Daniel Larimer(BM)对PoW优势的思考



  • 文章来源--------公众号:我药学习

    The Benefits of Proof of Work
    工作量证明的益处

    I have long been of the opinion that proof-of-stake is superior to proof-of-work. Today I would like to challenge my own beliefs against the lessons I have learned from over a year of experience with proof-of-stake. In the past I have argued that proof-of-work is wasteful and that the same end result can be achieved much cheaper with proof-of-stake. Today I would like to consider if there is some value to proof-of-work.

    我长期以来都认为PoS(Proof of Stake,权益证明,区块链技术共识机制的一种)机制优于PoW(Proof of Work,工作量证明,同为一种经典的区块链共识机制)。今天我想挑战下自己的观点,这些观点来自于我一年多的PoS相关经验所学。过去我一直认为PoW很浪费,且PoS可以用便宜的多的方式达成同样的结果。今天我要思考PoW会不会其实有它的价值所在。

    Mandatory Disclaimer
    免责声明

    What follows is an idea and perspective on proof-of-work and does not constitute a proposal to modify BitShares, its dilution rate, or its voting algorithms.

    以下只是对PoW的一些观点与想法,不够成对BitShares及其稀释率和投票算法的修改建议。

    Work Defined
    定义工作量(或功)

    If we are going to talk about the benefits of proof-of-work, then it helps to have a useful definition of work. In terms of physics, work is defined as power X time.

    如果我们要讨论工作量证明(PoW)的优点,那么就需要对【工作量】有一个有用的定义。在物理中,【功(工作量)】被定义为【功率】乘以【时间】

    W = Pt

    In more general terms, I would define work as something you must pay people to do. If it doesn’t cost anything then no work was done. Bitcoin satisfies both definitions of proof-of-work. No one would spend money building specialized hardware and consuming electricity if they weren’t paid to do so.

    在通常情况下,我把工作量定义为你必须【付钱】让人做的事情。如果没有付出任何代价,那就没有任何工作量被完成。比特币满足以上两种对工作量的定义。如果没有被支付报酬,没有人会花钱去打造专门的硬件并消耗电力。

    Therefore, Bitcoin’s block difficulty represents mathematical proof that about 25 BTC worth of work was performed. The corollary here is that if you pay a fixed amount of money into a competitive market then market competition will drive the work required to earn that money toward the value of the money.We can view Bitcoin Work as Difficulty or HashPower * 10 minutes.

    因此,比特币的挖矿难度事实上就是一种数学证明,表明大约有价值25BTC的工作量被完成了。这样推导下去的结果就是,如果你在一个自由竞争的市场中支付一笔固定的金钱,那么市场竞争就会使得获得这笔金钱的所需的工作量与这笔金钱本身的价值趋同。我们可以把【比特币工作量】看作【难度】或【哈希算力乘以10分钟】(注:这里哈希算力即对应物理学中的功率)

    W = D = HP * 10m

    In proof-of-stake systems we only have power, aka stake. I would like you to consider that committing to hold a token for a period of time in the future is work. No one gives up liquidity or locks funds up without expecting interest. The value of the interest paid is therefore proportional to the work required to earn it. In the same way, the value of the Bitcoin mining rewards is proportional to the work required to earn it.

    在PoS(权益证明)机制中,我们只有算力(功率),亦即权益。我希望你这样想:承诺在未来一段时间持有一个代币即是工作量。除非预期有利息,没人会放弃流动性或锁定资金。利息中包含的价值与为了获得利益所需的工作量是成比例的。同样的,比特币挖矿所获得回报的价值与获取这些收益所需的工作量也是成比例的。

    Based upon this understanding of work and the value of liquidity we can construct a proof-of-work system that doesn’t require a large amount of money to be transferred to the electric company and mining hardware manufactures. This proof-of-work system does require the money to be transferred to someone though.

    基于这种对工作量及流动性价值的理解,我们可以构建一种不需要把大笔的钱用于支付电费和硬件的工作量证明系统。不过这种工作量证明系统依然须要把钱转给某人。

    In this case, it is transferred to those who committed to hold the token by diluting those who demanded liquidity. This has the effect boosting the value of a token in the short term by postponing sell pressure.

