The significance of proof of work: creating an irrefutable history

Author: Gigi, Ethereum enthusiast

By definition, decentralized systems have no single source of truth. The breakthrough achieved by Satoshi Nakamoto was to create a system that allows all participants to independently focus on the same facts. It is the proof of work that makes this possible.

The significance of proof of work lies in its ability to create an immutable history. In the event of a dispute between two histories, the one with more work done prevails. By definition, the chain with the most work is the truth, which is what we refer to as Nakamoto consensus.

Why use work as a metric? In short, because doing work requires energy. You cannot fake, stall, or lie about it. Proof of work is reflected in the results of the work.

In the Bitcoin network, work refers to computation. Not all computation, but specifically computation that has no shortcuts: guessing. There are no shortcuts because this type of computation is devoid of progress. Each guess is a brand new attempt.

The best part is that the work itself is inherently included in the computation results. The data is self-evident: the computation result is the established fact, requiring no external entity to tell you what the truth is. Due to the probabilistic nature of guessing, the data implies the amount of work that has been expended.

Other mechanisms, such as proof of stake, do not possess this characteristic. You can never be sure that what you see is the truth because creating another truth incurs no cost outside the system.

Computation is the only bridge connecting the information world and the physical world. When processing information, all we have is information and its transformation: that is, computation. Computation requires energy. Energy is the bridge. Energy is tangible.

If you cut off one end of the bridge to the physical world, you will forever remain in a mirage: you cannot discern what really happened. You must trust others to tell you what happened. You cannot verify it yourself. You must rely on trust.

Proof of stake has many other issues, such as the fairness of validator elections (who decides how to elect?), the natural tendency towards centralization (more stake = more rewards = more stake), and the lack of inherent resistance to timestamp manipulation attacks.

Proof of work can address these issues. It can decentralize the election process and create physical proof of what has already occurred; it has real external costs that can decentralize the timekeeping work.

As for whether proof of work is a waste of resources, one must first understand what problem proof of work solves before answering this question. If you understand this problem correctly, you will know that there are no other trustless solutions.

Thus, the question becomes: what is the use of a trustless digital sound currency? Is it worth such a consumption of energy?

If it were producing refrigerators, cars, smartphones, and other items, the public would consider it worthwhile. If it produces Bitcoin, those who understand the social benefits brought by sound currencies with censorship resistance would also give a positive answer.

In summary, proof of work is not only useful but essential. Without it, trustless digital currencies cannot function at all. We cannot do without an anchor that connects to the physical world. Without this anchor, we cannot create a self-evident trustworthy history. Energy is the only thing we have that can serve as an anchor.

Proof of Work = Trust the physical to determine what happened

Proof of Stake = Trust people to determine what happened

I deeply sympathize with all those who think Bitcoin is a waste. I used to think the same way, but I have changed my perspective. Like most people, I knew nothing about (sound) money at that time.

In a potentially hostile decentralized system, achieving consensus on time cannot be perfectly resolved, even theoretically. The probabilistic practical solution of proof of work happens to address this tricky issue.

Ledgers take time, and time is a thermodynamic concept.