Governance in Blockchain Part I: The Bitcoin Experiment
The latest chapter in Bitcoin’s riveting drama occurred this past week. In the end, the efforts of the Bitcoin community were not enough to preserve a unified cryptocurrency, resulting in an official split in two: Bitcoin (BTC) and Bitcoin Cash (BCH). This move came largely as a surprise as it comes off the heels of a successful Bitcoin Improvement Proposal (BIP) to resolve short-term scaling issues and was thought to have largely satisfied community stakeholders.
The tumultuous and contentious history of Bitcoin since its inception exposes a flaw (or feature!) of the payment system — the neglect of a clear governance process in its original founding document. By governance I am referring to the decision-making process within the protocol that defines how rules are agreed-upon, created, and changed.
On the surface, the word “governance” seems to have a very negative connotation and may even seem antithetical to the ideas of decentralization touted by blockchain technology. While we have heard Bitcoin, Litecoin, and Ethereum compared to the physical, real-world assets of gold, silver, and oil respectively, we would do well to remember that these cryptocurrencies are ultimately software projects with developers and other stakeholders who influence their direction.
In my article on cryptoeconomics, I write about how incentive systems play a key role in the design of blockchain protocols. Governance is a key subset of this topic, encapsulating a system’s ability to adapt in the face of technical challenges, evolving use cases, and ideological differences.
So the million dollar question: How can we achieve this flexibility without compromising the integrity of a blockchain’s decentralized value proposition? How can we satisfy our hunger for progress and revolution when the very premise of blockchain — a set of rigid rules dependent on consensus — rebels against those very urges?
For now, we can learn a lot from Bitcoin’s near-decade of political firestorms , fierce debates, and competing clients. In this piece, I will dissect and discuss the motivations and characteristics of various Bitcoin stakeholder groups, focusing on their attitudes regarding scaling through block size increases and their roles in governance. In the end, does chaos turn out to be a pit or a ladder?
Bitcoin: Governance Pioneer in a Digital World
In 2008, the mysterious benefactor Satoshi Nakamoto conceptualized Bitcoin and implemented the first blockchain, the technology of distributed, trustless consensus. In the wake of this revolutionary invention, the online community rejoiced, reveling in the “automation of trust” and from the successful escape from the deficiencies of trust-based systems perhaps best exemplified by the unending scandals facing Wells Fargo.
Yet to say that Bitcoin is a “purely mathematical” solution and decentralized system unbeholden to governance is a statement far detached from reality. In practice, Bitcoin is subject to its own set of politics and the vagaries of human behavior. A simple search of Bitcoin-related topics will reveal a community enmired in ferocious factional strife with a delicious cherry topping of ad hominem attacks among Bitcoin Twitter personalities.
Three stakeholder groups exist in the Bitcoin ecosystem: miners, developers, and users. The governance “structure” that exists today evolved more organically. Typically developers of the core Bitcoin client make improvement proposals, some of which result in a completely new blockchain (e.g. Bitcoin Cash, Bitcoin Unlimited). Miners and users then choose whether or not to adopt them.
I will discuss in a bit more detail the perspectives of these three groups.
Miners
Miners perform the legwork of the Bitcoin system, adding transaction records to the blockchain through a proof-of-work (PoW) protocol. In this model of “one cpu, one vote”, miners with the greater processing power, or hash rate, are more likely to successfully post blocks and add credibility to their chain.
If the vast majority of miners does not recognize a chain as legitimate and refuses to post transactions, users are likely to reject it as well. In fact, increasing trends of miner consolidation in large mining pools actually grant certain figures, such as Jihan Wu (AntPool), tremendous influence in the Bitcoin community.
The mining community of Bitcoin presents a fascinating, primordial soup of competing interests both internally and externally. Large, “industrial-grade” mining outfits with warehouses of equipment have interests at odds with smaller miners, while the objectives of miners as a whole can conflict with the needs of users or the visions of developers.
