We’re looking for feedback on the Interop explainer. Takes <2min.

Would love to know: 
- What questions do you have after reading it?
- What was confusing about it?
- What do you want to know more about?
- What were you surprised to learn? 


https://docs.optimism.io/stack/protocol/interop/explainer

We know that Superchain interoperability is hard to explain, but we're too close to the problem to know why. 

We would greatly appreciate you taking the time to read this, and tag @zain to complain about what's unclear to you.

Contributing to OP Labs. Come join me on the track if you’re in /austin @ atxsprint.club

We know that Superchain interoperability is hard to explain, but we're too close to the problem to know why. 

We would greatly appreciate you taking the time to read this, and tag @zain to complain about what's unclear to you. 

https://docs.optimism.io/stack/protocol/interop/explainer

Pragmatism (Security) at OP Labs.
Prev: ConsenSys & Coinbase.

Attribs: Friendly, curious, introverted, mid-witted.
Chans: /eth-security /maurelian /stoicism




My biggest question isn't how it works, but what it means for future decentralization of sequencers – is this objective now off roadmap?

Hey Cassie! The initial release of Superchain interop does not impact the path towards sequencer decentralization

BDFL of Quilibrium, Eng @ Farcaster, ex-Coinbase, always opinionated, never hydrated

/quilibrium

Hey Cassie! The initial release of Superchain interop does not impact the path towards sequencer decentralization

https://docs.optimism.io/stack/explainer#configurable-sequencer-per-op-chain

How does it not? The modularity of sequencing only implies greater complexity for interop, but I don't see where or how this outcome is prevented in the docs.

Mind if I grab 15min of your time to chat sync? Want to make sure we're on the same page when talking about shared sequencing vs decentralized sequencing vs atomic / low latency interop so I can better respond to your q!

Probably won't have time in the near future, but the question can be restated if it helps:
When two OP chains have low latency interop, there is a degree of trust between those sequencers about state. With decentralized sequencing (where anyone can be a sequencer), under the optimistic rollup approach (not ZK proven blocks), how is low latency interop safely achieved? Or does it require L1 settlement (i.e. the "safe" number of blocks have passed) in that scenario?

gigameshmusic.com | formerly @optimism, @soundxyz, @farther

A successful challenge won't cause the L2 to reorg. 

I think you're alluding to, "What if the sequencer lies"

Reposting @soyboy's response:

"The sequencer and all of the nodes on the L2 must follow consensus rules, so there isn't a way to make an invalid transaction because all the nodes on the network just throw it out. The same way its handled on Ethereum, Bitcoin, etc. 

The challenge window is very specific to ensuring the withdrawals from the L2 to Ethereum are valid. Essentially the L2 txs and the L1 deposit txs into the L2 are easy to verify because the L2 is aware of both the L2 network state and L1 deposits into the L2 network state. "

Still consensus rules for the elected sequencer

Sure, I’m not asking about how sequencers are enforced to behave block by block, but rather how interop can't become a vector for attack by a malicious sequencer.

Simple example: sequencer on OP fork A sends sequencer on OP fork B some quantity of an ERC-20 that is valid for the block at the same time they also admit a transaction on fork A that spends that same quantity of ERC-20. How are conflicting actions derived from the same initial valid state such that A cannot lie to B without waiting for the L1?

Still consensus rules for the elected sequencer
https://warpcast.com/zain/0xa349e363