👋 Welcome to Modular Media! We cover news, updates, educational content, and more within the modular blockchain ecosystem.

Subscribe to get posts sent directly to your email every week, and follow us on Twitter for modular-related updates!

This week’s issue covers:

• Optimism Upgrades to Bedrock

• Caldera Launches Taro, an OP Stack Rollup Utilizing Celestia DA

• Avail Launches Kate Testnet

• More News & Announcements

• Best design for 1 billion TPS?

• More Discourse & Education

After months of development, Bedrock officially went live on OP mainnet on June 8th, in what was the first protocol upgrade voted on by the Collective’s governance process. This Bedrock upgrade introduced new features to the Optimism ecosystem, and also sets the foundation for the Superchain (a mesh of aligned L2s).

What is Bedrock

“Bedrock is the name of the first ever official release of the OP Stack, which is a set of free and open-source modular components that work together to power Optimism”. In other words, Bedrock represents Optimsim’s push toward a modular future.

To help elaborate, the OP Stack can be thought of as the actual code that is used to build these blockchain networks. This OP Stack was initially separated into three different layers: execution, settlement, and consensus, however, as mentioned last week, Celestia just added a “Data Availability (DA) interface for the OP Stack, adding an extra layer of customizability.

Because the OP Stack is separated into different layers, developers can mix & match different modules within each of these layers to ultimately build their own dedicated rollup, customized to the needs of your own specific solution. Also, developers are able to switch out their modules as new innovations arise. For example, Optimism will be able to switch from using fraud proofs to zk-proofs in the future when they’re ready.

Bedrock represents a complete rewrite of Optimism’s architecture into this new modular approach. In other words, Bedrock is a standalone implementation of the OP Stack, meaning that Optimism’s architecture is now separated into these different layers that make up the OP Stack.

Bedrock sets the foundation for the Optimism Superchain, which will be a mesh network of L2s. Developers will be able to leverage the OP Stack for a seamless deployment along this Superchain.

How will Optimism accrue value if each team operates its own independent network?

The Optimism sequencer.

Teams will be able to opt-in to using Optimism own in-house sequencer, which will allow the Optimism Foundation to earn extra MEV revenue. This revenue will go towards the Optimism Grant Rounds, which will then fund the development of more rollups, thus potentially increasing revenue even more and continuing this cycle.

What did Bedrock introduce

Some improvements that Bedrock introduced include:

• a reduction in fees thanks to new “batching & compression” techniques: Optimism transactions get bundled into “sequencer batches”, sequencer batches get grouped into “channels”, which are then submitted to Ethereum as “channel frames” (independent chunks of this channel data). Since the order of these frames has already been established, they can be submitted to Ethereum in parallel, ultimately increasing the efficiency.

• Bedrock will implement EIP-1559: “A transaction pricing mechanism that includes fixed-per-block network fee that is burned and dynamically expands/contracts block sizes to deal with transient congestion”. This is how $ETH issuance has gone negative during times of increased user activity.

• Lowered deposit time: It now takes about 2.5 minutes to deposit funds into Optimism, rather than the 10 minutes it would take before.

• increased security: “the Bedrock release introduces a 2-phase withdrawal process, doubling down on bridge security, which proved to be a primary risk factor for chains in 2022”.

What’s Next

Congrats to the entire Optimism Collective for this amazing achievement. We look forward to watching the growth of the Superchain over the coming year. To learn more about Bedrock you can check out the official docs.

In last week’s newsletter, we mentioned Celestia’s launch of a new interface for the OP Stack that’ll allow developers to start building their own OP Stack rollups that utilize a separate DA solution instead of posting to Ethereum. This past week, Caldera officially launched Taro, “the first live blockchain to make use of Celestia's newly released modular DA interface for the OP Stack”.

Specifically, the components of Taro can be broken down as follows:

• Execution: Taro (OP Stack rollup). This is where us end-users interact.

• DA: Celestia (Arabica testnet). This is where the data behind our rollup transactions are posted.