    在这种情况下,报酬被支付给了那些承诺通过持有代币来削弱流动性的人们。这样可以通过推迟卖压来在短期内推升代币价值。

    There are two ways to do proof-of-work with stake:
    1.Based on how long you held a coin in the past
    2.Based on how long you promise to hold a coin into the future

    有两种通过权益来做到工作量证明的途径:
    1. 基于你过去持有一个币多久
    2. 基于你承诺在未来持有一个币多久

    I will submit for your consideration that what someone did or didn’t do in the past is a risk-free sunk cost similar to renting mining equipment on a day to day basis. On the other hand, promising to hold a coin into the future is a high-risk cost similar to buying mining equipment that takes a year to earn back the initial capital. In terms of security, the future value of a coin is the only thing that matters and this is why option #2 is superior.

    我邀请你这样想——一个人在过去已做或未做的事情是一种无风险的沉没成本,类似于以日租的方式去租赁挖矿设备。另一方面,承诺在将来持有一个币则是一种高风险成本,类似于购买一年才能回本的挖矿设备。从安全角度来讲,唯一重要的流逝这个币在未来的价值,因此选项2更优(基于承诺在未来多久内持有这个币来衡量工作量)。

    Everyone Hates Politics
    人人都讨厌ZhengZhi

    As blockchains become more complex, politics becomes a bigger and bigger issue. Proof-of-work has this wonderful property of appearing to be free of politics. At a consensus level it is mathematically simple, clean, and fully decentralized. In theory anyone can join the network and start producing blocks.

    随着区块链越来越复杂,ZZ变成了一个越来越大的问题。工作量证明具有的美妙性质使其看起来似乎免受ZZ的影响。从共识的层面来说,它数学上足够简单、干净以及完全地去中心化。理论上来说任何人都可以加入网络并开始生产区块。

    In practice, proof-of-work has created a pay-to-play model where those who are willing to spend the most to control the network win. The security in such a system comes from the expense and opportunity cost of mining a particular block and the fact that those doing the mining have a long-term capital commitment and therefore act in a manner to maximize their return.

    但在实践中,工作量证明创造了一种【付钱才能玩】的模式,使得那些愿意花最多钱来控制网络的人成为赢家。这个网络的安全性来自于1/挖出一个区块的花费和机会成本,2/以及矿工的长期资本投入并因此以最大化回报为方向的行为方式。

    Politics is Unavoidable
    ZZ是不可避免的

    We have learned from the Bitcoin XT debate that politics is unavoidable. Proof-of-work doesn’t eliminate the politics it simply changes where the politics play out. Rather than a stake-weighted-vote on the blockchain it becomes a web poll or a vote at the Bitcoin Foundation. Mining pool operators end up voting on which fork to support and users vote with their CPU power.

    我们从围绕比特币XT区块扩容之争中了解到到ZZ是不可避免的。工作量证明并没有消灭ZZ,而只是让其换了个舞台而已。它变成了比特币基金内部的在线投票,而不是区块链上的权益权重投票。矿池最终投票决定支持哪条分叉,而用户则通过CPU算力来投票。

    Because politics is unavoidable, it makes the most sense to have the voting done directly on the blockchain. Bitcoin and Peercoin do this voting by signaling with each block that is produced, but this isn’t the only or most convenient way to do voting. Instead of voting by block production, voting can be done via transactions that indicate how “coindays” or “committed stake” should vote. This voting can even include voting on the set of active block producers and when they should be producing blocks like is done with delegated-proof-of-stake.

    既然ZZ是不可避免的,那么最合理的做法就是直接在链上进行投票。比特币和点点币通过向生成的区块发送信号来投票,但这并非最方便的投票方式。与其通过区块生产来投票,投票可以通过能表明【币天】或【投入权益】应当如何投票的转账来完成。这种投票可以通过【代理权益证明(DPoS)】来完成,其范围甚至可以包括活跃区块生产者的组建以及他们何时应该生产区块。

    The primary challenge with Bitcoin and Peercoin style proof-of-work systems is that only technically savvy individuals who can endure the costs of running a full node on the network may participate in the voting process. The Bitcoin economy solves this problem via cloud mining operations where non-technical users “vote-by-proxy” through hiring someone else to mine on their behalf.

