Crypto Decentralization Cannot Be An Afterthought
Don’t Forget What This Whole Is About
Naval recently posted on AirChat that, when you’re building a startup, there’s often that one thing at the beginning that you don’t know how to solve. It becomes the “we’ll figure it out later,” thing. He said, in his experience, that thing never gets figured out. I fear that decentralization is often the “we’ll figure this out later,” thing.
This article was originally published on my Substack
I think that decentralization is often an after-thought for crypto protocols because it’s really hard and there is major incentives to centralize. Decisions must be made, competitors beat, features shipped, headlines made, and users satisfied. These things are just easier with huge marketing budgets, centralized development teams, permissioned validators, and VC checks. Every startup wants to move fast and break things, but good luck with that if you rely on decentralized governance.
Ultimately, everyone wants things to be cheaper and faster and they want it to be the way that they want it, three things that are sacrificed with decentralized architecture. Thus, the commitment to decentralization must be just as firm.
If you want to see this fact in the wild, just check out your local zoo, err political system. Libertarians tend to frame political discussions around limiting the expansion of centralized government — (i.e. the question is, not what should the government do, but what role should the government play?). Liberals (in the classical sense) aren’t against government. Rather, they understand the power of it, and so they aim to distribute and decentralize it. Yet, there is always someone in favor of giving away power to a bigger government somewhere in the stack, especially when things aren’t going the way they want it to be going. And history has shown that those who have been given power, seldom surrender it.
The value of decentralization is kind of like the value of preventative health care. It’s not flashy and it doesn’t make big bucks. Why? Because you never get to know how many times you would have gotten sick if you hadn’t eaten well. You’ll never know how many accidents were prevented by your safe driving. You’ll never know how many lives were saved because the tyrant never rose to power. And you’ll never know what disasters were prevented by your slow, thoughtful, and uncompromising approach to building a decentralized protocol. Life saving surgeries are sexy. 10,000 tps are sexy. Decentralization is not.
I worry that we are forgetting this in the age of meme coins, NFTs, ETH-killers, and ETFs. We are forgetting the very values that these systems represented, what ultimately laid the foundation for what we are seeing now. I am not against these things. Like Evelyn Hall said, I might not agree with meme coins, but I will defend with my life your right to issue them. I don’t think meme coins are ruining crypto. Fads will come and go. But let’s not forget what’s more important.
What Is Decentralization?
Decentralization is important because decentralization is security. Vitalik published a great article years ago, exploring the definition of decentralization. According to him, decentralization is important because, decentralized systems are more fault tolerant (keep the network operational), attack resistant (prevent attack), and collusion resistant (prevent people working together to game the system in their favor). I like this definition. But you can also see how collusion resistance might be very annoying to the people who KNOW what’s best and what direction we should go. When the end justifies the means, everything becomes an obstacle — especially decentralization.
Bitcoin’s whitepaper writes in its introduction:
“What is needed is an electronic payment system based on cryptographic proof instead of trust, allowing any two willing parties to transact directly with each other without the need for a trusted third party.”
Ultimately, we can expand the term “transact” to include “interact” so that we can include turing-complete blockchains like Ethereum where network state changes include the transfer of information, not only financial transactions. This is what we are building. The timeline on this is long because the implications of such a system are immense. This is not about who can support a 3-minute NFT mint sellout without lagging today (ok, it’s also about that). It is about what system can survive the next 50 years without collapsing and without sacrificing decentralization. The protocol that can do that will be the most valuable asset in the world. That protocol will be capable of underpinning a global economy.
I find that people often underestimate the timeline on this and thus, put too much emphasis on short term achievements or failures. It is so long because I think true adoption will require generational change. It will require my generation (the millennials), and perhaps the one after, to reach roles of political and economic power, in order to truly rely on decentralized protocols. And which one do you think we’ll choose when we’re ready for that change?
