Understanding Proof of Stake through it’s Flaws (Pt. 1)

[Note: This article requires some background knowledge in cryptocurrency to appreciate. Understanding of the blockchain and Proof of Work are assumed. If you need help understanding them, a simple google search of ‘What is Blockchain’ will suffice.]

Recently there’s been a lot of buzz in the cryptocurrency community about a concept called Proof of Stake. Many in the community view it as the next step in the evolution of blockchain technology, while others view it as fundamentally flawed and think it should never become mainstream. I recently decided to educate myself about the technology, and found that by diving deep into it’s criticisms, I was able to understand Proof of Stake at a fundamental level. With blockchain, many details of an idea are left up to the person implementing them, and the decisions that person makes can have a marked impact on the way the technology can be used (see: the 1MB block size limit in bitcoin). However, there are also issues that will arise with a certain type of construction regardless of how it is implemented. By understanding these types of issues with proof of stake, we can get a much better understanding of the concept.

I’ll first briefly go over what Proof of Stake is and what problem it attempts to solve. This has been done before, and I’m lazy, so this will be brief. For a better treatment of this topic, see: https://hackernoon.com/what-is-proof-of-stake-8e0433018256

The problem that Proof of Stake attempts to solve is related to energy consumption. If you’re at all familiar with blockchain technology, you’ll know that Proof of Work is used to decide who gets to verify payments and send their results to the network, getting a nice ‘block-reward’ in return, as well as any transaction fees offered by the people submitting transactions. Proof of Work is the name for the highly difficult computational puzzle that miners must solve to get the block rewards. Miners go to great lengths to increase their computational power, with centers in China housing millions of dollars of equipment. Graphics-cards makers have even created specialized hardware, fine-tuned to solve the specific puzzle faster than other cards. According to some estimates, 32 TwH of energy are used each year to solve these puzzles, enough to power Denmark.

Now, the economics of this situation are such that miners will be willing to use more and more energy as the value of bitcoin becomes greater and greater. This situation contains a negative externality, whereby the more profitable bitcoin mining becomes, the worse the situation becomes for the environment.

And that’s not the only problem with proof of work. With the advent of expensive specialized mining hardware, there are economies of scale serving those with more capital who can buy more specialized mining hardware. The boogeyman of the bitcoin community is a situation in which a few powerful miners control all of the hardware, and can force us all to accept whatever rules they enforce.

So, then the question is, why even have Proof of Work in the first place? The answer is that Proof of Work is extremely practical. The fact that attempting to validate transactions has a monetary cost is Proof of Work’s greatest asset along with it’s greatest weakness. It means that there is a cost to sending messages to the network (at least those that other nodes would accept), and prevents all sorts of denial of service attacks that the network could suffer if it were free to send validly-looking yet meaningless/fake blocks.

Enter Proof of Stake. The great innovation, if one can call it that, of Proof of Stake is that it moves the mining power, that which grounds the network, away from something outside of the blockchain, and onto the chain itself. In proof of stake, the assets themselves become ‘simulated mining rigs’ which, if ‘Staked’, can passively generate income.

I won’t get into the details, but in essence, Proof of Stake is a system by which you can ‘Stake’ your coins through a special transaction on the network, and in doing so, you get a shot at mining the next block on the network and getting the mining reward. You’re discouraged from doing things like submitting a faulty block or trying to double-spend your coins because, while you’re validating the block, your coins remain in a ‘staked’ state, and you risk losing them all if someone can prove to the network that you cheated.

This system seems airtight. People won’t ever cheat because it can cost them, and denial of service attacks are still impossible, as you would only accept a block from someone who’s been selected by the network to stake the next block. And, since staking a coin doesn’t use any actual resources, only digital ones, it reduces the crypto’s toll on the environment to near-zero. When I first read this, it seemed like the perfect system, almost too good to be true.

And that’s because it is. There are going to be some serious flaws with any cryptocurrency that uses Proof of Stake. There are actually many coins currently being traded that use proof of stake, and you may have heard of the proposed update to Ethereum called Casper, and they all address these issues in markedly different ways. I’ll get to these issues as well how the different coins handle them in my next post.

Part 2 of this series is about the ‘Nothing at Stake’ problem and Part 3 is about Long-Range attacks.