Hello world. As a stake pool tester of a pool named after John Locke (Lockepool) I was wondering why we cant just silence Daniel Larimers critisism (https://twitter.com/bytemaster7/status/1181323723653222406) once and for all by making some form of social contract that makes sure we are indeed not 3 persons running all of the staking pools on different accounts. (It seems Daniel forgets the social aspects of PoS).
What if we made a stake pool alliance with a yearly conference where we showed up in person and shared experiences as stake pool operators and at the same time verified we are indeed different persons, with different hardware and different locations. We could even share how to be more secure and better our service as staking pools. For example if it is good to use laptops with sim cards internet so we can be up even when power shuts down. With now highly advanced laptops running on 35+ watts and staking in Cardano having minimal cpu/ram/harddisk costs I am sure creativity will blossom in finding cool ways! (Satellite launched staking pools anyone?!?)
In any case Lockepool wants to contribute to social contracts and ideas of governance / community that helps evolve the Cardano ecosystem. Not to lead but to be among a great community.
Anyone think this is a good idea feel free to discuss! Perhaps this can be evolved into a better idea as well.
You will not be able to respond to that criticism with a social contract. How does he know the contract is valid? Or that you really did meet up? Or that the pool operators are not lying? People can say “this is my node, this is my ip, this is my pool id”, but in the end it’s like any wallet. Even if you had a trusted third party validating the pools/nodes and their operators, there’s still trust involved. From what I can tell the closest thing you could get in a system like this would be a vote of confidence.
If you meet in person and show the different node operators and allow visitors to attend conference you could at-least have a lot of confidence. Much more than in say PoW where you have zero knowledge of the Chinese miners and who owns them (and anyone can buy rights of the network with computational power.) Also note that since for any person to fake a 51% attack they would still need 51% of ADA delegated very unlikely if there are say 500 persons on a conference all having a pool of they’re own.
If you claim out of 100 it is only 3 persons and I can show you with KYP (Know Your Pool) ID verification, location of my node, my hardware etc and more than 3 people can do that I do not see how this criticism does not fall to a crumbling pile.
This is for me main concern with most mathematical models. They forget the social aspect.
Oh for sure, that’s what I meant when I was referring to confidence. I am confident of the number of pool operators because I have talked to them in Telegram, I would be even more confident if I met them in person, but there’s still trust involved. I am also confident StakeNuts can spin up a hundred nodes at a moments notice because he said so in the channel :). But in the end, it doesn’t matter. The number of nodes is honestly pretty irrelevant at this point when the majority of ADA is owned by a very small minority. And since ADA is the equivalent of computing power in PoW, it doesn’t really matter if you have a million nodes or 10, it’s just cool to see people getting involved and planning for the decentralization of the network (and the decentralization of ADA itself).
Hmmm, tribalism, we never change. I meant we the majority of the humans.
Are you kidding right? Just have a look at the BTC’s hashrate distribution chart below, which is based on the mining power and compare to the ADA distribution
Also, it shows an average 0.01% per individual of the sold ~25bn, and
Cardano Foundation, Switzerland: 648,176,7613 ADA, means less than 2.1% of the ~31bn (1.4% of the 45bn)
IOHK: 2,463,071,701 ADA, means less than 8% (5.4% of the 45bn)
Emurgo: 2,074,165,644 ADA, means less than 6.7% (4.6% of the 45bn)
And of course they will sell a huge portion of it by the time.
You are comparing BTC pools to individual holders of ADA, it’s not an apples to apples comparison. You would need to compare individual owners of resourcing power to individual owners of ADA. The top 10 wallets contain 7.4 billion ADA. That is like saying there are 10 individual people who own computers networks that produce 25% of BTCs total hashrate. If you look at the top 100 wallets you will see it starts to become obvious that the majority of “hashing” power of ADA is owned by a very small majority of individuals. I am not talking about pools.
That being said I am a huge fan of Cardano obviously and I think their plan to distribute coins they set aside to support the network is a great play (it will directly support decentralization over a number of years). But right now it’s not even close to being decentralized as far as control is considered.
Keep in mind, everything is all about how to secure the Network and nothing else. But, how? Puzzle solving or staking with money or whatever? That’s a different question.
If you cannot get around this then we would misunderstand each other
Nope, it’s apples to apples. Just think of it.
In addition, it does not matter how it’s distributed at the begining. Think of the Bitcoin, it was only distributed in one hand and then in a very-very few hands.
Here comes a quick chart, only for 11th (as many as the bitcoin pools) of the most richer ADA holders that have power on the Network.
The PoS and PoW are very different, these riches will never have such a power than the bitcoin pools.
We’re talking abt the 51% (or higher, 75% or 90% e.g. for voting), and nothing else.
Why would I? As I said before, It’s all about the Network security and not about any small portion of stakes even that average 2.5% per indivitual you’re talking abt.
Bitcoin is not really secure anymore, but luckily there are economical incentives for the pool owners to defend the status quo.
Yes, I agree BTC started 100% centralized, one computer produced all of the hashrate at one time. The difference here is that the hashrate of BTC is not capped while the ADA “hashrate” does have a cap (total number of ADA). Once BTC was released, Satoshi had no control whatsoever over the hashrate of BTC, but few cared about controlling 51% because it had no value. With ADA it’s a bit inverted. There’s a lot of value on the line, the maximum “hashrate” is known and can be controlled.
