Jonas Theis 📜
@jonastheis_
Followers
6K
Following
144
Media
14
Statuses
158
Protocol Engineer @Scroll_ZKP. Previously @iota.
Joined February 2022
Having said that, decentralization is obviously a valuable tool to achieve all those and more properties. But it comes at a cost and for some use cases it might not be required or might even be the wrong tool.
1
0
5
Hot take: Decentralized sequencers are not required to provide good user guarantees. Decentralization is just a tool to achieve: - Trustless execution - Verifiable ordering - Censorship resistance - No single point of control TEEs + encrypted mempool can deliver these
probably the best reply to my "single sequencers" post yesterday decentralization is about control verifiability = no one controls the rules censorship resistance = no one controls who can use it without verifiability you have nothing without CR you can still have useful apps
5
0
6
Inspiring to hear @VitalikButerin talk with @TrustlessState and @RyanSAdams about Ethereum’s journey so far and the next 10 years ahead. This is what many of us joined and keep working for—making the world a better place. https://t.co/69cEyICaHo
1
2
14
How many of these properties does a rollup with sequencer in TEE + encrypted mempool have?
When talking about system properties, we need to do a better job at clarifying what we actually mean when mentioning the vague term decentralization. In this great article @mud2monarch breaks down some properties hidden within decentralization. https://t.co/CqvvB42kMO
2
0
3
Ethereum needs to scale but while doing so we also need to keep the complexity of the protocol itself and surrounding the protocol at bay. lean Ethereum is as important as it gets. @drakefjustin is spot on! And check out
Yesterday Ethereum turned 10. Today, lean Ethereum is unveiled as a vision—and personal mission—for the next 10 years. We stand at the dawn of a new era. Millions of TPS. Quantum adversaries. How does Ethereum marry extreme performance with uncompromising security and
1
0
11
Native yield sounds a bit like Blast. In theory a good idea but I'm wondering how this will be implemented in practice as it changes the security assumptions of the bridge significantly. Opt-in or all ETH deposits? "distributed to LPs" sounds like it doesn't accrue to ETH
Bridged ETH will generate DeFi rewards. ETH deposits will produce staking rewards to be distributed to LPs, fuelling Linea’s DeFi ecosystem. Linea LPs will accumulate this native yield in addition to returns from Linea DeFi activity, creating the best risk-adjusted returns and
3
0
7
Meditation. For me the tool of choice to practice awareness and attention (focus) is Vipassana meditation. Starting out a few minutes a day with an app like Headspace can get you quite far already. You don’t have to be hardcore and go to 10 day silent retreats though I’d
the reason why people become addicted to social media is lack of self-awareness by design they are meant to induce you into a reward based habit loop, with infinite scroll making it even harder to break, add to that better algos and notifications. you’re already aware of all of
2
1
5
When talking about system properties, we need to do a better job at clarifying what we actually mean when mentioning the vague term decentralization. In this great article @mud2monarch breaks down some properties hidden within decentralization. https://t.co/CqvvB42kMO
The call for decentralization of rollup sequencers is justified. But imo it is a call for properties and guarantees associated with decentralization (eg censorship resistance, verifiability, liveness, equivocation). It is possible to achieve these guarantees by other means.
1
3
13
The call for decentralization of rollup sequencers is justified. But imo it is a call for properties and guarantees associated with decentralization (eg censorship resistance, verifiability, liveness, equivocation). It is possible to achieve these guarantees by other means.
@bertcmiller @Data_Always haven't read the whole thing, but maybe a solution is to encrypt the txs so that a network level actor has no way to selectively censor txs going into the TEE
0
1
4
I'd argue single company vs decentralized isn't the right distinction either. It's more about trust model and control. You could build/run a rollup with centralized sequencer and strip away all operational control from the operator using eg TEE + encrypted mempool
L1 vs L2 is the wrong debate the actual distinction is this: single company operated vs decentralized the true power of the rollup design is that a company can operate a chain (1) with regulatory benefits of decentralized chains and (2) ship fast as a normal company if
0
0
4
Tbh I don’t really get the push for existing rollups to become native+based. 1. Technically some existing rollups that diverged too much/have different technology can’t become native. 2. Becoming native has benefits. But it essentially makes all native rollups equal in terms of
If Coinbase doesn’t say ‘Ethereum’ in their marketing language, that’s Ethereum’s problem not Coinbase’s Rational actors will act rationally Ethereum must fix its foundation to create stronger alignment among its ecosystem Integrate the rollups. Make them Native + Based.
1
0
4
We've been thinking about this idea for a while: How can you make a rollup safer (fast off-chain finality within few seconds) and cheaper (better compression) at the same time? The answer is subcommitments.
1/ 🔒 Subcommitments: Off-Chain Finality — Fast, Secure, and Cheap We introduce subcommitments, a new mechanism that boosts off-chain finality without sacrificing security or cost efficiency. Let’s break it down 🧵👇
1
0
5
Check out this deep dive by @jonastheis_, probably the most comprehensive piece on rollup design from first principles. It breaks down every component (inbox, sequencer, prover, bridge, L2 node) and explores key trade-offs like: tx data vs. state diffs, latency vs.
1/🧵 What is a rollup? And how do you make sense of inboxes, sequencers, bridges, and proofs? This thread breaks down "Rollups From First Principles" — a must-read by @jonastheis_ for anyone building or curious about L2s. 👇Let’s dive in. https://t.co/b7Cw8pmQ8N
1
1
16
Some thoughts on rollups
1/🧵 What is a rollup? And how do you make sense of inboxes, sequencers, bridges, and proofs? This thread breaks down "Rollups From First Principles" — a must-read by @jonastheis_ for anyone building or curious about L2s. 👇Let’s dive in. https://t.co/b7Cw8pmQ8N
0
1
7
I'm wondering what would be @megaeth_labs motivation and benefits of choosing an "execute-then-order" design? Since order determines execution outcomes this would allow to stream the order of tx even before (speculative parallel) execution is finished. An order could be
0
0
6
Agree with this. Kudos to OP for this + also doing a good job with their spec and docs! And yes, at Scroll we’re trying to simplify things massively. Not only contracts but the entire L2. Easier to understand and extend but also more performant, resilient, and secure.
@zengjiajun_eth @pcaversaccio probably the OP stack. the contracts are very simple and easy to understand which imo is one big factor that led to its success. scroll takes some inspiration from it so it’s also quite simple. taiko in comparison is very convoluted, and orbit stack is also more complicated
1
1
6
Getting to Stage 1 took more time and effort than anticipated. But consistent progress pays off — we’re here. Next stop: Stage 2. https://t.co/m4ebkEd84s
0
0
14