## Opening Segment ### Intro James, welcome to the Deeply Intents Podcast. Thank you for joining us. If you could please give a background on yourself and why you're still in it for the tech? What was the motivation for starting init 4 tech? ### Rollup Questions What is a rollup? Can you give a brief history of rollups? I think you've done an excellent job providing citations to Barry Whitehat and Jon Adler's work in writing, but it would be great to get a brief history lesson? In 2023, there was a significant amount of discussion about sovereign rollups, including heated twitter debates, long blog posts, and twitter spaces on the topic. To this day, there still seems to be confusion. What is a Sovereign rollup, and how does it relate to Velvet Forks? What are some of the biggest misconceptions about rollups that still exist today? ### Re(based) rollups In a previous post titled (Re)based rollups you spent some time breaking down what a Fork Choice Rule is in Ethereum and in Rollups. You mentioned - Totally host-following FCR (Based rollups) - follow the host fork-choice perfectly - Partially host-following FCR (Run-ahead rollups) - all rollup reorgs are triggered by a specific host reorg, but not all host reorgs cause a rollup reorg - Host-watching FCR -  only some rollup reorgs are triggered by a host reorg Perhaps we can unpack this a bit more? In the article, you also mention multi-host-following/watching rollups. Previously, you gave a presentation at ZK Summit 4 talking about this concept as a Rollout. If Would you be willing to explain what exactly this new type of construction is and how it relates to our discussion on FCR? ## Middle segment - Signet Questions This seems like a perfect segway to get into Signet. What is Signet? >Signet is a pragmatic Ethereum rollup with sustainable economic incentives. It offers a new set of ideas, aiming to radically modernize and streamline rollups. ### No Proofs! As you noted in your writing, many people have classified rollups by their proving system as ZK or fault proofs for a long time. Signet has neither of those. In particular, as noted in the docs: Signet has no proving system or state roots, which drastically reduces computation overhead. At first pass, this seems like a radical change from what we know of ZK and optimistic rollups. What led you to this design choice? What data does Signet actually post to Ethereum? In the trade-offs section of the docs, you note that;   > While removing proving systems and state roots offers significant benefits to most people, it does not support the following: > - Light clients > - Cryptographic proofs of state Do you suspect this may impact interoperability or any other applications in the future that is reliant on cryptographic state proofs? Do you plan to reconsider this in the future once the Ethereum main-net itself upgrades to a ZK-EVM or ZK-VM? -- ### Conditional and Cross-chain Transfers Signet offers market-based cross-chain transfers for instant asset movement and conditional transactions for secure cross-chain operations. Again, this seems at first glance like a radical departure from the status quo. Some things jumped out at me upon looking at the docs. - no time delay for deposits - users can send transactions directly from Ethereum to signet (corresponding to forced inclusion - USDC/USDT, ETH, and WBTC + assets created on Signet are all supported One obvious question is does then bridging (as in moving assets between Signet and Ethereum) rely on a market maker or filler? Let's spend a bit of time on Conditional Transactions. As I understand, Signet uses Conditional Transactions to enable both local trades and cross-chain transfers. > Conditional transactions succeed if and only if some condition is met at the end of execution. This condition can look at state within the rollup, or in limited cases on Ethereum. This allows same-block interaction and trading with Ethereum. This is really cool, and sounds like a flavor of what we have been calling intents for a while, and has some similarities to the flashbots bundle as well. What do you see as the biggest benefits w.r.t. local trades, (assuming this means trades which take place on signet)? Are their any applications you think are possible or could take advantage of access to both rollup and Ethereum state with Conditional Transactions? One thing that jumped out at me w.r.t. cross-chain Conditional Transfers, is the notion of no Lockup Period for withdraws from Signet to Ethereum, as long as there is a filler willing to fulfill the users order? This seems like it could give builders an edge in distribution, where ETH main-net users may be more willing to use this construction given available liquidity. Were you thinking about this from a product perspective here or was this just purely a technically motivated by improving the status quo of existing rollups? Cross-chain transfers I'm assuming could involve not just Ethereum but other rollups as well? I'm assuming nothing stops application developers from making arrangements with third party bridge providers to bring more assets to signet. Do you see a scenario where Signet picks an exclusive third party provider, which other Alt L1 teams have done in the past? For example, what happens if an application developer who already has an existing ERC-20 issued on Ethereum main-net wants the asset to be available on Signet? Would you recommend application or token deployers to launch their token on Signet first and then choose a bridge provider for their asset to Ethereum or issue the asset on Ethereum main-net first and then stand up their own filler/market maker for that token or pay one? -- ### Sequencing In your documents, you note how sequencing, which is defined as - how a blockchain orders transactions - is centralizing on Ethereum, as auctions have this effect, allowing specialized players to out-compete smaller players, and Rollup sequencing is centralized by default. > As a result, Signet takes an alternative approach. Signet uses a centralized sequencer that assigns block product rights to block builders in a round-robin fashion. That builder is responsible for creating a Signet block and for submitting that block to Ethereum. The Signet sequencer co-signs the builder’s block _without seeing its contents_. The Signet sequencer then just co-signs the block header without seeing its contents. So this means the sequencer cannot censor users, but the block is not valid unless it has the Sequencer's signature? This should be obvious to the listener by now, but explicitly how does this construction differ from existing run-ahead rollup sequencers? Who are the block-builders in this scenario, and how does the round robin work (Tendermint leader election?)? Do you expect that builders may in turn, auction off their blockspace to arbitrageurs and searchers? Also, I don't know how much your team has looked into BuilderNet or TOOL, which in my understanding, will be extended for rollup block building, do you suspect Signet builders will participate in that type of network? ## Last Segment ### Product You have built many products, tools, and protocols in this space, and probably more that go beyond crypto. What are some core things you've learned about building products? What do you think builders and startups in Ethereum get wrong about products? ### Ethereum This year, there has been a number of changes at the Ethereum foundation designed to answer many of the criticisms and position Ethereum to thrive now and into the future. As a builder, have you noticed any changes in the level of engagement with the EF? Along these lines, is there anything you think the EF should do better? (we have some EF listeners) What are some changes you think Ethereum should prioritize in the coming hard forks, if any? ## Closing Where do you see crypto in the next 5 years in terms of adoption, use cases, or social/societal impact?