阅读此篇论文受益匪浅，特此共享。接上一部分 坚定选择BCH的原因-3
Investigation of the potential for using the bitcoin blockchain as the world’s primary infrastructure for Internet commerce
调查将比特币区块链作为全球互联网商务主要基础设施的潜力
六. Removing the cap
六、取下盖子
Finney and Dillinger lobbied Satoshi Nakamoto heavily in 2009 to implement a block cap13.
Finney和Dillinger在2009年大量游说了Satoshi Nakamoto，以实施封顶。
The reason for this was that Satoshi’s vision of an economic market was heavily reliant on the value of transactions being able to restrict an attack.
其原因在于，satoshi对经济市场的看法严重依赖于交易能够限制攻击的价值。
In 2010 when the cap was implemented, it would have been possible to flood the network on a sustained basis for under USD 1 a day.
在2010年实施上限时，可能以每天不到1美元的价格持续洪水网络。
However, as the value of Bitcoin increased by over 15,000 times14 this attack was no longer viable.
但是，由于比特币的价值增加了15,000倍以上，这一攻击已经不再可行。
The addition of tiered fees, as was originally envisioned by Satoshi, was incorporated into his code to mitigate flooding attacks.
像Satoshi最初设想的那样，增加了分层收费，并被纳入他的法规中以减轻洪水袭击。
These fees would quickly render any flooding attack prohibitive and at the same time fund the miner into being able to economically profit from any attack.
这些费用很快就会使得任何洪水攻击都无法进行，同时使矿工能够从经济上受益于任何攻击。
In 2010, Satoshi determined that the block size would not be a limiting factor and that the cap was temporary.
2010年，Satoshi确定区块大小不会是一个限制因素，而上限是暂时的。
Shortly before he stopped actively developing code and direct interaction with the Bitcoin community, he implemented the change shown in Fig.
if (blocknumber > 115000)
maxblocksize = largerlimit
Fig. 11. Satoshi Bitcoin Talk post, October 4, 2010
图11. Satoshi Bitcoin Talk帖子，2010年10月4日
Satoshi’s original plan16 was to implement variable rate limiting to make it more expensive to attack Bitcoin whilst not impeding economic activity and allowing the market to determine the level of transactions that will be accepted.
Satoshi最初的计划16是实施可变利率限制，以便在不妨碍经济活动的情况下对比特币进行更为昂贵的投资，并允许市场确定将被接受的交易水平。
This change rewards all miners for accepting larger blocks and compensates them for the additional resources they require to mine larger blocks.
这一变化奖励所有的矿工接受更大的矿块，并补偿他们挖掘更大矿块所需的额外资源。
In alignment with Satoshi’s vision, the proposal is to implement a scaling process of:
为了与Satoshi的愿景一致，提案是实施一个缩放过程：
• 8MB
• 32MB
• No fixed limit with market-based fee and economic fee limits
以市场为基础的费用和经济费用限制没有固定的限制
In the seven years since the introduction of the cap, Moore’s law has held and the processing power of computers has increased more than 25 times.
上市七年来，摩尔定律已经持续了，电脑的处理能力增加了25倍以上。
By the end of the decade, the computing power will have increased by over 100 times relative to that of the systems that existed when the cap was introduced.
到十年末，计算能力将比引入上限时的系统增加100倍以上。
During this time, the cap has not changed and has become a severe limitation that is harshly restricting the growth of Bitcoin as a practical payment system.
在这段时间内，这个上限并没有改变，并成为一个严厉的限制，严格限制比特币作为实际支付系统的增长。
Clearly, the move from small home user systems to professional mining farms17 demonstrates the ability to safely move to larger block sizes as the fee system is developed and rigorously tested to enable the complete removal of any limits on the block size.
