2016年2月29日，在持续近十天关于香港圆桌会议的争论后，在巴比特论坛进行了一场主要面向中国社区用户的AMA。Adam back，比特大陆CEO吴忌寒，BTCC COO缪永权参与了这次AMA，在持续近4个小时的时间里耐心解答了中国社区用户提出的每一个问题。
Bitcoin is a user currency and operates by miners, exchanges, ecosystem companies and users collaborating. It is important to have consensus on changes to Bitcoin as it is necessary for the whole ecosystem to upgrade together and be in widespread agreement.
Bitcoin is a user currency and works by collaboration between miners, exchanges, ecosystem companies and users. Many miners and companies consider that the soft-fork with segregated witness is a safer and faster upgrade. Around 80% of miners by hashrate indicated support for this plan and the four largest bitcoin exchanges. You can read their joint statement on collaborating to upgrade bitcoin scale here
The new version included 7x faster signature validation which is important for miners and nodes and for bitcoin to scale to more transactions per hour.
Also much faster block assembly for miners which reduces orphan blocks and so improves profit.
You can read more about it here.新版本包括快7倍的签名验证，这对于矿工和节点扩展到更多的每小时比特币交易很重要。
No definitely not true. Bitcoin is a user currency, and so it is important that any changes should be in the interests of users. To ensure this, it is important that Bitcoin proposals are subject to independent review, and in Blockstream we created employment contracts guaranteeing Bitcoin developers independence very intentionally because we did not want any company to have control of Bitcoin (including Blockstream).
Blockstream President Adam Back为啥要改签名，隔离验证交给这公司大家放心吗？
This is written about here https://www.reddit.com/r/btc/comments/46s54r/why_has_blockstream_omitted_putting_their/d0d20b9
Segregated Witness is developed by several developers the main ones, only one of whom, Pieter Wuille, works for Blockstream. Johnson Lau (individual from HK), Eric Lombrozo (CEO ciphrex) and many other developers worked on integrated segregated witness into libraries and wallets and reviewing the code and design.
隔离见证是由几个主要的开发人员开发的。其中已有一个Pieter Wuille是为Blockstream工作的。Johnson Lau、Eric Lombrozo和很多其他的开发人员都在为代码的设计和Review、隔离见证对于库和钱包的继承而努力。
Blockstream’s mission is to extend Bitcoin to support different types of transactions for financial and other applications using sidechains. Sidechains offer a general extension mechanism allowing anyone to extend Bitcoin in an opt-in way. Examples of companies developing sidechains include rootstock and hivemind, both projects developed by other companies. Blockstream also developed the elements sidechain and the liquid sidechain. Blockstream has about 30 employees. Blockstream business model is to provide services and products delivering value to users and companies in the Bitcoin and finance areas.
What’s the weighing behind the July 2017 deadline? is it possible to make it happen by the end of 2016?
The plan involves 3 stages. The plan is that the first scaling upgrade using segregated witness will be ready by April 2016, and a further upgrade to reduce transaction sizes (to squeeze more transactions into a block) later this year. From the first two upgrades there should be enough excess capacity to accommodate transaction growth through to 2017 when the 3rd hard-fork stage will activate.
For safety it is important that there is widespread agreement from miners, exchanges, ecosystem companies and users who must all upgrade for hard-fork upgrades. There are many services, pools, users and ecosystem companies that must upgrade before activation.
In addition, Jul 2017 is the latest for the hard-fork, it maybe possible sooner, using a different type of fork, if there is widespread agreement from users, miners, exchanges and ecosystem companies to do that.
Segregated Witness is an alternative to a hard-fork and also provides similar scale. The reason to do SW first is that companies and users would like scale improvements to be available, and SW can be done sooner and more safely than a hard-fork. Also it becomes safer to do a second upgrade hard-fork after SW with the efficiency improvements that SW adds.
The plan is to increase scalability as fast and safely as possible so that many new users can enjoy the great features of Bitcoin. Scale involves many factors including network bandwidth, network latency, CPU overhead and memory. The Bitcoin 0.12 release includes 7x faster signature validation which allows bigger blocks and many other scale related performance improvements. The Bitcoin developers have been hard at work on scaling features for many months, the first of which is segregated witness. There multiple additional scale improvements that come after segregated witness, including Schnorr signatures, that can greatly compress transactions, to get more transactions in the same space as well as additional size increases to follow. Also we expect for some types of uses, lightning will be available later this year.
I think everyone committed to the future of Bitcoin will collaborate to see it succeed, we could see this at the HK workshop – everyone wants to scale and improve Bitcoin. I think our best way to scale and improve Bitcoin quickly and safely is to collaborate, for all miners, exchanges, users and ecosystem companies to work together. The questions are only about which technical way is best.
There were 80,000 lines of code written in the bitcoin 0.12 release, however this code was written by dozens developers from different countries, companies and individuals. Blockstream is a one part of this only and the code is part of the Bitcoin project. Bitcoin developers who work for Blockstream have employment contracts guaranteeing their independence for their work on Bitcoin, because Blockstream understands that it is important for no company to control Bitcoin (including Blockstream).
The 0.12 release brings many needed performance improvements that allow Bitcoin to scale to more transactions per hour, and bigger blocks. The segregated witness feature increases the effective block-size once users upgrade.
There are over 50 active Bitcoin developers from countries all over the world, some of whom work for the many Bitcoin companies, and some of whom are independent. Some Bitcoin developers work for Blockstream, however Blockstream truly is not and does not want to be in charge of Bitcoin. Bitcoin developers who work for Blockstream have employment contracts guaranteeing their independence for their work on Bitcoin, because Blockstream understands that it is important for no company to control Bitcoin (including Blockstream).
