There is no way you will keep any where near 50% of the users online and participating in staking.
So run the numbers with a more realistic participation percentage and you can see that this design can't possibly scale up to mass adoption. In the power-law distribution only about 33% of the wealth will be serious owners and of them many are not going to bother staking since that will end up pretty much an unprofitable activity due to competition. So in reality the math is going to look more like:
currently more than 40% of DCR are in the PoS mining and every voted ticket get the same reward when picked, it's a lottery system, only 1/3 off all ticket are in the PoS pools the majority is solo staking, like me. What part of this can't you understand? The Decred community manager pointed out that users MUST stake through pools (else their tokens are effectively debased) with lockup durations of 42 - 142 days, and I replied that delegating to pools is a form of centralized control.
...
You are forced to stake in a pool not because of any competitive advantage in mining (the entire thing is just a farce to obscure that whales control everything and extract the maximum that the market will bear), but because they are debasing everyone who doesn't delegate staking to pools. Thus it is just another scheme that forces the float towards 0 to make the price go up by highly disincentivizing trading. So this is Steem's model of locking up tokens. What happens is then you can't sell when the peak is reached (go look at Steem dropping from $4 to $0.10).
Hey hey, don't put words in my mouth. I never said users MUST stake through pools, only that many were. We do have several users that are staking on cloud servers or their own personal computers... edit to add: Also, unlike other PoS models, there is an incentive built into DCR that encourages 100% wallet uptime. If you are not online when your tickets are called to vote, you lose your subsidy! |
|
|
|
There was a guy (well I presume he was a guy but that really isn't important) on these forums in 2013 that had the name Decrits for his project. He disappeared. I doubt this is his project reincarnated.
Nope, we are only 98% attack proof  |
|
|
|
Decred locks your coins for a period of 28 days, on average. Worst case is 142ish.
What about coins on exchanges? Do we lose the staking benefits thus we are being debased if we HODL them on exchanges? I realize HODLing coins on exchanges is risky and stupid. (⊙_◎) I guess your implied policy is only use exchanges for trading them and get a wallet if you want to HODL? Yes, I am a strong proponent of HODLING. Exchanges cannot stake your coins because they would easily be caught with their pants down if people wanted to withdraw. People who keep coins on exchange are speculators though... you don't want short sighted speculators deciding the future of your blockchain, so it's ok that they are not able to vote with coins on exchange. |
|
|
|
Who else is copying the locking up the coins?
Every coin that pays an interest to staking, including every coin with masternodes such as PIVX. True, your coins are locked for a brief moment with traditional staking. But you can sell a DASH masternode at any time you desire. Decred locks your coins for a period of 28 days, on average. Worst case is 142ish. |
|
|
|
@dbt1033 I want to change the world. After you get enough $, you'll probably want to also. Or maybe that is already your priority focus. If so, we will cross paths again soon. I have a surprise.
Edit: thank you so much for the frank, apparently honest, and cordial discussion. I guess I am happy I looked and I understand what others are working on. And was able to get it straight from the developer. Met someone new. Got invited to Slack (which I may do in near future).
I really need to stop expending my time on these sort of activities (talking instead of coding). It had frustrated me that I hadn't looked at Decred nor PIVX and others were talking about it and asking me for my analysis.
I need to have my head in the programming cave sand.
So forum, you are going to be on your own for a while in terms of technical analysis. I really need to discipline myself.
I'm not a developer... just Community Manager. It was a fantastic discussion though, and we do hope to see you in slack sometime in the future. Thank you for your thoughts! |
|
|
|
- I currently think an ideal cryptocurrency coin emission curve includes a tail emission.
You don't prolly don't need it with PoS combined. Tail emission doesn't provide enough security any way, so you've go to get a handle on the fee market issues. Thus the reason to want PoS control. But democracy is always a winner-take-all power vacuum. -Proof of stake mining has a genuine penalty (coins locked up)
That is actually a good thing for making the price go up because the float is smaller. Dash's masternode scheme with compounding concentrating stake to Evan Inc is a perverse example. True and true. By having one's coins locked, it also requires people that wish to vote on Decred's future to have "skin in the game". It's not a perfect system, but it does incentive people to vote in Decred's best interest. |
|
|
|
To be honest, I just reached out to Charlie on twitter, and he took interest in the project. He had written a PoA proposal years ago, so I'm sure that seeing PoA in action intrigued him.
So LN is definitely coming. We've got Ethereum's Raiden, Decred prolly a lock to activate it, and Litecoin also appears to have strong incentives to activate SegWit as the prerequisite. So investing in Decred at this time is maybe diversification into the serious potential LN candidate blockchains. In addition the strong development team of Decred. The long-tail obtuse issues about the security are prolly secondary in speculators' calculation at this time. Well, Decred is evolving into a full scale decentralized autonomous organization. When the necessary mechanisms are in place, it woudn't hurt for users to vote on funding for such an audit. These mechanisms will be in place very soon. |
|
|
|
Lightning network implementation is currently on our roadmap for 2017, and we are on track to achieve all of our goals.