显然，从小型家庭用户系统到专业采矿农场的转变证明，随着收费系统的开发和严格测试，可以安全地移动到更大的块大小，从而能够完全消除块大小的任何限制。
This process will involve implementing the code in two stages. The first stage will see an increase in the block size that reflects the increase in computer power, after which a fee- and market- based system is planned, as seen in Fig. 12:
这个过程将涉及两个阶段的代码执行。第一阶段将会看到反映计算机功率增加的区块大小的增加，之后计划收费和市场化的系统，如图12所示：
if (blocknumber > 48600) //Around July 2017
maxblocksize =8000000 //8MB
if (blocknumber >538000) //Around Jan 2018
maxblocksize =32000000 //32MB
if (blocknumber >642500) //Around Jan 2019
maxblocksize =((block number - 642500) / 125) + 48000000
if (blocknumber >950000) //Around Jan 2025
maxblocksize = 1000000000000 // effectively uncapped and fee based有效地取消上限和收费
//Block size is configured to grow faster than Moore’s law and is controlled at this point forward with a market system of fees.
//区块大小被配置成比摩尔定律增长得更快，并且在这一点上被控制在一个市场收费系统中。
//The initial size is set and grows with time and is rate limited allowing Bitcoin to grow to a point where it can compete with Visa and PayPal
//最初的大小是随着时间的推移而增长的，而且是有限的，使得比特币可以增长到与Visa和PayPal竞争的地步
Fig. 12. Planned staged increase of Bitcoin block size
图12.计划中的比特币区块大小分阶段上涨
A dynamic system that is unrestrained and which implements a change responsibly and predictably followed by a fee-based system and the removal of all limits would enable the market to adjust to the changes and allow Bitcoin to become the system it should be.
一个动态的系统是不受限制的，它实现了一个负责任的和可预测的收费系统，并取消所有限制的变化，将使市场适应变化，使比特币成为它应该的系统。
As the size of a block is determined by fees, it would become uneconomical for an attacker to mount a sustained attack on Bitcoin, because any attempt at a sustained flooding attack would fund the systems needed to stop the attack.
由于区块大小是由费用决定的，攻击者对比特币发动持续攻击将变得不经济，因为任何持续的洪水攻击企图将资助停止攻击所需的系统。
Fig. 13. Transaction propagation in transactions/second with a proposed scaling network of specialized nodes (Tx denotes transactions)
图13.事务传播在交易/秒与专业节点的提议的缩放网络（Tx表示交易）
These changes would be implemented using the initial code shown in Fig. 14, and which is taken directly from Satoshi’s original code implementation.
这些改变将使用图14中所示的初始代码来实现，并且直接从Satoshi的原始代码实现中获取。
The implementation of a fee-based market structure will be created by allowing both community and miner interaction to ensure that the implementation is safe, tested, and is aligned to the market.
通过允许社区和矿工的互动确保实施安全，经过测试并与市场接轨，将创建收费型市场结构。
// Transaction fee requirements,mainly only needed to for flood control Under 10K (about 80 inputs) is free for first 1,000 transactions
//交易费用要求，主要只需要防洪10K以下（约80个输入），前1000次交易免费
// Base rate is 0.00001 per KB
//基本速率是每KB 0.00001
int 64 nMinFee = tx.GetMinFee(block->vex.size() < 1000);
Fig. 14. Code for implementing the change in the block size based upon Satoshi’s original code implementation
图14.根据Satoshi的原始代码实现来实现块大小变化的代码
The value in Fig. 14 is based on the original reference implementation by Satoshi.
图14中的值是基于Satoshi最初的参考实现。
The change is effectively a transition from 100 to 1,000 free transactions in a block to encourage growth. The minimum fee would be capped at USD 0.001 (one tenth of a cent) and this level can be referenced to the monthly average value of Bitcoin based on a three-month rolling average.
这个变化实际上是从一个街区100个到1000个免费交易的过渡，以鼓励增长。最低费用的上限为0.001美元（十分之一美分），这个水平可以参考比特币的月平均价值，以三个月的平均价格为基础。
Fig. 15. Fee structure based on uncapped volume.
图15.基于不限量的费用结构。
The fee-based rate controls will be implemented by adding two new variables derived from the original code:
基于费用的费率控制将通过添加两个从原始代码派生的新变量来实现：
int64 nStdFee for the next 10,000 Transactions
int64 nStdFee为下一个10,000交易
int64 nMaxFee for all other Transactions in a block
int64 nMaxFee用于块中的所有其他事务
nStdFee will encompass the majority of transactions at a rate of USD 0.005 (one half of a cent).
nStdFee将包含大部分交易，价格为0.005美元（1.5美分）。
nMaxFee is the flood limiter. Users who are willing to pay in times of congestion would be able to send transactions quickly at a rate of USD 0.05 (five cents).
nMaxFee是洪水限制器。愿意在交通拥堵时付费的用户将能够以0.05美元（5美分）的费率快速发送交易。
The fees are summarized in Table 1.
