Plasma: Scalable Autonomous Smart Contracts

The problem the paper names

A single chain that re-executes every contract call will not hold every application. Plasma's proposal is a child chain that posts a commitment upward, and a parent-chain exit game so users are not trapped when the child operator censors or lies.

What the design proposes

• The parent chain stores commitments, not every child transaction.
• Users must be able to exit to the parent with their assets, given enough data and time.
• Fraud is handled by challenges and withdrawals, not only by proving every transition up front.

How the mechanism is specified

• The operator can withhold data. The paper's response is an exit, which only works if the user has the data they need and watches the chain.
• Mass exits are a congestion event on the parent. The paper does not make that free.
• Map-reduce language in the paper is an analogy for nested chains. It is not an implementation.

What this page does not treat as proven

• Plasma is not an optimistic rollup and not a zk-rollup. Those publish different data and different proofs.
• An exit game that users do not watch is not a security mechanism. It is a brochure.
• Many Plasma variants were never deployed as specified. The paper is still the right citation for the idea.

Why a venture studio still reads it

Any venture pitching a 'child chain' owes this paper an answer: if the operator disappears tonight, what does a user post on the parent, and do they have the data. If the answer is 'trust the operator', it is not Plasma.