Learn You Some Crypto for Good
In the past few days, the market has crashed hard on basically everything, including Bitcoin. Financial losses naturally come with negative emotions and various theories of justification. Is coronavirus really the culprit? Is it the oil war? Or are they just excuses for something more fundamental? Perhaps nobody really knows since it might require a certain amount of higher dimensional insights which most of us frankly do not possess. Surprisingly, I wasn’t that much affected emotionally by the price this time like the roller coaster feeling I got from 2017 to 2018. Perhaps it has to do with the fact that my growing appreciation to the crypto space (primarily Bitcoin) over the years makes it much more than just a financial investment. As a technologist, it not only fascinates me on the technical side, but also serves as a forcing function for me to learn a bit more about other disciplines such as politics, economics, finance and philosophy, etc. In retrospect, I feel this learning experience is very rewarding and a journey well worth taking. In this post, I will try to explain its value proposition from three angles: Facinating Technologies, Importance of Alternatives and Self Sovereignty.
Blockchain is fascinating from the technical perspective. It was invented to solve the double spend problem in a decentralized setting without requiring a trusted third party. It achieves that with the interplay of two components: First, a modification resistent data structure for storing all the transactions. It turns out that an efficient implementation is to organize transactions into a sequence of blocks, each of which contains a cryptographic hash of the previous one, resulting a chain of blocks, hence the name blockchain. Second, a method to reach consensus among a set of networked computers of certain topology as to who should have the right to append the next block onto the blockchain, called a consensus algorithm.
A key contribution of Bitcoin as the first blockchain is that an incentive structure is built right into the consensus algorithm through the use of proof of work which coordinates the economic activities in the ecosystem of miners, merchants, engineers, exchanges, users, etc, resulting the world’s first dencentralized autonomous organization. This is a significant milestone in computer engineering since for the first time, engineers can program trust and incentives without delegating that to a centralized entity. Since the world at its base layer is fundamentally decentralized and governed by law of physics and different incentive structures, theoretically this gives engineers the power of genesis creation in whatever the “universe” they are creating. Consensus algorithm is an active area of research in the space with a lot of thought provoking exploration and debates, the most notable alternative to proof of work is various forms of proof of stake. At the end of the day, they are all supposed to help the blockchain achieve Byzantine fault tolerance.
There many other difficult technical problems in the crypto space. One of them is blockchain scalability, which is hard to achieve without also sacrificing its security and decentralization properties (see scalability trilemmas). Ethereum’s response to that seems to be leveraging clever constructions of proof of stake and sharding, while the answer from the Bitcoin community at the moment is mostly 2nd layer solutions such as lightning network. Since the value proposition of a blockchain is security and decentralization, trading off too much of these two properties is generally not desirable (e.g. blocksize debate). For blockchain to achieve mass adoption, a lot more research and development as well as real life battle testing remains to be done. Privacy is another important area of research & development in the space. Satoshi Nakamoto famously said in the Bitcoin white paper that The only way to confirm the absence of a transaction is to be aware of all transactions. There is always a tension between making everything publicly verifiable and the need for obfuscation. Bitcoin does not have good privacy built into the protocol layer, many technologies have been invented to improve that ever since its inception. Some blockchains try to introduce certain privacy features into the protocol layer. Monero, for example, leverages ring signatures to obfuscate the transaction graph and confidential transaction to hide transaction amount. Zcash leverages a type of zero-knowledge proof called zk-SNARKs to make shielded transaction completely private. Both of them sacrifices scalability to some extent since they make the transactions and thus the blockchain significantly heavier. Mimblewimble also uses confidential transactions to hide the transaction amount, but it goes a bit further by leveraging the encryption underneath confidential transaction (called homomorphic encryption) in a clever way to express coin ownership as well. Interestingly, through a mechanism called cut-through, mimblewimble blockchain can be made very compact thus achieves both privacy and scalability at the same time. Another approach to improve both privacy and scalability is through the use of zero knowledge proofs to generate proofs offchain and only perform verification onchain, as what Stackware and Coda aim to achieve. Privacy and scalability are so intimately related, another example is lightning network, even thought intended as a scalability solution, it also has privacy implications because it can potentially move most of the Bitcoin transactions offchain.
