| 1. | Voting is not decentralization, i.e. in democracy we are not sovereign because we can collectively regulate each others' outcomes and we are not independent w.r.t. to some a priori immutable protocol ground rules. Even if you argue that the choice of which altcoin to trade to is sovereign, the problem is that exchange rates can be painted and regulated by the ability to fool the greater fools, such as c.f. the upthread exposé on Dash, such that it becomes a collective Tragedy of the Commons, which is the antithesis of the sovereignty of the Inverse Commons— wherein selfishness can only improve everyone's outcome. It is very important you understand the Inverse Commons, because it is intimately involved with my upcoming proposal on how to make blockchains decentralized and impossible to regulate.
The tragedy of the commons predicts only three possible outcomes. One is the sea of mud. Another is for some actor with coercive power to enforce an allocation policy on behalf of the village (the communist solution). The third is for the commons to break up as village members fence off bits they can defend and manage sustainably (the property-rights solution).
sr. member
Activity: 336
Merit: 265
How can we replace hashrate with nodes  Exactly. Miners can far more easily spin up well-connected nodes to protect their position than nodes owners could build up hash power. Lost cause before it started. This can be argued more fundamentally. A Sybil attack is not a finite resource which can form the basis of a consensus rule. Consensus doesn't exist in a voting system that can be Sybil attacked, i.e. it is a Tragedy of the Commons power vacuum and to avoid divergence it requires a coercive power to have a monopoly on the unbounded resource. Core developer Luke-Jr didn't seem to understand reality: Feb 04 23:29:00 so.. if you have a majority of the hashing power, but not a majority of clients.. even if you were to change the bitcoin client rules, you couldn't get these new rules into the block chain
Feb 04 23:29:07 even if you had a majority of the hashing power
Feb 04 23:29:13 iz: no, you are wrong
Feb 04 23:29:17 Luke-Jr stop arguing like a stupid cunt. you say something, stand by it. where is the motherfucking unanymous community!
Feb 04 23:29:20 how would that work then, dub?
Feb 04 23:29:26 no maxblocksize is not there to keep the size of the blockchain down, i thought it was there so blocks can be verified and propogated sanely
Feb 04 23:29:26 iz: because bitcoin
Feb 04 23:29:36 dub: iz is correct
hero member
Activity: 770
Merit: 629
Well, if you change the RULES so that past transactions "don't count" any more, that's just as well killing immutability than when you change the data in the block chain. That's the point, past transactions weren't touched excepting the hacker's money that were locked. That is why there was 1 month deadline before he could withdraw money. The hardfork wouldn't have been possible past that point. The rule was that he could withdraw it. The fact that this rule was changed by the hard fork, means that the rule was not immutable. There is no distinction between changing the data and changing the rules, as I said before, because your "rights" in a system are a combination of both. If the rules change, the system is just as mutable as when the data change. If I give you a contract in which it is stated "I pay you 1000 dollars and you wash my car"
Let's go with this example. We make a smart contract that states every time i wash your car, the contract automatically pays me from your bank account 1000$. But then i find a way and i exploit the contract and for one car washing, i reactivate it over and over again and it gives me 1000$x 167 times ( somewhat similar exploit to what dao hacker did ). What will happen then ? Well, then you will go to court, and i will lose, might even go to jail. Because i exploited a weakness in the system with the purpose of stealing money and a weakness which wasn't in agreement with the intended purpose of the contract. Only a human judge can see that "I find a way" was not "intended". If it was in the contract, it was, by definition, the whole intention of the contract. If you need a judge, one shouldn't have used a smart contract. If this possibility exists, it is part of the outcome tree. If you sign a contract with someone, you should agree upon ALL possible cases treated by the contract, and then you should automatically have seen this case. If you do not consider all possible cases, that is equivalent to not reading the fine print in a human contract. For instance, you may claim that the fact that Facebook becomes owner of all your pictures, must be an error in intend. But it is in the fine print. The fine print is the law, even if you could not imagine this to be true. And, BTW, the reason why such silly errors are possible is because the language is Turing complete. No contract is ever written in a Turing-complete language. This is the *fundamental* flaw in ethereum. You cannot explore the full state tree. The fine print is potentially infinite in length. But, again: if you need human judgement to find out whether a certain behaviour of the contract was ill-conceived or was not the intend of the parties signing up, to distinguish between credible and incredible, you need a paper contract, and the concept of smart contract is dead. This last thing may very well be the case. I think this is what ethereum proved: smart contracts of Turing complete complexity are dead, because nobody can explore them entirely, so there is always room for errors/scam, and only humans can try to distinguish both. Because worse: let us take your example. Suppose that we DID write such a contract in ethereum. And suppose that you DID find the trick to get me 167 times $1000 out of my wallet for washing my car. Can I now ask for justice, and write an e-mail to Vitalik, as benevolent Supreme Judge, to see that I've been ruined because of a typo in a contract ? Or will I get "code is law" on my nose ? Will ethereum hard fork over my case, with my measly $167 000 lost, or is justice variable, and for small fish, code is law, and if they fat-finger their contracts, too bad for them, but important people can get a human judgement ? Should any smart contract system have a "popular jury" mechanism that can overrule any smart contract, and stop the unstoppable ? Wouldn't that be much fairer than having to ask Vitalik to write a hard fork, which only very rich and important people can ? Is that the idea of "immutability", that it is immutable, except when not ? That code is law, except when not ? That contracts are unstoppable, except when not ? What remains of all crypto if that is the case ? We already have such systems in place. It is called "court".
