Blockchain legos

The era of the monolithic blockchain is ending. The era of blockchain legos is dawning.

Broadly speaking, blockchains have three functions:

1. Execution: Collecting transactions from users, and executing them as efficiently as possible.
2. Consensus: Provide security and safety - eliminate double spends, reorgs or other fraud.
3. Data availability: Store transaction data and ensure it's always available.

For the last 12 years, blockchains have had to do all of this all by themselves. This has lead to significant inefficiencies and bottlenecks, leading to the blockchain trilemma. To have a highly decentralized and secure blockchain, the execution side is always underutilized. Ethereum is an example of this, of course. On the other hand, you can let the execution side off the leash, but you do this at the cost of an insecure consensus, and the rampant data availability required to keep execution fed makes these networks very centralized over time due to state bloat - Polygon PoS and Binance Smart Chain being examples of this trade-off.

But what if we could separate out these three functions, so three different protocols that are focused on one of these can do what they do best? Enter blockchain legos. This is why rollups are so special. They can focus exclusively on fast and cheap execution, while letting another chain do the hard work of security and most of data availability. But, it doesn't just stop at rollups! Here are some of my thought experiments on how this can proliferate. This is just a fun experiment, don't take these nomenclatures too seriously!

You can take this further... the sky's the limit as zk proofs mature. Some other ideas: adding the type of proof used. E.g. L2V would be zkR, L2F would be OR. Or even, L2P would be Aztec or zkSync as they use PLONKs, while L2T would be StarkNet. Similarly, L212V would be validium, L212F would be Plasma. And before anyone asks, sidechains are just L1. Or, one could argue L222, but that's once again, just L1.

So how does this impact users? It's simple - over the next couple of years, most users will find themselves on execution-exclusive chains like rollups, volitions and validiums, instead of L1s. These will offer lower fees, higher security, and higher decentralization than any L1 ever could. Today's L1s will either need to become a rollup, volition or validium, or become a security and/or data availability chain. The latter is precisely Ethereum's pivot. Yes, there'll still be an execution layer, but it'll mostly be used to settle rollup transactions and larger financial institutions, governments, whales etc. Ethereum's greatest strength will be industry leading security and decentralization, and massive data availability with data sharding - making it the perfect home for rollups. Going forward, most users will use rollups, and rollups will then settle on Ethereum. Of course, there'll be other data availability only or security + data availability solutions emerging, starting with Celestia and Polygon Avail. Architecturally, these are actually the only competition for Ethereum going forward, till other L1s start making their pivots to the new era of blockchain legos.

I believe 90+% of all transactions in the blockchain industry will happen on rollups or other execution-exclusive chains. This is a paradigm shift in how blockchains function, and with Arbitrum One's launch imminent, this would be a great time to reorient your perspective from being L1-centric to being rollup-centric!

If you want to learn more about these various types of blockchain lego solutions, I go in-depth into them on my Reddit profile or my blog at polynya.medium.com.