Quarantine witness _blockchain

Brief introduction

In fact, the quarantine witness contains a lot of content, specific content and the corresponding benefits are as follows:
https://bitcoincore.org/zh_CN/2016/01/26/segwit-benefits/#section-1

More:
"Talking about block chain (21): The isolation witness of the bit currency" background the chunk size is 1M bytes, the number of blocks to be validated is increasing and the efficiency of transaction processing is getting lower. Ching design Bitcoin system, did not put two parts of data processing, resulting in transaction ID calculation mixed transaction status and witness. Because the witness itself includes the signature, and the signature cannot be signed on its own, the witness can be changed by anyone without the consent of the parties (two different signatures are valid), resulting in the so-called transaction ductility (malleability). More
Therefore, after the transaction is issued, the transaction ID before confirmation can be arbitrarily changed, so transactions based on unacknowledged transactions are absolutely unsafe.
In February 2014, the attackers exploited the loophole to MtGox the world's largest Bitcoin trading platform, leading to its demise, but the problem has not been resolved yet. That is, the Exchange initiates a transaction TX and tracks its TXID, while the hacker's signature to modify TX is tx_hack and therefore has a different txid. Tx_hack and TX are essentially the same. And Tx_hack was first packaged into the block, the money has been diverted away. And the exchange because of tracking txid, so that the money has not been sent successfully, so resend. The soft fork content signature of the quarantine witness is moved to the witness section

In the bitcoin trading structure, the signature information occupies a large number of bytes, which means that more space is occupied. In addition to the miners need to verify the signature, ordinary users need only know whether the currency is available, that is, many people do not need this information.

Therefore, there is a proposal: to isolate this signature information, input to put a pointer to the signature information on the line.

Benefits

1) The maximum block capacity can be increased with a soft fork:
Because the old nodes simply do not see the quarantined witness, even if the real chunks are over 1MB, they will still assume that the blocks are accepted without exceeding the limit. SW can provide about 2MB of valid chunk space without any hard fork risk.
2) From then on, only the person who issued the transaction could change the transaction ID, and no third party could do so. In the case of multiple signature transactions, only a few signatories agree to change the transaction ID. This ensures the validity of a chain of unacknowledged transactions, which is necessary for two-way payment channels or lightning networks. With two-way payment channels or lightning networks, two or more people can actually do unlimited transactions, without having to put a large number of fragmented transactions in the block chain, greatly reducing the space pressure in the block.
3) Lightweight purses can become lighter because they do not need to receive witness data again. The witness data needs to be stored only when the miners need to verify.
4) can significantly improve the signature structure. On the block chain, there was once a transaction of more than 5,000 inputs, because it took half a minute to complete the inspection because the design was flawed. The proposed SW fork will solve the problem. More Disadvantages

1) If the Segwit is active, those unsigned signatures will not be able to receive transactions from the Segwit wallet. If someone sends you a bitcoin through the Segwit wallet, you won't see the deal until a miner confirms the deal in the block. This essentially destroys the 0 confirmation payment for all the wallets that do not want to upgrade segwit. Complex new payment types can cause bugs in wallets that have poor code quality. Segwit wallets need to be smarter about how to handle sending money to newer and less-updated wallets. If something goes wrong, it can cause a loss of funds.
2) The quarantine witness actually lowers the handling charge for the miners. Miners deal with a block that contains more transactions, effectively obtaining the same cost as the 1MB block data value. Quarantine witnesses sacrifice transaction fees to allow the network to achieve higher transaction throughput per second. May later increase the bit currency transaction handling fee. The
3) Quarantine witness cannot be rolled back because, for Segwit clients, the transaction looks like everyone pays (technically, all people who can spend the output). Segwit activation, if through the majority of miners software voluntarily downgraded back and forth Segwit, then the isolation of all the funds to witness the output can be taken away by unscrupulous miners. As more and more funds are locked in segwit output, the incentive to collude with miners is higher. The
4) Segwit does not actually increase the chunk size, it simply calculates the chunk size in a different way and discounts the quarantine witness data. This means that normal, non-segregated witness transactions will not be counted. This also means that only the Segwit transactions are used by everyone in the Bitcoin network to achieve effective chunk size increases.
for those segwit outputs, Segwit does not actually fix a ductility error or a two-time hash problem. Yes, this means that as long as there is a segwit output on the block chain (for example, Ben Cong bits in the middle of the time), these problems still exist on the network. Just as the pseudo block size increases, Segwit does not actually provide block size increases unless the entire network upgrades and converts their segwit bitcoin into Segwit bitcoin. If you combine this with the previous concept-as more and more bits are put into segwit output, the incentive for miners to collude to steal Bitcoin is greater-I don't think the full use of Segwit is the same as the current bit-currency network. For example, the current risk of 51% miners conspiring is equal to the risk that a transaction can be censored or that miners can repeatedly pay for their own transactions. At present, 51% of the Union cannot steal your bitcoin. But if everyone uses segwit, then this can become a reality.