legendary
Activity: 2674
Merit: 3000
Terminated.
And Satoshi was mining silently way before I heard about Bitcoin for the 1st time (in 2013). I consider it almost like pre-mine. So I don't care either way.
In other words, every coin that was mined by anyone besides you (until you discovered it) means that it was premined. It looks like logic isn't one of your strengths. Or the centralized MtGox and Bitfinex heists waiting to profit on the decentralization-minded Bitcoin "investors". Relevance? I was talking about ETH. What truth? That you had interactions with the chain that was heisted on MtGox and Bitfinex. (And don't tell me that there is nothing that can by done on chain to prevent losing funds on centralized exchanges, because there is something that can be done but your stupendous Core hasn't apparently realized or proposed it afaik)
What should be done? Can you guarantee that?
I can guarantee nothing. Just like when you mods banned me for posting too much about China's mining cartel.
Firstly, I couldn't have banned you anyways as I've never had that privilege. Secondly, I'm not a moderator anymore. You're preaching to the choir.
hero member
Activity: 532
Merit: 500
Well, if you change the RULES so that past transactions "don't count" any more, that's just as well killing immutability than when you change the data in the block chain. That's the point, past transactions weren't touched excepting the hacker's money that were locked. That is why there was 1 month deadline before he could withdraw money. The hardfork wouldn't have been possible past that point. They were only *temporarily* locked and according to the rules that should have remained immutable, and according to the contract that should have been unstoppable, he should have gotten them. The only way he didn't get what he was supposed to get according to the execution of the unstoppable contract and the immutable rules, was to stop the unstoppable contract, by changing the rules (hard fork). You're one of those that go with "code is law". A code can aim to be immutable as long as it works as developer intended. Claiming "immutability" should take priority over everything else in any given scenario is pure hypocrisy and stupidity, and that's not even debatable. It's common sense. And you're a hypocrit, because in real life you wouldn't do that. Go do something considered "wrong" and then interpret the laws to fit and excuse your action. And tell that to a judge. If you really go with code is law. But you won't because we both know i am right. If I give you a contract in which it is stated "I pay you 1000 dollars and you wash my car"
Let's go with this example. We make a smart contract that states every time i wash your car, the contract automatically pays me from your bank account 1000$. But then i find a way and i exploit the contract and for one car washing, i reactivate it over and over again and it gives me 1000$x 167 times ( somewhat similar exploit to what dao hacker did ). What will happen then ? Well, then you will go to court, and i will lose, might even go to jail. Because i exploited a weakness in the system with the purpose of stealing money and a weakness which wasn't in agreement with the intended purpose of the contract.
hero member
Activity: 770
Merit: 629
And the fact that not even the most primitive blockchains, those that aim to become payment systems, cannot work in a perfect decentralization manner should make you even easier to understand why such a complex project like ethereum cannot be possible with a perfect decentralization ( aka no developers, no leadership, as dinofelis naively understands it ).
The importance of the "pure form of decentralisation" (the only meaningful definition of that word in fact) is that it is the only guarantee of freedom of corruption. If the rules are graved in stone and cannot evolve, then nobody can put them to their hand once you're in it, and no government or other power structure can corrupt the leadership of the system, because there isn't any. Only a perfectly decentralized system has no "single point of failure" and can survive in a truly trustless and dangerous environment in which most of the others are your enemies, and only a few people, far in between, can be counted on. I see the most important reason for decentralized systems, as those that remain operational in "occupied territory by the enemy" (the state and all its servants and worshippers in my case), even if largely infiltrated. From the moment there is "leadership", you can forget about such properties. The only reasons why I care about decentralized systems.
hero member
Activity: 770
Merit: 629
From my POV and almost everyone else's, the immutability wasn't broken. The hackers money were locked thanks to the mechanisms the fat guy from dao put in place (atleast he got that right ). So, ethereum could hardfork without messing with any past transaction.
Well, if you change the RULES so that past transactions "don't count" any more, that's just as well killing immutability than when you change the data in the block chain. Except for the hacker's who lost all he took, because that was the point, to not let him get away. All that was doable because the tokens were locked, otherwise the hardfork wouldn't have been possible.
