Welcome to Modular Media, a weekly newsletter covering news, updates, educational content, and more within the modular ecosystem.

Subscribe to get issues sent directly to your email every Tuesday, and also, make sure to follow us on Twitter for modular-related updates!

You can check out our previous newsletter issue here.

📣 News & Announcements

Ethereum Upgrades to Shapella

Ethereum’s highly anticipated Shapella hard fork officially took place on April 12th at around 6:30 PM PST. This was the first hardfork since Ethereum’s merge to proof of stake. While there were several changes that came along with this new upgrade, the main focus was on $ETH withdrawals.

As you may already know, anyone can run their own Ethereum validator node and start contributing to the network's security. The incentive for providing your ETH stake is the ETH rewards that come from both transaction fees (user gas fees) and block rewards (newly minted ETH).

In order to run a validator however, the user needs to stake his/her own ETH into the Beacon chain (Ethereum’s PoS network) smart contract. Each validator node requires a stake of 32 ETH.

Those of us who were not able to meet this 32 ETH requirement have simply headed over to a liquid staking solution, such as Lido or Rocket Pool, and taken part in contributing towards Ethereum’s security this way. What these solutions do is pool together user funds until the 32 ETH requirement is met. For every pool of 32 ETH they create, they spin up a new validator node for the Ethereum network.

Whenever users deposit their ETH into these liquid staking solutions, they receive a corresponding token in return (stETH, rETH, etc.) that represents the original user deposit. So for example, if I deposit 10 ETH into Lido, I get 10 stETH in return. Holding this token allows me to accrue validator rewards while giving me the opportunity to exit at any time by simply swapping the stETH (or rETH, etc.) back for regular unstaked ETH.

This is why the Shapella upgrade was a much bigger deal for solo stakers (those who run their own home validator node).

That being said, there are some interesting questions that come to mind when it comes to liquid staking solutions, such as:

Will we see more users withdraw their assets from liquid staking solutions in order to run their own solo setup, now that they can rest assured knowing they can withdraw at any time
We may soon see these solutions develop a mechanism for the automatic staking of issuance rewards. Will most users opt in for this?

Numbers so far

As we can see from the chart above, total staked ETH is down since withdrawals were activated, coming down about 1M ETH, now settling at 17,401,375.

Total staked ETH was of course expected to come down following the hard fork, considering the pent-up demand spanning two years now. What was not as expected was for deposits to immediately come back up and outweigh withdrawals so soon, as we can see below:

You can check out the stats for yourself here.

In regards to the price of ETH, many expected this to fall dramatically, however, this was not the case at all. In fact, ETH has been up only ever since:

What’s next for Ethereum?

Now that Shapella is out of the way, the focus for the Ethereum Foundation is now set on Cancun-Deneb, which is the next hard fork and will introduce EIP-4844, or proto-danksharding. Of course, this topic is for future discussion.

It’s amazing to see Ethereum reach all these milestones. Congrats to the entire team of people who helped bring this upgrade to life!

Introducing the Stackr SDK

Last week, Stackr Labs officially introduced the Stackr SDK, which is a framework that enables developers to launch their own application-specific “micro-rollups” using traditional general-purpose programming languages such as Python and JavaScript, Go, C, Rust, and more.

As a reminder, rollups are a type of off-chain scaling solution that takes a bunch of user transactions, bundles them up (rolls them up), computes them off-chain, and then the results of the computations and data are stored on-chain (on the base layer the rollups is built on top of).

Most rollups we’re accustomed to today are general-purpose rollups, such as Arbitrum and Optimism. The problem with general-purpose blockchains however is that dapps are sharing computational resources and blockspace with one another. Aside from this, dapps under a general-specific rollup have to abide by the rules of that rollup, for example, having to use the EVM within Arbitrum. That being said, there has been a big pivot towards favoring application-specific rollups now. With app-specific rollups, each dapp is its own rollup, meaning it gets its own blockspace, it can set its own rules, and much more, rather than being tied to the rules of a general purpose rollup.

The problem however is having to build your own rollup from the ground up. This is where rollup framework solutions like Stackr come into play.

What is Stackr

Stackr provides developers with different modular components that can be put together to build a rollup that suits the specific needs of each individual team. In other words, each team can build its own customized, dedicated rollup (execution environment) in minutes, rather than in months or years. This means you only focus on your dapp, not on how the underlying cryptography works.

Over time, the Stackr team will add more and more modules for developers to choose between, such as, do you want to use fraud proofs or zk-proofs, or do you want a sovereign rollup or a settled rollup, etc.

Each rollup within the Stackr ecosystem will be able to have seamless interoperability with each other thanks to the use of a “shared aggregator”. You can think of the aggregator as a bundler. In short, this shared aggregator receives a bunch of data from different apps (rollups), bundles that data, and submits it on-chain. So, a dapp written in Python for example, will be able to communicate with a JavaScript written dapp, or with a C written dapp, etc. You can see a visual example of how this will look down below:

How are they different than others?