    比特币和点点币类的PoW系统所面临的主要挑战在于,只有技术上懂行且能够负担得起在网络上运行一个全节点的成本的人才可能参与投票过程。比特币经济解决这个问题靠的是云挖矿,非技术用户通过雇佣别人为他们挖矿来进行“委托代理投票”。

    Voter Apathy
    投票者无感

    Under delegated proof-of-stake one of our major challenges is voter apathy. Many people vote once and then forget to change their vote or they vote for a proxy and then forget to follow up. The usual end result is that it becomes difficult to vote out incumbents or vote in new individuals. The presence of voter apathy is a sign that incentives are not properly aligned in the system.

    在代理权益证明机制(DPoS)下我们的主要挑战是投票者无感。人们投过一次票后忘记修改他们的投票,或者他们给一个代理投票后却忘了跟进。最终结果就使得把现任投出局或者把新人投上位变得困难。投票者无感的存在标志着激励与系统不一致。

    This is where migrating from delegated-proof-of-stake to a delegated-proof-of-work system has some serious benefits. When power is in the hands of users who have a long-term capital commitment to a project they have greater incentive to vote because they cannot sell for months or years. Under the existing delegated-proof-of-stake system most users opt to passively accept the decisions made by others and then vote with their feet by selling the token if they don’t like the result.

    此时从委托权益证明(DPoS)向委托工作量证明(DPoW)转变就有一些非常重要的好处。当权力在那些对一个项目有长期资本投入的用户手里时,他们有更大的动机去投票,因为他们在数月或数年内都不能抛售。在现存的DPoS系统下,大部分用户都选择被动地接受其他人的决定,然后如果他们不喜欢这个决定的结果,就通过抛售手头的代币一走了之。

    When users have the ability to “vote without commitment” the whole community plays a heads-I-win, tails-you-lose game where winning the vote means losing the minority who disengage in the process which then undermines the value of the token.

    当用户可以“无约束投票”时,整个社区就会变成一场“正面我赢,背面你输”(注:即不管怎样社区都会受到伤害,正面背面指硬币的两面)的游戏,赢得投票就意味着那些少数未参与到这个过程的人会流失,致使代币的价值受损。

    Eliminating Exchange Risk
    消除交易所风险

    Stake-commitment based proof-of-work has the added benefit of reducing the risk posed by centralized exchanges having a large percentage of network stake. In BitShares 25% of the stake is held in the top several exchanges. These exchanges have opted to “not-vote” and thus do not help secure the network, but the risk remains that at any moment they could start voting and have complete control over the network.

    基于权益约束的工作量证明可以带来额外的好处——它可以减少中心化交易所持有大比重的网络权益所带来的风险。在Bitshares,25%的权益是由几家最大的交易所持有的。这些交易所选择不投票,因此未对网络安全起到帮助,但是在任何时候,它们开始进行投票并彻底控制整个网络的风险都是存在的。

    By adding stake-commitment based proof-of-work to the mix, the exchange would be unable to both vote and guarantee withdraw on demand. In the best case, the exchange could keep their funds divided into a ladder of commitments such that they could maintain withdraw requests.

    通过混合加入基于权益约束的工作量证明,交易所就无法在投票的同时保证提现(注:如同比特币里的【币天】机制引入了时间因素,如果在投票机制中加入时间因素,例如投票后代币被锁定,那么投票后交易所就可能无法及时满足用户的代币提现需求,增加了持有大量代币的中心化交易所投票作恶的成本)。在最佳情况下,交易所可以把它们的资金划分为一个义务梯度,以保持提现需求的稳定。

    Is Proof-of-Work Wasteful
    工作量证明是否浪费

    The $1 million dollar a day question is whether or not money spent on proof-of-work is wasteful. To answer this question requires making a comparison among several alternatives. For the sake of this analysis I am going to assume that there is a fixed cost for the proof-of-work and then look at the value of what is solved. Bitcoin and other similar coins use proof-of-work to solve several different problems at the same time:
    1.Deciding when the next block should be produced
    2.Deciding who should produce the next block
    3.Making it expensive to produce an alternative chain
    4.Deciding on the best blockchain
    5.Distribution
    6.Side Benefits (such as Primecoin)

