Blockchain Killer: The “Impossible Triangle” that plagues transaction quality

In October 2021, I wrote an article Speculative article on the "Trade Quality Trilemma". Since then, almost all chains have implemented minimum fees in the range of $0.01 to $0.50, but there are still some chains that are reluctant to comply like Immutable X, Solana or Arbitrum Nova, which gives us some data.

As you know, I am no longer talking about scaling and infrastructure, what I wrote in 2021 is already deeply understood and proven, so I don’t need to talk about it anymore Write them. I think at a time when we have nearly infinite scaling, one-click rollups, and massive data layers within our horizons, it's even more critical to delve into the basics of blockchain and write about applications and governance for application developers. EigenDA claims to process 100 MB per second. With stateful compression, this is 5 million "TPS" on EigenDA, just settling the data on Validiums/Optimiums. But it also means that as scaling goes too far, the deal quality trilemma will once again become an important topic.

Once your transaction fees drop significantly below $0.01, the chain and its infrastructure become vulnerable to spam, DDoS, and tiny MEV attacks. Especially if the chain also has a financial ecosystem. Let’s consider two chains with fees in the sub-$0.01 range – Arbitrum Nova and Solana. For strategic reasons, Arbitrum Nova is promoted more as a gaming and NFT chain, while Solana is more of an avid casino chain. Obviously both have more features, but as far as I'm concerned this is their main use. The latter comes with a financial ecosystem that results in a large number of tiny MEV and spam transactions - the "low quality" transactions I described earlier.

But first, tiny MEV. The lower your fees, the more likely low-value MEV opportunities will be. People rush to seize these opportunities, resulting in many deals falling through. Even Solana and Arbitrum Nova have the lowest fees, but other chains may try lower fees, which could end up causing over 99% of transactions to fail, be garbage, and be worthless.

Now, some may say, what's the problem? Let them spam attack chains to their heart's content. There are two big problems:

Sustainability and cost: spam is shared across the web, and historical data can accumulate into petabytes in a short period of time. Innovations like validity proofs can significantly reduce the computational burden, but are still a linear cost for sequence generators/builders, and historical data remains the ultimate bottleneck. By simply not processing worthless transactions, the marginal cost of legitimate transactions becomes cheaper - by orders of magnitude for chains where >90% of transactions are garbage, tiny MEV, or failed. It's easy to push these issues aside in the short term, but in the long term they can become Achilles' heel.

$0.01 is too expensive for some use cases: some use cases require free transactions. If you make a $100 DeFi trade, the difference between $0.01 or $0.001 is insignificant, but if you take action in an on-chain game or social network, the difference between $0.001 and $0 is everything.

So, it is clear that we need a solution to the transaction quality trilemma. This brings us to Immutable X and Sorare, which are by far the best solutions. They offer free transactions but use a "Web2 style" approach to spam mitigation. The key is to provide an alternative path to censorship resistance, which comes at a cost. This way, you get the best of both worlds - free transactions for those who need them, but censorship resistance for edge cases. Ideally, you might want to offer free transactions while being censorship-resistant, but this is harder to solve.

The key to achieving this goal is:

a) State isolation

b) Spam mitigation.

a) is critical, otherwise you'll end up with tiny MEV and irrelevant spam. In theory, this could be done within a chain (L2 or L1), but currently it's best implemented on L2, as shown by Immutable X and Sorare.

The obvious drawback here today is the lack of composability, but the solution may be restricted composability with other L2/L1s. It's also worth considering that almost every conceivable use case that requires zero-fee microtransactions simply requires composability - social, gaming, etc.

Spam mitigation is an open problem - I feel there is a lot of research and engineering work required to develop the best solution. I suspect this will require a tailor-made solution to fit the specific application.

Finally, for chains oriented toward high-value financial transactions, a lower fee limit of around $0.01 is acceptable. This is enough to include almost all valuable transactions while reducing the smallest spam or tiny MEV. I don't know what the optimal number is, but it's probably in the range of $0.01 to $0.10.