They were only *temporarily* locked and according to the rules that should have remained immutable, and according to the contract that should have been unstoppable, he should have gotten them. The only way he didn't get what he was supposed to get according to the execution of the unstoppable contract and the immutable rules, was to stop the unstoppable contract, by changing the rules (hard fork). Whether morally, legally, etc... right or wrong, according to the principles of an immutable set of rules on which an unstoppable contract runs, he should have gotten them. The denial of his having obtained his coins is a form of breaking permissionlessness, while ethereum was permissionless before, so this WAS a modification of the rules, and hence their immutability was not a matter of consensus, but of a (large) majority denying the use of the system according to the former rules by one single individual. Even EF agreed with that. Wether you think it was for investor's sake or to not let the hacker get away, that doesn't really matter. The point is, it was doable without messing with the transaction history, therefore immutability.
Immutability is both of course: the data and the rules. They modified the rules. I agree they didn't modify the data. If I give you a contract in which it is stated "I pay you 1000 dollars and you wash my car" and then I redefine "wash" into "handing over property" ; I redefine "my" by "possessive adjective of the second person" and I redefine "car" as "the land and the house in which one lives", then I can say that the contract, without changing a word, now specifies that I bought your house for 1000 dollars. Even though I didn't change a word of the "data". I changed the rules that tell what the data mean. And yes, central development is needed in public blockchain scene. Every project has it, can't be done otherwise. Bitcoin only is alive today because of central development ( remember that HF when shitloads of btc were created ? ).
There is a difference between developing software implementing a protocol, and modifying the protocol. My idea is that many people can develop software, change the user interface, do things on top, and so on, but that a crypto system, if it is to be decentralized, is laid down once and for all, and if you want to modify it, you are actually making another decentralized crypto system. If a system's protocol can be modified, it is not decentralized. It is then just a system like any other, with "leadership", "voting" and so on, and there's nothing special about it. Remember that before ethereum happened people didn't even agree with hardforks as necessary upgrade steps.
I'm still of that opinion, UNLESS regular hard forks are part of the protocol. That is, you could have block chains which have only a finite number of allowed blocks. At a certain point, the chain stops. You can then use this chain as a genesis block for another chain (or different other chains). If it can hardfork by a centrally planned authority, I wonder why one goes through the hassle of finding consensus algorithms: the central authority decides. Vitalik's private key could sign all blocks on the ETH chain, and you wouldn't need miners, or Casper. Proof of emperor. Much simpler. About immutability, read above. Nobody would agree with a hardfork that would break the immutability, aka change the transaction history.
No. Immutability means also that all rules remain eternally the same. That the rules are laid down with the genesis block, and that they will never ever change again, so that, when you buy your very first token, you have a contract with all other users of those tokens, that the rules are for ever graved in stone and will not change. Immutability is not just keeping the old blocks as they were, if you are allowed to change the meaning of those data, or what you can do with it. If the rules can change, and you don't know how, the system is not immutable, but arbitrary. If there are leaders that decide about these changes, the changes are not only arbitrary, but centralized. You only have customers that play according to the ever-changing rules and try to adapt their strategies along the constantly newly dictated rules.
hero member
Activity: 532
Merit: 500
@iamnotback yes and no, because BU-core war is different. Everyone knew ETH is official even before ETC was born. The "problem" was this: barry and chandler dumped ETH and pumped ETC, and obviously miners followed the price. Even though most mined ETC just to convert into ETH because everyone knew it is the official one,
This is why ETH is a centralized coin too. We're not going to go over this again, but remember that ETH was the fork to save the DAO investor's stake, and ETC continued the "unstoppable and immutable" chain in the spirit of the original idea. Y it fueled the illusion created by barry sillbert and chandler gao that a flippening was happening and that Ethereum foundation was going to switch aswell to ETC. But the illusion broke when vitalik officially stated that the vote was already cast out, the result both before and after the fork was clear ( like 95%+ of the hashrate followed the fork updating their client ). So, the result of the manipulation wouldn't have mattered because the EF was going to develop on ETH no matter what.
This is indeed when the market appreciates much more central leadership than decentralization and the principles that go with it. It is something to keep in mind in crypto: principles are OK, until they aren't, and big money is at stake. From my POV and almost everyone else's, the immutability wasn't broken. The hackers money were locked thanks to the mechanisms the fat guy from dao put in place (atleast he got that right ). So, ethereum could hardfork without messing with any past transaction. Except for the hacker's who lost all he took, because that was the point, to not let him get away. All that was doable because the tokens were locked, otherwise the hardfork wouldn't have been possible. Even EF agreed with that. Wether you think it was for investor's sake or to not let the hacker get away, that doesn't really matter. The point is, it was doable without messing with the transaction history, therefore immutability. And yes, central development is needed in public blockchain scene. Every project has it, can't be done otherwise. Bitcoin only is alive today because of central development ( remember that HF when shitloads of btc were created ? ).