    “每日一百万美元”问题在于花钱在工作量证明上是否属于浪费资源。要回答这个问题,我们需要在一些替代选项之间做出比较。为了便于分析,我假设工作量证明机制具有一个固定成本,然后去分析被解决问题具有的价值。比特币和其他类似使用PoW的币同时解决了以下几个不同问题:
    1/决定【何时】生产下一个区块
    2/决定【何人】生产下一个区块
    3/让生产替代链的成本高昂
    4/选出最好链(即确定共识链)
    5/分发
    6/附加利益(如质数币,又叫素数币)

    Of these six things, mining is a terribly inefficient/slow way to determine when blocks should be produced and who should produce them. A lot of work is wasted due to a failure to reach consensus on when and who in advance. This leaves only solving the last 4 problems as a means to justify the cost.

    这六项当中,挖矿在决定区块的生产时间及生产者上极其低效。因为没有提前就此达成共识,大量的工作量被浪费了。这就使得后四项问题的解决其实仅仅只是使成本合理化的手段而已。

    The capital commitment that bitcoin mining enforces creates a kind of self-reinforcing lock-in. People commit to the project, invest in equipment and then are forced to market and promote Bitcoin over a long-term basis in order to recover their costs and make a small profit. Distributing the new coins via mining provides an easy way to get new users on board without having to buy in. Both of these things have non-zero value.

    被比特币挖矿行为推高的资本投入创造了一种自我强化的闭环。人们投入这个项目,投资设备,然后被迫长期去营销与推广比特币以收回成本并从中获利。通过挖矿分发新币则给新用户提供了一种不用非要买币的简单上车方式。所有这些都是具有价值的。

    If BitShares diluted to pay people who produce difficult hashes built off of recent blocks then this work would end up securing the network with the same kind of cost and objective proof as Bitcoin. This would give BitShares the benefits of solving problems #3, #4, #5, and #6 while still retaining the benefits BitShares enjoys in terms of consistent 3-second blocks and the elimination of the need for mining pools.

    如果Bitshares通过稀释(增发?)来付报酬给那些在新区块后进行困难哈希运算的人们,那么这最终会以与比特币同类的成本及客观证明来保证网络的安全。这样会赋予Bitshares解决#3#4#5#6问题后所带来的收益,并同时保有Bitshares原本具有的3秒出块及无须挖矿的优势。

    So the question of whether or not a particular kind of proof-of-work is wasteful depends only on the relative value of side benefits to the value spent. In the case of Bitcoin or Peercoin, the side benefits include the generation of specialized hardware. In the case of stake-commitment proof-of-work the side benefit is that a large percentage of a token are kept out of circulation which helps increase the value of the liquid tokens. In other words, instead of investing money outside the ecosystem in hardware and electricity, money gets invested into the ecosystem in terms of long-term investment support.

    所以一种工作量证明机制是否属于浪费主要取决于附加利益与被消耗价值之间的相对价值。在比特币和点点币的例子中,附加利益还包括特殊硬件的产生。在有权益义务的工作量证明中,附加收益则在于大比例的代币没有流通,因此促进了流通部分代币的价值。换言之,相比其被投入到生态之外的电费及硬件上,金钱以长期投资支持的形式被投入到了生态系统内部。

    When stake-commitment proof-of-work is used, value is transferred from those who need liquidity but choose to hold anyway to those who give it up for fixed periods of time. This creates an investment vehicle that is more than just speculation.

    当有权益义务的工作量证明被使用时,价值从那些需要流动性但却选择持有的人(注:短期持有的投机者)那里转移到了那些在固定时间内放弃流动性的人(注:长期持有的投资者)。这样就创造了一种不仅仅只是用于投机的投资工具。

    Investor vs. Speculator
    投资者与投机者

    Sometimes it is helpful to consider the difference between a speculator and an investor. An investor puts money into a company in exchange for equity that may not be liquid for years. The money he put in goes directly toward funding the company and contributes to the long term value of the company. A speculator on the other hand could deal entirely in derivatives because they only care about changes in the price. In other words, speculators buy and sell frequently and have no loyalty. The most that can be said for speculators is that they provide liquidity that allows people to enter and leave the market at smaller spreads.