If you’ve been following Modular Media for a while now, you’ll know that there are several teams working on making rollups accessible to all.

That said, there are generally two different approaches teams are taking. There are teams like Caldera and AltLayer who are working on rollups-as-a-service. This approach compares to a hosting company - you just deploy your contract in a single click (usually) and they take care of managing your rollup and everything. On the other hand, there are teams like Rollkit and Stackr who are building Rollup SDKs. These are not 1-click solutions that take care of managing your rollup or hosting, but rather, they provide developers with a set of modules that developers can choose between (a fraud proof module or zk-proof module for example). This does require a bit more developer experience, however, the reward is more control over your rollup.

However, Stackr separates itself from the rest of the pack even further by being the only solution focused on bringing traditional web2 programming languages to web3, such as Python and JavaScript.

As we can see from the images below, there’s a big discrepancy between the number of web2 developers vs web3. Rather than continuing to try and bring web2 to web3, Stackr is taking the approach of taking web3 to web2, by using the tools most developers are already familiar with.

What’s next

According to the team, “the SDK will also be as flexible as possible regarding proofs (fraud and validity), rollup types (classic and sovereign), ordering (centralized and decentralized), and base layers”. Of course, not all these modules will be rolled out from day one, as these things take time to build.

Stackr is a relatively new project. If you want to stay up to date with everything going on within this ecosystem, make sure to give them a follow on Twitter, and keep up with their Mirror page. Of course, we’ll make sure to keep you updated on all of Stackr’s latest milestones along the way.

They are also hiring! Join Stackr

Intmax raises $5M to build a stateless zk-rollup

On April 13th, Intmax announced a $5 million seed round in order to build a stateless zk-rollup. This round included several capital fund investors, such as HashKey Capital, Bitscale Capital, and others, as well as angel investors like Justin Drake from the Ethereum Foundation, and Scott Moore, founder of Gitcoin.

Intmax is a relatively new project that is being developed by a company located in Switzerland called Ryodan Systems.

Over the past year, CEO Leona Hioki has been publishing posts on the Ethereum Research forum where he shares details on some of the mechanisms behind what is now Intmax, including:

A post about a zk-rollup with no transaction history data
A post about the pre-consensus mechanism
& several others, which you can find here (of course, keep in mind that not all of them may be related to Intmax specifically).

Intmax intends to bring hyper-scaling and privacy to the Ethereum community. What makes Intmax such an exciting project to many prominent people in the space, such as Justin Drake, is the fact that Intmax is one of, if not the only, team working on a hybrid zk-rollup/plasma solution.

Intmax is a stateless rollup

While in general, zk-rollups do provide us with increased scalability for Ethereum, there still lies a bottleneck with data costs. In fact, over 90% of Ethereum rollup costs come from posting data to the L1.

Intmax uses a new architecture, where the end-users and node operators communicate with each other under the hood. This is what makes them stateless.

Essentially, when a user submits a transaction on Intmax, a node operator returns a “Merkle proof” of their asset. In short, a Merkle proof is a way to prove that a particular transaction is included in a block without having to provide the entire block. Using a Merkle proof, a user could calculate a “Merkle root” for a specific block and confirm that it matches the one in the block header provided by the node operator.

“Thanks to this online communication, no transaction history is required to be stored on-chain for contract execution, thus the gas cost on Intmax is reduced massively (95% cost savings over typical zkRollups!).”

Privacy by default

Because the system doesn’t have to post data on-chain, users are able to have privacy by default. We should note that while zero-knowledge proofs do provide privacy-enabling features, not all zk-rollups employ this privacy feature. Most still use zk-rollups strictly for increased scalability.

This means features like customized privacy for NFTs, potentially introducing new forms of private authentication for Ethereum users.

What’s next

Intmax is expected to launch its alpha version on Ethereum mainnet in Q2 this year, with a full mainnet launch expected to happen sometime in Q4 2023.

One thing to note is that the team mentioned their sequencer will be decentralized from day 1 of the full mainnet launch. Specifically, the team mentioned that anyone will be able to participate and get rewarded for block mining as soon as this becomes available. Of course, benefits that come with a decentralized sequencer include censorship resistance, increased interoperability between dapps, MEV decentralization, and more.

Intmax is still a very young project. If you wish to stay updated with everything they are building, make sure to follow them on Twitter or on their Medium account, where they seem to post most of their attest updates. Of course, we’ll make sure to keep you updated with any big updates that come from Intmax as well.

