What You Can Do By The End

• Explain why capacity is limited
• Describe how a rollup works
• Compare optimistic and ZK rollups
• Explain data availability
• Judge the risk of a bridge

Recap Of Weeks 7 And 8

• Tokens are ledgers inside contracts.
• DeFi chains those contracts together.
• Every action costs gas on Ethereum.
• Demand drives gas prices up.

Why Capacity Is Limited

• Every full node re-executes every transaction.
• Blocks have a fixed size.
• More demand than space means a bidding contest.
• Bigger blocks make nodes harder to run.

The Trilemma

• Scalability: many transactions per second.
• Security: attacks stay expensive.
• Decentralization: many people can verify.
• Improving one tends to strain another.

Layers

• Layer 1: the base chain that settles results.
• Layer 2: networks that run the activity.
• Layer 2 posts data and results to layer 1.
• Users pay a share of one base fee.

users -> sequencer -> L2 block
                |
         batch of data + proof/claim
                |
         contract on Ethereum (L1)

What A Rollup Does

• A sequencer orders transactions quickly.
• Compressed data goes to Ethereum.
• A contract on Ethereum records the new state.
• One L1 transaction carries many L2 transactions.

Optimistic Rollups

• State is accepted unless challenged.
• A challenge period gives watchers time.
• A valid fraud proof discards the bad state.
• Runs ordinary Solidity with little change.

Zero-Knowledge Rollups

• Each batch comes with a validity proof.
• Ethereum checks the proof, not the transactions.
• Checking is far cheaper than re-running.
• Examples: zkSync and Starknet.

The Cost Of Proving

• Generating proofs needs heavy computation.
• A zkEVM must prove Ethereum rules exactly.
• Withdrawals can finalize once the proof lands.
• Complex proving software is a trust surface.

Compare The Two

• Optimistic: challenge window, simpler, longer exits.
• ZK: proofs, harder to build, faster finality.
• Both depend on data being available.
• Both can have centralized sequencers today.

Data Availability

• Data must be public so anyone can rebuild state.
• Hidden data blocks fraud checks and exits.
• EIP-4844 blobs made data cheaper in 2024.
• Outside data layers add a trust assumption.

Reading The L2 Landscape

• Who runs the sequencer, and is there a forced-exit path?
• Who can upgrade the contracts, and with what delay?
• Where does the data live?
• L2BEAT summarizes these risks.

Bridges

• Lock on one chain, mint on the other.
• Burn to go back and unlock the original.
• Native bridges inherit the rollup's security.
• Third-party bridges add their own trust.

chain A: lock 10 tokens in bridge contract
chain B: mint 10 wrapped tokens
...
chain B: burn 10 wrapped tokens
chain A: unlock 10 tokens

Bridge Safety Habits

• Prefer native bridges for large amounts.
• Send a small test transfer first.
• Check the official site address independently.
• Practice on test networks.

Recap

• Capacity limits drive fees.
• Rollups execute off chain and settle on Ethereum.
• Optimistic uses challenges, ZK uses proofs.
• Data availability and bridges decide safety.

Next Week: Wallets, Security And Custody

• Seed phrases and hardware wallets.
• Phishing, approvals and common contract bugs.
• Audits and what they can promise.

Questions And Exercise

• Deploy ClassToken on a rollup test network.
• Compare fees on both networks.
• Write your network choice with reasons.