LSANKAR

Whenever something changes suddenly, people start wondering what's next for the systems that we've put in place.

The rise of DeFi on Ethereum has brought that sort of questioning into the smart contract platform space. Ethereum gas prices are as high as they've ever been. All of this congestion has users and application developers asking the obvious question: is Ethereum the place to be?

It turns out that that's an incredibly nuanced question with a number of variables to take into consideration:

impending mainnet launch of a number of new L2 solutions

Eth 2.0+sharding progress

new, well-funded layer 1 chains like Solana/Near

community/tooling inertia in Ethereum

app-chain + interoperability technology like Cosmos/Polkadot

the tradeoff b/w interoperability with DeFi primitives and low fees for new applications

It's a lot to keep track of! It doesn't help that many of the narratives in the ecosystem are driven by the economic incentives of the narrators.

I want to discuss one possible future that I'm really excited about. Using Eth 2.0 shards as a data availability layer for L2s.

That's an incredibly jargon-y, technical phrase and that's why I think it hasn't gotten the love in general internet writing that other ideas have.

My hope is that this is one more mental model of the future that you'll be able to reference as we all navigate the next 5-10 years.

What's the deal with data availability?

'Data availability' is high on the list of important concepts that aren't widely understood. It's commonly used in conversation among researchers but I rarely see it discussed outside of that.

Rather than explain the how, I'll discuss the what of data availability. For more on the technical specifics (the how), there are other excellent resources.

Data availability is a property a blockchain might have. It's helpful to understand what it's not in addition to what it is.

It's not any guarantee around the integrity of data on the chain. For example, knowing that the transactions in a block are being sent from senders who can afford it asks for more than data availability.

It is a guarantee that any user (including a light client) can find all the data in the blockchain.

Here's a visual representation of what a data availability blockchain gives us

Just an ordered sequence of data elements without any notion of what they represent. It's perhaps easy to think about as an ordered, permissionless, append-only database.

We'd need more for a blockchain like Bitcoin or Ethereum; we need a way to know which transactions are valid and which aren't.

Rather than handle this in the blockchain, we could interpret the data availability blockchain within some 'execution context' that tells us which chunks are valid and which aren't:

If we're interpreting our blockchain with 'Execution context 1', chunk 1 represents valid data and chunk 2 represents invalid data. This execution context can live somewhere else, like layer 2.

The beauty of this approach to blockchains is that it completely separates data from execution. We can construct one chain that's really fast at ordering data and handle execution. We can then interpret that data through a layer 2 execution context, like Optimistic rollup.

Eth 2.0 shards as a data availability layer

One possible future of Eth 2.0 sees all but one of the shards becoming high-throughput data availability blockchains for L2s. The exception shard is the same shard that eth1 will transition into and thus is the only shard that actually has an 'opinion' of what data represents at the consensus level.

Here's what that might look like:

Any worries that cross-shard transactions won't be possible are allayed in this future. Layer 2 can make use of all shards in its transactions. Ultimately what economic separation exists between shards is completely up to the L2s.

And we won't have to wait for Phase 2 of Eth 2.0 to get there. I'm hesitant to give time estimates, but this makes the scalable, executable Ethereum dream much more imminent.

This is, of course, just one vision. It's entirely possible that this world doesn't materialize, but it'll be a big deal if it does.

The new protocol wars are afoot!

We're about to enter a world of unknown unknowns as blockchains and scaling technologies try to self-actualize in this new high congestion DeFi regime.

The savvy person will consider many possibilities on a distribution and follow the space closely to understand what's going on as it develops. I hope this brief exploration of one possible future adds to the portfolio of futures that you're considering.

In future posts, I'll dig into a few more.

Thanks to Brian Flynn and Aditya Asgaonkar for comments/review.