Does your application need strict global consensus?

Over the last year and a bit, I have written extensively about blockchain applications in general, and strict global consensus in particular. While there’s certainly a niche audience interested in this topic, it’s unpopular, and I hadn’t seen anyone mention or talk about it. @randomishwalk said it best ! I have written at this point over a dozen blog posts about this topic because I continue to believe it’s the most important in crypto currently. Infra and scaling is “solved” - with limitless “TPS” on the horizon with new technologies like validity proofs and data availability sampling. The real question is what will actually use the exponential increase in scale? To answer that, understanding strict global consensus is critical.

A couple of days ago, I saw a mention from @randomishwalk again , someone else had mentioned “global consensus” . It’s difficult to find the motivation to keep writing about these “realistic” topics, which have such a niche audience, when 99% of the crypto sphere is completely obsessed with infra and degeneracy. But having seen the sentence “Are you building something that benefits from global consensus?” (which is basically the topic of most of my blog posts in 2022-23, alongside Governance) start a flowchart see 300,000 views, I’ve found something to write about again.

Nothing here is new I haven’t talked about before, but may help those who have seen that flowchart and are wondering if their applications require global consensus.

I’m not Vitalik and have my own opinion on the matter which may diverge from what Vitalik meant. I call it “strict global consensus” for clarity. You can achieve rough global consensus, but Ethereum (or public blockchains in general) cannot help you with them.

The best example is Ethereum itself. The network is run by thousands of nodes, and developed by a dozen different client teams. They achieve rough global consensus on exactly what Ethereum is, through highly subjective means. Ethereum itself cannot help with this process, of Ethereum existing.

The “strict” also captures that the fact that public blockchains can only do objectivity - another big theme of my blog posts this year. So, “strict global consensus” can thus be defined as - you need everyone running the network globally to agree strictly on a set of objective outputs (some set of numbers).

So, here are an example of things that do not require strict global consensus:

1. Data storage: This is a vast field with many usecases, but in almost all cases, you don’t need everyone in the world to agree on said data. There are tons of options available, from self-custody with tapes and hard drives, to choice of thousands of data storage providers, to distributed options like IPFS or BitTorrent. Indeed, blockchains themselves will prune data and rely on solutions like BitTorrent to remember pruned data.

2. Governance of any kind: governance is a largely subjective and complex endeavour. Public blockchains can help in certain aspects, but trying to shoehorn complex variables into restrictive objective outputs is dangerous.

3. Law and contracts: Likewise, law is immensely complex and subjective, and an evolving discipline. Except for the simplest and dumbest contracts, blockchains cannot help.

4. And really, pretty much everything… Strict global consensus is a very specific feature and there are very few usecases where they make sense.

Now, for where you might require strict global consensus:

1. Objective money: 15 years in, public blockchains are in the maturity phase, and this remains the dominant usecase for very good reason. An alternative store-of-value that is globally accessible and controlled by a set of global noderunners does absolutely require strict global consensus. Of course, “objective money” or value comes in various forms, and are used by other primitives like DeFi, NFTs, DAOs etc. It should be noted that subjective money - like credit - cannot be done on public blockchains.

2. Objective identity: First, I’ll point out that most of identity is subjective. While there have been efforts like attestations, at the end of the day identity and reputation are complex variables, just like we as people are complex and multifaceted. That said, you can do a limited form of identity on public blockchains - objective identity - such as ENS or POAP.

3. Evading laws, filling regulatory gaps: While for most applications public blockchains are wasteful, there are certain usecases which are not legal or there exists gaps in regulatory infrastructure. Today, USDT and USDC are the top consumer usecases for crypto outside of SoV in BTC ETH, and both are defacto centralized. Public blockchains enabled filling in regulatory gaps. However, it’s important to note that these gaps are not permanent - a well-designed democratic USD CBDC can easily replace this usecase, and do so in a more efficient and decentralized manner. Likewise with cross-border payments in certain countries - indeed, this has gotten a lot better between certain countries. Then there’s also downright sketchy stuff, like ponzis, ranging from transparent gambling to deceptive scams. I do think there’ll eventually be regulations for stuff like memecoins and NFTs, now that the pandora’s box is open. This will see this activity move from public blockchains to CEXs - which is where most of the trading happens today anyway.

4. Other niche stuff, such as collaborative storytelling . But the dominant usecases will remain objective money, objective identity and filling regulatory gaps, as it has for 15 years now.

Of course, the real magic comes when you mix and match a bit of the above. Farcaster is a great example of this, using public blockchains for objective money and objective identity but accomplishing everything else off public blockchains. This is where innovation may come from, but for said innovation to happen, we need to think deeply about what strict global consensus actually enables.

As Vitalik says in the flowchart, there’s ton of cool tech outside of public blockchains that can help if strict global consensus is not required.

Lastly, there’ll be applications deployed on public blockchains that don’t actually require strict global consensus or public blockchains but may make sense for chasing incentives, which is fair enough. I want to focus on applications that have long-term, sustainable usecases with potential product-market fit that leverage the unique properties of public blockchains.