Programmability is also a point of contention in the crypto space. Bitcoin’s scripting system allows the creation of smart contract and is purposefully not turing complete to avoid the halting problem. The progress made in lightning network development demonstrated how much can be achieved despite of this limitation. This design choice is very much in line with the philosophy where blockchain is not for computation but for verification. What nodes really care about is a proof that certain computation was performed, repeating the same computation is just one way to achieve it. With this philosophy, many proposals have been made to improve the existing Bitcoin scripting system. Schnorr signature, a digital signature scheme that offers the ability to perform algebra on signatures, is hopefully soon coming to Bitcoin. This paves the way for technologies such as signature aggregation, Taproot and Graftroot which are great enhancements to the privacy and efficiency of the scripting system. It also opens up doors for things like Scriptless Script, which leverages Schnorr signature’s algebra capability to capture the semantics of a class of offchain protocols and only use the blockchain for verification. Zero knowledge proof could also be a promising future direction since in theory it can be constructed for arbitrary computation offchain and then get verified onchain, one demonstration is Greg Maxwell’s ZKCP. On the other hand, Ethereum lead the way to an array of Blockchains that offer turing complete programming capabilities. The idea is to express any sort of programs easily and get them executed on all the nodes in the network which make them hard to stop. This expressiveness does come with many tradeoffs: A fee model now needs to be developed to measure the computational resources that a program consumes and charge it accordingly. As a general purpose platform for deploying dApps, the blockchain is usually much bigger, affecting its scalability and decentralization properties. Turing complete languages can also potentially open up much larger attack surface than a more limiting language which is relatively easier to reason about. Neverthelss, dApps platforms have certainly attracted a lot of the developer mindshare and we start to see applications especially in the DeFi space start to gain momentum.
Crypto also offers a set of unique challenges when it comes to software engineering. Deploying software updates in a decentralized p2p network is very difficult, especially when consensus over a shared state needs to be maintained by different versions of the software at the same time (see Forkology). When money is quite literally on the line and the cost of introducing and fixing bugs could be extremely high, move fast and break things is definitely not an option. In fact, the development of systems such as Bitcoin should go through the same level of scrutiny and testing as aerospace software and its deployment should be analogous to shipping hardware. Formal verification, a technique that hasn’t seen much adoption in many software projects due to its formidable cost, becomes much more economically feasible in crypto space considering the amount of value at stake. MakerDao, for example, formally verified all the core contracts in its MCD system.
Of course, the basic research of cryptography, which is one of the building blocks of blockchain, has also seen a lot of progress because of the emergance of the crypto industry in the past 10 years. More types of zero knowledge proofs and their efficient applications are being developed, aiming to improve both the privacy and scalability of the blockchain system. Multi-party computation is also an vibrant research area which could be leveraged to design secure protocols that encode complex interactions among many participants offchain and simplify the corresponding smart contracts onchain. Recent advancement to Post-quantum cryptography is also something worth looking into.
This is just a tip of an iceberg. Other difficult challenges include the oracle problem, the design of governance models and more. Crypto is inherently multi-disciplinary, which means that even a seemingly pure technical change might butterfly effect its way into causing significant consequences, which makes the problem extra interesting and challenging.
Importance of Alternatives
People have vastly different perspectives when it comes to crypto space. For some it is just another vechicle for financial speculations. For others, it is an emerging financial infrastructure that will ultimately disrupt and replace the dysfunctional status quo. Many view it as a promising tool that can help disintermediate third parties and improve efficiency of a range of economic activities. Others treat it as an asymmetric weapon to protect their privacy in the world of massive corporate and government surveillance, or even a path towards anarchism.
It is fair to assume that for many dynamic and complex issues, there is no solution that will always be the most optimal in all scenarios (aka a “silver bullet”). This applies to things from system designs, business models to environmental issues or even the forms of governments. One explanation is that complex issue most likely has a structure that’s comprised of sub-components interacting with each other in a nuanced and subtle way, solving it requires carefully and constantly balancing different kind of tradeoffs given that the structure might always be evolving. However, at a given point in time one solution might seem to monopolize the solution space. An example is the business model of most of the internet companies today. By owning users’ identity and harvesting users’ data through massive surveillance, those companies are able to somehow monetize on that and provide “free” or “freemium” services to their users. This is actually a pretty ingenious model that is often time underappreciated since it is incentive compatible for most of the participants in the system with the current state of mainstream technologies. Majority of the people will always value convenience and immediate economic benefits over eliminating the perceived low probability risk that one day their privacy will be breached. Internet companies will always be incentivized to accumulate more user data. What we end up with are companies that have amassed so much power that they need to remind themselves “don’t be evil” and most users are effectively left with little to no alternatives.
