I have an alternative thesis to crypto, and I think what tends to throw people off when I talk about it is trying to frame it in terms of interactions with crypto today: account abstraction, opt-in privacy, bridge-and-interact – all of these things don't need to exist when rethinking from first principles.

BDFL of Quilibrium, Eng @ Farcaster, ex-Coinbase, always opinionated, never hydrated

/quilibrium

Why do we need account abstraction?

Interactions are complicated, require strong key management, & different types of accounts make this a series of logistical hurdles to overcome. But when we interact with the classic web, we typically don't pop open a terminal, fire up openssl and connect with a client-side cert.

Why do we need opt-in privacy?

Interactions on most L1s involve a public ledger – the separation of users from a network has many layers of de-anonymization:
- the indexer apis your wallet relies on
- the RPC that received your transaction
- the transaction record itself which shows your address

In order to reconcile this, we can add layers of privacy, or even holistic solutions, but this incurs greater costs and processing.

When we interact over Signal, the only privacy link unsolved is the network traffic analysis between users and Signal itself.

Why do we need bridge-and-interact?

L1 transaction costs are high and slow, transactions on an L2 require an operation to bring funds over, consolidating these steps into one makes the user experience slightly easier.

When a normie runs a program on their computer, they generally do not have to consider L1/L2/L3 cache of the CPU because it is managed automatically.

Why need opt-in privacy when we can utilize MPC and ZK approaches that already satisfactorily handle the problem in a way that guarantees privacy for all?

Why need L2s at all when the network can leverage both CPU and distributed database research from _decades_ ago to make a singular network that can scale globally?

This is my vision for the future of crypto. It's not a geeky 8088 breadboard you interact with via a PS/2 keyboard and tap some BASIC into. It's _the web_. You use your browser. You don't use a wallet. You use TLS, interacting with an MPC protocol that handles TLS frames to transparently operate with the underlying net

I don't think about wallets, I think about browsers. I don't think about a distributed ledger, I think about operating systems and databases.

This isn't 50 years away. This is happening right now. It's time to transcend to a new web.

Instead, we can take these lessons learned and build to use the hard work and research that has already been done in these discrete cross sections of computer science.

Why need AA when browsers already support WebAuthN natively and do the hardware enclave work for you and even better – segregated to the domain!