• Settlement: Ethereum (Goerli testnet). This is where disputes are settled. If there is a faulty block, you use the data on Celestia to submit a proof.

Why would a rollup want to use Celestia over Ethereum?

As some of you may already know, rollups are an off-chain scaling solution that takes a bunch of user transactions, bundle them together into a single block, and posts the end state on the L1. For anyone (you, me, or anyone else) to check that the transactions are being processed correctly, rollups are required to post data on-chain and make that data available for anyone to gather. This data verifies a given block.

Before Celestia built this DA interface, OP Stack rollups could only post this data on Ethereum. There were two issues with this:

• high gas cost: The gas fees we pay on the Ethereum L2s mainly consists of data costs (the cost to post the data on the L1 - Ethereum). In fact, Optimism has paid over $20M in fees for posting data to Ethereum so far.

• low throughput: “Even if used completely for data availability, Ethereum calldata can only support a few hundred transactions per second”. Every 13 seconds, one block is posted on Ethereum. Each block consumes around 15M in gas, with a typical transaction consuming around 3200 gas. If we deduce this to the second, then we get to around 360 transactions per second.

So as we can see, the advantage of posting data to Celestia as opposed to Ethereum is lower operational costs for teams, while the end-users get to benefit from even lower fees. Also, we don’t have to worry about random gas spikes during times of increased congestion on Ethereum (a lot of apps & rollups bid to post on Ethereum).

How to get started on Taro (include a short guide)

(1) Head to the Taro Hub page

(2) Click on the “add chain to Metamask” button (you can also add the chain manually if you prefer)

(3) Next, click on “request faucet funds on Taro rollup” (simply enter your wallet address after clicking, and you should receive $GETH within a couple of seconds)

(4) Once you receive the testnet funds, there are two options for you to choose from. You can either create a post on the GM portal, or mint a generative NFT.

(5) Once you submit a transaction, you can view it for yourself on the Taro Block Explorer. So far, the testnet has seen close to 88K transactions, from across nearly 28K wallet addresses:

What’s Next

Caldera is a rollup-as-a-service (RaaS) solution that developers can use to build and deploy their own custom OP Stack rollup “in one click”. Following Celestia’s launch of the DA interface for the OP Stack, Caldera added Celestia DA as a feature of Caldera chains, now allowing developers to use Caldera to deploy their own custom OP Stack x Celestia rollup in a single click. You can reach out to the team if you’re interested.

For developers interested in experimenting, you can start deploying your own smart contracts on top of Taro today.

On June 7th, Avail announced the launch of their “Kate” testnet (Phase 2), bringing the data availability (DA) solution one step closer to mainnet launch. According to the team, “our objectives for Phase 2 include exhaustive testing, iterative system enhancements, and the continued expansion of our validator network”.

Phase 1, which was first deployed in June 2022, marked the initial testnet launch of the Avail network. This initial testnet was deployed with a select group of internal validators (had to make sure everything was running smoothly before opening up the network to external actors), whereas phase 2 (Kate) will expand the validator set to include external members.

Specifically, phase 2 plans to onboard 200 more external validators. For anyone interested, you can indicate your interest in becoming a validator by filling out the “validator interest form”.

Here are some of the new enhancements in Kate:

• Complete migration from Polygon (Avail is now officially its own independent solution)

• increased performance boost for light clients

• seamless upgrades for validators moving forward

• a discord faucet for testnet tokens (for participating validators to use for onboarding)

• revamped docs (help with onboarding for validators)

You can learn all about the technical details through Avail’s official blog post.

What is Avail

Avail is a data availability solution (similar to Celestia or Eigen DA). Avail was initially another branch of the Polygon team, however, just a couple of months ago spun out to become its own independent entity.

Data availability solutions take in transaction data from other chains, store that data, and make it available for anyone to retrieve. Teams can trust that their data is being held by Avail thanks to the use of light clients.