More News & Announcements

The AltLayer Launchpad is now open to all for a free trial
Phase 3 of Celestia’s Blockspace Race officially begins
Uniswap wallet is now available for download on Apple store in most countries
StarkWare announces their investment in Unstoppable Games, a web3 game studio, just officially announced that “Influence”, a space-themed MMO game that is already quite popular within the StarkNet community. The game’s first NFT sale took place on April 11th, and the game is expected to launch this summer.
StarkWare announces investment in “Influence”, a space-themed MMO game that is already quite popular within the StarkNet community.
Speaking of StarkNet games, the Realms team just announced Realms Autonomous Worlds, something they have been working on for 18 months now. Learn all about these digital worlds in the thread linked above.
“Expectium Protocol is the first decentralized information market project on Starknet”. This dapp essentially allows users to vote on real-world outcomes (a predictions market).
Last week, Vitalik joined members of the StarkWare team to discuss “Why Cairo” (StarkWare’s native programming language”). Here’s an article that goes over key points from that conversation.
StarkTokyo conference was held this past weekend. For those that couldn’t attend, @Nurstar wrote this thread that summarizes everything
DIA has integrated its oracle solutions with zkSync Era testnet
Cielo Finance, an analytics platform for EVM chains, is now live on zkSync Era
Symbiosis has integrated zkSync Era into its platform, allowing users to start swapping and farming on the network directly from the Symbiosis app.
Lodestar, Arbtitrum’s native money market is now live, with initial support for $MAGIC, $plvGLP, & $DPX.
Rebase is expanding its AR/NFT technology to the Arbitrum network, as it looks to start its multi-chain journey. Rebase describes itself as a web3 “Pokemon Go” like application.
Pocket Buff announces that the upcoming launch of their MMORPG Sword & Magic World web3 game will take place on the Arbitrum network
Fuel, a modular execution network, just launched its dApp template, “a toolkit to build decentralized apps on Fuel” using built-in components and configurations, while also allowing developers to integrate different out-of-the-box features, like a “Fuel browser wallet connection”.
A couple of members from the Fuel community just launched this community-led ecosystem directory, allowing you to check out the 30+ projects live on the network so far.
The Scroll team launched a new Twitter account that is entirely dedicated to all project and ecosystem activity (project launches, etc.). The Scroll team also added a new ecosystem page on its website, allowing you to check out some of the dapps that have been deployed on the network so far.
Biconomy helped bring the first Account Abstraction application to Mantle’s L2 network. “Devs can now leverage Biconomy’s full Web3 UX stack to unlock the full power of AA for their end users”.
Saga announces a partnership with Ava Labs, for the automated deployment of Subnets. Saga focuses on making blockspace easily available to teams across different networks.
Last week we mentioned Saga’s launch of the Cassio testnet, Saga’s first Chainlet deployment. Cassio has already managed to achieve great success - you can check out the numbers here.
Users can now bridge between Caldera rollups, and other connected L1s, following a partnership with Hyperlane’s interoperability protocol. Teams can start reaching out to Caldera to integrate a bridge into their Caldera chain.
Caldera announces a partnership with Giant Leap, a web3 gaming studio, to bring the launch of their “Underverse” game to its own application-specific rollup.
For anyone interested in all the latest developments across the Caldera ecosystem, check out this Q1 report the team just published.
@Colludingnode and Rollkit introduce “lazy sequencing”
Members of the Rollkit team introduced lazy sequencing. Here’s an eight-minute video that details the mechanism.
Celestia’s first “Modular Fellows Cohort program” has just concluded. There were ten fellows total in this three-month program, and each had to build their own application on a modular chain. Here’s a recap of what they built.
The Informal Systems team just published this March update for the Cosmos Hub.
The Stride team published a proposal on whether or not the protocol should adopt interchain security. The proposal has been published on both the Stride and Cosmos Hub governance forums, for community members to start voting on.
The Interchain Foundation published its Annual report for 2022. Learn about everything that has transpired across the Cosmos ecosystem this past year.

📚 Discourse & Education

Rollup fees vs. Solana fees

Logan Jastremski from Frictionless Capital suggests that Ethereum and rollup communities are delusional in thinking EIP-4844 and Danksharding will drop rollup fees below Solana fees.

Why is he claiming this, and is he correct?

His argument focuses on two suboptimal aspects of Ethereum relative to Solana:

Lower data availability (AKA data throughput)
Lack of parallelization (AKA compute throughput)

Data throughput, in a rollup context, refers to how much data can be posted to the base layer. If rollups can process a lot of data but the base layer can only handle a little data, there’s a bottleneck and fees will spike. A similar dynamic applies to compute throughput.

So how does Ethereum handle compute and data throughput to keep fees down in a rollup world? Is it possible?