这些费用总结在表1中。
Table 1
Fee structure and example breakdown
费用结构和例子分解
Currently, the fees are flexible and decided by the market.
目前收费灵活，由市场决定。
The change will be implemented by January 2018 to allow the miners time to change to the new code and to prepare for the introduction of a variable block size.
这一改变将在2018年1月实施，以便矿工有时间更改新的代码，并为引入可变块大小做准备。
Miners can stop accepting transactions at any time knowing that other miners will accept these and incorporate a growing fee base that will be lost to a miner not accepting larger block sizes when the transactions are at a level the miner chooses not to process.
矿工们可以随时停止接受交易，因为他们知道其他矿工会接受这些交易，并将不断增长的收费基础纳入矿工选择不处理的交易水平，而不会接受较大规模的矿工。
Over time, miners will become increasingly incentivized to incorporate as many transactions as they can.
随着时间的推移，矿工会越来越激励，尽可能多地进行交易。
The arguments against nodes have misconstrued the nature of a node.
反对节点的论点误解了一个节点的本质。
Rather than being a full implementation of the Bitcoin software, a node consists of software with mining enabled.
比较完整的比特币软件，节点由启用了挖掘的软件组成。
Nodes have been defined in the original readme.txt18 file distributed with the reference client. This document states:
已经在参考客户端分发的原始readme.txt文件中定义了节点。该文件指出：
To support the network by running a node, select:
Options->Generate Coins and keep the program open or minimized.
要通过运行节点来支持网络，请选择：选项 - >生成钱币并保持程序打开或最小化。
It runs at idle priority when no other programs are using the CPU.
当没有其他程序正在使用CPU时，它以空闲优先级运行。
Your computer will be solving a very difficult computational problem that is used to lock in blocks of transactions.
您的计算机将解决用于锁定交易块的非常困难的计算问题。
The time to generate a block varies each time, but may take days or months, depending on the speed of your computer and the competition on the network.
生成块的时间每次都不相同，但可能需要几天或几个月，具体取决于计算机的速度和网络上的竞争。
It's not a computation that has to start over from the beginning if you stop and restart it.
如果您停止并重新启动计算，则不必从头开始计算。
A solution might be found at any given moment it's running. As a reward for supporting the network, you receive coins when you successfully generate a block.
在运行的任何时刻都可能找到解决方案。作为支持网络的奖励，当您成功生成一个块时，您会收到硬币。
The implication is that the reference client is not a node unless it is also mining blocks.
其含义是参考客户端不是一个节点，除非它也是挖掘块。
七. Concluding remarks and recommendations
七.总结发言和建议
Our analysis of the Fast Payment Network entailed simulating a variety of scenarios based on existing node distribution models.
我们对快速支付网络的分析需要根据现有的节点分布模型来模拟各种情况。
The growth of the network takes away certain possible scenarios such as the widespread distribution of home nodes.
网络的增长带走了一些可能的情况，例如家庭节点的广泛分布。
The primary reason for such a variation has its origins in the need to process many transactions.
造成这种变化的主要原因是由于需要处理许多交易。
External transactions alone account for over 5,000 TPS across the major card processing companies.
仅主要卡处理公司的外部交易就占了超过5,000 TPS
When we incorporate the full range of electronic transactions we obtain a figure of the order of 100,000 TPS for the current state of the world financial system.
当我们纳入全部电子交易时，我们获得了当前世界金融体系的10万TPS数量级的数字。
The rate of 5,000 TPS at the level of the existing card processing companies provides for a level of transactions that equates to a block size of 240 MB.
现有卡处理公司级别的5000TPS的速率提供了相当于240MB的块大小的交易级别。
This is the minimum block size Bitcoin would need to achieve to be competitive.