As you may already know, most blockchain networks have their own validators that run full nodes in order to keep the network running 24/7. That said, not everyone is able to afford the luxury of running their own full-node. For one, it costs money to buy the hardware, and second, it’s tough to keep your node running 24/7 so that you don't risk losing your validator stake. This is where light clients come in.

Light clients are able to check that the data is being appropriately held, without having to run a full node and download the entire block. Instead, light clients simply download a single chunk, or a portion, of the block and together create the entire block. Best of all, you can run a light node on almost any device, from your tablet to your smartphone. The more people running light nodes, the more people there are to download these chunks of data, meaning the more blocks can scale.

You can follow Avail on Twitter to keep track of the team’s journey to mainnet.

• Electron Labs launches its permissionless zk-bridge in closed beta on Near & Ethereum. As of now, the bridge is only accessible for “Tinkerers NFT” holders. The birdge is expected to go live for the public on June 12th.

• Argus, a web3 game developer and publishing platform, announced a $10M seed round, which included investors such as Haun Ventures, Alchemy Platform, Balaji, and more. Argus will sue the funds to continue development of “World Engine”, a sharded rollup SDK built for on-chain games. Teams are able to start applying to become private beta testers for World Engine.

• Base, Coinbase’s L2 rollup launched on Optimism’s Superchain, recently made some scalability improvements after witnessing increased gas fees due to data costs (most gas is spent paying for data publication on the L1, Ethereum). You can learn about the two new changes they made here.

• On June 7th, Arbitrum experienced some issues on its sequencer that caused batches not to be posted on the L1 (Ethereum). In short, “When the Sequencer tried to post a batch on-chain, the bug hit, and the transaction reverted”. Arbitrum’s sequencer wallet only receives funds (to use for gas when posting batches to the L1) only if it posts bacthes successfully. Because the bug caused an issue, the sequencer wallet was not able to be replenished, adding more delay to L1 batch posting. You can learn more about what happened here.

• @anuragarjun, Avail founder,

Viktor Bunin from Coinbase Cloud and Credibly Neutral presents an interesting framing for scaleable blockchain systems. If you assume crypto reaches 1 billion TPS over time, how would you design the system? Viktor argues that in this future it does not make sense to optimize for L1 throughput.

Many members of the Solana community had opinions, as Solana is of course a blockchain that heavily optimizes for L1 throughput. Here are some of their responses:

• Hardware might keep up with demand

• Solana will still be relatively more scaleable

• You cannot do 1bn TPS without high throughput

• The goal is to minimize number of layers

Let’s address them:

• It’s unlikely that hardware keeps up with demand. There probably won’t be many orders of magnitude in hardware improvement in 10 years, not to mention that developers have historically always pushed the boundaries of what’s possible with computing resources (e.g. internet bandwidth, cloud compute, GPUs).

• While it’s true that throughput must come from somewhere, this does not conflict with Viktor’s take as the (specifically compute) throughput is expected to come from rollups. It’s easy to poke fun at 100 TPS rollups, but rollups are far from limited to this. In fact, they can support much higher TPS - Fuel and Solana rollups are two examples. Of course, the underlying base layer must support high enough data throughput, but this is very doable.

• There’s seemingly easy arguments to make such as that Solana will always be a relatively more scaleable base layer, even in a modular world, and that it’s better to

• Empire discusses sovereign rollups and the inevitability of modular tech stacks with Mustafa from Celestia and Preston from Sovereign 🎙️🔥

• Unchained Podcast discusses whether modular blockchains will win and ELI5s rollups 🎙️🔥

• @ShivanshuMadan explains the recent rollups ≠ bridges debate in super simple terms ✍️💎

• Bankless explores all things restaking with Sreeram from EigenLayer 🎙️

• @khushii_w walks through the history of intents in web1 and web2, as well as how users will interact with them in web3 💬

• @yuxiao_deng unpacks the sequencer decentralization toolbox 💬💎

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

If you found this issue useful, please share this tweet so more people can see it 🙏

https://twitter.com/modularmedia_/status/1668667254144397314?s=20

👋 Welcome to Modular Media! We cover news, updates, educational content, and more within the modular blockchain ecosystem.