People in the Bitcoin ecosystem have a responsibility to be fair, balanced and honest in their comments. Confidence in the ecosystem can be weakened by unnecessarily dramatic statements that could be misleading, particularly when made by those perceived to be influential. While Bitcoin is decentralised, it is still important for people to act prudently. Samson’s tweets were often humorous in a way intended to remind people of their responsibility.
I think collectively we can more safely and quickly upgrade Bitcoin by working together. Bitcoin changes are normally made via the long standing open development process which reaches agreement on features by consensus amongst developers and the ecosystem of miners, exchanges, users and companies through widespread agreement. Many version upgrades have been safely made with excellent results during the last 5 years using this process. It is only with XT and now Classic that someone tried to bypass the collaborative upgrade process. Classic is a copy of Bitcoin with some rushed and incompatible changes. The main problem is not the features, but that Classic was not done in a spirit of collaboration, and so creates an “us” vs “them” mindset which makes it harder for people to agree on an outcome – this then creates unwanted drama, negative news and loss of confidence in Bitcoin’s future. We can easily achieve consensus if we gather requirements, make and review proposals and work together.
The reason it is safer to do Segregated Witness first is because there is no chance for the network to split with a soft-fork like SW, while there are additional risks of network split with hard-forks, if they are not planned carefully and with reasonable time for everyone to upgrade. Soft-forks have been done many times and are safe because miners protect people during the upgrade process. In a hard-fork everyone must upgrade before it can be activated. It is common in other non-Bitcoin protocol network upgrades for there to be a period of backwards compatibility. Hard-forks are not backwards compatible so they take longer.
Bitcoin has had many upgrades with more lines of code than Segregated Witness and deployed multiple complex features. For comparison, the most recent release of Bitcoin, 0.12 Bitcoin has 80,000 lines of new code in it, and no one is saying it is not safe to run. The Bitcoin development team has lots of experience over 5 years of making improvements to Bitcoin. Segregated Witness has undergone extensive testing, and people who have looked at the code say that while the idea is innovative, the change is relatively simple and elegant and not that complicated to code. In hindsight it is the way that transactions should have been formatted from the beginning.
There are several factors that determine the largest supportable block-size including bandwidth of miners, latency of the network, CPU overhead to validate transactions and the network protocol used to transfer the block. Testing and simulation has been done by a range of people concluding that 4 MB blocks are safe but much larger blocks start to have problems. One result was by academics http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf from 5 leading international universities.
Lightning is a layer2 mechanism to greatly increase Bitcoin scale by more efficient network usage, estimates range from 1,000x to 10,000 higher scale. Each lightning transaction is a real Bitcoin transaction that is sent to the recipient directly, rather than being broadcast to the Bitcoin network. This is achieved via a more advanced version of payment channels using smart-contracts and uses the transaction malleability fix from segregated witness. Lightning also makes Bitcoin transactions provide instant secure confirmation, whereas Bitcoin direct transactions take 10 minutes or more to confirm.
I have just one question here. As we all know, the transaction volume of bitcoin has been increasing dramatically in recent years. And the block size is reaching the limit of 1M. Several companies will have to postpone their’s development on bitcoin. At the current situation, you guys still refuse to increase the block size quickly. Why do you guys make such a choice besides the safety concerns you guys have been talking about?
The Bitcoin core developers are happy to collaborate with all developers. The core development team is very large, from many countries, companies and individuals. Over 50 active developers, and many more infrequent developers.
The core developers spent many months working on scaling features in Bitcoin 0.12 this year and last year. Bitcoin 0.12 which took about 8-months of work includes 80,000 lines of new code and scaling features including 7x faster signature validation, much faster block assembly. These features are needed to increase block-size. If not for this work increasing the block-size would not have been possible.
Bitcoin core evaluated two different methods to increase the block-size: via segregated witness soft-fork and via hard-fork. The soft-fork method is safer and faster because during a soft-fork upgrade miners protect users that have not yet upgraded. With hard-forks everyone has to upgrade before activation, which is slower.
The scale provided by segregated witness is essentially the same once users upgrade as a hard-fork. In addition segregated witness allows a new, cleverer type of scale with upgrade to Schnorr signatures that can reduce transaction sizes by up to 50% (in theory even up to 60% smaller) so we should ultimately be able to get up to 2x as many transactions from the same block-size.
With these features, it will be better to start expressing scale in terms of transactions per hour rather than block-size because as technology improves we can get more transactions per block. This is especially true for later layer2 mechanisms like lightning.
I think the future of Bitcoin is very bright. Bitcoin offers attractive properties for users. During the next 12 – 18months many exciting new innovations should become available. Initial scale upgrades in the short term, and then high scale from lightning. We should also expect improved extensibility and features for Bitcoin from side-chains allowing innovative extensions like elements sidechain and liquid sidechain from blockstream, and from other companies and developers innovations like rootstock (ethereum contracts in Bitcoin), hivemind (prediction markets) and other new ideas like zerocash, snarks and experimental ideas.
Sidechains are more about extensibility than scale or speed, but the fees on a pegged sidechain are in BTC like with Bitcoin, so transactions on sidechains are a new source of fee revenue for miners, who secure sidechain transactions via merge-mining. Sidechains support extended functionality including issued assets such as CNY by issuers like QQ, or gold ETF or shares and bonds. Interestingly with smart-contracts you can even make options where the option contract is executed by the block-chain without having to trust an underwriting company.