Because ETH is implicitly "honest" about the BDFL being Vitalik and that its raison d'être (at least for the current juncture) is ongoing R&D experimentation (or at least that is the geekcool idealism marketing to drag in the greater fools to part them from their money), so it is clear that it is not intended to be immutable at this time. My mistake when I originally sided with ETC was I was projecting my idealism (protocol immutability) on the project, but the project's community has a different priority (also a form of idealism) which is about decentralizing experimentation in blockchain apps.
Thus ETH's overriding mandate isn't "code is the law" and instead is "decentralized community governance" is the law.
Remember that before ethereum happened people didn't even agree with hardforks as necessary upgrade steps. Because, they knew about bitcoin, litecoin, and every other shitcoin. And what all of them had in common ? Well all of them aimed to do the most basic things a blockchain can do, transfer tokens representing value. And with that mindset they looked at ethereum and they thought it sucks because it hardforks, because if BTC doesn't need it then why does eth ? Because it sucks ! None of them took into consideration what ethereum means, the fact that today it can do what top 50 projects can do combined and better is something still many can't even process. Ironic is the fact that soon it will be better than all of them combined with raiden's implementation, even as payment system, the only thing they can poorly do. About immutability, read above. Nobody would agree with a hardfork that would break the immutability, aka change the transaction history. And the fact that not even the most primitive blockchains, those that aim to become payment systems, cannot work in a perfect decentralization manner should make you even easier to understand why such a complex project like ethereum cannot be possible with a perfect decentralization ( aka no developers, no leadership, as dinofelis naively understands it ). And that "code is law" thing was mostly promoted by idealists and that is the finish point, where we all hope this trip will end. At this very moment we don't even have simple decentralized payment systems working properly after 8 years and we're so naively going with "code is law" ? That's like not understanding water but aiming for the stars. "Code is law" will never be the case, because code will be law as long as it will do what the developer intend it to do. Is that simple.
hero member
Activity: 770
Merit: 629
I just don't see how it is a "bankster cartel" when you have devs like gmaxwell constantly pushing for anonymity features like confidential transactions, mimblewimble etc, and guarantee that the network remains decentralized by letting the small man have a role by being able to run nodes.
I think you are fundamentally missing the expected dynamics of LN. In order for you, as a modest bitcoin user to be able to use the LN, you will have to open a channel to a LN hub. If you do 5 bitcoin transactions a year, to buy shoes, a book and two smartphones on Open Bazaar, you will need to keep your channel open to that hub for all that time ; if you close it, you have an on-chain transaction - which you could have used directly to pay the counter party and not use the LN ; but that's HUGELY expensive with a small "settling" chain block and A LOT OF TRAFFIC on the LN. Essentially, the on-chain traffic is soo difficult and expensive, that only VERY BIG AND SPORADIC settlements are going to be possible on chain. So you have to connect to a hub that will not settle your channel before you've done what you wanted to do with your bitcoins during a year. And it needs to have sufficient other channels so that you will be able to get through THIS partner to your destination (your coins are locked up in this channel, so you cannot put them in another channel, again, without a transaction on chain). --> so you better have a big, reliable hub partner, that is himself connected to other, big reliable hub partners, so that you can reach your destination *without settling*. If ever it settles along the path, you're in for an on chain transaction, which is too expensive and in any case, there's no room on chain. The only way this can reliably function is if there's a back bone of BIG hubs with A LOT OF COINS in their backbone channels, that reach *everybody* and to whom you REMAIN connected. In other words, these hubs are banks, and you are a customer, opening an account with them. They know ALL YOUR TRANSACTIONS, and who you are ; they put their "terms and conditions" and if you don't like it, you can't remain a customer. They may refuse certain transactions (they pull the plug and make you settle). They may require monthly payments from you to remain their customer. They may report all of your transactions to tax authorities. Maybe "being an LN hub" will become regularised ; people would enjoy the legal framing of LN hubs. The ONLY difference is that they cannot steal your money. But they work like a normal bank. Most normal banks don't steal your money. They put up conditions, fees, ... for you to remain their customer. The same Poisson statistics that pushes people to pool hash rate, is also the reason that big hubs have statistical advantage over small hubs: small hubs have to settle their small channels often, that is very expensive. Big hubs can keep their channels with a lot of coins open a long time, and divide the settling cost over many more transactions through that channel. In fact, the economies of scale go almost linearly with the invested amount of coins. I'm not saying that the rewards go linearly, no the *economies of scale* go linearly with invested amount of coins (even more so, quadratically, because of Poisson statistics). That's WAY WAY worse than the economies of scale in mining. Only big hubs can survive in the fee competition on the LN.
|
