeping the network's security autonomous without external oversight.

Falling back on a centralized authority just because PoW based cryptocurrencies have mining pools is throwing the baby out with the bath water.

This.

It is interesting how new forum members here are advocating for authority, while all the more senior members highlight how that is kind of pointless on a blockchain.  Assuming the newbies are not just shills for PoH/GeeqCorp, this probably means that people new to the blockchain space typically need quite some time to actually understand what blockchain and decentralisation is and isn't about.

If you think it is better to use VISA it feels like you see more drawbacks in crypto rather than benefits. Maybe you don't want mass crypto adoption and use it for your everyday payments.

Looks like there are a lot of shills in this thread.

I can see many cryptocurrencies that try to be more decentralized as being centralized in the beginning so this may be a direction that it is perceived in a later stage of a centralized strategy.


there are many different opinions about blockchain and decentralization, this is why i came to this place and learn more of that and evolve my critical thinking.

I think monsterer2 is pointing out the absurdity of centralized cryptocurrencies rather than claiming that decentralized cryptocurrencies are non-viable.



There's no hint of a later decentralized stage though. Even if such a stage would be advertised, its feasability would be rather questionable seeing how centralization lies at the very heart of this project. Centralized token issuance, centralized consensus... heck, even the name "Proof of Honesty" is trademarked apparently.

But what would the Proof of Honesty participants be risking if they collude and control the network? If it is very cheap to collude and cheat the network then it would be very easy to do.


Then would you agree that this is a collective for of centralization that would cost nothing to cheat the network? How would such a coin be valuable?



I stopped reading at "users decide". That would be a consensus nightmare.

The critical problems that you pointed out are valid for this consensus.
Hence, PoH has a long way to go in establishing their niche here in the crypto world.
It would take time to validate their objectives for this consensus.
With how things are being handled in crypto, there's high chance that being labeled as "dishonest" may not be true at all and vice versa.
The criteria should be very precise on how they identify as "dishonest" or "honest" nodes, as their explanation is still hazy at the moment.

seems pretty interesting but how would you know a dishonest node over an honest node what evidence would there be, and also as a miner would you only get a reward if people choose your node?

From what I have read Dishonest and honest nodes are selected based on unanimity of users who has inserted their transactions for validation.A miner still gets reward if he is honest but still not selected. It feels like you are rewarded for keeping the system "Honest".

That suggests that the network should trust other users to act in good faith which is already a security hole itself in my opinion. A decentralized network should never require third party trust. This is why Bitcoin robustly works.

I think a decentralized reputation system can sort-of work as a proof-of-honesty. You can see my post about a potential sidechain that I want to work on regarding this: https://bitcointalk.org/index.php?topic=4919679.msg44289606#msg44289606

There are tons of problems with this approach. First of all, a user has no idea if the validator is indeed honest even if he was honest in the previous block. Second, a malicious entity can create an infinite amount of malicious nodes to create an illusion of the selected validator, to be honest. In the end, this consensus will inevitably lead to having a handful of trusted validators thus making this system controlled by them.

On the side note, your idea reminds me of how EOS (and it's deligated proof of stake) works. This is a step away from the trustless and decentralized system.