Subscribe to get posts sent directly to your email every week, and follow us on Twitter for modular-related updates!

This week’s issue covers:

• Optimism Upgrades to Bedrock

• Caldera Launches Taro, an OP Stack Rollup Utilizing Celestia DA

• Avail Launches Kate Testnet

• More News & Announcements

• Best design for 1 billion TPS?

• More Discourse & Education

After months of development, Bedrock officially went live on OP mainnet on June 8th, in what was the first protocol upgrade voted on by the Collective’s governance process. This Bedrock upgrade introduced new features to the Optimism ecosystem, and also sets the foundation for the Superchain (a mesh of aligned L2s).

What is Bedrock

“Bedrock is the name of the first ever official release of the OP Stack, which is a set of free and open-source modular components that work together to power Optimism”. In other words, Bedrock represents Optimsim’s push toward a modular future.

To help elaborate, the OP Stack can be thought of as the actual code that is used to build these blockchain networks. This OP Stack was initially separated into three different layers: execution, settlement, and consensus, however, as mentioned last week, Celestia just added a “Data Availability (DA) interface for the OP Stack, adding an extra layer of customizability.

Because the OP Stack is separated into different layers, developers can mix & match different modules within each of these layers to ultimately build their own dedicated rollup, customized to the needs of your own specific solution. Also, developers are able to switch out their modules as new innovations arise. For example, Optimism will be able to switch from using fraud proofs to zk-proofs in the future when they’re ready.

Bedrock represents a complete rewrite of Optimism’s architecture into this new modular approach. In other words, Bedrock is a standalone implementation of the OP Stack, meaning that Optimism’s architecture is now separated into these different layers that make up the OP Stack.

Bedrock sets the foundation for the Optimism Superchain, which will be a mesh network of L2s. Developers will be able to leverage the OP Stack for a seamless deployment along this Superchain.

How will Optimism accrue value if each team operates its own independent network?

The Optimism sequencer.

Teams will be able to opt-in to using Optimism own in-house sequencer, which will allow the Optimism Foundation to earn extra MEV revenue. This revenue will go towards the Optimism Grant Rounds, which will then fund the development of more rollups, thus potentially increasing revenue even more and continuing this cycle.

What did Bedrock introduce

Some improvements that Bedrock introduced include:

• a reduction in fees thanks to new “batching & compression” techniques: Optimism transactions get bundled into “sequencer batches”, sequencer batches get grouped into “channels”, which are then submitted to Ethereum as “channel frames” (independent chunks of this channel data). Since the order of these frames has already been established, they can be submitted to Ethereum in parallel, ultimately increasing the efficiency.

• Bedrock will implement EIP-1559: “A transaction pricing mechanism that includes fixed-per-block network fee that is burned and dynamically expands/contracts block sizes to deal with transient congestion”. This is how $ETH issuance has gone negative during times of increased user activity.

• Lowered deposit time: It now takes about 2.5 minutes to deposit funds into Optimism, rather than the 10 minutes it would take before.

• increased security: “the Bedrock release introduces a 2-phase withdrawal process, doubling down on bridge security, which proved to be a primary risk factor for chains in 2022”.

What’s Next

Congrats to the entire Optimism Collective for this amazing achievement. We look forward to watching the growth of the Superchain over the coming year. To learn more about Bedrock you can check out the official docs.

In last week’s newsletter, we mentioned Celestia’s launch of a new interface for the OP Stack that’ll allow developers to start building their own OP Stack rollups that utilize a separate DA solution instead of posting to Ethereum. This past week, Caldera officially launched Taro, “the first live blockchain to make use of Celestia's newly released modular DA interface for the OP Stack”.

Specifically, the components of Taro can be broken down as follows:

• Execution: Taro (OP Stack rollup). This is where us end-users interact.

• DA: Celestia (Arabica testnet). This is where the data behind our rollup transactions are posted.