As we know, Ethereum tackles compute throughput by offloading the work to rollups. You can run a highly parallelized rollup on Ethereum, with local fee markets to solve the compute throughput problem just as Solana does. In fact, you can even run a Solana rollup on Ethereum that has less overhead than L1 Solana. The extra overhead comes from L1 Solana having to pass around a bunch of messages for consensus, as L1s do. A Solana rollup, even a decentralized one, likely wouldn’t need to pass around as many messages.

Of course, this only scratches the surface of what’s possible with rollups. You can innovate on the Solana rollup, for example enhancing the local fee market mechanism or trying new mechanisms to address state contention. Or you can experiment with different approaches to rollup leader election / consensus. Better yet, multiple developers can try multiple experiments in parallel.

Ok, what about data throughput? This is the other piece of the puzzle, and a separate problem to solve.

Dankrad points to the answer in his response, but we’ve also covered the topic in previous issues:

The bottom line is that Ethereum and other modular base layers like Celestia will treat data throughput as a governable parameter and increase hardware requirements as needed, while retaining trust-minimization and security properties.

How do we reconcile Logan and Ethereum community POVs?

Like many Solana ecosystem discussions, Logan’s post assumes Ethereum’s very near-term data throughput - that is Ethereum’s data throughput within the next few months. His post also assumes the execution layer being used to address compute throughput is the EVM. There are ways to parallelize the EVM, but it’s almost besides the point. There are countless ways to innovate with different execution environments and state contention mechanisms. The beauty of rollups, open innovation and a massive design space.

More Discourse & Education

0xResearch releases an episode “Why Crypto’s Future is Modular”, covering modular interoperability, light clients, settlement assurances and more with Nick from Celestia 🔥
@barnabemonnot writes an article exploring how to engineer protocol credibility
@0xJim writes an article explaining how bridging changes from the monolithic to modular world, soft confirmations, shared security and more 💎
Bell Curve releases an episode discussing MEV economics and relayer monetization with Matt from Blocknative
@chainyoda hosts a Twitter Space discussing what zkIBC is and why it matters for zkEVMs with Garvit from Electron Labs and Jack from Strangelove
@_bfarmer explains why optimistic rollups don’t work well for L3s and how 3rd party bridges don’t work well during periods of volatility 🧠
@analyticalali creates an MEV market map, segmenting the space into three main categories (infrastructure, solutions, applications) and many helpful subcategories - read the full article here 💎
@bkiepuszewski explains the difference between two rollup roles: Sequencers and Validators
Blockworks Research breaks down different approaches to MEV, including Skip Protocol’s Skip Select / Skip Secure as well as @EffortCapital’s sovereign MEV vision
@Data_Always writes an article exploring how successful staking protocols could negatively impact Ethereum’s monetary policy
@ezwillstarr writes an article exploring the math behind ETH staking and potential consequences
@BecauseBitcoin hosts a Twitter Space discussing liquid staking derivatives, data availability and restaker privacy with Sreeram from EigenLayer and @SnoringPokemon
Ether.fi hosts a Twitter Space discussing the Ethereum Sheppella upgrade and liquid staking with Sreeram from EigenLayer
@ShivanshuMadan explains sovereign rollups and how Sovereign Labs unlocks cross-rollup composability
@0xShivani explains the speed and efficiency benefits of sovereign rollups, referencing the recent 0xResearch episode with Sovereign Labs
Celestia writes an article on “How modular blockchains create a world of abundance”, creating a flywheel between apps, blockspace and light nodes
@0xRainandCoffee summarizes their recent article on “Modular MEV”, covering where the input for MEV comes from
@doganeth_en explains the differences between Account Abstraction and MPC wallets, including helpful examples and comparisons
@AlexGuuu breaks down the benefits of rollup-as-a-service solutions and Caldera
@expctchaos highlights user and adoption stats after two weeks of Polygon zkEVM
@odin_free creates an updated map of the Starknet ecosystem
@0xidanlevin explains how culture can be a strategic differentiator, even for zkEVMs
@JoshCStein walks through the ideas behind “lazy sequencing”
@0xjaypeg explains EIP-4844 in simple terms
@zarboq walks through the basic building blocks of Cairo programming
@nickwh8te proposes the idea for a modular blockchain wallet 💡
Eclipse breaks down rollups for beginners
CryptoSapiens hosts a Twitter Space discussing the basics and state of L2s with Bartek from L2Beat, @ddwchen from Rehash and Isaac from Token Engineering Commons
@seunlanlege writes an article explaining polynomial commitments as the foundation for ZK-SNARKS (not for beginners lol)
Taiko writes an article explaining PLONK, SNARKS and more
@SalomonCrypto explains how EigenLayer enables a new category of services
@adietrichs suggests that “danksharding” might be a misleading name (due to using the word sharding), rather than something like “data availability scaling” 🤔

That's all for this week! Thanks for reading 🧱🎬

If you found this useful, please like and retweet this tweet so more people can see it!

Weekly Rollup #10