    有时候思考一个投机者与一个投资者之间的区别是有帮助的。投资者把钱投入一家公司来交换可能数年内都不会有流动性的股东权益。他的钱直接用于对公司进行资助,对公司的长期价值做出贡献。反过来,一个投机者可能完全只交易衍生品,因为他只关心价格的变化。换句话说,投机者频繁买卖,没有忠诚度可言。关于投机者最多只能说他们为市场提供了流动性,让人们能以最小价差进出。

    A startup company that has 100% of its outstanding stock on the open market has a harder time attracting new investment because newly created shares are competing with existing shares for today’s liquidity. For this reason almost all startups prefer to tie up investor funds and deny liquidity until they reach IPO or similar liquidity event.

    一个把100%的已发行股票都放在公开市场上的初创公司会很难吸引新的投资,因为新创造的份额会与目前已有的份额争夺当前的流动性。由于这个原因,几乎所有初创公司都会偏向捆住投资者的资金,以降低流动性(注:不让投资者自由抛售、套现)直到IPO或类似的流动性事件。

    By diluting speculators to compensate investors with a long-term outlook you can have the best of both worlds. If all investors lock up their funds for an equal period of time then there is in effect no dilution faced by investors. On the other hand, if some investors demand liquidity then the other investors profit.

    通过稀释投机者,并以长期的市场前景来补偿投资者,最终结果对双方都是最好的。如果所有投资者把资金锁定同样长的时间段,他们就不会面临稀释效应。另一方面,如果有些投资者需要套现流动性,那么其他的投资者就会获利。

    Distribution
    分配

    When a new company is created shares are allocated to those who contributed to the company proportional to their contribution. If two partners each put in equal work in the first year they should each own 50% of the company. If in the second year one of the partners quits, then at the end of the second year one partner should have 67% and the other 33% based upon the amount of work they contributed.Based upon this theory, then all dilution to pay for work the company wants done is legitimate.

    当一家新公司被创立时,它的股份会根据贡献度来按比例分配给贡献者。如果两个合伙人在第一年投入同样的工作量,他们应该每人拥有公司的50%。然后如果其中一个合伙人在第二年退出了,那么根据两者贡献的工作量,一人应该拥有67%的股份而另一人应该拥有33%。基于以上理论,这种为了支付公司所需的工作量而进行的稀释是合理的。

    The only question that matters is whether the work being done is worth having done. To use the partner example, if in the second year the remaining partner dug holes and filled them back in then the original partner may be upset because that work did not bring any additional value to the shares.
    From this perspective, paying individuals to commit funds for longer periods of time and/or using the rate of pay as a means of regulating price volatility provides enduring value where as paying for miners might have less and less value as the amount of work increases.

    唯一要紧的问题就是被完成的工作量是否真的值得做。引用前面的合伙人问题,如果第二年中留下的那个合伙人只是在反复挖坑然后再把这些坑填上,那原来离开的那个合伙人就会比较郁闷,因为这些工作量并没有为股票带来附加价值。从这个视角来看,给那些承诺更长时间里锁定资金的投资人付报酬以及/或者通过报酬率来规范价格波动的做法,将能够带来持久的价值。而付给矿工报酬所带来的价值则可能随着工作量上升而下降。

    Conclusion
    结论

    After a year gaining experience with delegated-proof-of-stake (DPOS) I have concluded that certain kinds of proof-of-work are actually incredibly valuable and useful to an ecosystem’s development and that using dilution as high as 15% on short-term speculators to compensate long-term investors can be very helpful in both securing the network, building loyalty, and creating a profitable system.

    经过在DPoS机制上积累了一年经验后,我得出了这样的结论:某些特定形式的PoW对生态的发展其实是非常有价值且有用的,并且通过对短期投机者进行高达15%的稀释来补偿长期投资者是非常有帮助的,不论是对于保证网络安全,建立忠诚度还是打造一个具有盈利的系统。


Log in to reply
 

Looks like your connection to Asch was lost, please wait while we try to reconnect.