• Settlement: Ethereum (Goerli testnet). This is where disputes are settled. If there is a faulty block, you use the data on Celestia to submit a proof.

Why would a rollup want to use Celestia over Ethereum?

As some of you may already know, rollups are an off-chain scaling solution that takes a bunch of user transactions, bundle them together into a single block, and posts the end state on the L1. For anyone (you, me, or anyone else) to check that the transactions are being processed correctly, rollups are required to post data on-chain and make that data available for anyone to gather. This data verifies a given block.

Before Celestia built this DA interface, OP Stack rollups could only post this data on Ethereum. There were two issues with this:

• high gas cost: The gas fees we pay on the Ethereum L2s mainly consists of data costs (the cost to post the data on the L1 - Ethereum). In fact, Optimism has paid over $20M in fees for posting data to Ethereum so far.

• low throughput: “Even if used completely for data availability, Ethereum calldata can only support a few hundred transactions per second”. Every 13 seconds, one block is posted on Ethereum. Each block consumes around 15M in gas, with a typical transaction consuming around 3200 gas. If we deduce this to the second, then we get to around 360 transactions per second.

So as we can see, the advantage of posting data to Celestia as opposed to Ethereum is lower operational costs for teams, while the end-users get to benefit from even lower fees. Also, we don’t have to worry about random gas spikes during times of increased congestion on Ethereum (a lot of apps & rollups bid to post on Ethereum).

How to get started on Taro (include a short guide)

(1) Head to the Taro Hub page

(2) Click on the “add chain to Metamask” button (you can also add the chain manually if you prefer)

(3) Next, click on “request faucet funds on Taro rollup” (simply enter your wallet address after clicking, and you should receive $GETH within a couple of seconds)

(4) Once you receive the testnet funds, there are two options for you to choose from. You can either create a post on the GM portal, or mint a generative NFT.

(5) Once you submit a transaction, you can view it for yourself on the Taro Block Explorer. So far, the testnet has seen close to 88K transactions, from across nearly 28K wallet addresses:

What’s Next

Caldera is a rollup-as-a-service (RaaS) solution that developers can use to build and deploy their own custom OP Stack rollup “in one click”. Following Celestia’s launch of the DA interface for the OP Stack, Caldera added Celestia DA as a feature of Caldera chains, now allowing developers to use Caldera to deploy their own custom OP Stack x Celestia rollup in a single click. You can reach out to the team if you’re interested.

For developers interested in experimenting, you can start deploying your own smart contracts on top of Taro today.

On June 7th, Avail announced the launch of their “Kate” testnet (Phase 2), bringing the data availability (DA) solution one step closer to mainnet launch. According to the team, “our objectives for Phase 2 include exhaustive testing, iterative system enhancements, and the continued expansion of our validator network”.

Phase 1, which was first deployed in June 2022, marked the initial testnet launch of the Avail network. This initial testnet was deployed with a select group of internal validators (had to make sure everything was running smoothly before opening up the network to external actors), whereas phase 2 (Kate) will expand the validator set to include external members.

Specifically, phase 2 plans to onboard 200 more external validators. For anyone interested, you can indicate your interest in becoming a validator by filling out the “validator interest form”.

Here are some of the new enhancements in Kate:

• Complete migration from Polygon (Avail is now officially its own independent solution)

• increased performance boost for light clients

• seamless upgrades for validators moving forward

• a discord faucet for testnet tokens (for participating validators to use for onboarding)

• revamped docs (help with onboarding for validators)

You can learn all about the technical details through Avail’s official blog post.

What is Avail

Avail is a data availability solution (similar to Celestia or Eigen DA). Avail was initially another branch of the Polygon team, however, just a couple of months ago spun out to become its own independent entity.

Data availability solutions take in transaction data from other chains, store that data, and make it available for anyone to retrieve. Teams can trust that their data is being held by Avail thanks to the use of light clients.

