Topic: Obyte: Totally new consensus algorithm + private untraceable payments - page 100. (Read 1234271 times)

Ok, so still talking about Bitcoin rather than Byteball. What's up with that?



DPoS is one of the most broken consensus mechanisms out there.

The network would very rapidly notice whenever a witness stops posting units. Since witnesses are publicly known, they can be contacted, and if no satisfactory explanation is given (or the problem is not solved) the community at large will decide to replace that particular witness address for someone else's.

So no matter what the state power does, or a hacker, or just a witness turned malicious by himself, Byteball is not vulnerable.

So in that sense I'd argue we don't even have to "trust" these witnesses.

You know, when you need infinite energy to redo the PoW, transaction finality happens. But even then, you can sure of transaction finality with only 1 confirmation because no miner will shoot to their foot. If he tries to, it will be prohibitely expensive.

I read in Byteball whitepaper that bitcoins transactions have probabilistic finality, not deterministic. Well, that is the theory. The practice is another thing. It is the same as miners. Miners get specialized, and entry barriers to mining appears. But the protocol itself does not impose restrictions on who can mine, and that is what matters. Equally, theorically any transaction can be reverted (probabilistic finality). In practice, only 2 days.

Theory -> Anybody can be a miner, protocol says.
Practise-> Mining gets specialized, so entry barriers appears. Miners also has something to lose: their mining investments. Incentives balanced.

Theory-> Anybody can be a witness, protocol says.
Practise-> In order to be a witness (I mean, receiving payload bytes, not only paying them to actual witnesses), every node (or at least the nodes which add the most quantity of units) has to write your address in theirs witnesses lists AKA ask for permission to these nodes. Anybody could not be a witness. If you want to remain anonymous, you can't -> entry barrier. Sure you can try to, but it will be useless. Would you choose an anonymous witness? Of course not.
A witness also has something to lose -> their reputation. But I am not sure of the value of that.

I am confused. If witnesses are so powerless, why not impose them? (and now I am not trolling). Community claimed a lot for going back to airdrops, but you imposed your view. Why not doing the same now that we all want to see witnesses distributed?

Don't you realize that we are coordinating through byteball team for changing witness, because we can't do it in a decentralized way?

Very interesting post. In the case of a small group of miners, they has a lot of incentives to keep up the system. If they are not doing what they have to do, we change PoW and they lose all mining investments. I am quite sure they will do what they have supposed to do even with that emergent centralization.

In the case of a small group of witnesses... someone say I don't have to trust them. In fact, this is what I am doing: I don't care about witnesses selection.

Troll-mode: Choose them and advance few years to the future, please. I trust your choices since 25-dec-2016, and I think all actual bb users would if they were asked.

Way better. I understand that you are now the center, but when you distribute the witnesses left, you will have moved the centers, no destroyed them. Decentralization means "without center", at least in theory.

Bitcoin is not run by miners, it's run by a community. (If miners were in charge, we would have big blocks right now.)
This in the beauty and strength of bitcoin. It is multi-leveled, > full nodes, miners, devs and community all able to check each other and keeping each other honest.

I like byteball as a concept but critisism must be taken seriously if you want to improve.
Imo, not being able to run a Byteball full node on a desktop computer is a massive problem.

Is Bite ball protected against arbitrary legislation?

How can one prevent the state power from simply eliminating the witnesses?

As long as there are other Crypto currencies this danger probably will not exist. However, the more secure projects will still be chosen as value stores.
Then why should one still deal with bite ball? Especially since so far nothing is offered that is clearly more attractive from a consumer perspective.

Bitcoin will always be nr 1.

I understood it as Bitcoin becomes more and more centralized as bigger miners get more powerful. You don't know when it is happening because these few people are decentralizing their mining operations, so it would not be too obvious that they have the most power. Sounds nuts, but hashrate of Bitcoin network is steadily going up. Who are they, we don't know, but I doubt they are Venezuelans or Iranians who don't have enough money even for food, let alone buying ASICs. https://news.8btc.com/bitcoin-hashrate-increases-in-a-bear-market-investors-are-still-optimistic-in-the-long-term
Byteball on the other hand, becomes more decentralized as more witness candidates and better witness candidates will join and since they are public, they can't be the same rich person.

Think of this way, you kind of know who are the rich guys today, but you can't do anything about it, they rule the world and the banks and the real-estate and the companies. You can't even vote them off from lobbying to the government because they will lobby to whoever is currently in parliament. Just check what's happening in EU with upload filters, that's Facebook and Google lobbying result.
With Bitcoin, you don't know who the rich guys are and you don't know how many rich guys are there, there could be 10, but there could also be only 1. And you can't do anything about it, you could fork and make their hardware useless, but you probably don't get the miners because the rich guys are becoming the majority of the miners.
Byteball moves in the opposite direction, eventually there will be so many witness candidates that if they breach your trust, you replace them and if enough people feel the same, better witness gets elected.

