According to the official terminology of Bitcoin Core, "abandoned" or "stale blocks" are those that are not part of the main chain. They can be created in a natural way when two miners produce two blocks at almost the same time, or they can be generated by an attacker in order to attack a blockchain. And "orphan blocks" refer to those blocks, the "parent block" of which is still unknown, that is, the node has not yet processed them. The community, however, often confuses these concepts, mistaking an "orphan" block for one that did not enter the main chain, because it was ahead of the other. This designation is used by Blockchain.info in particular.

Let us return to the traditional terminology, whereby the "parent" links of the orphaned block are usually installed later, the block is confirmed and takes its place in the main network. The abandoned, or the "backdoor" blocks, remain unnecessary and never enter the main chain of Bitcoin.

This forms a number of problems, and one of them concerns the miners. If the miner accidentally produced his block at about the same time as another and the network did not choose that block, then he will not receive a reward. In addition, too many blocks of this nature can make the network more vulnerable to attacks, namely to the 51% attack and double-spending.

"If a malicious attacker controls a large fraction of the hashing power, he will always have the advantage of mining the next block and he can choose not to transfer the block immediately [to the main blockchain], but to do this at about the same time as mining the [other] new block,” writes Bernard Peh, the leading developer of Blockbid. He also notes that it is necessary to be extremely precise with the implementation of one’s own block, always staying one step ahead of the competing miner. Thus, an attacker creates a parallel controlled network, and when it becomes sufficiently long, the Bitcoin protocol begins to consider it as having priority. Having obtained such power, an attacker can "roll back" transactions, that is, recognize already verified transactions as invalid and reuse the funds within their "own" chain.

Such centralization of mining can be used not only by hackers, but also by any network members who want to receive the greatest reward. The concept of "selfish mining" was first described in 2013 by researchers of Cornell University Emin Gun Sirer and Ittay Eyal. They proved that the miners can get more by combining into groups and hiding the newest of the generated blocks, thus secretly creating their own branch (which for the first time will technically be a chain of "abandoned blocks").

Initially, it will be shorter than the main network. The "selfish miners," however, can determine the time for "showing" the new blocks, making them public only when the size of their network becomes commensurate with the size of the main one. Thus, they know that their network (with a high degree of probability) will become the main one, and the honest miners, not knowing this, are working in vain. When the "private network" reaches a competitive size and will be presented openly, the simple miners can also join it in search of greater rewards. And this development of events can also lead to the scenario of a 51% attack depending on the wishes of the pool controlling the network.

In 2015, the problem of abandoned blocks was discussed in the context of block size and scalability (in the same year, the developer of Bitcoin Core Pieter Wuille put forward the proposal of SegWit). At the height of the 2015 debate, Rusty Russell, a member of the Blockstream team currently working on software development on the basis of the Lightning Network, noted that increasing the size of the Bitcoin block will increase the percentage of abandoned blocks, as "large blocks need more processing time." "Blocks that do not contain transactions are the smallest, and therefore they are processed more quickly. They still provide a reward of 25 Bitcoins [after yet another halving, the reward in the Bitcoin network is 12.5 Bitcoins. — DeCenter], although they do not really help Bitcoin users," as Russell wrote. Even then, he noted that the growing level of abandoned blocks negates the hypothesis that the miners deliberately refuse to centralize so as not to drop the reputation of Bitcoin and its price. "The miners behave very badly. Pools organize attacks on each other with surprising regularity . . . Large mining pools use their power for double-spending and steal thousands of Bitcoins from the gaming service. [Russell refers to the attack carried out by GHash.IO on BetCoin Dice. — DeCenter] . . . If the large miners can use large blocks as weapons against smaller ones [miners], most likely they will do it," Russell warned.

According to Blockchain.info statistics, the largest number of abandoned blocks was observed in the period from March 2014 to June 2017, as then a few such blocks were produced in a week. According to the chart, however, before this period and after it, the number of abandoned blocks was equal to zero, which looks implausible. Members of the community have already noted that the chart is working incorrectly. "They [abandoned blocks. — DeCenter] have been significantly reduced in number, but some remain," wrote one of the users of Reddit. "This chart is broken, my nodes got abandoned blocks every 500 to 1,000 blocks in the last 6 months," wrote the user under the nickname statoshi in January, also mentioning that he was talking to the developers Blockchain.info on this topic and "they are aware that it is broken." On another page of Blockchain.info, there are two cases of abandoned blocks from this year dated January 12 and June 4. The participants note that there are "several abandoned blocks" that are not reflected in the first chart (and, perhaps, in the second one too), but at the same time, the community does recognize a "sharp reduction" in their number. Many assume that this is the result of the accelerated distribution of blocks due to the FIBRE relay network created by Matt Corallo. The protocol was not implemented in the Bitcoin Core client but can be applied by users who have a full Bitcoin node on their devices. FIBRE "transmits blocks on the node network with virtually no delays at superluminal speed through the fiber." "For anyone who wants to really understand why (many) technical experts are so judgmental against large block sizes and Bitcoin Cash, a good start will be understanding the FIBRE network, why it exists, and how it works," as one of the Reddit posts reads. It also suggested that the rise of abandoned blocks was triggered by the growing popularity of Bitcoin, whereas until 2013, there were not enough miners to create controversial situations with simultaneously generated blocks.