It is also no surprise that miners will promote governance actions that will protect their incentives and the significant fixed cost of their equipment. Through the PoW protocol, miners seek to maximize their chances of finding and posting the next block to receive the block reward as well as that block’s total transaction fees.
Broadly speaking, large miners seek to dominate block discovery by overwhelming their smaller competitors by sheer hash power. In fact, bitcoin mining operations have scaled up to the extent of adopting application specific integrated circuit (ASIC) chips, exploiting a Bitcoin design oversight to mine 30% faster. This heavily favors large-scale mining and makes it largely unprofitable to participate as a mere CPU or GPU miner.
Miners are also careful to examine how Bitcoin improvement proposals might affect transaction fees. As you may know, the Bitcoin community has been struggling to find the right path forward to speed up the service’s transaction speeds. Two short-term scaling solutions have dominated the news over the past couple months: SegWit and block size increases (SegWit2x is combination of the two). Both allow for faster transaction processing times, the mechanism for each could not be more different.
SegWit (Segregated Witness)
SegWit works to increase the efficiency of the PoW system by eliminating something known as transaction malleability while better utilizing the space within each block. In other words, the solution allows slightly more transactions to be crammed into each 1MB block by separating out certain data from each transaction. Given the recent, positive signaling among a requisite majority of Bitcoin nodes to this proposal, SegWit is actually set to be activated this week at the time of writing.
Many vocal members of the mining community see SegWit as a relatively complex solution to a simple problem that could be solved with block size increases. Moreover, they contend that this piece of software is not as well-tested as the developers claim and, as it cannot be reverted, could have unforeseen consequences for the future of Bitcoin. It is also worth nothing that SegWit also makes ASIC mining incompatible.
As such, despite the benefits to the user and widespread encouragement from the developer community, it is for these reasons that many miners have sought to delay the implementation of this particular proposal for so long.
Block Size Increase
The block size debate centers around speeding up transactions by increasing the size of each block (currently capped at 1MB), and also has its fair share of nuance. While larger blocks equate to more transactions posted and consequently more transaction fees, larger block sizes demand greater expenditure of processing power.
At large block sizes, this effectively sidelines many of the smaller miners and further leads Bitcoin down the path of hash power centralization. The question of when and how large the increase should be have led to dramatic actions in the community, from the proposed 2MB hardfork in November to the recent fork that created Bitcoin Cash, housing 8MB blocks.
The centralization of mining power is seen as a positive by those in the Bitcoin community who believe “Bitcoin is for the miners”. The line of thinking goes that preserving the incentive system for miners is of the utmost importance and that transaction fees may necessarily need to remain high as the issuance of Bitcoin declines over time. Since they are the “workhorses” of the entire system, miners necessarily should have an out-sized role in governance and heavy decision-making power with regard to Bitcoin’s future.
Developers
The developer “caste” shoulders a great deal of Bitcoin’s ideological burdens. What should be understood is that the Bitcoin system is a protocol, or a set of rules that dictate how computers in the network communicate with each other. Anyone can write software that follows the protocol, and it is up to the users to choose which software implementation they would like to use.
In this way, we can think of developers as advisers or consultants; they propose a specific vision or direction for Bitcoin, while miners and users choose whether to adopt it (e.g. mining or transacting on Bitcoin vs. Bitcoin Cash).
Several books or a laughably long Egyptian scroll could probably be written regarding the formidable cornucopia of developer opinions on what Bitcoin should be. If you’re relatively new to the Bitcoin world, I realize it can quickly get very confusing with the many different implementations such as Bitcoin Core, Bitcoin Classic, Bitcoin Cash, Bitcoin Unlimited, Bitcoin XT, and Bitcoin ABC. Don’t panic.
For the purposes of this article I will focus on Bitcoin Core and the ideological differences that resulted in several of these alternative implementations
Bitcoin Core
Standard Bitcoin (BTC), the largest implementation, is maintained by the developers behind the Bitcoin Core reference client. This is the Bitcoin you’re hearing about in the news. As you can see in the chart below, Bitcoin Core is the most popular implementation by far, while its many competitors manifest the complex web of different thought leadership and philosophies of how Bitcoin should be governed and how the protocol should be improved.