Ultimately, we can find ways to optimize for speed later on (lightning, L2s, zero-knowledge proofs). But I would argue that we seldom find ways to decentralize later on. This is why, personally, I don’t worry so much with short-term scaling issues with Bitcoin and Ethereum. The question is whether they ever can be solved. If they can, then I’m fine with the slow road. Scaling feels to be more of a technical issue which feels solvable with time while decentralization is more of a philosophical issue which tends not to be “solved” but protected.
While I wanted to make the above point, obviously, scalability is still important, and the possibility of scalability is critical. So, is there a path forwards for decentralized architecture to be able to scale to support billions of users? I will tackle this article in the future.
The Three Forms Of Decentralization
Decentralization is a lovely buzz word, but it’s critical to explore what actually makes something decentralized. Again, I want to reference Vitalik’s article here again. He refers to three different forms of decentralization:
- Architectural Decentralization (how many computers)
- Political Decentralization (how many owners of those computers)
- Logical Decentralization (if the interfaces and data structures were cut in half, would it continue to operate?) — i.e. 51% attack for Bitcoin.
In blockchain terms, we are speaking about:
- How many nodes and what’s the barrier of entry?
- How many entities operating those nodes?
- How many of the nodes can shut off or be compromised while the network still operates?
What’s interesting to consider is that, ultimately, total consensus regarding motives is a form of centralization. Vitalik cites how Bitcoin happened into a happy situation where developers tend to be english speakers from the West while miners tend to be Chinese speakers from China, creating an additional barrier to collusion.
We reach a funny sort of paradox when we speak about decentralization as a diverse marketplace of ideas and trials, yet then I say that the one thing everyone in this marketplace should value is decentralization. I remember a quote here that said that you haven’t found truth if you haven’t found paradox. To value decentralization is to protect the marketplace. It is a blank canvas which gives freedom of expression…and immutability.
I will mention here that I think it’s very important to take this framework and apply it to existing protocols and evaluate their decentralization levels. That is a task for tomorrow.
The Long Road Forward
I am not bashing platforms that opt for performance over decentralization. If I was launching an NFT today, I would probably choose the performance-based protocol. I think it’s even possible that such a protocol becomes the preferred option for low-value economic activities. And I think that’s great. But I would argue that it simply will not be sufficient for securing high value transactions and interactions.
Politics should be framed first around, what do we want government to do and then, is it government’s role to do it? Decentralized protocol design should be — what do we want this protocol to do and what is being sacrificed to make that happen?
We must allow for the possibility that we don’t know the right thing. What is most important is protecting the institutions (and now, protocols) that allow for that public discourse: for a trial of ideas and designs. I will not even fault those protocols choosing tps over decentralization. We must even allow for them as part of this experiment. The more the merrier. But I do invite a deeper awareness into the public discord about the critical importance of the freedom of expression and immutability.
We see all too well the corrosive capacity of group-think to dissolve the decentralized systems that, to them, represent an obstacle. Ultimately, even code will falter to protect us if too many join their ranks. It is a more formidable defense than anything we have seen before…but if in the end, decentralization is no longer valued, even code can be changed.
When the axe came into the forest, the trees said “The handle is one of us”
What interests me now is how these developing protocols respond to the masses chanting, “faster.” What I care about is whether the loudest voices in the development space are willing to take the hard road. I care about who will have compromised least in 20 years, not which protocol is pumping today. I’m not saying not to chase short term gains. But I am pleading that you not mistake short term financial aspirations with core values. And often, those who chase the next shiny thing miss the forest through the trees and end up hodling Fool’s Gold.
Thanks for reading. I would love for you to share your comments and ideas. Please leave me a comment or hit me up on AirChat or Twitter to discuss this article (I like AirChat the most!)
As always, none of this is financial advice. I am not telling you to buy or sell anything, just sharing my underlying research and conclusions.
I do hold a portfolio of cryptocurrencies. I was not paid by anyone to write this article.
If you would like to support this writing, please, leave me a comment (I love your feedback), pledge on Substack for when/if this newsletter ever goes paid-mode (you won’t be charged, only if/ever we make this paid):
…or you can buy me a cup of coffee or send some ERC20 tokens my way!
ERC-20 Address: 0x4F9960B2A808e8083C2D12420Aba4496D10e6D37
