Why BitcoinXT Must Never Gain Consensus
In order for bitcoin to reach its potential as a technology, scaling the current blocksize needs to occur within the near future. Hardly anyone is in disagreement that the bitcoin blocksize needs to be increased. Yet, what we are currently seeing in the bitcoin community today, as the blocksize debate rages on, is a dispute about what users envision the bitcoin payment system to become in the months and years ahead. In this article, we will examine the recently proposed BitcoinXT fork (BIP101), and how exactly developers should proceed in scaling the bitcoin technology with adherence to its inventor’s original vision – an apolitical, distributed consensus system.
BitcoinXT: Bitcoin Improvement Proposal 101
A hard fork of the bitcoin protocol would carry dramatic implications. One of which would be an incompatible division of blockchain records. Because we are dealing with a protocol technology for the transfer of value, in this scenario it is winner takes all. If a split in the blockchain would occur, one would absorb the entirety of the network purchasing power while the other would be sapped of all relevance. This is a scenario which should be avoided at all costs simply because the alternatives that exist for accomplishing the same end (achieving scalability of the bitcoin payment system) are numerous and less of a security risk.
Putting into effect BIP101 would result in an increase to 8MB after January 11, 2016. In order to gain consensus, the number of nodes running the bitcoinxt client would need to reach a supermajority of 75%. After that point, there would be a 2 week window to transition to the new fork as the old blockchain becomes incompatible with bitcoinxt. From there, the blocksize limit is set to increase linearly to a maximum of 8GB in 2036. Once started, the block limit doubling schedule cannot be stopped until 8Gb is reached.
Visa currently has a transaction processing capability of 50,000 per second. If this 8GB blocksize were to be instantiated, then bitcoin would potentially be able to compete with the likes of traditional payment systems, but is that what it has been originally designed for?
If Bitcoin comes to depend on bureaucrats rather than protocol, you might as well use Visa. They know how to do that far better.
– Nick Szabo, August 21, 2015
The current blocksize limit stands at 1MB – a limit which carries with it the risk of transaction backlogging and reducing the overall effectiveness of the payment system as previous patterns of growth are expected to continue. As we have seen in years past, the largest periods of user growth tend to be seasonal and happen in the winter months. Currently, the bitcoin network is operating at 30% of this capacity. If we remain at the current blocksize, we can expect problems to arise in 2 – 8 months due to the constraints on transaction throughput. In a situation where the transaction queue becomes backlogged, you can expect payment confirmations to become less reliable, as each node would require more memory to process this growing list of delayed transactions.
In the event of the blocksize limit being hit consistently, with confirmations times becoming an issue, emergency limit increases are something which can be implemented quite seamlessly. This type of software adjust has already been demonstrated reasonably well. If the transaction backlog becomes large enough, the network will suffer over-saturation and larger fees will be needed to confirm new transactions in a reasonable time frame. Therefore, with a few caveats, the tradeoff between urgently switching to bitcoinxt and waiting for more time to come to a consensus on a more robust and ideologically-aligned consensus is between absorbing the risks of a hard fork and temporarily tolerating higher fees and transaction bloating. Eventually, the blocksize needs to be increased, but rushing to implement a new proposal may prove to be the most callous decision over the long term.
Bitcoin Is Not a Purely Public Good
Higher transaction fees due to transaction bloating are technically possible today. Yet, wallet providers have not developed the infrastructure necessary to operate a market mechanism where the software is capable of predicting how much a user would need to spend in order to have their payment confirmed in a particular span of time. Although most everyone would like to use a completely free payment system, the economic realities of doing so do not align with decentralization.
When block space is readily available, miners suffer from a “tragedy of the commons” problem in which the only rational behaviour is to accept all transactions, no matter how small the fee. With scarcity, however, there’s suddenly a competitive market for block space. Those paying fees may not appreciate the extra costs, but a transition to a system that allows mining fees to become a larger fraction of mining income is actually a good thing for the long term security of the network.
– David Hudson, Bitcoin Traffic Bulletin
If users are to favor permissionless, decentralized payment networks, they must be willing to pay for such a service. In the case of bitcoin, there seems to be an underlying misconception that it represents a pure public good, much like the internet protocols of TCP/IP. In the case of bitcoin, this is not so similar. Because the bitcoin network is largely non-excludable, yet a rivalrous mining environment exists, we can see that this utility represents a publicly provided private good – one where pay-for-performance is implemented into the system by default. Bitcoin is not a purely public good because it incorporates the rivalrous relationships of miners and user competing for transaction confirmations (based on less or more of a transaction fee). Aiming to mold bitcoin into a pure public good with the attributes of instant transactions and negligible fees is a disservice to its inventor(s) because they aimed first and foremost for an trustless, consensus system. Increasing the blocksize reduces the economic incentives for decentralization.
The tradeoff is clear: convenience comes with the cost of centralization while decentralization comes with the cost of a rivalrous, pay-for-performance system. Pick your poison.
The current system where every user is a network node is not the intended configuration for large scale. That would be like every Usenet user runs their own NNTP server. The design supports letting users just be users. The more burden it is to run a node, the fewer nodes there will be. Those few nodes will be big server farms.
– Satoshi Nakamoto, July 2010
A Disagreement of Ideologies
What the debate around bitcoin is currently focused on is not what is sustainable or how urgent the need to increase the blocksize is. The debate is around what the precise use case of bitcoin is – an ideological crisis, not an engineering crisis. A vision crisis, not a management crisis. Herein lies the rub – it is the users who must decide which direction they want the development of bitcoin to proceed. Do they want near-instant transaction confirmations and lower fees? Or do they want an incorruptible digital ledger to which they can entrust their wealth?
Is bitcoin destined to become a competitor to VISA and MasterCard? Something with which we can use to pay for everything from small retail purchases to large one-time payments?
Introduction of Tor Blacklisting
One of the less talked about implications of bitcoinxt is that the implementation would add the ability to block Tor exit nodes – something which is highly controversial among a userbase who place a high priority on privacy. This is contrary to previous implementation of not revealing any IP information. Certainly, nodes could opt to use a VPN. However, the stated reason for implementing blacklists (mitigation of DDoS attacks) is a thinly veiled front for making the system less private overall).
If bitcoin were hard forked with BIP101, there would be no option to revert from an eventual 8,192MB blocksize by the year 2036 except through another fork in the protocol.
Bitcoin technology enthusiasts and core developers need more time to come to a consensus on the best course of action for scaling the payment network in a responsible and ideological manner. In doing so, they can retain the robustness of the technology while allowing the blocksize to scale with adherence to the core principles of decentralization and privacy.
Combined with the slippery slope of blacklisting certain nodes which grant improved privacy, consenting on BIP101 is a step in the direction of bitcoin becoming a more controlled and centralized form of money. Bitcoin needs to be scaled up via increasing the blocksize. However, there is no immediate urgency. Bloating the network is far better than forking the technology with a risky implementation. BIP101 is not the long-term solution.
The blueprints for scaling the bitcoin are beginning to emerge and the Lightning Network may prove to be the...
Ethereum today is where bitcoin was in 2010 – raw infrastructure, lack of developers, and plenty of skeptics.
The financial landscape in developing economies such as Africa is well positioned to leapfrog traditional banking and move directly...