Bitcoin Core evolved from Bitcoin’s original implementation by its creator, Satoshi Nakamoto, and is maintained by a group of volunteer developers. Through Bitcoin Improvement Proposals (BIPs), these developers reach a general consensus through extensive peer review to determine modifications to the existing protocol.
Bitcoin Core developers tend to be risk-averse, highly valuing the security and integrity of Bitcoin at the expense of increasing accessibility and speed of the network for users. To that end, they have resisted pushes to vastly increasing the block size limit for various technical and political reasons, including the necessity of a hard fork (a split into a new blockchain that forces all users to update), the fact that it is a short-term solution that does not solve the scaling problem long term, and damage to mining decentralization that I’ve discussed above.
Hard forks are generally considered very risky as it could have untold consequences for not only Bitcoin’s market share but also impact its value proposition of “digital scarcity”. The do-or-die assumption that a majority will accept this 2MB hard fork is very concerning as the lack of widespread consensus would essentially lead to a doubling of the 21M coin cap of BTC and lead to further brand confusion.
Consider the emotional and financial rollercoaster over the past few days brought on by the Bitcoin Cash hard fork, which the market is still busy digesting. The November hard fork is undeniably the next major test for Bitcoin Core, and it will certainly not be the last.
Historically Core developers have promoted only SegWit since it is a soft fork proposal that is “backwards compatible” for all users. While SegWit2x, a compromise of sorts that combines SegWit activation with a block size increase to 2MB, is set to occur in November, the recent outcry portends yet another upcoming battle in the ongoing civil war among developers.
Dissenting Developers
Many developers in the community have taken issue with the Bitcoin Core team’s decisions and conservative approach toward improving transaction speeds. In particular, proponents of block size increases believe them to be true to the original spirit of Bitcoin and are dismissive of SegWit, citing the quote from Nakamoto below as proof.
“The eventual solution will be to not care how big it gets.” — Satoshi Nakamoto
This bloody struggle to determine the block size limit compounded with the lack of a clear governance model has led to many developer groups forging their own paths. We’ll take a quick walk through Bitcoin history over the past couple years.
Bitcoin XT
The ideas to increase block size began in August 2015 with the Bitcoin XT hard fork that proposed an eight-fold blocksize increase from 1MB to 8MB and doubling every two years. This very aggressive increase never received more than 10% support from the community.
Bitcoin Classic
Bitcoin Classic can be considered its successor with a 2MB block size proposal in 2016, but it too failed to implement any meaningful progress and has since been in a steady state of decline. It’s worth noting that Bitcoin Classic attempted to institute a more democratic governance model through consider.it.
Bitcoin Unlimited
These two experiments led to the genesis of Bitcoin Unlimited (BTU), which takes the blocksize scaling solution and governance changes to the next level. BTU removes the hardcoded 1MB block size limit and allows miners to set their own block sizes through an “emergent consensus” algorithm, which essentially means they will mine block sizes on the longest chain. The thinking goes that an ideal block size or equilibrium would eventually be reached through this consensus.
Notably supported by controversial angel investor Roger Ver and super-miner Jihan Wu, the most striking proposition in BTU deals with governance. Many BTU proponents are suspicious of the influence of large blockchain tech companies such as Blockstream that fund Bitcoin Core development. While this distrust may not be warranted, Bitcoin Unlimited heavily focuses shifting decision-making power away from the central group of developers around Bitcoin Core. Through their Articles of Federation, BTU outlines a governance process by which all members, who enter the community through a public vote, can propose, discuss, and vote for specific improvement proposals.
As the node implementation chart above suggests, Bitcoin Unlimited has had very limited success due to its block size proposal and small development community. Nonetheless, it is still worthy of note in that it is the only somewhat-mainstream implementation of Bitcoin thus far with an in-depth focus on governance processes and issues.
Bitcoin Cash
The frustration with the lack of progress regarding block size increases from the Bitcoin Core development community came to a head this past week with the hard fork of Bitcoin Cash (BCH).
