One article analyzes the new data economy of Ethereum

Ethereum’s core developers are already working on another major upgrade. **This upgrade will be centered around Ethereum Improvement Proposal #4844 (EIP-4844). They designated a new portmanteau "Dencun" to refer to this upgrade (combining "Deneb" and "Cancun" for consensus and execution layer updates, respectively). **

EIP-4844 may reduce transaction costs on mainnet, but its focus is on lowering Ethereum Layer 2 fees). To achieve this goal, the approach of this EIP is all about data. EIP will improve the way L2s encodes data on mainnet. L2s currently spend most of their fees on writing to the Ethereum mainnet to validate their ledger (using transaction call data). This also increases the fees on mainnet. You can see this in Etherscan's "gas consumption" ranking list where 5%-10% of mainnet fees are usually related to L2, such as zkSync and Arbitrum.

Example of a large gas spender during July 2, 2023, with zkSync and Arbitrum near the top

Therefore, EIP-4844 is significant. In this upgrade, users of ethereum, such as L2s, will be able to encode so-called data blobs. These blobs will be cheaper as part of the new transaction type because the data will only be kept for 30 days. There will be a second fee market on mainnet for the cost of submitting blobs on the beacon chain (the consensus layer). Blob fees will have dynamics similar to how EIP-1559 manages supply and demand. All of these complexities, including the fascinating details about the blob data itself, are by design to bring Ethereum closer to future scaling upgrades. **L2 can use these cheaper blobs to validate their ledger. **

But EIP-4844 was the first to introduce an important idea in a future update of Ethereum: transient data. 1 This upgrade made me think about its implications. Other planned protocol changes also have the property of on-chain ephemeral data. Looking at the planned upgrades, we see that data is an important part of Ethereum's future. Or, to put it another way, the absence of data is a big part of the future. **

Let's consider some other examples. I'll focus on NFTs to illustrate what data temporality means for the future. Despite the ephemeral issues, this series of upgrades represents ethereum's growing data economy.

1. Historical data processing: EIP-4444

I'm particularly curious about the impact on applications that use on-chain data. In particular, there are more and more NFTs using on-chain data storage. On-chain NFTs store their data on-chain because the asset (artwork, PFP, etc.) is said to be permanent - you can retrieve it on-chain at any time.

But the temporary nature of these upgrades and chain data raises important questions. There are legitimate concerns about the storage and availability of data.

Consider another major improvement proposal: EIP-4444. This EIP will likely be implemented within the next year or two. The idea behind this proposal is simple: Ethereum nodes will no longer need to keep transaction histories longer than a year. This will include block headers, call data, etc. This may affect applications that use historical data, such as market analysis or economic research. It also affects some NFT projects. For example, some well-known NFT projects store their code or data in call data. You can also see this on Etherscan. This is the C code that generated Synth Poems, one of the 0xDEAFBEEF prototype projects. It's in the calldata used for this transaction (its hash is recoverable from the contract function here):

This code is needed to recreate the hypnotic audio-visual experience of 0xDEAFBEEF's works. EIP-4444 prompts nodes to delete this call data because it is 2 years old. (This means that even if you spin up a node yourself in the future, you won't be able to access this data.

An important distinction here is between memory and storage. Since the code for 0xDEAFBEEF is in the call data, there is a risk in EIP-4444 upgrades - it is not accessible in the EVM, and the call data is only in memory when the transaction occurs. Thus, calldata is a historical transaction record accessible to full nodes of the syncchain (but not in the EVM itself). EIP-4444 means it will be pruned after a year.

In contrast, projects using storage retain data in their contracts for access by the EVM. ** On-chain NFTs store data in the contract storage itself. These are part of the Ethereum state, so EIP-4444 is not at risk. **Examples of this storage model are Avastars and CyberBrokers. These NFT items feature a beautiful and layered complex set of functions to assemble SVG artwork.

Other planned upgrades mean contract storage is not completely secure either. It may be subject to later Ethereum upgrades involving state expiration.

