Thread by @TrustlessState

0/ @VitalikButerin's ZK Rollup presentation () is under-appreciated

In this video, we see:
1a. A piece of technology implemented on Bitcoin
1b. But Bitcoin is limited, it didn't work
2. The same concept, without those restrictions, implemented on Ethereum

1/ Vitalik introduces the concept of a "meta-protocol", and the attempted Bitcoin implementation of it: Mastercoin

A Mastercoin tx makes a Bitcoin tx, but it defines a new set of rules for how BTC txs operates. Its uses Bitcoin as data store, but not as a state execution engine

2/ Mastercoin transaction on Bitcoin have a 'flag' on them, that identifies them as transactions that should be sent through the Mastercoin protocol, so that they can behave by Mastercoin's new set of rules. An EVM of sorts, but not turing-complete. (calculator_vs_computer.jpeg)

3/ But it has some problems that prevent its usage on Bitcoin.

Problem 1. Not lite-client friendly. If you want to learn what the state of the Mastercoin protocol is, an immense amount of calculating is needed by lite-clients that make it infeasible.

4/ Problem 2. Activity inside of Mastercoin protocol cannot change state of anything outside of the Mastercoin protocol. Mastercoin works with Mastercoin flagged Bitcoin txs, but not others.

This is what non-composability looks like. No money legos for Bitcoin (sadface).

5/ Introducing ZK-Rollups:
1. Bundles transactions using Merkle trees
2. Uses ZKsnarks to reduce size of Txs
3. Without using any off-chain data, so is therefore trustless.
-This is like plasma, but more rudimentary... but also on-chain.

6/ With ZK-Rollups:
Gas required for an individual ETH-transfer transaction: 21000
Gas required for a ZK-Rollup transaction, per individual: 884

Expands theoretical on-chain transaction throughput from 15 Tx/S to 500 Tx/S (if all blockspace is allocated to ZK-Rollups)

7/ With ZK-Rollups:
- More complex state transition functions
- Multiple tokens
- Privacy preserving computation

Possible things inside of ZK-Rollups: A mini @Uniswap, for whatever reason. A high-throughput exchange, for whatever reason. ZK-Roll Ups = 1 Money lego.

8/ But wait! There's more!
ZK-ZK-Rollups!

This is really stretching my ability to comprehend this, but basically...
You can use ZK-Snarks to manage transactions that use the data and state transitions inside the Merkle trees inside of a ZK-Rollup. ¯\_(ツ)_/¯

9/ Gas per individual tx inside a ZK-ZK-Rollup:

7100 gas per transaction. 1/3 the gas of a normal Ethereum transaction, but now fully privacy preserving, and with access to a more complex state-execution engine. ZK-ZK-Roll Ups = 1 Money lego. That's 2 money-legos so far!

10/ The Beacon chain introduces 2.8Mb/sec of data availability
One ZK-ZK-Rollup = 105Kbs
Maxed out, the beacon chain could hold up to 27,000 *programmable* tx/second, *with* privacy

11/ But wait, there's like 4 more applications of this! Check out the video to see them, but I want to get back to my main point of this thread:

Cryptographic objects / mechanisms are abound, but their implementation will be enabled, or limited by, the base chain they operate on

12/ Mastercoin illustrates a piece of technology that tried to find it's fit on Bitcoin, but Bitcoin has maximized for simplicity, and it couldn't find a home there

Z-Cash's ZK-Snarks have also made the migration...

https://twitter.com/trustlessstate/status/1090885771353935872

13/ This also illustrates why the criticism "Everyone in Ethereum just follows @VitalikButerin" is shortsighted

It's because he's finding the right solutions! What a silly thing to criticize, when he's putting forth viable solutions to the hard problems of crypto.

14/ As I see it, the crypto community will converge into two separate camps:

1. Maximizing for enabling complexity: putting all features into the same design space, and enabling composability between them.

15/

2. Maximizing for simplicity: reducing the chain to a minimal set of functions, while trusting that other layers will recreate the functionality pushed out of the main chain.

However, this lacks composability, as all functions are executed siloed from eachother.

16/ I am here in cryptocurrency to see what we can build with our growing set of money legos, and this is why composibility is important.

This is one of many examples of a theoretical cryptographic mechanism that found it's death on other chains, only to find a home on Ethereum

The chances I got something wrong about ZK-Rollups and how they operate is non-0, just fyi.