BCH developers prioritize Bitcoin’s utility as a transactional currency and payment service for users as opposed to a conservative store of value. To that end, they have increased the block size to 8MB with plans to automate future increases in the spirit of Bitcoin XT. No significant governance changes were made.
Overall, the multi-faceted conflicts and ideologies are brought to life on developer whiteboards. In the great chaos of online forums and lines of code, developers struggle to chart what each believes to be the optimal course and use case for Bitcoin. Without any widely-accepted system of governance beyond the vying and lobbying of different developer factions, the creative destruction and forking of Bitcoin is likely to continue.
Users
While miners codify Bitcoin history and developers implement changes, both groups’ influence is tempered by the choice of which Bitcoin implementation users decide to accept (and of course developers and miners are subsets of users themselves). If the majority of users do not believe in the integrity of a chain and refuse to acknowledge or transact on it, the chain will steadily decline in value and its set of defining protocol will simply fade into obscurity.
It’s also worth noting that different use cases for Bitcoin users have a significant impact on improvement proposals. Libertarians and investors that look to Bitcoin as a secure and reliable store of wealth will react differently from consumers who prefer a convenient, transactional currency for small purchases that is as easy to use as Venmo or Paypal. More different still are embattled and persecuted citizens who live in authoritarian countries and are looking for a secure avenue to move funds or make payments internationally.
Some limited governance experiments aside, users vote with their wallets, impacting the market price (market manipulation aside) by buying or selling off their Bitcoin. In this way, developers and miners theoretically have a vested interest in not carelessly “rocking the boat”, as poor public reception of Bitcoin can only negatively impact price and adoption. In practice, however, the insular, tripartite community of Bitcoin has never shied away from airing its dirty laundry in spectacular and hilarious fashion.
Working as Intended?
The current state of Bitcoin and its rival implementations largely boil down to a chaotic balancing act. Miners seek to extract maximum value from transaction fees and efficient mining. Users look to satisfy their own use cases and profit from each Bitcoin implementation’s rise or fall. Developers have the thankless task of developing solutions that will align these interests while staying true to their own ideologies.
Akin to the feudalistic landscape of the Dark Ages, the governance vacuum in Bitcoin has led many dissenting developers from Bitcoin Core to simply propose ideological improvements and “launch-and-see”. Amid great fanfare and Twitter drama, stakeholders of Bitcoin vie for their voices to be heard and struggle to attain more influence.
However, some members of the Bitcoin community would argue that the limited guidelines surrounding governance are a feature and safeguard.
The politics are intense, there are ways that bitcoin governance is like government, and proposals to fork the software are kind of like constitutional amendments. But I’m increasingly comfortable thinking of bitcoin governance as a market phenomenon.
Specifically, groups with differing visions are competing to win the favor of bitcoin miners and nodes, so that their vision, if it prevails, can carry the bitcoin project forward. — Jim Harper, Cato Institute
Yet, can this limited governance model, decentralized as it is, address the necessary balance between security, scalability, and speed that will address the myriad of Bitcoin use cases? Should Bitcoin simply prioritize some over others? How sure are we that a clear, accepted solution will rise to the top and that the Bitcoin community will not continue to fracture and factionalize, with each new idea spawning a separate implementation?
Overall, the main takeaway regarding governance in Bitcoin is that there really isn’t that much of it. And how could there be when the entire system was born out of a genius, unproven concept described in a nine-page white paper that was rightfully concerned more about its initial implementation than its future scalability a decade or a century down the road?
In the end, no matter where Bitcoin as a whole ends up, the fact that the Bitcoin Core system has persisted for almost ten years on the back of a transparent, ostensibly consensus-driven, and caustic community is a tremendous achievement.
The Bitcoin experiment and its highly political development climate have opened a lot of eyes toward what worked and what didn’t. In Part II of this series, I will discuss some key issues in blockchain governance and how some other cryptocurrencies, particularly Ethereum, are working to tackle these problems.