If one day, all Byteball current witnesses are replaced then your only hope that Bitcoin is more decentralized is because you think that the big miners are not the same people, but you can't be sure. Also, if some of those people become so powerful that they could pull off 51% attack, your hope is that they will not do it because it will destroy the value of their holdings, but you don't know how much they hold BTC by that time, they could have been planning for exit or fork for very long time.

Hmm, dPOS actually totally makes sense. I am not saying don't invest into BTC, it will still for sure have nice returns, but it will not be nr1 forever.

Yes, economy of scale is a thing. But why are you so focussed on talking about Bitcoin? I mean, as the inventor of Byteball (if that is the appropriate term), I would be much more interested in your views regarding Byteballs inherent problems, not Bitcoins.

Like the DPoS comparison I made in the post. Do you have any thoughts on that, especially with Lisk and EOS as rather gloomy outlooks? I mean, I realize they are not comparable on a 1:1 basis, but there are similarities, don't you think?


Yes, "independent witness" is way better. "Decentralized witness" is misleading.

You are right about incentives.  But there is something else.  PoW mining is an economic activity.  You have capex, you have opex, almost all your profits are burned in your huge costs, and you have to care about efficiency in order to stay in the game.  And in any economic activity, you have economy of scale.  As your mining farm gets bigger, it becomes more efficient and stands better against the competition.  You get volume discounts for miners, you will be the first in line for the new generation of miners, you can get cheaper electricity when you buy wholesale, it becomes worthwhile to relocate to a more favorable location, and you can hire the best talent in the industry.  It doesn't matter if the PoW algorithm is ASICable or not, the economy of scale is universal.  It might be not very important in the growing market when everyone, even a smaller and weaker miner, gets a portion of the growing pie, but nothing lasts forever and as the market matures and gets more competitive, the weaker players are darwinized, and the hash power gets concentrated in the hands of a few bigger miners.  Miners, not mining pools.  This looks like an end-game of any PoW based blockchain, because it has a heavy "hook" into the economy, and the economy favors scale.  You can call it emergent centralization.  Incidentally, Byteball moves in the opposite direction.

Of course you have to trust them (as a group), it's clearly stated everywhere.  The whole point is about leveraging the trust in 12 people/orgs to enable random strangers to trust each other, which they wouldn't do otherwise.  It's only in a trustworthy environment that value can be created.

Looks like you are still trying to make fun of the phrase "decentralized witness".  Maybe not the best way to say it, "independent witness" would sound better but I bet you understood it anyway.

Maybe it got lost in translation, but what I meant by "directly" was that you can't take thousands of miners vs. 12 witnesses and say: "more is more decentralized" if you do not take account what miners can do and what witnesses can do. Of course they are comparable, but only if you actually take account all the facts. It is more easier to compare Bitcoin with some Bitcoin fork or some dPOS blockchain because there is less difference. Maybe I should have put NOT DIRECTLY in all caps, but I was hoping you get the idea from rest of the text.



Well, I am not sure why would somebody brag about fact like that, sounds pretty bad if somebody could rewrite 2 days of transactions. It sounds pretty bad if somebody could rewrite even last 1 hour, so kind of ridiculous to waste so much energy on something so uncertain.

I have also some facts you might not know, on Byteball, it doesn't matter how much fees you pay, your transactions won't be censored because of that or for any other reason. And the best part, you can be certain about the finality of the transactions, no need stop your business for 2 days or even hour.

In what way do we need to trust a witness? A witness just sends transactions which (by definition of the protocol) cannot be fake or wrong or invalid in any way. And these witness txs are simply our way of attaining consensus on which transactions are final.

With Bitcoin, you don't really have to trust miners: if there are conflicting txs pending, then it doesn't matter which one they confirm. Whichever one they  choose to put in their block becomes final. The only risk is with enough hashing power they could roll back and restart mining from an earlier block, thus changing recent history in a way.

Likewise, you really don't have to trust witnesses with Byteball: if there are conflicting txs pending, then it doesn't matter which one they choose. Whatever they choose to append their tx to becomes final. And the roll back risk does not even exist here.

So, Byteball is even MORE trustless than Bitcoin.

All users in the Byteball network follow the same rules, they are all equal, but to be able to resolve doublespends you need some order in the transactions, witnesses provide this order. They follow the same rules as everybody else. The difference is that you know who they are so you can kick them out if they misbehave. This gives YOU the power and not the witness. You can't do that with miners.