Okay so you're the fallback if Litecoin can't get SegWit activated? That is why Charlie Lee was in on this, in case the miners of his Litecoin don't cooperate? Intrigue. Or maybe just putting your bets in many hats to diversify? Another issue with Decred may be that it doesn't appear to solve massive TPS scaling and real-time instant transactions. So all the new risky (i.e. not extremely well vetted) changes to consensus algorithm without significant gains at least in those stated areas afaik.
So the PoS control enables "the community" (of whales, lol) to overrule miners when they need to, e.g. on activating LN. So that is perhaps the big gain (if you consider that a gain). Actually, PoS enforces consensus. PoS voting decides how the blockchain will evolve. Features are activated through an automatic hard forking mechanism. They are precoded, so this happens once "yes" votebits reach critical mass. Two validation systems are integrated into the Decred blockchain, Proof-of-Work (PoW) and Proof-of-Stake (PoS). The PoW system (miners) create and validate new blocks added to the blockchain. The PoS system allows ticket holders (voters) to participate in verifying a block's validity. The PoS system acts as a 2nd authentication factor on the chain. If 3 of the 5 randomly selected ticket holders approve, the block is added. The PoS tickets are also used to vote on a change to the protocol (hard fork) over a period of time. Without support of the super-majority vote, the minority chain will become unsustainable due to invalidated blocks. |
|
|
|
Lightning network implementation is currently on our roadmap for 2017, and we are on track to achieve all of our goals.
Okay so you're the fallback if Litecoin can't get SegWit activated? That is why Charlie Lee was in on this, in case the miners of his Litecoin don't cooperate? Intrigue. Or maybe just putting your bets in many hats to diversify? To be honest, I just reached out to Charlie on twitter, and he took interest in the project. He had written a PoA proposal years ago, so I'm sure that seeing PoA in action intrigued him. |
|
|
|
Interesting discussion. A question Iamnotback if I may. How well does "game theory"hold up when applied to systems...or just in general.
I remember seeing an Adam Curtis documentary where Adam Curtis looked at the way the ideas of Nash were applied to American foreign policy. In that case he argued that it the theory did not work very well. So..what can we say about how it might work in this situation?
Well it is a good point because game theory can be considered in a narrow context wherein there are externalities which change the likelihood. For example, the security issues I was discussing upthread are long-tail events so you guys speculating on near-term gain overwhelm any relevance of the potential obtuse security issues. However if you want to take over the world as Bitcoin is attempting to do, then the game theory weaknesses have to be analyzed in every plausible context. Because it will get attacked. So I would say that American foreign policy is a highly manipulated thing that has externalities we can't even see, so a model of it is prolly intractable to define. Another issue with Decred may be that it doesn't appear to solve massive TPS scaling and real-time instant transactions. So all the new risky (i.e. not extremely well vetted) changes to consensus algorithm without any significant gains. So for me it doesn't feel like it could possibly be the next big thing unless I am not fully aware of the possibilities of that PoS+PoW design. Lightning network implementation is currently on our roadmap for 2017, and we are on track to achieve all of our goals. However, the community will have to vote on it before it activates. |
|
|
|
@dbt1033 seems like a nice guy. Apologies for being so hard on you. I am just tired of all the scammy coins like Dash and PIVX.
Any way, you can prove you guys are really serious by hiring some serious researchers to analyze it. I don't trust your white paper. You guys are no doubt very good programmers. But being a good programmer doesn't necessarily mean you are an expert at this particular area of game theory and economics. There is no math backing up the claims in your white paper about selfish mining. It is just a claim with no proofs.
Never stop questioning man; that's how we will change the world. I really do hope you take me up on my offer to come by slack (if you're an IRC type, we are bridged). Decred isn't going anywhere, and we would love to have more people like you in our community as we continue to push forward. |
|
|
|
What I am saying is that mixing PoW and PoS sounds very cool. But if you actually do a very deep analysis of all the game theory, you'll realize it is really much worse than PoW alone.
Do you have a whitepaper or other published work discussing this? I'd be interested to read more. I do but it is not published yet. If I have some time, I'll try to present some analysis on the selfish mining aspect of PoA, but the problem is the market won't listen. And I have more important work to do. Really you guys should spend some $ to actually dig deep into the game theory. You have money now. You should hire someone impartial and pay them regardless of what they find out. If you guys aren't afraid. Any way you've already pocketed a lot of money. So why not try to research more. Spend your money on valuable activity.Cool! It will be an interesting read, I'm sure. When do you plan on publishing? edit to add: This isn't a bad idea... something we will consider. I know you think that we are a bunch of Bozos at Decred, but I think that if you found the time to come by and hang out, you would find that we are quite open to intellectual discourse / suggestions. |
|
|
|
What I am saying is that mixing PoW and PoS sounds very cool. But if you actually do a very deep analysis of all the game theory, you'll realize it is really much worse than PoW alone.
PoW is also centralizing too, so it is not like we really have any solution. At least PoW is a stalemate w.r.t. to protocol immutability as we see now with Bitcoin.
