Why Web3 is a good thing

Apr 21, 2022

I see a lot of heated religious arguments when it comes to web3, and here’s why I think it’s a thing of the future.

It’s basically a single sign-on technology (“connect wallet”) that comes batteries included with a payment system. Moreover, you own your credentials—so your SSO and payment system access doesn’t depend on a corporation.
In the future, you can provably ask the app you’re using to “forget” about you. So it’s not only what data is stored (or proofs thereof) that’s public, but also what data is not stored.
Provability is also built in. When the app computes anything for you (all the way from delivering you news content to doing non-trivial compute such as ML inference) it’s easy to verify what “backend” computation has been run on it. It may even be easy in the future to efficiently verify the computation itself with succinct proofs—like a checksum but not just on storage but also on compute. It’s the trustless/verifiable cloud.
“Code is law.” Beyond provability, code on a smart-contract blockchain maintains the order between transacting parties (is entered into as a contract / law) until parties mutually agree to change the law. A good application is …
DAOs. I don’t think DAOs are a great idea right this day: they’re just companies made with tokens that act as digital share certificates distributed in highly dubious ways. But as the technology matures, many of the constraints one sees in a shareholders agreement or the commercial code can be encoded into smart contracts cutting out the middlemen (ridiculously expensive lawyers and arbiters).
Efficient markets. For just about everything from digital art made by 13 yr olds to stocks issued by 16 yr olds, there’s someone who’s ready to make a market for it. Or, in other words, everything is liquid and basically anything can be securitized efficiently.
It incentivizes the community at large to focus on the hard problems around security and privacy. If I hear about a cryptographic technology that “computes on a piece of data it doesn’t know about, with code it doesn’t know about, in an untrusted environment it doesn’t know about, and write the result to a place it doesn’t know about” I don’t know whether to be surprised anymore. All thanks to people throwing around thousands of dollars on ape pics.

I think any subset of these benefits is enough to make it a groundbreaking technology and a paradigm shift.

Things I’m less than enthusiastic about are NFTs (the technology is great, the instances thereof are mostly overhyped snake oil), decentralization of infra (someone still needs to buy servers in bulk and run them in an iceberg to save costs), and web3 as a replacement for web2 (it’s very early to make a call, but they can probably coexist nicely. ).