2. Data clear status

At this point, you might be asking why the lack of data is so important to the future of Ethereum. A compelling case is made in the episode Unbanked with Vitalik. The interview is a bit dated, but the content is very dated and still a clear discussion of many of the roadmap features.

Around the 40:00 mark of this interview, Vitalik summarizes the challenges data will present to those who wish to participate in Ethereum security - for example by running a node. When Ethereum scales, it will generate petabytes of data per year under the current data model. This is prohibitive for most participants, as they need to be fully in sync with the ever-growing blockchain data.

3. New Data Economy

EIP-4844 blobs are transient. This bridge between L1 mainnet and L2 lasts about a month, after which validators on the Beacon chain do not need to keep them. Where will the blob go? Are they needed in audit or analysis? In EIP-4444, historical data is deleted after one year, and state expiration would involve some similar state deletion schedule. The future of "temporary data".

To observers, this may seem worrisome, especially if you're in a project that makes good use of historical data or contract storage (arguably, that's everything; perhaps most notably on-chain NFTs).

But this transient data approach is necessary. Otherwise the chain becomes too heavy. It is, as Vitalik describes it, "the heavy burden of history". But this presents new challenges for data preservation, recovery, analysis, etc. With challenges comes opportunities. With EIP-4844, we've incorporated a new fee market in the Blob transaction type. EIP-4444 and status expirations are also opening up new opportunities in other markets. Here are some new ideas:

1) Centralized service

The obvious choice for maintaining historical and state data is a centralized service. Vitalik also mentioned Etherscan and other methods (including Beaconscan) in his interview. There is an incentive to maintain these data sources as they are monetized as a service. For Ethereum, this will become even more important, beyond so-called "cleansing", EIP-4444 and state expiration. Tools like Etherscan have been routinely mentioned as critical infrastructure. Their importance will grow in the future era of transient data

2) Encourage distributed data storage

Another way to store history and state data is to create a distributed system (similar to IPFS) built on top of Ethereum. The portal network aims to create a peer-to-peer system that allows lightweight clients to distribute data loads so that history can still be accessed in a similar way to the current API. The Graph is a prominent data infrastructure that many hope will approximate a fully decentralized storage system that can be incentivized through participation in governance and paid data usage. **

Graph for subgraph browser; chain data heap

3) Status maintenance service

The next two raise more interesting possibilities and are related to state expiration. A storage slot can be kept active on a contract to maintain its presence on the chain when the state expires. One can imagine a new contract function periodically "pinging" another contract to maintain some state. **Clients can register with a state maintenance server that uses emerging standards to "ping" all contracts created by a given wallet. For a small fee, subscriptions can be "loaded" for decades to come (similar to an ENS registry). **It can also be decentralized, using a system of contracts that clients can check periodically to ensure the system is functioning properly. If not, they can seek other services or set up a booking system themselves to call a "maintenance" contract.

State maintenance more fully monetizes the "state tree". Some may worry that this is an additional cost to the user, and like the sad "Apple Peripherals", it may lead to higher distributed costs. But the argument against this is that data preservation is costly, especially if there is some tension with securing the blockchain. **Thus, data maintenance services let users pay for the privilege of this data preservation, and let validators and other participants focus on consensus and security. **

4) State restoration service

In his bankless discussion with Vitalik, he emphasized that history is unlikely to be lost. Through the above services we can expect multiple more or less centralized robust tools for persisting historical and state data. But even without these tools, assuming you have information about the storage in the contract, you can still recover them. State restoration can also be a service. It can provide point-and-click tools and some standards and practices for preserving the history that matters to you. You can then bring the data held by the individual to the service, upload it and establish proof of restoration of those states.

Recovery can be fun and satisfying, source MoonCats! 4 Conclusion

Ethereum has to accommodate the security and efficiency of its consensus mechanism, which we expect to massively increase in future use. This goal is at odds with the wonderful and rich data blockchains create. The upcoming upgrade will usher in a new era of "ephemeral data", but it will also bring new and interesting economic possibilities for the maintenance, recovery and management of blockchain data.