When staking with Cardano a lottery is performed and leaders are elected to produce blocks. The more ADA delegated, the more likely you are to produce a block. Stakes with the most Cardano are going to produce the most blocks. I can assure you that IOHK (and friends) are going to protect this blockchain from an attack by delegating their stake to trusted pools (their own). This also means the majority of rewards that are distributed will be awarded to them initially. Initially this is the safest way (not just for them right now, but for anyone who has invested in ADA). But this is absolutely a trust issue right now because it is centralized. You say it’s all about voting, but if you have 51% of the ADA staked (staked, not owned) then you absolutely control the blockchain. IOHK is not going to delegate a single ADA less than what’s required to keep the blockchain secure.
I hear what you are saying, I see your charts, but time will tell when pools are staked. When Billions of ADA are on the line, you think it’s going to be decentralized when staking is permitted on mainnet? No way. It’s too risky for them and for all honest players. Decentralization is the goal, but it is not where it is today, it will take time. There is a lot of conjecture right now in this conversation, but we will see a clearer picture once staking is implemented on mainnet.
So a couple points of clarification from today’s AMA. The IOHK team stressed their interest in not competing with others pool operators, which I think the community will really appreciate. My perspective on this is purely about decentralization and security, not so much around profitability. They also stressed that at some point delegating to a given pool will start to become unprofitable. The system is really designed to support x number of pools (ex 1000), but again this question/response seemed to focus on profitability. The only reason people are going to stake is to make a return (or in less common scenarios to attempt to control or secure the blockchain), so certainly it will cause holders to delegate to pools they both trust and can profit from (will force decentralization in the network) - as long as the pool network itself is decentralized (vs a small group of people creating 1,000 pools) which would be against everything everyone is working towards.
In the end the AMA was very positive. I do believe in the end everything will work out fine and I know the IOHK team is doing everything to make sure decentralization occurs at a successful pace. I do want to see it decentralized, but I also want to see it move at a pace that is safe for the community.
Tthe decentralisation should be irrelevent to the number of Pools or individual human behaviours.
The k will be specified for the Network and it’s going to be 1000 at the beginning.
The pools will be profitable/attractable only until they reach 1/k*All stake and have some amount of stake pledged.
Everybody can create pool provided they have the registration fee + optional pledge, it does not mean they will be attractable (incentives to delegate).
For example IOHK has 2.4bn of the initial 31bn, means in simplest scenario they should have 80 pools, but nothing prevent them to create 800, and CF for creating 200 while Emurgo 700, or anybody to create 1000 pools priovided they have enough ADA for registration + pledge.
This is much more than the expected k=1000.
Will the security of the Network be screwed?
Based on the Ouroboros Genesis-Praos it’s not, if the initial bootsrapping of the Network is done properly, as the function of Genesis-Praos converges to the k while the function of the Bitcoin converge to 1.
That means, it does not matter how many pools anybody creates the number of pools will converge to the 1000 (Nash-equilibirium).
But which thousand of the all?
This is only related to the bootstrapping of the Network, as it does not matter how many pools are created later if the bootstrapping is done properly.
It means, the initial KYC is required and necessary, but later it’s irrelevant as the pools will come and go by the time, but the whole network will be stable and resilient for any adversary behaviour.
The stake pool alliance with a yearly conference or similar thing would not help at all, as that means we’ve put the weakest chain in the system us the humans.
So, I am not afraid, cos if the Ouroboros Genesis-Praos works as it’s proved in the paper then it will reach a certain point (network-effect) sometimes in the future, when it will be and must be fully self-sustainable, that means it won’t be sensitive to our behaviours. For example, creating 600 new pools and attracting the 51% of the stake-holders by some Ponzi schemes or similar.
But, if it does not work then, that’s fine. We will loose our money, but that’s ok cos the system would not be viable for long term anyway, despite how hard we would like to force it.
It’s like an evolutionary game, in which the Nature does not give a F*ck about us the humans, it’s all about the life and not one form of the Life.
This in my view is a misunderstanding. We are not putting the trust in humans any more than the game theory behind pool distribution toward 1000 pools. It is not what the system relies upon but could enhance it and make it easier for delegating stakers to say be more certain a given pool really only is running on rockpi’s. Social contract for social choices.
Don’t be too idealistic, we humans are fallible. What would prevent this alliance to do cartelisation?
First rule: Centralisation -> Power -> Corruption, with NO exception.
They should not know each other.
In ideal case, we should not have any pools, (maybe one which is fully integrated in the system) as it makes the system much more complex, but individual devices running 24/7 (phones as an example) with some stakes on them, in which if a stake leader device is not online for its block creation time, then the other online devices would do a lottery (or decided already in the previous epoch) who will create the block of behalf of that leader. As that’s the future, and not individual nodes, servers etc.
But, it’s just a dream until proper checkpoint, shrading and side channel features are invented/implemented, as the data reducnadcy in the system is enormous as the state of the system is based on the concensus of the state of individual nodes who store the state of the system individually.
I am missing something fundamental here. The assumption is the chain with the most blocks is the correct and honest chain right? If 51% of block producers are producing dishonest blocks then their chain will be the longest chain moving forward correct? I feel like there’s something you are saying that is very important that I am not grasping.
Edit: I am watching a video on Ouroboros Gensis by Prof Aggelos Kiayias and clearly this is more complex than I understood. I am going to have to watch this video a few times I think before I grasp everything that he is saying.
