Scatterbrain

On Blockchain

01/04/23

// disclaimer: the following post assumes the reader has basic knowledge of crypto and at least partially understands fundamental blockchain concepts such as Proof-of-Work and Proof-of-Stake.

Last night, a friend and former co-worker at Snap messaged me, asking for my opinion on two offers. One was for a well-known tech company, the other for a popular decentralized exchange platform. My friend knew of my reputation as a “crypto skeptic” which I made clear (probably too clear, knowing myself) in conversations which took place on the subject around Snap. I proceeded to bombard their DMs with somewhere around 15 lengthy messages detailing my distain for the field. Towards the end of my rant, I had the unequivocally brilliant idea to publish a slightly more cohesive and constrained version of said rant here on my blog so not only my friend, but the entire world could pause to laugh at my ice cold takes.

To begin to understand my criticisms, one first must understand why cryptocurrencies and blockchain exist in the first place. Decentralized finance (DeFi) is a concept which has been discussed in academia and finance sectors for some time, but as a phrase was only recently coined by Ethereum developers a few years ago. DeFi as a concept is simple: unregulated currencies and exchanges which rely on no singular government or entity for distribution and trading. There are many advantages to a DeFi system: financial transparency, transaction speed, immutability, and no reliance on governing bodies are among the most appealing to me personally. In general, I am strongly in favor of DeFi (not an unpopular opinion to have). However, I firmly believe blockchain and crypto are not the DeFi solution everyone claims they are.

Blockchain is inherently inefficient, both as a raw data structure (the literal blockchain) and when considering its protocols. Two main cryptographic protocols are used in crypto: Proof-of-Work (PoW) and Proof-of-Stake (PoS). For a brief reminder, PoW currencies function by enabling miners to complete ever-increasingly complex computational hashing tasks (normally involving appending blocks to the chain or sometimes minting new coins) in return for a reward of crypto, while PoS systems rely on individual validator nodes to validate and append blocks. For the remainder of this post, I will assume the reader understands the differences between the two protocols, as well as blockchain basics such as hashing/mining difficulty, how transactions are stored in a block, and why validation is a necessary evil.

Bitcoin is inarguably the most popular PoW cryptocurrency (and the OG, a rare combo). It is also a perfect example of the shortcomings of PoW systems. Back in early 2014, five years after the first Bitcoin was minted, hashing and appending a new block (which these days generally records about 2k transactions¹) to the blockchain required solving a task with a hashing difficulty of 10 million.² The hash of the first (genesis) block of Bitcoin had a difficulty of one (1) and is used as the reference for hash difficulty since. To put this rather arbitrary unit system into perspective, a hash difficulty of 10 million would take 102.35 days³ to solve with a mining rig comprised of a single NVIDIA RTX 3090 optimally hashing at 4.86 GH/s.⁴ At the time, finding this block would reward 25 Bitcoins (research Bitcoin halving), which if held today would be worth ~$425k. Today, the hash difficulty is 35 trillion, which results in the same rig taking 981517.18 years to find a block (which only rewards 6.25 Bitcoins today). Now, obviously only novices would even consider mining Bitcoin with a GPU these days, and modern ASIC rigs are significantly more powerful. However, my point is it takes an obscene amount of processing to append a single block (~2k transactions) to a PoW blockchain once the currency matures.

This computational complexity issue highlights the scalability trilemma of blockchain: security, scalability and decentralization (the trilemma states one can generally only have two of the three). Bitcoin (and PoW protocols in general) are secure and decentralized. However, they aren’t scalable. On the other hand, PoS protocols present their own problems. Instead of endless hashing/mining, PoS protocols rely on validators when adding blocks to the blockchain. These validators are selected randomly, but the percentage of stake in the currency a validator has linearly increases their odds of being selected to validate a block. While this process is significantly less computationally complex compared to a PoW system, it leads to two issues. First, if a validator obtains a majority stake in the currency, they will essentially gain full control (a security risk). Currently, the only way to avoid this issue is by regulating the currency against bad actors and attacks, leading to centralization. Therefore, while PoS systems are more scalable than PoW, they have to decide between increased security risks or some degree of decentralization, both of which violate DeFi fundamentals. Regardless, the lack of scalability of PoW protocols is generally regarded as worse than PoS’s issues and has led a trend towards PoS, highlighted by Ethereum’s swap from PoW to PoS earlier this year.

Two other major problems exist in crypto outside of blockchain protocol failures: system complexity and environmental impact. I will not spend much time deep diving into the system design of PoS systems (although it is fascinating), but I will link this article from Jump Crypto which provides a solid overview. The number and complexity of layers needed for the leanest PoS systems rival the tech stacks of mainstream Big Tech platforms. PoS systems pale when compared to the comparatively simple, more purely cryptographical design of PoW protocols. Additionally, in all crypto (but especially PoW coins), the energy requirements to verify transactions are astronomical. The environmental ramifications of crypto have been discussed ad nauseam, but I will highlight a statistic from this article from Columbia: Bitcoin alone generates 22.9 million metric tons of CO2 emissions each year (equivalent to the power requirements for ~2.7 billion homes for the same year) in power consumption costs. Process the fact for a minute.

Fundamental issues plague blockchain tech and, by extension, crypto as a whole. Until they are addressed, I firmly believe the excessive wealth, brainpower, and energy poured into the field is wasted and could be better spent on other ventures, even just other DeFi experiments.