There is no difference between one witness and another in any kind of practical way so it doesn't matter that you can't choose them freely, they're all the same. They don't *do* anything, they *witness*. The only reason to replace one is because you think some candidate has more to lose in the real world so their bond is stronger and they have even less reason to misbehave than another one. "A" witness doesn't matter, it's about all 12, so no you never trust a single witness and you don't have to. I'm afraid you just don't understand how the mechanism works and what a witness can and can not do, even though we've been trying to explain it you. Instead of asking more questions you jump to conclusions that are just plain wrong.

This means users need external coordination outside the system in case of witnesses misbehaviour, quite impossible case seeing what witnesses you are presenting.
Of course not. Who are the ones of the list putting real resources to work so anybody can be sure of real resources spent without having to know who are he/she/whatever?
More distributed fits better I think. And byte waste too.

Of course, you can slice their head, and you have sucessfully decentralized a witness.

A bitcoin node doesn't have to care about power a miner have in order to use bitcoin network, like with witnesses in Byteball.

It checks if signatures and proof of work are valid. If so, this means it has received the data he has to receive. Nobody has to know another, nobody has to reveal the identity, removing failure points of the system.

In byteball, you have witnesses. If you haven't got a witnesses list, you have to get the last unit one in order to use the platform. If you have already  one, it must be compatible.
They are publicly, known and well behavioured. Should have to trust them in order to use the platform? Please answer.

I don't have to care about that. Am I sure they spent resources to get that block? That spent resources is where all nodes form a consensus.
In byteball, the witnesses as a whole is where all nodes form a consensus.

It is not hard to understand, I think.

I tell you a fact that maybe you don't know: even if 100% of hashrate belongs to same bad actor, he could only rewrite 2 days on the past. So if you have to be 100% sure of transaction finality in bitcoin, simply have to wait +288 valid blocks and done.

It is, is exactly what all cryptocurrencies which are not bitcoin are doing. And is what we are talking about.

*compares Byteball with Bitcoin* --> "Byteball isn't directly comparable with other blockchains"

This doesn't add up, does it.


Btw., this whole "it's more decentralized than Bitcoin" is factually wrong. I am not interested in theories about the future, but in facts. Currently, Byteball is more centralized and I doubt that spreading the 12 witnesses geographically and over different entities will not change that much. But since you seem to like theories, let's talk about that:

The way Byteball selects witnesses, as well as their functions, is roughly comparable to DPoS systems. I mentioned that before, but I'll state it again. So, if you want to take a look into Byteballs future, you can take a look at DPoS projects, such as Lisk, EOS, Tron, Ark and so on.

DPoS has a pretty big problem: due to its structure, it is

a) not as permissionless as mining is (and to a lesser degree staking in traditional PoS; both obviously have financial barriers) and

b) it invites collusion (see EOS and Lisk for prime examples), while PoW invites competition.

What that means in practice is that is way more profitable for delegates/witnesses to work together and vote for each other than it is for miners, because miners are basically stuck in a big prisoners dillemma. There is no inherent incentive to work together as miners, and if there is, a group of miners has no in-system way to punish miners who do not behave as the group wants them to.

Which sounds good at first glance ("they are working together, great!") is a huge problem for a trustless and permissionless system. In a trustless system, you don't want people to work together; in fact, it should not play a role whether people work together or not, at least on a protocol level that is. If the protocol layer itself cares about people working together (or not), it most likely means this layer is not as trustless/permissionless as it claims to be.

You know what also doesn't invalidate something being decentralized?
* The fact that witnesses can be only replaced gradually 1-by-1, this doesn't mean that it is not decentralized. This rule is there not just for fun.
* Only 12 witnesses doesn't also mean that it is not decentralized because 12 witnesses are not same as 12 validators or 12 miners or 12 representatives or 12 stakers or 12 coordinators.
* More than 12 witnesses doesn't make it more decentralized either, only thing that makes it more decentralized is replacing current witnesses with other more reputable persons and companies. Also, running a witness full node will make it more decentralized, even if you are not picked as a witness on anyone's list https://github.com/byteball/byteball-witness

And that is the process that Byteball is currently on, getting more decentralized with every new replaced witness.
When half of the Byteball witnesses get replaced then Byteball will be more decentralized than Bitcoin because these witnesses don't have the power that Bitcoin miners have and because you will never know how much Bitcoin hashpower belongs the same people because these are just some anonymous machines in mining pools, that Bitcoin full node that you might be running doesn't make it more decentralized because your input is minuscule compared to those who have big serverfarms with almost free electricity.

Byteball consensus is not directly comparable with other blockchain consensuses because Byteball is not just another Bitcoin fork.

Netflix is not decentralized. But Netflix uses decentralized technologies in order to reduce bandwidth costs.

Byteball is not decentralized. But Byteball uses decentralized technologies in order to coordinate admision of new units in the database and distribute it to all full nodes.

You can use decentralized technologies without forming a decentralized consensus. But that does not make platform consensus "magically decentralized".