We are living in the world of capitalism where wealth has the tendency to get more concentrated, organizations and state are getting bigger and more powerful often at the expense of the individual rights. This trend is certainly not sustainable, as a Chinese proverb says: “that which is long divided must unify; that which is long unified must divide” (分久必合，合久必分). The contention between centralization and decentralization is the underlying forces behind so much human social development and will continue to be so for a long time. At a time when centralization is the dominating theme of the society, what blockchain technologies brings on to the table, the decentralization of trust, could potentially offer valuable counterbalance to that. We as a society needs real incentive compatible alternatives to fix our privacy issues than just empty slogans. We as a society needs to have serious experiments of financial systems that do not follow the school of keynesian economics which offers little to remediate the economic problems we suffer today. We as a society needs more starfish rather than spiders building bazaars rather than cathedrals where it makes sense. For many people, perhaps the idea of anarchism is never about actually reaching the state of anarchism but always about alternatives and counterbalances since if the road ahead is not clear, the best bet is to keep an open mind, maintain diversity and zigzag our way forward knowing that we always alternatives.
Self sovereignty is the idea that people should have ownership and control over their own bodies, time, properties and responsibilities. It is at the core of many political philosophies that underlines individualism. Self sovereignty is important since it is essential to personal freedom, self determination and liberty. On the other hand, the forming of societies or any other artificial combinations means that individuals need to surrender part of their self sovereignty to the collective goal of the group, which in many case are actually beneficial both personally and as a whole. After all, humans are by nature incentive driven social animals that try to survive as a race. There are many literatures about the relationship between individualism and collectivism, but it is fair to say that it’s a spectrum rather than binary.
Even though the absolute individualism is perhaps neither desirable nor possible, it has become much more difficult to exercise one’s self soverignty as governments and corporations are making great strides to capture it for their own benefits. Nowadays, we need trust and permissions to use our own identity, our own data and to transact with our own money. Consumerism and the culture of debt also lure people into sacrificing their long term freedom for instant gratification, therefore perpetuating the deprivation of self sovereignty and social stratification.
As digital bearer asset, Bitcoin offers a technology and network that enables and encourages self sovereignty of money. Andreas Antonopoulos famously coined the term “not your keys, not your coins”. Holding your own private keys is the first step towards self-sovereignty in Bitcoin since it cryptographically ensures your property rights to your coin. This of course comes with a lot of responsibilities as well since if the key is lost, your coin will be lost forever. Running a full node is the next step. It not only contributes to the overall consensus of the network, but also gives you the ultimate self-sovereignty within the network since you don’t need to trust any one else when it comes to what happened to your coins or ask for permission as to what will happen to your coins. Even though Bitcoin and other cryptocurrencies were painted as magical internet money that’s used by criminals to buy drugs on the darknet, or a ponzi scheme that will burst at any point in time, what really happened in the past 10 years is that it managed to create a global wealth pool of hundreds of billions of dollars that are borderless and censorship resistent, which helps dissidents in rogue states take a piece of their self sovereignty back and be more likely to fight against the corruption and tyranny knowing that they can take their wealth and move somewhere else by just remembering a bunch of English words.
Bitcoin also encourages a culture that promotes low time preference and delayed gratification. As a sound money and store of value, the long term price of bitcoin is generally expected to rise over time against the constantly inflating fiat currencies despite of the short term volatilities. When money actually keeps or increases in value, it encourages saving and re-investment instead of debt and immediate spending. This shift in mindset might be the key to get us out of the current situation where the economy is powered by over-consumption, debt with artificially low interest rate and quantitive easing.
Self sovereign identity and the processing of encrypted data also attract a lot of research and development to combat the encroachment of privacy. I do think a lot more work needs to be done in this area to really change the consumer behavior, not only in terms of technology but also inventive engineering, to really put the sovereignty of identity and personal data back to the users.
In summary, learning crypto to a certain degree changes the way that I experience and interact with the world. I feel it is an experiment that could be leveraged to potentially do good for the society by providing a balancing force to a lot of the social economical problems that we have right now.