Decentralization. A core value of the blockchain. Though probably it is a distant future rather than the current situation. Or the opposite? This is a 🧵 about professional validators and why things are better with decentralization than it seems from the first sight.
Systems with a single managing actor is a norm for web2. A norm that in some cases is extremely harmful for the people who use these systems. It is enough to recall Cambridge Analytica or Wirecard cases.
en.wikipedia.org/wiki/Facebook%E2%80%93Cambridge_Analytica_data_scandal
en.wikipedia.org/wiki/Wirecard_scandal
Same corresponds to Governments. Anywhere where there’s a single person in charge, democracy is doomed to become autocracy or dictatorship, that makes only harm. Just look at Russia.
wsj.com/articles/taking-dictators-literally-and-seriously-xi-jinping-vladimir-putin-ukraine-russia-china-iran-11646421800
Only weighted approaches where any action may be questioned and rejected can make a sustainable move towards better future for all.
Unfortunately, this comes at cost. Decentralized democratic procedures are slow and might seem non-suitable for the current world order. This is a major topic and if you want me to share the thoughts about it in a separate thread, please indicate.
Though democracy was invented and successfully utilized for centuries, until 2009 there were no ‘democratic’ software approaches developed. Bitcoin changed it, marking one of the most important innovations in the 21st century to date.
en.wikipedia.org/wiki/Bitcoin
Satoshi Nakamoto designed a system where anyone would be able to participate as a maintainer, not just a user. And if the number of maintainers is high, this system would become decentralized – a digital democracy.
Nowadays we have many blockchains that *can* be decentralized. However, many people are claiming that their decentralization is a myth because of a small number of the validators – nodes that maintain the system.
A common practice is to measure Nakamoto Coefficient – a minimum number of validators that is enough to execute incorrect transactions on the blockchain (often referred to as 51% attack on PoW or 67% attack on PoS/BFT networks).
minima.global/post/the-nakamoto-coefficient-an-attempt-to-quantify-decentralization
High NC means that the network is decentralized, while low -- the opposite. However, NC, being a simple metric, unfortunately is not capturing all the intricacies of the decentralization.
Validators may be run by the highly connected entities (to intentionally increase the NC). They might be using the same cloud providers or protocol implementations, which dramatically increase the risk of an attack.
Besides that, the number of *delegators* (in PoS consensuses) capable of performing an attack through re-delegating to a single validator, may be lower than the NC.
A special case would be the social attack vectors, where KOLs, usually, protocol founders, are de-facto unilaterally governing the direction of the development of the protocol.
The list of centralization threats can grow for a while.
But besides the negative factors, blockchain space got itself into an intriguing position that unites different ecosystems. And this is possible because of the professional validators (PVs).
PVs are the organizations that are specialized in running servers. They don’t develop dApps. They don’t onboard millions of users. They run servers and get paid according to the consensus algorithm. Any blockchain needs such participants.
And if there’s a need, the open market suffices it. Companies like BisonTrails, @everstake_pool and @dsrvlabs have been created to run many blockchains simultaneously. And now such validators are usually responsible for a large chunk of operations of any major blockchain.
This brings us straight to a surprising outcome: blockchain decentralization is not something separate for each blockchain. There is a connection through PVs between them.
To put is simple, imagine a collusion of validators in NEAR. It must include PVs because of the stake distribution. And there’s no point to break only NEAR, since the same companies are operating Solana, Arbitrum and Optimism.
If one network is hacked, many of them would be hacked at the same time.
Wait, isn’t it bad? Yes and no. The total amount of maintainers in the space is much less than just a sum of the validator numbers, indeed. But the amount of money *lost* by these maintainers in case of the collusion is an order higher than they own in a single ecosystem.
Importantly, this world order helps smaller blockchains to enjoy the decentralization of the bigger brothers: PVs are the same and their attack on a small blockchain would mean destroying their reputation everywhere.
The only other ‘technical’ thing that blockchains should nowadays worry about is the wide distribution of the base token – so the delegators are not the threat to the decentralization. Thus, introducing a limit of tokens sold to a single party. Good figure would be 1-2%.
Social aspects of the decentralization is a much more complicated question. Is it OK that a tiny set of people can drive masses? State developed branches of government to keep any influence contained. Maybe it’s time for us to adopt these ideas?
The only thing that is clear now is that the number of validators or Nakamoto Coefficient or any other technical metric is not able to fully describe the decentralization of the protocol. But perhaps we don’t need to focus on this term for a *single* protocol?
Open market unites different blockchain ecosystems through PVs. Let’s stop fighting because together we are stronger. Blockchain adoption is non-existent comparing to web2 adoption. It’s time to change this.