You are obviously in slack... why don't you come by? Our development team is willing to have a more technical discussion with you if you are interested. |
|
|
|
What I am saying is that mixing PoW and PoS sounds very cool. But if you actually do a very deep analysis of all the game theory, you'll realize it is really much worse than PoW alone.
Do you have a whitepaper or other published work discussing this? I'd be interested to read more. |
|
|
|
Well $18+ million dollars is a lot of money to attack a $50 million dollar network... come on.
You are not looking at it properly. The economics and design forces it to centralization so you'll have a few big whales running the whole thing eventually. And $18 million is nothing because they paid nearly nothing for it by the time they have it. They extract it over time for free due to the game theory. Correct, but you are making a lot of assumptions here. |
|
|
|
However, it is prohibitively expensive to do what you describe. Can we agree on that?
No we can't agree, because if you took the time to actually analyze deeply what happens with selfish mining strategies and stake combined, the PoW is going to become rapidly centralized but you will have no way to actually know it. I am not really concerned about double-spending. More so I am concerned about attacks when it comes time to make protocol designs. The controlling stake+PoW will extract the maximum fees and they will monopolize everything. It is really about milking the thing to take all the value out of it. I hate shit like this, so it is difficult for me to agree. Sorry. Well $18+ million dollars is a lot of money to attack a $50 million dollar network... come on. You are 100% correct though, Decred, like bitcoin, relies on a majority of honest hash power and stakers. |
|
|
|
You always know when someone has lost when they resort to name calling. Have a nice day brother.
Tell that to the girls by the xerox machine. Meanwhile those who understand math and game theory know that the adjective (not name) prescribed to you is accurate. You guys are faking some level of competency that you don't have. There is no free lunch in consensus designs. You don't go merging PoS and PoW without introducing crazy game theory weaknesses. You guys are just lazy to actually find the weaknesses. And you even deny the math that was presented in the PoA paper. Pretending that pools aren't a form of centralization. They are a form of centralization... that was never denied. We would like better decentralization of pools / better distribution of pool users. I'm not sure why you are so angry man. I'm not arguing with your math; you're obviously not a fool. However, it is prohibitively expensive to do what you describe. Can we agree on that? |
|
|
|
LOL, come on man, just crawl back in your cave. A little FUD is nice; we don't get much at Decred, but show up with better material next time  You're an idiot. You deny the PoA math. And who ever wrote your white paper has not properly analyzed the vulnerabilities when combined with selfish mining. You always know when someone has lost when they resort to name calling. Have a nice day brother. |
|
|
|
This is why Decred has stake pools to allow for offline staking.
Apparently you don't know what the word centralization means. LOL, come on man, just crawl back in your cave. A little FUD is nice; we don't get much at Decred, but show up with better material next time |
|
|
|
Yes, those "bozos" are the developers... they happen to be well respected in the bitcoin community for their work on btcsuite.
Being good at copying things doesn't mean we should trust them (you) with knives, i.e. consensus algorithm design. They (you) should stay near the girls at the xerox machine. That vulnerability is discussed in our whitepaper, which you can read here (you obviously didn't): https://decred.org/dtb001.pdfConducting such an attack would require one to hold over 50% of active staking tickets (which would cost upwards of 18 million usd at this point in time assuming all tickets were purchase at the 30 day vwap [ once again... extremely difficult to do without significantly increasing ticket prices / the price of the attack]) It would also require a large percentage of the PoW hashrate to be controlled by the attacker... again, prohibitively expensive. Obviously you didn't read what I wrote. There is no way you will keep any where near 50% of the users online and participating in staking.So run the numbers with a more realistic participation percentage and you can see that this design can't possibly scale up to mass adoption. In the power-law distribution only about 33% of the wealth will be serious owners and of them many are not going to bother staking since that will end up pretty much an unprofitable activity due to competition. So in reality the math is going to look more like: ((1/0.33 - 1) × 0.20)3 = 0.06Where the attacking whales only need 6% of the hashrate to double-spend. And I didn't even get into the math of selfish-mining attacks which are much much more damning. The security is absolute dogshit.The only way this design makes any sense is for a very centralized stake where we trust that overlord. Which is precisely what you guys are counting on and you damn well know it. Else you are really really bozos if you can't do some simple math. This is simply not true! If users are not online when their tickets are called to vote, they lose the subsidy. It's not like traditional PoS where you only receive subsidy when you are staking. There is a heavy monetary incentive to be online 100% of the time. This is why Decred has stake pools to allow for offline staking. From the whitepaper (that you still did not read): "It should also be noted that many well known mining attacks, such as selfish mining [26] and stubborn mining [27], will no longer function advantageously in a system where there is effective decentralization of stake mining and no PoW-PoS miner collusion. This is simply because it is impossible to generate secret extensions to blockchains without the assistance of stake miners." I'm not here to argue with you about vulnerabilities. Every system has them. Hell, there are several ways to attack bitcoin. But people just don't want to spend 18 million usd to destroy decred (and lose their investment). Over time, it will only become more expensive. |
|
|
|
|
Show Posts