这是比特币需要达到的竞争力的最小块大小。
This assumes the use of the Blockchain for only electronic commercial usage.
这假定仅将区块链用于电子商业用途。
A complete change of the financial system would require the ability to scale to over 100,000 TPS and a block size of the order of 10 to 12 GB for each 10-minute period.
财务系统的彻底改变将需要能够扩展到超过10万TPS，并且每10分钟的时间内需要10到12 GB的块大小。
We can increase the block size at Moore’s Law for a single host node and this could be achieved slightly faster by creating specialized systems, which would enable us to revert to following exponential Moore’s Law.
我们可以增加单个主节点的摩尔定律的块大小，通过创建专门的系统可以稍微快一些，这将使我们能够恢复遵循指数摩尔定律。
Traffic that incorporates the Internet of Things and the proposed growth rate of what could become trillions of devices needs to scale further to be effective.
包含物联网的流量和可能成为万亿设备的增长速度需要进一步提高。
This would require the incorporation of two aspects.
这将需要纳入两个方面。
First, we would need to investigate the time taken to distribute block sizes as large as those proposed above.
首先，我们需要研究分配与上面提出的那些块一样大的块的时间。
Second, we would need to consider the time taken to process the transactions within this block size.
其次，我们需要考虑在这个区块大小内处理交易的时间。
A miner can reduce the necessity of validating all information they receive prior to starting to analyze and solve the block puzzle if they are able to rely on the information they have received.
矿工可以在开始分析和解决块拼图之前，减少验证所有收到的信息的必要性，只要他们能够依靠他们收到的信息。
A market for information can be provided where a miner can reduce their infrastructure cost by specializing in the solution of the hash puzzle and allowing for a separate infrastructure dedicated to the validation of information contained within the block.
在矿工可以通过专门解决散列拼图来降低基础设施成本并允许单独的基础设施专用于验证块内所含信息的情况下，可以提供信息市场。
This distribution function leads to a scenario where the miner specializes in the solution of a hash problem associated with previously validated information.
这种分配功能导致了矿工专门解决与先前验证的信息相关联的散列问题的情况。
The merchant node not only assumes the role of validating the system to ensure that the signatures are correct and that the hash function is effective but also provides it to the miner in a timelier manner.
商家节点不仅承担验证系统的角色，以确保签名是正确的，并且哈希函数是有效的，而且还以更及时的方式提供给矿工。
In time, different node functions are expected to become more specialized.
随着时间的推移，不同的节点功能将会变得更加专业化。
Rather than maintaining all aspects of the Bitcoin Blockchain, a processing node would be able to utilize the benefits of the Merkle tree structure to hold only a prior selection of data.
处理节点不是维护比特币区块链的所有方面，而是能够利用Merkle树结构的好处来仅保留先前选择的数据。
Specialized nodes that hold complete copies of the Bitcoin Blockchain would be able to fulfill the supporting roles that could be offset from both mining and processing nodes.
持有比特币区块链完整副本的专业节点将能够完成可以与采矿和处理节点相抵消的支持角色。
This distribution function within the Bitcoin Blockchain would enable the infrastructure to scale far more than has been envisioned.
比特币区块链中的这种分配功能将使基础设施的扩展能力远远超出预想。
In doing this, we would also provide a means of growing the distribution of node entities.
在这样做的时候，我们也会提供一种增加节点实体分布的方法。
The current system has changed such that a small number of mining conglomerates control the majority of hash power within the network.
目前的系统已经发生了变化，使得少数矿业集团控制网络中的大部分散列能力。
This has led to a distributed non-aligned cartel-like system with much of the power held within a limited group and outside the control of most user nodes.
这导致了一个分布式不对等的卡特尔式的系统，大部分的权力被限制在一个有限的组中，并且在大多数用户节点的控制之外。
The belief that 6,000 nodes presently control the Bitcoin network is a fallacy based on a lack of understanding of how the voting system operates.
现在6000个节点控制比特币网络的信念是基于对投票系统如何运作缺乏了解的谬误。
The system we are proposing allows far more competition and overcomes many of the flaws introduced by centralized mining organizations that rely on a collection of free nodes.