As you may already know, most blockchain networks have their own validators that run full nodes in order to keep the network running 24/7. That said, not everyone is able to afford the luxury of running their own full-node. For one, it costs money to buy the hardware, and second, it’s tough to keep your node running 24/7 so that you don't risk losing your validator stake. This is where light clients come in.

Light clients are able to check that the data is being appropriately held, without having to run a full node and download the entire block. Instead, light clients simply download a single chunk, or a portion, of the block and together create the entire block. Best of all, you can run a light node on almost any device, from your tablet to your smartphone. The more people running light nodes, the more people there are to download these chunks of data, meaning the more blocks can scale.

You can follow Avail on Twitter to keep track of the team’s journey to mainnet.

• Electron Labs launches its permissionless zk-bridge in closed beta on Near & Ethereum. As of now, the bridge is only accessible for “Tinkerers NFT” holders. The birdge is expected to go live for the public on June 12th.

• Argus, a web3 game developer and publishing platform, announced a $10M seed round, which included investors such as Haun Ventures, Alchemy Platform, Balaji, and more. Argus will sue the funds to continue development of “World Engine”, a sharded rollup SDK built for on-chain games. Teams are able to start applying to become private beta testers for World Engine.

• Base, Coinbase’s L2 rollup launched on Optimism’s Superchain, recently made some scalability improvements after witnessing increased gas fees due to data costs (most gas is spent paying for data publication on the L1, Ethereum). You can learn about the two new changes they made here.

• On June 7th, Arbitrum experienced some issues on its sequencer that caused batches not to be posted on the L1 (Ethereum). In short, “When the Sequencer tried to post a batch on-chain, the bug hit, and the transaction reverted”. Arbitrum’s sequencer wallet only receives funds (to use for gas when posting batches to the L1) only if it posts bacthes successfully. Because the bug caused an issue, the sequencer wallet was not able to be replenished, adding more delay to L1 batch posting. You can learn more about what happened here.

• @anuragarjun, Avail founder,

Viktor Bunin from Coinbase Cloud and Credibly Neutral presents an interesting framing for scaleable blockchain systems. If you assume crypto reaches 1 billion TPS over time, how would you design the system? Viktor argues that in this future it does not make sense to optimize for L1 throughput.

Many members of the Solana community had opinions, as Solana is of course a blockchain that heavily optimizes for L1 throughput. Here are some of their responses:

• Hardware might keep up with demand

• Solana will still be relatively more scaleable

• You cannot do 1bn TPS without high throughput

• The goal is to minimize number of layers

Let’s address them:

• It’s unlikely that hardware keeps up with demand. There probably won’t be many orders of magnitude in hardware improvement in 10 years, not to mention that developers have historically always pushed the boundaries of what’s possible with computing resources (e.g. internet bandwidth, cloud compute, GPUs).

• While it’s true that throughput must come from somewhere, this does not conflict with Viktor’s take as the (specifically compute) throughput is expected to come from rollups. It’s easy to poke fun at 100 TPS rollups, but rollups are far from limited to this. In fact, they can support much higher TPS - Fuel and Solana rollups are two examples. Of course, the underlying base layer must support high enough data throughput, but this is very doable.

• There’s seemingly easy arguments to make such as that Solana will always be a relatively more scaleable base layer, even in a modular world, and that it’s better to

• Empire discusses sovereign rollups and the inevitability of modular tech stacks with Mustafa from Celestia and Preston from Sovereign 🎙️🔥

• Unchained Podcast discusses whether modular blockchains will win and ELI5s rollups 🎙️🔥

• @ShivanshuMadan explains the recent rollups ≠ bridges debate in super simple terms ✍️💎

• Bankless explores all things restaking with Sreeram from EigenLayer 🎙️

• @khushii_w walks through the history of intents in web1 and web2, as well as how users will interact with them in web3 💬

• @yuxiao_deng unpacks the sequencer decentralization toolbox 💬💎

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

If you found this issue useful, please share this tweet so more people can see it 🙏

https://twitter.com/modularmedia_/status/1668667254144397314?s=20