On June 9, Craig Wright, the main adept of Bitcoin Cash, joined the discussions about the abandoned blocks, having published a post in his blog titled "Iron and Steel." He argues that such blocks (in Wright's terminology these are "orphan blocks") are not a defect that needs to be eliminated. Drawing an analogy between forging steel and processing Bitcoin transactions, Wright compared the abandoned blocks to the iron that is needed for steel production. "Many developers think that orphan blocks represent the main problem that needs attention and elimination. Orphans are not a disadvantage, it is carbon, which is smelted into iron to create Bitcoin steel," Wright writes. He argues that such blocks are a necessary part of the system, while the miners want to get rid of them for obvious reasons, as they do not yield any rewards. But this is only an individual benefit, whereas globally, these blocks do not cause harm or damage to the system. "The number of produced orphaned blocks does not in any way affect the structure of the rewards or the system as a whole," Wright says, noting that if there is a theoretical reduction in the number of such blocks or their total extermination, there will be no more rewards. "Miners . . . do not understand that this is [the possibility of the appearance of abandoned blocks. — DeCenter] the desired state, and they were accustomed to the fact that it is undesirable, and to the fact that if these blocks are removed, then they somehow earn more. This is a lie with which we must fight," believes Wright.

He also notes that if you "repair" such blocks, then it can destroy the entire system. "The simple truth is that orphaned blocks and what is understood as their shortcomings are a key element that ensures the work of Bitcoin. This is not something that needs to be eliminated to make Bitcoin more efficient, it is a critical aspect that, if eliminated, will lead to the collapse of the system." Wright believes that "Bitcoin requires a certain proportion of inefficiency" and "orphan blocks unite the miners,” since they oblige them to constantly invest in resources that will help them stay in the game. "Orphaned blocks contribute to competition within Bitcoin as a system. Bitcoin is not designed for equal distribution. Its purpose is to create a competitive environment in which miners and businesses compete for any advantage, competition is at the very center of this system and it cannot work in any other way . . . This helps all miners. As always in business, however, most people think about the short term and see what they are losing, not the system as a whole," Wright writes.

He notes that in the game theory, the relationship between miners corresponds to the "stag hunting" model, which describes the conflict between personal and public interests. This type of interaction was described by Jean-Jacques Rousseau in 1755 in the treatise "Reasoning about the Origin and Grounds for Inequality between People.” "When hunting a stag, everyone understood that they must remain on their posts. But if a hare ran by any of the hunters, it was not necessary to doubt that this hunter would follow it without a twinge of conscience and, after catching the prey, he will not lament that he deprived his comrades of their quarry." That is, the greatest profit simultaneously requires greater cooperation between the participants. A person, both a hunter and a miner, however, can easily give in to the idea of ​​a small but personal gain in the form of a hare. At the same time, as Wright notes, if a hunter decides to go hunting a stag alone, "the chance of success is small and suboptimal," and a more global problem is the need for "too much trust between the players.”

Bitcoin solves this problem with asymmetries. As Wright notes, Bitcoin's incentive system makes it more profitable to "hunt stag,” rather than chase after small prey. The main strategy involves choosing between increasing the hash (computing power) and increasing the "connectivity" (that is, the ease with which new players enter the system, which increases decentralization). In this case, the first option that is more selfish and materially profitable is the "pursuit of the hare," while the formation of a decentralized network (albeit with a lower individual profitability) is a collective "stag hunt." The key weapon in pursuit of the hare is the ASIC miners, who have already declared war on Monero and Ethereum. According to Wright, however, the desired prospect is a combination of increasing the hash and maximizing the "connectivity" to the network. He believes that one of the key aspects for the success of the mining system is the "tight connection" between the miners and the "signal system.” "The signal will allow players to unite to create a more optimal strategy. A safer system (more tightly connected) in the interests of both the user and the miner. Orphans mean significant material losses for an individual miner . . . and a miner who has a stronger connection with the rest of the participants may expect a lower proportion of orphans and a higher proportion of successfully extracted blocks."

Wright emphasizes that "mining is not about mining blocks, but about passing the extracted block to other miners,” because the block enters the main network only when another miner "connects" the next block to it, and so on.

In addition, Wright, who claims that Bitcoin Cash is the only "real Bitcoin" reflecting Satoshi's original ideas, again refers to the Bitcoin white paper, saying that abandoned blocks are an organic part of the system providing "the only way miners can honestly vote." The miners "vote with their CPU power, expressing their acceptance of valid blocks by working on extending them and rejecting invalid blocks by refusing to work on them. Any needed rules and incentives can be enforced with this consensus mechanism," as Wright quoted the Bitcoin white paper.