我们提出的系统允许更多的竞争，并克服依赖于一系列免费节点的集中式采矿组织引入的许多缺陷。
Our research demonstrates the feasibility of a propagation system that can take generalized connectivity between any system incorporating both free nodes and paid nodes allowing for the propagation of anything from single Satoshi transactions up to including large transactions that are guaranteed across the network.
我们的研究证明了传播系统的可行性，该系统可以在任何包含空闲节点和付费节点的系统之间进行通用连接，从而允许从单一的Satoshi事务到包括在整个网络中保证的大事务的传播。
What we have shown is the need not so much of increasing the number of nodes but the amount of processing power, where we take the data going up to 10,000 individual nodes to where we expect 1000-1200 enhanced nodes in the future.
我们所展示的不是增加节点的数量，而是增加处理能力的数量，在那里我们把数据上升到10000个单独的节点，我们预计将来有1000-1200个增强节点。
The implementation of a DAC in the system would allow users to invest in the network and would allow merchants who pay this to fund the creation of autonomous systems that propagate information and act in a provably secure manner.
在系统中实施DAC将允许用户投资网络，并允许商家支付这个资金来创建传播信息并以可证实的方式行事的自治系统。
Each of these nodes would be verifiable independently outside of the network.
这些节点中的每一个都可以独立于网络之外进行验证。
This study, in which we modeled double spend attacks and network scalability, was primarily concerned with the implementation of paid specialized nodes can enable the fast dissemination of data to render double spending unviable.
本研究中，我们模拟双重支出攻击和网络可扩展性，主要关注付费专业节点的实施可以使数据的快速传播，使双重开支不可行。
This would mean that, even in large networks, merchants would be able to detect from their connected nodes that an attempted double spend had occurred within seconds enabling a merchant to cancel a transaction and stop the transfer of goods to a client.
这意味着，即使在大型网络中，商家也能够从其连接的节点中检测到在数秒内发生了尝试性的双重花费，从而使商户取消交易并停止将商品转移给客户。
The ability to link transactions consisting of multiple transactions for the transfer of goods and payment simultaneously means that there are more effective solutions for the integrated provision of digital rights and other exchanges that cannot be conducted using a double spend.
将多个交易组成的交易同时进行货物和付款转移的能力意味着，对于数字权利和其他交换的综合提供，双重花费是无法实现的更有效的解决方案。
For instance, if a transaction for the sale of property was linked to the payment of goods in a single transaction, a double spend would violate this principal and violate the transfer of the goods itself.
例如，如果出售物业的交易与单一交易中的货物支付挂钩，双重支出就会违反这一原则，并且违反货物本身的转移。
Therefore, we did not concentrate on this and the solution to these issues will be provided in a subsequent paper.
因此，我们并没有把注意力放在这个问题上，这个问题的解决方案将在后续的论文中提供。
As competing networks seek to optimize their losses, a merchant will not necessarily seek to achieve zero losses if the cost of stopping or mitigating that loss was to exceed the cost of the loss itself.
由于竞争网络寻求优化其损失，如果停止或减轻损失的成本超过损失本身的成本，商人不一定会寻求实现零损失。
This ability of the proposed network to encourage competitive growth is expected to allow it to become a global payment system.
拟议的网络能够鼓励竞争性增长，有望使其成为全球支付系统。
The basis of the Bitcoin network is derived through epidemic propagation.
比特币网络的基础是通过传播传播。
Rather than proposing to change this, we seek to introduce an additional layer into the network that improves the efficiency of the propagation of transactions in a manner that allows neighboring systems to interact more rapidly.
我们不是试图改变这种情况，而是试图在网络中引入一个附加层，以提高交易传播的效率，使相邻系统之间的交互更加迅速。
Competing node networks would still have to broadcast transactions that are sent to them across the remainder of the network.
竞争的节点网络仍然需要广播在网络的其余部分发送给它们的事务。
Where there is no trust between these entities a gossip network remains the preferred solution.
在这些实体之间没有信任的地方，八卦网络仍然是首选解决方案。
However, the efficiency of the network will also be maintained and increased not only through competing nodes but through an amalgamation of both competing and cooperating nodes.
然而，网络的效率也将不仅通过竞争节点而且通过竞争和合作节点的合并来保持和增加